MEDICAL PROGRESS Uremia - anaesthetics.ukzn.ac.za · uremia but were removed from this category as their causes were discovered. Today the term “uremia” is used loosely to describe

review article

T h e n e w e ng l a nd j o u r na l o f m e dic i n e

n engl j med 357;13 www.nejm.org september 27, 20071316

MEDICAL PROGRESS

UremiaTimothy W. Meyer, M.D., and Thomas H. Hostetter, M.D.

From Stanford University School of Med-icine, Veterans Affairs Palo Alto Health Care System, Palo Alto, CA (T.W.M.), and the Albert Einstein College of Medicine, New York (T.H.H.). Address reprint re-quests to Dr. Hostetter at the Albert Ein-stein College of Medicine, Rm. 615, Ull-mann Bldg., 1300 Morris Park Ave., Bronx, NY 10461, or at [email protected].

N Engl J Med 2007;357:1316-25.Copyright © 2007 Massachusetts Medical Society.

Medical progress has altered the course and thus the defini-tion of uremia, which once encompassed all the signs and symptoms of ad-vanced kidney failure. Hypertension due to volume overload, hypocalcemic

tetany, and anemia due to erythropoietin deficiency were once considered signs of uremia but were removed from this category as their causes were discovered. Today the term “uremia” is used loosely to describe the illness accompanying kidney failure that cannot be explained by derangements in extracellular volume, inorganic ion con-centrations, or lack of known renal synthetic products. We now assume that uremic illness is due largely to the accumulation of organic waste products, not all identified as yet, that are normally cleared by the kidneys.

No specific time point demarcates the onset of uremia in patients with progressive loss of kidney function. The features of uremia identified in patients with end-stage kidney failure may be present to a lesser degree in people with a glomerular filtra-tion rate that is barely below 50% of the normal rate, which at 30 years of age ranges between 100 and 120 ml per minute per 1.73 m2 of body-surface area. Thus, in the United States alone, uremic symptoms may be present to some degree in an esti-mated 8 million people who have a glomerular filtration rate below 60 ml per min-ute per 1.73 m2 of body-surface area.1 However, early symptoms of uremia, such as fatigue, are nonspecific, making the condition difficult to identify. At present, moreover, we can slow progression to kidney failure but can treat uremia only by replacing kidney function. Thus, the question of whether a patient has uremia comes down to whether dialysis or a transplant would be beneficial.

Treatment of uremia is now dominated by dialysis, in large part because donor kidneys are in short supply. In the United States in 2004, approximately 100,000 people began receiving kidney-replacement therapy for end-stage renal disease, and 335,000 people were receiving ongoing treatment with dialysis.2 In some cases, pa-tients are treated with dialysis for decades, but overall outcomes are disappointing. The 5-year survival rates between 1995 and 1999 were under 35% for both hemodi-alysis and peritoneal dialysis. Patients treated with dialysis are hospitalized on aver-age twice a year, and their quality of life is often low.

Not all of the illness of a patient undergoing dialysis can be ascribed to uremia. Indeed, the evolution of dialysis has made the effects of uremia more difficult to dis-tinguish, since the severity of classic uremic symptoms is attenuated. Instead, pa-tients undergoing dialysis now have a new illness, which Depner3 aptly named the “residual syndrome.” This illness comprises partially treated uremia; ill effects of di-alysis, such as fluctuation in the extracellular fluid volume and exposure to bioincom-patible materials; and residual inorganic ion disturbances, including acidemia and hyperphosphatemia. In many patients, the residual syndrome is complicated by the effects of advancing age and systemic diseases that were responsible for the loss of kidney function.

Although patients undergoing dialysis have a complex illness, there are compelling reasons to believe that inadequate removal of organic wastes is an important con-

Copyright © 2007 Massachusetts Medical Society. All rights reserved. Downloaded from www.nejm.org by RENI ROUNCIVELL on October 6, 2007 .

MEDICAL PROGRESS

n engl j med 357;13 www.nejm.org september 27, 2007 1317

tributor. Dialysis is initiated when uremic symp-toms, among which anorexia and lethargy are usually the most prominent, advance to the point at which treatment is expected to effect an im-provement. The glomerular filtration rate at this point averages about 7% of the normal value. As compared with such a low glomerular filtration rate, conventional dialysis provides only slightly better removal of many solutes and inferior re-moval of some. Renal replacement therapy does keep patients alive, but because of these limita-tions, it does not completely relieve uremic symp-toms. The fact that transplantation reverses this residual syndrome constitutes strong evidence for the ill effects of toxic solute accumulation despite dialysis. Successful transplantation, which can re-store the glomerular filtration rate to more than half the normal value, markedly improves the over-all quality of life and enhances specific functions, including sleep, sexual function, cognition, exer-cise capacity, and, in children, growth.4-7

Solu tes Cle a r ed by the K idne y a nd R e ta ined in Ur emi a

Although retained uremic solutes cause symptoms, identifying the responsible solutes has proved dif-ficult because of the multiplicity of retained solutes (reviewed by the European Uremic Toxin Work Group8) as well as the variety and subtlety of the symptoms. Advances in chromatography and spec-

troscopy continue to lengthen the list of implicated solutes. In general, solutes that accumulate in the highest concentrations, and were therefore iden-tified first, have been the most studied. But only a few compounds have been linked to specific toxic effects.9-11 Plasma concentrations of several com-pounds correlate more closely with altered men-tal function than do concentrations of urea. Some of these compounds, including certain guani-dines, accumulate in the cerebrospinal fluid, which is consistent with their proposed effects on the brain.12 However, we lack experiments showing that uremic signs and symptoms can be replicated by raising solute levels in people or animals with normal kidney function to equal those in patients with uremia. Uremic solutes are therefore usually categorized on the basis of their structure. Selected examples are shown in Table 1.

Urea is quantitatively the most important solute excreted by the kidney and was the first organic solute detected in the blood of patients with kid-ney failure. Both hemodialysis and peritoneal di-alysis are currently prescribed to achieve target values for urea clearance. Yet early studies indi-cated that urea itself causes only a minor part of uremic illness.13,14 One study showed that uremic symptoms were relieved by initiation of dialysis, even when urea was added to the dialysate to maintain the blood urea nitrogen level at approxi-mately 90 mg per deciliter (urea, 32 mmol per liter).14

Table 1. Uremic Solutes.*

Solute Group Example Source Characteristics

Peptides and small proteins

Beta2-microglobulin Shed from MHC Poorly dialyzed because of large size

Guanidines Guanidinosuccinic acid

Arginine Increased production in uremia

Phenols p-Cresol sulfate Phenylalanine, tyrosine Protein bound, produced by gut bacteria

Indoles Indican Tryptophan Protein bound, produced by gut bacteria

Aliphatic amines Dimethylamine Choline Large volume of distribution, produced by gut bacteria

Furans CMPF Unknown Tightly protein bound

Polyols Myoinositol Dietary intake, cell synthesis from glucose

Normally degraded by the kidney rather than excreted

Nucleosides Pseudouridine tRNA Most prominent of several altered RNA species

Dicarboxylic acids Oxalate Ascorbic acid Formation of crystal deposits

Carbonyls Glyoxal Glycolytic intermediates Reaction with proteins to form advanced glycation end products

* Uremic solutes may have multiple sources, although only one is listed. MHC denotes major histocompatibility complex, and CMPF 3-carboxy-4-methyl-5-propyl-2-furanpropionic acid.




Other simple nitrogen-containing solutes that accumulate in uremia include the aliphatic amines monomethylamine, dimethylamine, and trimeth-ylamine. These compounds are produced by both gut bacteria and mammalian cells. They are posi-tively charged at physiologic pH, and their removal during intermittent hemodialysis may be limited by their preferential distribution within the rela-tively acidic intracellular compartment.15 The ure-mic fetor, or fishy breath, of patients with uremia is attributable to trimethylamine, and amines have been associated with impaired brain function in both patients and animal models.16-18

Many uremic solutes contain aromatic rings. The uremic phenols derive from the amino acids tyrosine and phenylalanine and from aromatic compounds in vegetables. Indoles arise in analo-gous fashion from tryptophan and vegetable in-doles. In both cases, metabolism of parent com-pounds by methylation, dehydroxylation, oxidation, reduction, or conjugation produces a bewildering array of solutes. In contrast to methylamines, con-jugated phenols and indoles are often negatively charged and contribute to the increase in the an-ion gap observed in kidney failure. The structural similarity of waste phenols and indoles to neuro-transmitters has encouraged speculation that these compounds interfere with the function of the central nervous system, but evidence of the tox-icity of individual compounds is weak. The most extensively studied phenol is p-cresol, which is formed by colonic bacteria from tyrosine and phe-nylalanine and then circulates conjugated with sulfate.19,20 High p-cresol levels have been associ-ated with poor outcomes in patients undergoing dialysis.19,21

The kidney clears not only small molecules but also proteins with molecular weights between 10 and 30 kD. Plasma levels of these low-molecular-weight proteins rise as the kidney fails, but only beta2-microglobulin, which causes dialysis-related amyloidosis, has known toxicity.9,22 The extent to which kidney failure increases the levels of pep-tides with molecular weights between 500 D and 10 kD is less well defined,23 although it is expected that proteomic techniques may ultimately yield a more complete picture.23,24

Solu te R emova l by Di a lysis

Uremia could theoretically be treated by reducing solute production, but this is not part of current practice. High protein intake increases the produc-

tion of many solutes, including various guanidines, indoles, and phenols. Patients with kidney failure tend to reduce their protein intake spontaneously, and before dialysis became available, physicians found that marked protein restriction relieved ure-mic symptoms.25,26 Protein restriction can have ill effects, however, and it is now recommended that patients undergoing dialysis receive 1.2 g of pro-tein per kilogram of body weight per day, which is nearly the amount provided by an average diet in the United States. Since a number of the best-known uremic solutes — such as aliphatic amines, d-amino acids, methylguanidine, hippurate, and many indoles and phenols — are produced entirely or in part by gut bacteria, the use of sorbents to reduce the load of such solutes has been considered but has not been systematically studied.27

At present, most patients with end-stage renal disease undergo hemodialysis three times per week. The dialysis prescription is adjusted to re-move about two thirds of the total-body urea con-tent during each treatment (Fig. 1). This standard was adopted after clinical trials showed that to a point, patient outcomes improve with increasing fractional urea removal.3,28 The Hemodialysis (HEMO) Study29 showed that outcomes are not improved by increasing fractional urea removal above the current standard. Treatment that re-moves the majority of urea should also remove more toxic solutes if, like urea, they diffuse rela-tively freely throughout body water. But removal of many solutes is more limited, owing to large molecular size, protein binding, or sequestration within body compartments. Plasma levels of such solutes therefore remain much higher than normal urea levels in patients undergoing conventional dialysis (Fig. 2). Removal of such solutes can be increased by various modifications of the stan-dard dialysis treatment. If a specific modification of dialysis reduced illness, this might reveal the characteristics of important uremic toxins.

Large Solutes

Hemodialysis was initially performed with the use of dialysis membranes that provided limited clear-ance of solutes with molecular weights above 1 kD. Treatment with the use of these membranes wak-ened comatose patients, relieved vomiting, and partially reversed other classic uremic symptoms. This provided evidence, which remains convincing, that some important uremic toxins are small. But clinical observations led pioneering investigators to speculate that other important toxins were “mid-


MEDICAL PROGRESS


dle molecules,” with molecular weights between 300 D and 2 kD.32 The original middle-molecule hypothesis was never carefully tested. Although the phrase “middle molecules” remains in use, its meaning has gradually shifted to include larger sol-utes. The adoption of new, more permeable mem-brane materials essentially ended investigation of the relative toxicity of solutes in size ranges below 1 kD. However, the toxicity of solutes with mo-lecular weights above 1 kD remains under inves-tigation. Increasing the removal of large solutes, such as beta2-microglobulin, with the use of “high-flux” as compared with “low-flux” membranes had no significant benefit in the HEMO Study.29 How-ever, the clearance of large molecules can be in-creased further by adding ultrafiltration to the di-alysis process (Fig. 3). Ongoing multicenter trials may reveal whether the combination of ultrafil-tration with dialysis, called hemodiafiltration, im-proves outcomes.33

Solutes Bound to Albumin

Solutes that bind to albumin are poorly removed by conventional dialysis3,20,34,35 not because they are large but because only the free, unbound sol-ute concentration contributes to the gradient driv-ing solute across the dialysis membrane. When ex-

pressed as multiples of normal levels, the levels of these compounds are therefore higher than those of unbound solutes, such as urea, in patients un-dergoing hemodialysis. There is reason to suspect that at least some protein-bound solutes are toxic. The normal kidney clears many protein-bound sol-utes by active tubular secretion. Presumably, the combination of protein binding and tubular secre-tion represents an evolutionary adaptation that al-lows the excretion of toxic molecules and keeps the extracellular fluid concentrations of their unbound fraction very low. The aggregate toxicity of protein-bound solutes could theoretically be assessed by comparing the effects of different modifications of dialysis, but this has not been attempted in practice.36

Sequestered Solutes

Other solutes are sequestered, or retained, in com-partments where their concentration does not equilibrate rapidly with that of the plasma.37 In-

16p6

50

Plas

ma

Con

cent

ratio

n(m

ultip

le o

f nor

mal

val

ue)

30

40

20

10

0Blood urea

nitrogenp-Cresolsulfate

β2-micro-globulin

Solutes

Guanidino-succinic acid

AUTHOR:

FIGURE:

JOB: ISSUE:

4-CH/T

RETAKE

SIZE

ICM

CASE

EMail LineH/TCombo

Revised

AUTHOR, PLEASE NOTE: Figure has been redrawn and type has been reset.

Please check carefully.

REG F

Enon

1st

2nd3rd

Hostetter

2 of 3

09-27-07

ARTIST: ts

35713

Figure 2. Time-Averaged Plasma Solute Levels in Patients Undergoing Conventional Thrice-Weekly Hemodialysis.

Conventional hemodialysis is prescribed to remove blood urea nitrogen effectively, so that the average urea level in a patient undergoing hemodialysis is only about four times the normal value. But dialysis is much less effective in controlling the levels of other solutes. Binding to albumin limits the dialysis of p-cresol sul-fate, and large molecular size limits the dialysis of beta2-microglobulin; as a result, the average levels of these solutes in patients undergoing hemodialysis are about 10 times and 20 times the normal levels, respectively. The plasma level of guanidinosuccinic acid is even higher, averaging more than 40 times the normal val-ue. Guanidinosuccinic acid levels rise this high largely because the production of guanidinosuccinic acid in-creases in patients with uremia; sequestration within cells impairs the efficiency of dialysis and contributes to the elevation of plasma levels of related guanidines. Solute ratios are approximations based on data from Martinez et al.,20 Raj et al.,30 and Eloot et al.31

Patient 1

Patient 2

16p6

120B

lood

Ure

a N

itrog

en (m

g/dl

)

80

100

60

40

20

0Monday Wednesday Friday Sunday

AUTHOR:

FIGURE:

JOB: ISSUE:

4-CH/T

RETAKE

SIZE

ICM

CASE

EMail LineH/TCombo

Revised

AUTHOR, PLEASE NOTE: Figure has been redrawn and type has been reset.

Please check carefully.

REG F

Enon

1st2nd3rd

Hostetter

1 of 3

09-27-07

ARTIST: ts

35713

Figure 1. Blood Urea Nitrogen Levels in Two Theoretical Patients Undergoing Conventional Thrice-Weekly Hemodialysis for 3 Hours on Monday, Wednesday, and Friday.

Urea nitrogen levels fall precipitously as urea is rapidly removed during treatment and then rise gradually be-tween treatments, with the highest levels observed after the 3-day interdialytic interval (from Friday after dialysis until Monday before dialysis). Both patients were re-ceiving the same dose of dialysis, as evidenced by the 68% drop in urea levels for both patients with each treatment. This drop constitutes adequate dialysis, according to the current U.S. standard. Patient 1, who had higher absolute plasma urea levels than Patient 2, was presumably eating more protein. To convert the values for blood urea nitrogen to millimoles of urea per liter, multiply by 0.357.




termittent dialysis may rapidly lower the plasma concentration of such solutes but removes only a small portion of the total-body solute content. Se-questration only slightly impairs the removal of urea, which is currently used to assess the ade-quacy of dialysis.3 But other solutes may equilibrate more slowly than urea between body compart-ments, such as between cell water and plasma, ef-fectively sequestering them from dialysis. Only a few solutes have been carefully studied, but clin-ically significant sequestration has been demon-strated for phosphate, creatinine, uric acid, several guanidines, and beta2-microglobulin. Theoretical-ly, the contribution of sequestered solutes to ure-mic toxicity, like the contribution of large solutes or protein-bound solutes, could be assessed by comparing the efficacy of different dialysis pre-scriptions. When treatment is intermittent, the re-moval of sequestered as compared with rapidly equilibrating solutes can be increased by length-ening each dialysis session or by increasing the number of sessions per week. Retrospective stud-ies have shown that longer sessions are associated with better outcomes, but these studies were not sufficiently well controlled to confirm that length-ening treatment is beneficial.38 In the United States,

the combined preference of patients and providers for short dialysis sessions has reduced treatment time toward the minimum required to achieve the target urea removal, making the average duration of each hemodialysis session about 3.5 hours.

Signs a nd S ymp t oms of Ur emi a

Frequently identified signs and symptoms of ure-mia are listed in Table 2. That a fundamental met-abolic disturbance such as uremia should have such a wide variety of consequences is not remarkable. The complications of untreated diabetes and hy-perthyroidism are similarly extensive. But uremia is different in that we cannot trace all its compli-cations to dysregulation of a single, key compound. And with the exception of kidney transplantation, current therapy for uremia is less successful than insulin and thyroid hormone replacement in re-storing normal function.

Given the signs and symptoms listed in Table 2, it is not surprising that the quality of life declines in people with chronic kidney disease. A panel preparing treatment guidelines concluded that well-being was reduced when the glomerular fil-tration rate was less than 60 ml per minute per

A Dialysis B Hemofiltration

Blood Blood UltrafiltrateDialysate

Urea

Low-molecular-weight proteins Adjustable pressure

08/23/07

AUTHOR PLEASE NOTE:Figure has been redrawn and type has been reset

Please check carefully

Author

Fig #Title

ME

DEArtist

Issue date

COLOR FIGURE

Rev4Dr. Hostetter

09/27/2007

4

InglefingerDaniel Muller

Figure 3. Dialysis versus Hemofiltration.

In dialysis (Panel A), solutes diffuse through a thin membrane separating the blood and dialysate, which flow in opposite directions. Small solutes such as urea (small yellow spheres) diffuse readily. Larger solutes, including low-molecular-weight proteins (large green spheres), diffuse less readily and are not cleared as effectively when blood passes through the dialyzer. In hemofiltration (Panel B), fluid is forced through the same membrane by pressure, and solutes are carried with the fluid by convection. As compared with diffusion, convection removes larger solutes at almost the same rate as small solutes. Standard dialysis treatments include some hemofiltration in order to remove the fluid that accumulates with daily intake. The removal of large solutes can be augmented by increasing the amount of ultrafiltration. The combined process of dialysis and high-volume ultrafiltration, which requires the provision of intravenous replace-ment fluid to offset the ultrafiltration rate, is called hemodiafiltration.


MEDICAL PROGRESS


1.73 m2 of body-surface area.39 Subjects in the Modification of Diet in Renal Disease study who had glomerular filtration rates of less than 55 ml per minute per 1.73 m2 reported fatigue and re-duced stamina that correlated with this rate.40 In a study using the Medical Outcomes Study 36-item Short-Form General Health Survey, people who had a glomerular filtration rate of less than 50 ml per minute per 1.73 m2 but who were not yet being treated with dialysis scored lower than the general population on 8 of the instrument’s 10 scales.41 Physical functioning is also often im-paired in patients with kidney failure. In those undergoing dialysis, exercise capacity has been reported to average only about 50% of normal capacity. Treatment of anemia improves but does not normalize exercise capacity. The most detailed studies suggest that susceptibility to fatigue is attributable to both muscle energy failure and neural defects.42 The degree to which the qual-ity of life is impaired by the uremic environment, by the deconditioning of the patient, and by the effects of coexisting conditions has not been com-pletely analyzed. A recent study indicated, however, that even selected, highly functional patients undergoing dialysis have notable physical limita-tions, including impairment of balance, walking speed, and sensory function.43

A particularly interesting group of uremic signs and symptoms reflect altered nerve function. Clas-sic descriptions emphasized that patients with uremia could appear alert despite defects in mem-ory, ability to plan, and attention.13 Cognitive im-pairment has been detected in people with glo-merular filtration rates between 30 and 60 ml per minute per 1.73m2 of body-surface area.44 As kidney function worsened, patients progressed to coma or catatonia that could be relieved by dialy-sis.13 Recent studies have shown that cognitive function remains impaired in patients receiving standard treatment with dialysis. Neuropsycho-logical testing has revealed defects, particularly in attention, memory, and the performance of higher-order tasks. Central nervous system chang-es due to vascular disease and other processes often contribute to these defects. But it has gen-erally been concluded that patients undergoing dialysis have cognitive impairment that cannot be ascribed entirely to coexisting conditions.45,46 Another reflection of altered central nervous sys-tem function in patients with uremia is impaired sleep.5,47 Sleep is fragmented by brief arousals and apneic episodes, which are often associated

with bursts of repetitive leg movement. When awake, patients may have the restless legs syn-drome,48 a condition in which they continually feel the need to move their legs. In the past, se-vere sensorimotor neuropathy developed in pa-tients with untreated uremia, and remission of neuropathy was a major goal of dialysis. With current dialysis protocols, neuropathy is usually subclinical, but it can still be detected in nerve-function tests.49,50

Me ta bol ic Effec t s of Ur emi a

Among the metabolic effects of uremia (Table 2), the most extensively studied is insulin resistance, which may contribute to the accelerated vascular disease that is the major cause of death in patients with kidney failure.51,52 When the glomerular fil-tration rate falls below 50 ml per minute per 1.73 m2, insulin resistance occurs.53 The reason

Table 2. Signs and Symptoms of Uremia.

Neural and muscular

Fatigue

Peripheral neuropathy

Decreased mental acuity

Seizures

Anorexia and nausea

Decreased sense of smell and taste

Cramps

Restless legs

Sleep disturbances

Coma

Reduced muscle membrane potential

Endocrine and metabolic

Amenorrhea and sexual dysfunction

Reduced body temperature

Altered amino acid levels

Bone disease due to phosphate retention, hyperparathy-roidism, and vitamin D deficiency

Reduced resting energy expenditure

Insulin resistance

Increased protein–muscle catabolism

Other

Serositis (including pericarditis)

Itching

Hiccups

Oxidant stress

Anemia due to erythropoietin deficiency and shortened red-cell survival

Granulocyte and lymphocyte dysfunction

Platelet dysfunction




it occurs is unclear. Insulin binds normally to its receptor, and receptor density is unchanged. The accumulation of hormones derived from fat and in-creased levels of glucagon and fatty acids also seem insufficient to account for insulin resistance.52,54 The observations that insulin resistance can be transferred by uremic serum and that it is improved by dialysis and protein restriction suggest that ac-cumulation of nitrogenous solutes causes insulin resistance.52 In addition, physical inactivity dimin-ishes the action of insulin, and in patients with uremia, insulin resistance may develop in part be-cause of deconditioning.

Another effect of uremia that has been the ob-ject of recent attention is oxidative stress.55 In-creases in the levels of primary oxidants have not been documented, presumably because of the eva-nescent nature of substances such as superoxide anion and hydrogen peroxide. Accumulation of oxidant reaction products is therefore taken as evidence of increased oxidant activity. Among the markers of oxidation are proteins containing oxi-dized amino acids.56,57 Some proteins are modi-fied by direct oxidation, and others by combi-nation with carbonyl compounds. The modified proteins can form compounds identical to the ad-vanced glycation products that were first identi-fied in patients with diabetes. The carbonyls ini-tiating protein modification have not been fully characterized but include glyoxal and methyl gly-oxal, which are produced by oxidation of both sugars and lipids.58 Modified proteins presumably do not contribute to the effects of uremia that are rapidly reversible, such as confusion or nausea, but might cause gradual changes in tissue structure. The loss of extracellular reducing substances pro-vides additional evidence of oxidant stress in ure-mia. The case of albumin, which undergoes oxi-dation at its single free cysteine thiol group, is particularly interesting. Plasma albumin is more highly oxidized in patients with uremia than in normal subjects, and it is rapidly converted to its reduced form by hemodialysis.59

A third recently identified concomitant of ure-mia is systemic inflammation. Some patients have elevations in inflammatory markers, including C-reactive protein, interleukin-6, and tumor necro-sis factor α, that are unexplained by coexisting conditions.60 Inflammation may interact with in-sulin resistance and oxidant stress to promote vascular disease in these patients. Indeed, insulin

resistance, oxidative stress, and inflammation ap-pear before renal disease progresses to the point at which dialysis is required, and these factors may thus contribute to the high rates of cardiovascu-lar morbidity and mortality among patients with chronic kidney disease.61

Cel lul a r F unc tions

Several effects of uremia are transferable by blood and plasma, underscoring the role of retained sol-utes in generating toxicity. An often-described ab-normality has been the inhibition of sodium–potassium ATPase. Decreased sodium–potassium ATPase activity was first described in red cells from patients with uremia.62 Subsequent reports noted the same effect in other cell types and showed that the inhibition is attributable to one or more factors in uremic plasma.63 The evidence for a circulating inhibitor includes the findings that dialysis reduces the inhibitory activity and that uremic plasma can suppress sodium–potassium ATPase activity in the short term. A number of candidate factors have been considered, and digitalis-like substances have received the greatest attention. Several such com-pounds, including marinobufagenin and telocino-bufagin, have been identified in excess in patients with kidney failure.64 However, confirmation that these substances cause the inhibition of sodium–potassium ATPase is lacking.

W h y Is the Gl omerul a r Filtr ation R ate Nor m a l ly

So High?

Presumably, the kidney evolved to rid the extracel-lular fluid of the solutes that cause uremic illness when kidney function is reduced. Glomerular fil-tration, the initial step in urine formation, proceeds at an extremely high rate; a volume equaling that of the entire extracellular f luid is filtered every 2 hours. At rest, approximately 10% of the body’s energy consumption is devoted to reabsorption ne-cessitated by this high filtration rate. The rate of fluid processing clearly exceeds that required sim-ply to rid the body of the daily intake of water and inorganic ions. But uremia is not detected until the glomerular filtration rate is less than half the nor-mal rate. One explanation for the apparent super-abundance of kidney function is that it constitutes a safety factor, like the capacity of bone to with-


MEDICAL PROGRESS


stand greater-than-usual mechanical loads. In the case of the kidney, the ingestion of toxins may con-stitute an analogous increase in load. If so, we would expect patients with uremia to have limited tolerance for certain foods, just as they have lim-ited tolerance for certain medications. Patients un-dergoing dialysis have become comatose after in-gestion of star fruit.65 However, given the variety of chemicals found in plants, there are remarkably few reports of this kind. Perhaps toxin intake was greater before civilization improved the selection and preservation of foods, making a persistently high renal clearance rate worth the metabolic cost.

Alternatively, the glomerular filtration rate may appear to be excessive because our clinical criteria are too crude to detect the consequences of mild impairment in kidney function. Fitness in an evo-lutionary sense may require that concentrations of some solutes be maintained below the levels at which disease is detected. It is possible, for in-stance, that a sensitive process such as fertility, growth in children, or peak mental performance might be disturbed by a small increase in the levels of retained toxins. A particularly interesting find-ing has been the identification of similar trans-port systems in the kidney tubule and the blood–brain barrier.66 This finding suggests that the kidney, together with the liver, may be designed to keep organic waste levels in the extracellular fluid sufficiently low to allow a second-stage pumping system in the blood–brain barrier to keep the brain interstitium exquisitely clean.

F u t ur e S t udies

Dialysis for acute kidney failure presents a partic-ularly difficult problem. Symptoms that trigger the initiation of dialysis in patients with chronic kid-ney failure often cannot be distinguished in pa-tients who are in intensive care. Mortality rates are high, and some speculate that more aggressive therapy than is now prescribed for end-stage re-nal disease is needed in these very sick patients. However, trials of continuous treatment with he-modiafiltration or hemofiltration have shown no advantage over intermittent hemodialysis.67,68 An ongoing multicenter trial is investigating wheth-er increasing urea removal beyond the standard for long-term hemodialysis will be helpful.69

Efforts are also under way to improve long-term

dialysis. The HEMO Study showed no benefit of increased removal of urea or low-molecular-weight proteins during thrice-weekly hemodialysis.29 The decreases in time-averaged solute levels, however, were modest. As previously noted, ongoing Euro-pean trials will establish whether more efficient removal of low-molecular-weight proteins by he-modiafiltration is beneficial.33 Nightly home he-modialysis allows for particularly large increases in solute removal.70 Patients receiving this treat-ment report improvement in well-being and have partial reversal of the sleep disturbances that oc-cur with conventional treatment.47 Ongoing tri-als by the Frequent Hemodialysis Network are comparing nocturnal home hemodialysis or in-center hemodialysis six times a week with stan-dard thrice-weekly in-center hemodialysis.71 If more intensive dialysis proves beneficial, major issues of practicality and cost will loom. Relatively few patients are likely to be willing and able to per-form hemodialysis at home, and frequent in-center hemodialysis would be both burdensome and costly. Peritoneal dialysis, now used by only about 10% of patients in the United States, is easier to perform at home than hemodialysis and is less expensive as well. Although a randomized com-parison of peritoneal dialysis and hemodialysis has not been possible, the two approaches appear to be associated with a similarly high overall burden of residual illness.

Summ a r y

Maintenance of life in patients without kidney function is a remarkable achievement of modern medicine. But current treatment with dialysis car-ries a high price and leaves a persistent burden of disability. Although both the side effects of dialy-sis and the coexisting conditions in patients re-ceiving this treatment contribute to the residual illness, retained solutes that are poorly cleared by standard treatment are an important part of the problem. A better understanding of uremic solutes and their toxic effects would place dialysis on a more rational basis and should lead to more ef-fective therapy.

Supported by grants from the National Institutes of Health (R33 DK71251, to Dr. Meyer, and R21 DK077326, to Dr. Hos-tetter).

No potential conflict of interest relevant to this article was reported.




References

Coresh J, Byrd-Holt D, Astor BC, et al. Chronic kidney disease awareness, preva-lence, and trends among U.S. adults, 1999 to 2000. J Am Soc Nephrol 2005;16:180-8.

USRDS 2006 annual data report: atlas of end-stage renal disease in the United States. Bethesda, MD: U.S. Renal Data Sys-tem, 2006.

Depner TA. Uremic toxicity: urea and beyond. Semin Dial 2001;14:246-51.

Valderrábano F, Jofre R, López-Gomez JM. Quality of life in end-stage renal dis-ease patients. Am J Kidney Dis 2001;38: 443-64.

Perl J, Unruh ML, Chan CT. Sleep dis-orders in end-stage renal disease: ‘Markers of inadequate dialysis’? Kidney Int 2006; 70:1687-93.

Griva K, Thompson D, Jayasena D, Davenport A, Harrison M, Newman SP. Cognitive functioning pre- to post-kidney transplantation — a prospective study. Nephrol Dial Transplant 2006;21:3275-82.

Fine RN. Growth following solid-organ transplantation. Pediatr Transplant 2002;6: 47-52.

Vanholder R, De Smet R, Glorieux G, et al. Review on uremic toxins: classifica-tion, concentration, and interindividual variability. Kidney Int 2003;63:1934-43.

Dember LM, Jaber BL. Dialysis-related amyloidosis: late finding or hidden epi-demic? Semin Dial 2006;19:105-9.

Ravani P, Tripepi G, Malberti F, Testa S, Mallamaci F, Zoccali C. Asymmetrical dimethylarginine predicts progression to dialysis and death in patients with chron-ic kidney disease: a competing risks mod-eling approach. J Am Soc Nephrol 2005; 16:2449-55.

Boccardo P, Remuzzi G, Galbusera M. Platelet dysfunction in renal failure. Semin Thromb Hemost 2004;30:579-89.

De Deyn PP, D’Hooge R, Van Bogaert PP, Marescau B. Endogenous guanidino compounds as uremic neurotoxins. Kid-ney Int Suppl 2001;78:S77-S83.

Biochemistry of uremia. In: Schreiner G, Maher J. Uremia. Springfield, IL: Thom-as, 1960:55-85.

Johnson WJ, Hagge WW, Wagoner RD, Dinapoli RP, Rosevear JW. Effects of urea loading in patients with far-advanced renal failure. Mayo Clin Proc 1972;47: 21-9.

Smith JL, Wishnok JS, Deen WM. Me-tabolism and excretion of methylamines in rats. Toxicol Appl Pharmacol 1994;125: 296-308.

Simenhoff ML, Saukkonen JJ, Burke JF, et al. Importance of aliphatic amines in uremia. Kidney Int Suppl 1978;Jun(8):S16-S19.

Pirisino R, Ghelardini C, De Siena G, et al. Methylamine: a new endogenous modulator of neuron firing? Med Sci Monit 2005;11:RA257-61.

Simenhoff ML, Burke JF, Saukkonen

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

JJ, Ordinario AT, Doty R. Biochemical profile of uremic breath. N Engl J Med 1977;297:132-5.

Bammens B, Evenepoel P, Keuleers H, Verbeke K, Vanrenterghem Y. Free serum concentrations of the protein-bound reten-tion solute p-cresol predict mortality in hemodialysis patients. Kidney Int 2006;69: 1081-7.

Martinez AW, Recht NS, Hostetter TH, Meyer TW. Removal of p-cresol sul-fate by hemodialysis. J Am Soc Nephrol 2005;16:3430-6.

De Smet R, Van Kaer J, Van Vlem B, et al. Toxicity of free p-cresol: a prospective and cross-sectional analysis. Clin Chem 2003;49:470-8.

Verroust PJ, Birn H, Nielsen R, Ko-zyraki R, Christensen EI. The tandem en-docytic receptors megalin and cubilin are important proteins in renal pathology. Kidney Int 2002;62:745-56.

Norden AG, Sharratt P, Cutillas PR, Cramer R, Gardner SC, Unwin RJ. Quanti-tative amino acid and proteomic analysis: very low excretion of polypeptides >750 Da in normal urine. Kidney Int 2004;66:1994-2003.

Weissinger EM, Kaiser T, Meert N, et al. Proteomics: a novel tool to unravel the patho-physiology of uraemia. Nephrol Dial Transplant 2004;19:3068-77.

Kopple JD, Greene T, Chumlea WC, et al. Relationship between nutritional sta-tus and the glomerular filtration rate: re-sults from the MDRD study. Kidney Int 2000;57:1688-703.

Giovannetti S, Maggiore Q. A low-ni-trogen diet with proteins of high biologi-cal value for severe chronic uraemia. Lan-cet 1964;37:1000-3.

Niwa T, Emoto Y, Maeda K, Uehara Y, Yamada N, Shibata M. Oral sorbent sup-presses accumulation of albumin-bound indoxyl sulphate in serum of haemodialy-sis patients. Nephrol Dial Transplant 1991; 6:105-9.

Gotch FA, Sargent JA. A mechanistic analysis of the National Cooperative Di-alysis Study (NCDS). Kidney Int 1985;28: 526-34.

Eknoyan G, Beck GJ, Cheung AK, et al. Effect of dialysis dose and membrane flux in maintenance hemodialysis. N Engl J Med 2002;347:2010-9.

Raj DS, Ouwendyk M, Francoeur R, Pierratos A. beta(2)-Microglobulin kinet-ics in nocturnal haemodialysis. Nephrol Dial Transplant 2000;15:58-64.

Eloot S, Torremans A, De Smet R, et al. Kinetic behavior of urea is different from that of other water-soluble com-pounds: the case of the guanidino com-pounds. Kidney Int 2005;67:1566-75.

Babb AL, Ahmad S, Bergström J, Scrib-ner BH. The middle molecule hypothesis in perspective. Am J Kidney Dis 1981;1: 46-50.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

Canaud B, Morena M, Leray-Moragues H, Chalabi L, Cristol JP. Overview of clin-ical studies in hemodiafiltration: what do we need now? Hemodial Int 2006;10: Suppl 1:S5-S12.

Lesaffer G, De Smet R, Lameire N, Dhondt A, Duym P, Vanholder R. Intradia-lytic removal of protein-bound uraemic toxins: role of solute characteristics and of dialyser membrane. Nephrol Dial Trans-plant 2000;15:50-7.

Bammens B, Evenepoel P, Verbeke K, Vanrenterghem Y. Removal of middle mol-ecules and protein-bound solutes by peri-toneal dialysis and relation with uremic symptoms. Kidney Int 2003;64:2238-43.

Meyer TW, Leeper EC, Bartlett DW, et al. Increasing dialysate flow and dialyzer mass transfer area coefficient to increase the clearance of protein-bound solutes. J Am Soc Nephrol 2004;15:1927-35.

Schneditz D, Daugirdas JT. Compart-ment effects in hemodialysis. Semin Dial 2001;14:271-7.

Chertow GM, Kurella M, Lowrie EG. The tortoise and hare on hemodialysis: does slow and steady win the race? Kidney Int 2006;70:24-5.

National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis 2002; 39:Suppl 1:S1-S266.

Rocco MV, Gassman JJ, Wang SR, Ka-plan RM. Cross-sectional study of quality of life and symptoms in chronic renal dis-ease patients: the Modification of Diet in Renal Disease Study. Am J Kidney Dis 1997;29:888-96.

Perlman RL, Finkelstein FO, Liu L, et al. Quality of life in chronic kidney dis-ease (CKD): a cross-sectional analysis in the Renal Research Institute-CKD study. Am J Kidney Dis 2005;45:658-66.

Johansen KL. Physical functioning and exercise capacity in patients on dialysis. Adv Ren Replace Ther 1999;6:141-8.

Blake C, O’Meara YM. Subjective and objective physical limitations in high-func-tioning renal dialysis patients. Nephrol Dial Transplant 2004;19:3124-9.

Hailpern SM, Melamed ML, Cohen HD, Hostetter TH. Moderate kidney dis-ease and cognitive function in adults 20-59 years of age. J Am Soc Nephrol 2007;18: 2205-13.

Kurella M, Chertow GM, Luan J, Yaffe K. Cognitive impairment in chronic kid-ney disease. J Am Geriatr Soc 2004;52: 1863-9.

Murray AM, Tupper DE, Knopman DS, et al. Cognitive impairment in hemodialy-sis patients is common. Neurology 2006; 67:216-23.

Hanly PJ, Pierratos A. Improvement of sleep apnea in patients with chronic renal failure who undergo nocturnal hemodi-alysis. N Engl J Med 2001;344:102-7.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.


MEDICAL PROGRESS


Mucsi I, Molnar MZ, Ambrus C, et al. Restless legs syndrome, insomnia and quality of life in patients on maintenance dialysis. Nephrol Dial Transplant 2005;20: 571-7.

Krishnan AV, Phoon RK, Pussell BA, Charlesworth JA, Bostock H, Kiernan MC. Altered motor nerve excitability in end-stage kidney disease. Brain 2005;128: 2164-74.

Brouns R, De Deyn PP. Neurological complications in renal failure: a review. Clin Neurol Neurosurg 2004;107:1-16.

Shinohara K, Shoji T, Emoto M, et al. Insulin resistance as an independent pre-dictor of cardiovascular mortality in pa-tients with end-stage renal disease. J Am Soc Nephrol 2002;13:1894-900.

Rigalleau V, Gin H. Carbohydrate me-tabolism in uraemia. Curr Opin Clin Nutr Metab Care 2005;8:463-9.

Kobayashi S, Maesato K, Moriya H, Ohtake T, Ikeda T. Insulin resistance in patients with chronic kidney disease. Am J Kidney Dis 2005;45:275-80.

Axelsson J, Bergsten A, Qureshi AR, et al. Elevated resistin levels in chronic kidney disease are associated with de-creased glomerular filtration rate and in-flammation, but not with insulin resis-tance. Kidney Int 2006;69:596-604.

Himmelfarb J, Stenvinkel P, Ikizler TA, Hakim RM. The elephant in uremia: oxidant stress as a unifying concept of cardiovascular disease in uremia. Kidney Int 2002;62:1524-38.

48.

49.

50.

51.

52.

53.

54.

55.

Himmelfarb J, McMonagle E, McMe-namin E. Plasma protein thiol oxidation and carbonyl formation in chronic renal failure. Kidney Int 2000;58:2571-8.

Capeillère-Blandin C, Gausson V, Descamps-Latscha B, Witko-Sarsat V. Bio-chemical and spectrophotometric signifi-cance of advanced oxidized protein prod-ucts. Biochim Biophys Acta 2004;1689: 91-102.

Miyata T, Sugiyama S, Saito A, Kuro-kawa K. Reactive carbonyl compounds re-lated uremic toxicity (“carbonyl stress”). Kidney Int Suppl 2001;78:S25-S31.

Himmelfarb J, McMenamin E, Mc-Monagle E. Plasma aminothiol oxidation in chronic hemodialysis patients. Kidney Int 2002;61:705-16.

Avesani CM, Carrero JJ, Axelsson J, Qureshi AR, Lindholm B, Stenvinkel P. In-f lammation and wasting in chronic kid-ney disease: partners in crime. Kidney Int Suppl 2006;1:S8-S13.

Zoccali C. Traditional and emerging cardiovascular and renal risk factors: an epidemiologic perspective. Kidney Int 2006;70:26-33.

Welt LG, Sachs JR, McManus TJ. An ion transport defect in erythrocytes from uremic patients. Trans Assoc Am Physi-cians 1964;77:169-81.

Kahn T, Thomas K. Na+-K+ pump in chronic renal failure. Am J Physiol 1987; 252:F785-F793.

Komiyama Y, Dong XH, Nishimura N, et al. A novel endogenous digitalis, telo-

56.

57.

58.

59.

60.

61.

62.

63.

64.

cinobufagin, exhibits elevated plasma lev-els in patients with terminal renal failure. Clin Biochem 2005;38:36-45.

Chang JM, Hwang SJ, Kuo HT, et al. Fatal outcome after ingestion of star fruit (Averrhoa carambola) in uremic patients. Am J Kidney Dis 2000;35:189-93.

Ohtsuki S. New aspects of the blood-brain barrier transporters: its physiologi-cal roles in the central nervous system. Biol Pharm Bull 2004;27:1489-96.

Mehta RL, McDonald B, Gabbai FB, et al. A randomized clinical trial of con-tinuous versus intermittent dialysis for acute renal failure. Kidney Int 2001;60: 1154-63.

Vinsonneau C, Camus C, Combes A, et al. Continuous venovenous haemodiafil-tration versus intermittent haemodialysis for acute renal failure in patients with mul-tiple-organ dysfunction syndrome: a mul-ticentre randomised trial. Lancet 2006;368: 379-85.

Palevsky PM, O’Connor T, Zhang JH, Star RA, Smith MW. Design of the VA/NIH Acute Renal Failure Trial Network (ATN) Study: intensive versus convention-al renal support in acute renal failure. Clin Trials 2005;2:423-35.

Pierratos A. Daily nocturnal home he-modialysis. Kidney Int 2004;65:1975-86.

Suri RS, Garg AX, Chertow GM, et al. Frequent Hemodialysis Network (FHN) randomized trials: study design. Kidney Int 2007;71:349-59.Copyright © 2007 Massachusetts Medical Society.

65.

66.

67.

68.

69.

70.

71.

powerpoint slides of journal figures and tables

At the Journal’s Web site, subscribers can automatically create PowerPoint slides. In a figure or table in the full-text version of any article at www.nejm.org, click on Get PowerPoint Slide. A PowerPoint slide containing the image, with its title and reference citation, can then be downloaded and saved.


Acute kidney injury in the intensive care unit: An update andprimer for the intensivist

Paula Dennen, MD; Ivor S. Douglas, MD; Robert Anderson

Acute kidney injury (AKI), pre-viously termed acute renal fail-ure, refers to a sudden declinein kidney function causing dis-

turbances in fluid, electrolyte, and acid–base balance because of a loss in smallsolute clearance and decreased glomeru-lar filtration rate (GFR). The nomencla-ture shift to AKI more accurately repre-sents the spectrum of disease fromsubclinical injury to complete organ fail-ure. This review focuses on key questionsfor the intensivist faced with AKI in theICU.

Epidemiology of Acute KidneyInjury in the Intensive Care Unit

AKI in the ICU is common, increasingin incidence (1–4), and is associated with

a substantial increase in morbidity andmortality (5, 6). AKI occurs in approxi-mately 7% of all hospitalized patients (7)and in up to 36% to 67% of critically illpatients depending on the definition used(6, 8–11). Based on �75,000 critically illadults, more severe AKI occurs in 4% to25% of all ICU admissions (6, 8, 9, 11).On average, 5% to 6% of ICU patientswith AKI require renal replacement ther-apy (RRT) (6, 8–11).

Reported mortality in ICU patientswith AKI varies considerably betweenstudies depending on AKI definitionand the patient population studied(e.g., sepsis, trauma, cardiothoracicsurgery, or contrast nephropathy). Inthe majority of studies, mortality in-creases proportionately with increasingseverity of AKI (6, 10 –13). In patientswith severe AKI requiring RRT, mortal-ity is approximately 50% to 70% (9,14 –16). While AKI requiring RRT in theICU is a well-recognized independentrisk factor for in-hospital mortality(17), even small changes in serum cre-atinine (SCr) are associated with in-creased mortality (18 –21). Notably,multiple studies of patients with AKIand sepsis (22–24), mechanical ventila-tion (25), major trauma (26, 27), car-diopulmonary bypass (17, 28 –30), andburn injuries (31) have consistently

demonstrated an increased risk of deathdespite adjustment for comorbiditiesand severity of illness.

Morbidity, a less appreciated conse-quence of AKI in the ICU, is associatedwith increased cost (18), increased lengthof stay (6, 14, 18, 26), and increased riskof chronic kidney disease (CKD), includ-ing end-stage kidney disease (9, 15, 16,32–37). The true incidence of CKD afterAKI is unknown because epidemiologicstudies do not routinely or consistentlyreport rates of renal recovery and thosethat do use variable definitions (38).

Definition of Acute KidneyInjury in the Intensive Care Unit

More than 35 definitions of AKI cur-rently exist in the literature (39). TheAcute Dialysis Quality Initiative convenedin 2002 and proposed the RIFLE classifi-cation (risk, injury, failure, loss, end-stage kidney disease) specifically for AKIin critically ill patients (Table 1) (40).Using SCr and urine output, the RIFLEcriteria define three grades of severityand two outcome classes. The most se-vere classification met by either criterionshould be used. Of note, patients withprimary kidney diseases such as glomer-ulonephritis were excluded from this def-inition.

From Divisions of Nephrology and Critical CareMedicine (PD), Division of Pulmonary Sciences andCritical Care Medicine (ISD), Division of Medicine (RA),Denver Health Medical Center and University of Colo-rado, Denver, Colorado.

Dr. Douglas is supported by HL070940,HL088138, and N01-HR-56167.

The authors have not disclosed any potential con-flicts of interest.

For information regarding this article, E-mail:[email protected]

Copyright © 2009 by the Society of Critical CareMedicine and Lippincott Williams & Wilkins

DOI: 10.1097/CCM.0b013e3181bfb0b5

Objective: Acute kidney injury is common in critically illpatients and is associated with significant morbidity and mor-tality. Patients across the spectrum of critical illness haveacute kidney injury. This requires clinicians from across dis-ciplines to be familiar with recent advances in definitions,diagnosis, prevention, and management of acute kidney injuryin the intensive care unit. The purpose of this concise review,therefore, is to address, for the non-nephrologist, clinicallyrelevant topical questions regarding acute kidney injury in theintensive care unit.

Data Sources: The authors (nephrologists and intensivists)performed a directed review of PubMed to evaluate topicsincluding the definition, diagnosis, prevention, and treatmentof acute kidney injury in the intensive care unit. The goal of

this review is to address topics important to the practicingintensivist.

Data Synthesis and Findings: Whenever available, preferentialconsideration was given to randomized controlled trials. In theabsence of randomized trials, observational and retrospectivestudies and consensus opinions were included.

Conclusions: Acute kidney injury in the intensive care unit is aclinically relevant problem requiring awareness and expertiseamong physicians from a wide variety of fields. Although manyquestions remain controversial and without definitive answers, aperiodic update of this rapidly evolving field provides a frameworkfor understanding and managing acute kidney injury in the inten-sive care unit. (Crit Care Med 2010; 38:000–000)

KEY WORDS: acute kidney injury; intensive care unit

1Crit Care Med 2010 Vol. 38, No. 1

More recently the Acute Kidney InjuryNetwork (AKIN), an international multi-disciplinary organization composed ofnephrologists and intensivists, furthermodified the RIFLE criteria recognizingthat even very small changes in SCr(�0.3 mg/dL) adversely impact clinicaloutcome (6, 7, 10, 11, 19, 21, 41). Accord-ing to AKIN, the most current consensusdiagnostic criteria for AKI is “an abrupt(within 48 hrs) reduction in kidney func-tion currently defined as an absolute in-crease in serum creatinine of more thanor equal to 0.3 mg/dL (�26.4 �mol/L), apercentage increase in serum creatinineof �50% (1.5-fold from baseline), or areduction in urine output (documentedoliguria of �0.5 mL/kg per hr for �6hrs)” (42). Importantly, the AKIN defini-tion and classification system incorpo-rates creatinine, urine output, and time(Table 1). Both the RIFLE and AKIN cri-teria were developed to facilitate clinicalinvestigation and comparison acrossstudy populations. Epidemiologic datacomparing the RIFLE and AKIN criteriahave demonstrated concordance in criti-cally ill patients (43, 44).

Diagnosis of Acute KidneyInjury in the Intensive Care Unit

Traditional tools to diagnose AKI(SCr) and determine etiology of AKI(clinical history, physical examination,renal ultrasound, fractional excretion ofsodium [FeNa], fractional excretion ofurea, blood urea nitrogen [BUN], andurine microscopy) remain the corner-stone of diagnostic tools available to the

clinician in the ICU. The use of SCr toestimate GFR is limited, however, by thelack of steady-state conditions in criti-cally ill patients. Determinants of the SCr(rate of production, apparent volume ofdistribution, and rate of elimination) arevariable in the ICU setting (6, 8–11, 45,46). Medications (e.g., trimethoprim,cimetidine) impair creatinine secretionand therefore may cause increases in SCrwithout reflecting a true decrease inGFR. Finally, SCr lacks sensitivity andunderestimates the degree of kidney dys-function in a critically ill patient. In-creases in SCr substantially lag behind a

reduction in GFR (Fig. 1) and thus do notprovide a useful real-time assessment ofGFR.

AKI spans the continuum from prere-nal azotemia to acute tubular necrosis,from functional to structural injury. Ef-forts to differentiate between these twoentities have classically included FeNaand urine microscopy. Urine microscopycan be helpful in differential diagnosis(e.g., granular casts and renal tubularepithelial cells in acute tubular necrosis,cellular casts in glomerular injury, eosi-nophiluria in acute interstitial nephritis,or atheroembolic AKI). Of clinical note,

Figure 1. Relationship between GFR and SCr. Large changes in GFR (e.g., 50% decrease from 120mL/min to 60 mL/min) are reflected in only small changes in SCr (0.7 mg/dL to 1.2 mg/dL).

Table 1. Classification/staging systems for acute kidney injury

RIFLE SCr Criteria UOP CriteriaAKINStage SCr Criteria UOP Criteria

R 1 SCr � 1.5 �0.5 mL/kg/hr � 6 hrs 1 1 in SCr �0.3 mg/dL or 1�150% to 200% frombaseline (1.5- to 2-fold)

�0.5 mL/kg/hr for �8 hrs

I 1 SCr � 2 �0.5 mL/kg/hr � 12 hrs 2 1 in SCr to �200% to 300%from baseline(�2- to 3-fold)

�0.5 mL/kg/hr for �12 hrs

F 1 SCr � 3, or SCr �4 mg/dLwith an acute rise of at least0.5 mg/dL

�0.5 mL/kg/hr � 24 hrsor anuria � 12 hrs

3 1 in SCr to �300% (3-fold)from baseline or SCr �4mg/dL with an acute rise ofat least 0.5 mg/dL

�0.5 mL/kg/hr � 24 hrs oranuria � 12 hrs

L Persistent loss of kidney functionfor �4 wks

E Persistent loss of kidney functionfor �3 months

RIFLE, risk, injury, failure, loss, end-stage kidney disease; AKIN, acute kidney injury network; UOP, urine output.RIFLE criteria adapted from Bellomo et al (40). AKIN criteria adapted from Mehta et al (42).

2 Crit Care Med 2010 Vol. 38, No. 1

nephrologist review of urine microscopyhas been demonstrated to be superior toclinical laboratory interpretation (47).Using a proposed scoring system, micro-scopic examination of the urine sedimentis a highly predictive method for differ-entiating prerenal azotemia from acutetubular necrosis (48). However, the pres-ence of muddy brown casts and renaltubular epithelial cells are usually seenrelatively late and thus are not sensitivefor early detection of AKI (49, 50). FeNa isfrequently useful for differentiating “pre-renal” (diminished renal perfusion, FeNa�1%) from “intra-renal” (ischemia ornephrotoxins, FeNa �2%) (50, 51). Urinemicroscopy and FeNa can be valuabletools in determining the cause of AKI buthave no current role in early detection ordiagnosis of AKI. Furthermore, “prere-nal” and “intra-renal” causes of AKI com-monly coexist in the ICU patient.

Prerenal azotemia, in the absence ofvalidated new diagnostic biomarkers, of-ten remains a retrospective diagnosis,made only after response to a volumechallenge. Whereas it is important to ap-propriately identify and treat prerenalazotemia, fluid administration is notwithout consequence in the critically illpatient. A complete assessment of the pa-tient’s overall volume status is pivotalbefore aggressive resuscitative efforts toenhance renal perfusion. This is of par-ticular importance considering data dem-onstrating adverse effects of volume over-load in critically ill patients (52, 53).Because of the limitations of traditionaltools, novel candidate biomarkers of AKI(discussed separately) are being activelyinvestigated.

Common Causes of AcuteKidney Injury in the IntensiveCare Unit

The cause of AKI in the ICU is com-monly “multi-factorial” and frequentlydevelops from a combination of hypovo-lemia, sepsis, medications, and hemody-namic perturbations (Table 2). It is fre-quently not possible to isolate a singlecause, thereby further complicating thesearch for effective interventions in thiscomplex disease process. The pathophys-iology of AKI varies according to the un-derlying etiology and is beyond the scopeof this article.

Sepsis is the most common cause ofAKI in a general ICU, accounting for upto 50% of cases (6, 8–11, 23, 45, 54). AKIis common after cardiac surgery, occur-

ring in up to 42% of patients withoutpre-existing kidney disease, and is associ-ated with increased morbidity and mor-tality with elevations in SCr as small as0.3 mg/dL (19). Trauma associated AKI ismulti-factorial (e.g., hemorrhagic shock,abdominal compartment syndrome,rhabdomyolysis) and occurs in up to 31%of adult trauma patients (55). The kid-neys are early sensors of intra-abdominalhypertension and abdominal compart-ment pressures �12 mm Hg may be as-sociated with AKI (56). A sustained intra-abdominal pressure �20 mm Hg inassociation with new organ dysfunctionwill be associated with AKI in �30% ofcases (57, 58). Rhabdomyolysis accountsfor 28% of trauma-associated AKI requir-ing dialysis (59).

Medications are a common cause ofAKI and, according to Uchino et al (9),account for nearly 20% of all cases of AKIin the ICU. The mechanism of medicationinduced AKI is variable and includesacute interstitial nephritis, direct tubulartoxicity (e.g., aminoglycosides), and he-modynamic perturbations (e.g., nonste-roidal anti-inflammatory agents, angio-tensin-converting enzyme inhibitors).Acute interstitial nephritis is likely anunder-recognized etiology of medication-associated AKI in the ICU because of therelative paucity of clinical findings andneed for high index of suspicion. Table 3lists common nephrotoxins encounteredin the care of critically ill patients.

Prevention and Management ofAcute Kidney Injury in theIntensive Care Unit

Primary prevention of AKI in the ICUis limited to those conditions in whichthe timing of injury are predictable, suchas exposure to radiocontrast dye, cardio-

pulmonary bypass, large-volume para-centesis in a cirrhotic patient, or chemo-therapy. In contrast to most cases ofcommunity-acquired AKI, nearly all casesof ICU-associated AKI result from morethan a single insult (6, 8–11, 45, 50, 60,61). In the critically ill patient, the firstkidney insult is often not predictable.Therefore, prevention of AKI in the ICUoften means prevention of a secondaryinsult in an “at-risk” patient. For exam-ple, in a retrospective study of �5000 ICUpatients, 67% of patients had AKI de-velop, and 45% of AKI occurred after ICUadmission (6). It is in these patients thatthere is a potential role for prevention.

General principles of “secondary” AKIprevention include: (1) recognition of un-derlying risk factors that predispose pa-tients to AKI (e.g., diabetes, chronic kid-ney disease, age, hypertension, cardiac orliver dysfunction); and (2) maintenanceof renal perfusion, avoidance of hypergly-cemia, and avoidance of nephrotoxins inthese high-risk patients. Specific clinicalsituations in which there is evidence forpreventive strategies (e.g., contrast expo-sure, hepatorenal syndrome [HRS]) arediscussed.

Preventing Contrast-Induced Ne-phropathy. The primary strategies for con-trast-induced nephropathy (CIN) preven-tion include hydration, N-acetylcysteine(NAC), and use of low-volume nonioniclow-osmolar or iso-osmolar contrast. Nostrategy has been effective in completelypreventing CIN. Risk factors for CIN in-clude diabetes, CKD, hypotension, effec-tive or true volume depletion (including

Table 2. Common causes of AKI in the ICU

Five Most Common Causes of AKI in the ICUa

● Sepsis (most common)● Major surgery● Low cardiac output● Hypovolemia● Medications

Other Common Causes of AKI in the ICU● Hepatorenal syndrome● Trauma● Cardiopulmonary bypass● Abdominal compartment syndrome● Rhabdomyolysis● Obstruction

aThe five most common causes of AKI in theICU based on nearly 30,000 patients (9).

Table 3. Common nephrotoxins that cause AKI inICU patients

Exogenous● Medications

● NSAIDS● Antimicrobials

● Aminoglycosides● Amphotericin● Penicillinsa

● Acyclovirb

● Chemotherapeutic agents● Radiocontrast dye● Ingestions

● Ethylene glycolEndogenous

● Rhabdomyolysis● Hemolysis (HUS/TTP)● Tumor lysis syndrome

NSAIDS, non-steroidal anti-inflammatorydrugs; HUS, hemolytic uremic syndrome; TTP,thrombotic thrombocytopenic purpura.

aAcute interstitial nephritis (AIN); bcrystalnephropathy.


cirrhosis and congestive heart failure),and concurrent use of nephrotoxic med-ications. Critically ill patients intuitivelyrepresent a patient population at highrisk for CIN given frequent hemodynamicinstability, multiple organ dysfunction,use of nephrotoxic medications, and mul-tiple underlying comorbidities (e.g., dia-betes, CKD). However, despite the largenumber of randomized controlled trials(RCT) published on prevention strategiesfor CIN, there has been only one RCTperformed specifically in critically illadults (111). The true incidence of andrisk for CIN in critically ill patients isthus unknown.

Adequate hydration is a well-estab-lished measure to decrease the risk ofCIN, whereas the choice of fluid remainscontroversial. Trials comparing the use ofsodium bicarbonate and sodium chloridefor the prevention of CIN have yieldedconflicting results. Five meta-analyses ofsodium bicarbonate suggest a beneficialrole of isotonic sodium bicarbonate overisotonic saline (112–116); however, thereis considerable heterogeneity and somepublication bias confounding these find-ings. The most recent RCT of bicarbonatevs. normal saline showed no difference inthe primary outcome of �25% decre-ment in GFR within 4 days (117). Basedon currently available evidence, there is astrong suggestion that sodium bicarbon-ate may be superior to isotonic saline todecrease the risk of CIN.

NAC is a free radical scavenger shownto decrease the risk of CIN compared toplacebo (118). Since 2003, �10 meta-analyses published on the role of NAC inCIN have yielded conflicting results likelyattributable, in part, to heterogeneity in-patient populations. In a recent meta-analysis of 41 studies, NAC plus salinereduced the risk for CIN more effectivelythan saline alone (119). A previous meta-analysis in 2007 by Gonzales et al (120)did not support the efficacy of NAC toprevent or decrease the risk of CIN. Fur-thermore, there are conflicting data as towhether NAC, itself, may decrease SCrmeasurement without affecting GFR(121, 122).

Low-volume nonionic low-osmolar oriso-osmolar contrast preparations areclearly associated with a decrease in CINwhen compared to high osmolar agents.The data regarding nonionic low-osmolarcontrast media vs. iso-osmolar contrastmedia (currently only iodixanol) is con-troversial. Two meta-analyses report con-flicting results (123, 124). McCullough et

al (123) found that use of iso-osmolar con-trast media resulted in a lower incidence ofCIN when compared to low-osmolar con-trast media. However, Heinrich et al (124),in the most recent meta-analysis, re-ported no significant difference betweenthe two unless the low-osmolar contrastmedia was iohexol, suggesting that alllow-osmolar contrast media preparationsmay not be the same.

Both small observational and prospec-tive studies have shown an increase in therisk of CIN with peri-procedural use ofangiotensin-converting enzyme inhibi-tors (125–127). However, a recent ran-domized prospective trial performed instable outpatients did not show any dif-ference in incidence of CIN between pa-tients who did or did not discontinueangiotensin-converting enzyme inhibi-tors or angiotensin receptor blockers be-fore contrast (128). Angiotensin-convert-ing enzyme inhibitors have not beenprospectively studied in the critically ill.Therefore, although there is currently in-sufficient evidence to support discontin-uation of these medications in criticallyill adults, further study is warrantedgiven the widespread use of these agentsin clinical practice.

Whereas the use of peri-proceduralhemofiltration in patients undergoingpercutaneous coronary intervention wasshown, in two studies, to decrease therisk of AKI (5% vs. 50%; p � 0.0001)(129, 130), this has not been widelyadopted into clinical practice. In a sys-tematic review of extracorporeal thera-pies for prevention of CIN, analysis of thehemodialysis studies alone (including fiveRCT), there was no benefit of hemodial-ysis and, in fact, there was a trend favor-ing standard therapy compared to pro-phylactic hemodialysis (131). Asubsequent RCT of prophylactic hemodi-alysis in 82 patients with advanced CKD(baseline SCr 4.9 mg/dL) demonstratedimproved outcomes (shorter length ofstay and lower rate of long-term dialysisdependence after hospital discharge) withprophylactic hemodialysis (132). A criti-cal limitation of all of these studies is thatthe clinical end point SCr was directlyimpacted by the intervention itself (he-mofiltration or hemodialysis).

Fenoldopam and theophylline are twoadditional agents that have been consid-ered for their potential role in the pre-vention of CIN. None of the four RCTcomparing fenoldopam to either salinealone (133, 134) or NAC (135, 136) dem-onstrated any beneficial effect in the pre-

vention of CIN. The role of theophyllinefor CIN prevention is inconsistent acrossstudies. Although two meta-analyses sug-gest that prophylactic theophylline mayprovide some benefit, the studies wereperformed in primarily low-risk patients,and clinically relevant outcomes were notconsistently reported (137, 138). There-fore, we cannot currently recommend theuse of theophylline for prevention of CINin critically ill patients.

The majority of these studies were notperformed in critically ill patients andtherefore provide no definitive guidanceas to how the risk of CIN in the criticallyill should be ameliorated. Because of theabsence of sufficient data in the patientpopulation of interest, clinicians must ex-trapolate from the best available evidencefrom other patient populations. There-fore, our recommendations include: (1)avoid use of intravenous contrast in high-risk patients if alterative imaging tech-niques are available; (2) avoid preexpo-sure volume expansion using eitherbicarbonate or isotonic saline; (3) al-though of questionable benefit, use ofNAC is safe, inexpensive, and may de-crease risk of AKI; (4) avoid concomitantuse of nephrotoxic medications if possi-ble; and (5) use low-volume low-osmolaror iso-osmolar contrast. Future studiesare needed to determine the true role ofthese preventive measures in critically illpatients.

Preventing Acute Kidney Injury inHepatic Dysfunction. AKI is a commoncomplication of critically ill patients withhepatic failure. Pentoxifylline decreasesthe incidence of AKI attributable to HRSin acute alcoholic hepatitis (139). Use ofintravenous albumin in patients with cir-rhosis and spontaneous bacterial perito-nitis significantly reduces both the inci-dence of AKI (33% to 10%) and mortality(41% to 22%) (140). Albumin decreasesthe incidence of AKI after large-volumeparacentesis (141), and when used incombination with splanchnic vasocon-stricting agents (e.g., terlipressin) maydecrease mortality in HRS (142, 143).However, definitive therapy for AKI as aconsequence of HRS remains liver trans-plantation in appropriate candidates. Fiverandomized trials of vasoconstrictingagents (terlipressin or noradrenalin) plusalbumin in the treatment of HRS all dem-onstrated improved renal function inHRS (144–148). A mortality benefit wasonly demonstrated in responders to ther-apy (145). Terlipressin is not available inthe US. In a retrospective study per-


formed in the US, patients treated withvasopressin had significantly higher re-covery rates and improved survival whencompared to octreotide alone (149). Fur-thermore, findings from three small ob-servational and retrospective studiesdemonstrate improved outcomes withmidodrine and octreotide (HRS reversaland decreased mortality) (150 –152).These findings justify a larger RCT toappropriately evaluate this treatmentmodality.

Management of AKI in the ICU re-volves around optimizing hemodynamicsand renal perfusion, correcting metabolicderangements, providing adequate nutri-tion, and mitigating progression of in-jury. These management considerationsare discussed.

Maintain Renal Perfusion. Optimiza-tion of renal perfusion may require vol-ume resuscitation, inotropic, or vasopres-sor support. Extrapolated primarily fromanimal studies (62, 63), the human kid-ney has a compromised ability to auto-regulate (maintain constancy of renalblood flow and GFR over a wide range ofrenal perfusion pressures) in AKI. There-fore, as a priority, prevention or manage-ment of AKI should include maintenanceof hemodynamic stability and avoidanceof volume depletion. A mean arterialpressure of �65 mm Hg is a generallyaccepted target; however, the data arelimited (64, 65) and do not include pa-tients with established AKI (loss of auto-regulation). The level at which renalblood flow becomes dependent on sys-temic arterial pressure varies signifi-cantly based on age, underlying illness(e.g., hypertension), and the acute illnessor condition (AKI, sepsis, and cardiopul-monary bypass). After volume resuscita-tion, blood flow should be restored towithin autoregulatory parameters. Thisfrequently requires vasopressor or inotro-pic support in the setting of septic shock,the most common cause of AKI in theICU. There are currently no RCT compar-ing vasopressor agents; therefore, there isno evidence that, from a renal protectionstandpoint, there is a vasopressor agentof choice to improve kidney outcomes.

Decreased renal blood flow (attribut-able to either hypotension or high renalvascular resistance, from an imbalancebetween renal vasoconstriction and vaso-dilation) is a common feature in manyforms of AKI. Consequently, there hasbeen considerable interest in renal vaso-dilators to maintain renal perfusion forprevention or treatment of AKI. Whereas

dopamine infusion may cause a transientimprovement in urine output (66), “re-nal” dose dopamine does not reduce theincidence of AKI, the need for RRT, orimprove outcomes in AKI (66–71). Fur-thermore, “low-dose” dopamine mayworsen renal perfusion in critically illadults with AKI (72) and is associatedwith increased myocardial oxygen de-mand and an increased incidence of atrialfibrillation (73). There is additional con-cern for extrarenal adverse effects of do-pamine, including negative immuno-modulating effects (74). Thus, there isbroad consensus that dopamine is poten-tially harmful and without evidence ofclinical benefit for either prevention ortreatment of AKI. Therefore, its contin-ued use for putative “renal protection”should be avoided.

Fenoldopam is a selective dopamine-1receptor agonist approved for the treat-ment of hypertensive crisis (75). Paradox-ically, the lowest doses of fenoldopam(�1 �g/kg per min) are purported toincrease renal blood flow without sys-temic effects. Despite encouraging datafrom pilot studies, (76–78) a prospectiveplacebo-controlled study of low-dosefenoldopam in sepsis failed to decreasemortality or need for RRT despite asmaller increase in SCr (79). Larger stud-ies to validate the meta-analytic observa-tion that fenoldopam both reduces theneed for RRT (OR, 0.54; p � 0.007) anddecreases mortality (OR, 0.64; p � 0.01)(80) are currently ongoing in cardiac sur-gery patients (clinicaltrials.gov ID:NCT00557219).

Fluid Choice in Acute Kidney Injury.The primary physiologic intention of vol-ume resuscitation is the restoration ofcirculating volume to prevent or mitigateorgan injury. The kidneys normally re-ceive up to 25% of the cardiac output andare exquisitely sensitive to hypoperfusionattributable to true or relative hypovole-mia. For this reason, the question ofwhether a particular type of fluid influ-ences development of AKI is of pivotalimportance.

Whereas crystalloid solutions remainthe preferred treatment in usual care, thedebate over whether colloid solutionsprovide any additional benefit remains anarea of active investigation (81–85). In alandmark trial evaluating the impact offluid choice on clinical outcomes, theSAFE study investigators randomizednearly 7000 patients to volume resuscita-tion with saline or albumin. They dem-onstrated no difference in survival or

need for RRT between the two groups(86). In post hoc subgroup analysis, re-suscitation with albumin was associatedwith increased mortality in critically illpatients after traumatic brain injury (87).In contrast, there was a trend towardimproved survival in septic shock patientsreceiving albumin (30.7% in albumingroup vs. 35.3% in saline group; p � 0.09)(86). Based on currently available litera-ture, there is no evidence of a mortalitybenefit supporting the preferential use ofalbumin over crystalloids in a heteroge-nous critically ill patient population (84).

Synthetic colloids (e.g., hydroxyethylstarches, dextrans) are still widely useddespite multiple reported safety concernswith regard to renal outcomes (88–90).An increased risk of AKI with the use ofhydroxyethyl starches has been demon-strated in multiple small studies, andmost recently a systematic review of 12randomized trials demonstrated an in-creased risk of AKI with the use of hy-droxyethyl starches among patients withsepsis (91). In contrast, the largest indi-vidual retrospective analysis (SOAP studycohort92) explored the effects of hydroxy-ethyl starches on renal function and didnot find the use of hydroxyethyl starchesto be an independent risk factor for AKIor need for RRT (93). The dose and prep-aration varied between studies. The ad-verse event profile has been linked, inpart, to the individual preparation, withthe lowest molecular weight offering thebest side effect profile.

The question of fluid managementdoes not end with the choice of fluid;careful consideration of the amount offluid administered is also important. Crit-ical illness is a dynamic process requiringfrequent assessment of and adjustment tofluid status. In a prospective RCT of pa-tients with ARDS, a fluid conservativestrategy decreased ventilator days and didnot increase the need for RRT (53). Fur-thermore, an observational study of�3000 patients demonstrated an associ-ation between positive fluid balance andincreased mortality in patients with AKI(52). However, the question remainswhether this is simply a marker of sever-ity of illness or true causation; this ob-servation warrants further investigation.

Avoid Hyperglycemia. Although thebeneficial effects of intensive insulintherapy on mortality in critically ill pa-tients remains controversial (94–96), twolarge RCT demonstrated a decreased in-cidence of AKI and a decreased require-ment for RRT with tight glucose control


(95, 96). Furthermore, a more detailedsecondary analysis strongly suggests thattight blood glucose control may be reno-protective in critically ill patients (97).Two smaller retrospective studies re-ported similar results (decreased inci-dence of AKI and decreased need for post-operative dialysis) in nondiabetic cardiacsurgical patients (98) and in patients re-ceiving TPN (99). However, in contrast,in the largest and most recent prospec-tive RCT of intensive vs. conventionalglucose control in �6000 critically ill pa-tients, there was no difference in thenumber of patients requiring RRT (94).The overall incidence of AKI, however,was not reported in this study. It there-fore remains unclear if there is a reno-protective role for tight glycemic controland, if present, whether any such effect isattributable to the avoidance of glucosetoxicity or a beneficial effect of insulin.These findings warrant further study, es-pecially in view of the fact that intensiveglycemic control may be associated witha higher frequency of clinically relevanthypoglycemia.

Avoid Nephrotoxins. Nephrotoxicmedications are a contributing factor inup to 25% of all severe AKI in critically illpatients (8, 9; Table 3). Identification ofat-risk patients is pivotal. Aminoglyco-sides, although less commonly used forsevere Gram-negative infections thanpreviously, are associated with significantnephrotoxicity. Although once-daily dos-ing of aminoglycosides has been shown,in some studies, to decrease the inci-dence of AKI (100, 101), published meta-analyses support comparable efficacy anddecreased cost but do not consistentlydemonstrate a significant reduction innephrotoxicity (102–106). Extended in-terval dosing should not be used in pa-tients with CKD. Standard amphotericinB has been associated with AKI in 25% to30% of patients (107). The lipid formula-tion of amphotericin B is preferred be-cause of reduced nephrotoxicity of 19%vs. 34% (108). Caspofungin, a newer an-tifungal agent, is associated with an evensafer renal profile (109). The use of apro-tinin, a serine protease inhibitor used todecrease blood loss during cardiac sur-gery, has been associated with increasedrisk of AKI and need for dialysis (110).

ICU patients frequently have fluctuat-ing renal function and a variable volumeof distribution. Standard estimates of re-nal function are poor in critically ill pa-tients. Therefore, medications must becarefully dose adjusted because of varied

pharmacokinetics in critically ill patientswith and without underlying CKD.

Diuretics in Acute Kidney Injury. Useof diuretics in the prevention or treat-ment of AKI has physiologic merit but itsuse is not supported by prospective clin-ical study. Diuretics can increase urineoutput but have not been found to have aconsistent impact on mortality (153–157). Mehta et al (157) demonstrated thatfailure to respond to diuretics was asso-ciated with an increased risk of death andnon-recovery of renal function. Subse-quently, in a large, prospective, multina-tional study, Uchino et al (158) did notdemonstrate an increased mortality, thusleaving unresolved the therapeutic role ofdiuretics in critically ill patients with re-nal dysfunction. Although oliguric AKIhas been associated with worse outcomesthan nonoliguric AKI (159), there is noevidence supporting efforts to convertnonoliguric AKI with diuretics. Diureticshave not been found to shorten the du-ration of AKI, reduce the need for RRT, orimprove overall outcomes (160). Further-more, a recently published RCT compar-ing the use of furosemide vs. placebo inthe recovery phase of AKI requiring con-tinuous renal replacement therapy(CRRT), furosemide was found to in-crease urine output and sodium excretionbut did not improve renal recovery (161).In a multinational survey, nephrologistsand intensivists reported clinical uncer-tainty about the use of diuretics in AKI,thus justifying the need for a definitiveRCT (162).

Because diuretic use in AKI has notbeen shown to decrease mortality, thereis no role for diuretics to convert oliguricAKI to nonoliguric AKI. However, regard-ing an increased appreciation for the po-tential detrimental downstream effects ofvolume overload, it may be reasonable totry diuretics for control of volume over-load. The clinician should, however, becareful not to delay initiation of RRT forvolume overload in the critically ill pa-tient with AKI.

Nutritional Considerations. Malnutri-tion in hospitalized patients is associatedwith increased mortality (163). Assess-ment of the nutritional status of criticallyill patients is limited by the unreliabilityof traditional markers of nutritional sta-tus in critical illness in general, and AKIin particular. Prealbumin is excretedmainly by the kidneys and hence may befalsely elevated in patients with AKI(164). Patients with AKI are hypercata-bolic with a negative nitrogen balance

(165), resulting from both increased pro-tein catabolism and impaired protein syn-thesis. Therefore, according to the asso-ciated degree of catabolism, caloricsupport with 25 to 35 kcal/kg per day isrecommended in AKI.

The impact of CRRT on nutrition inthe ICU is two-fold. Because protein ca-tabolism is markedly increased in mostpatients requiring CRRT (165–167), theuse of CRRT enhances the clinician’s abil-ity to provide adequate nutrition becauseof an improved ability to manage volume.Unfortunately, the recommendedamount of protein in this population re-mains controversial and recommenda-tions are based solely on expert opinion,because there are no data available fromRCT. Although there are no studies dem-onstrating a benefit in outcomes (e.g.,survival or dialysis-free days), consensusrecommendations include nonprotein ca-loric intake of 20 to 30 kcal/kg bodyweight per day and a protein intake of 1.5g/kg per day (168). However, several stud-ies have demonstrated a less negative oreven positive nitrogen balance in thosepatients receiving up to 2.5 g/kg per daywhile receiving CRRT without evidence ofadverse effects (169–171). An increase innonprotein calories in critically ill pa-tients with AKI does not improve nitro-gen balance (172).

Renal Replacement Therapy forAcute Kidney Injury in theIntensive Care Unit

Despite decades of clinical trials inves-tigating potential pharmacologic inter-ventions in AKI, current treatment op-tions are primarily limited to RRT.Practice patterns vary widely regardingtiming of initiation of RRT, dose deliv-ered, and choice of modality as evidencedby international surveys (173–176).There is no current consensus on theindications for RRT for AKI. With agreater appreciation for and understand-ing of the role of the kidney in distantorgan injury (177), it may be more appro-priate to consider renal replacementtherapy as renal supportive therapy (178).For the purposes of this review, we reviewthe most up-to-date evidence availableaddressing timing, dosing, and modalityof RRT.

Timing of Renal Replacement. Thereis little prospective data regarding theappropriate timing of initiation of RRTand that which are available are incon-clusive. The “absolute” indications for


initiation of dialysis (severe hyperkale-mia, clinically apparent signs of uremia,severe acidemia, and volume overload, in-cluding pulmonary edema complicatedby hypoxia or cardiogenic shock) arebroadly accepted usual care standards.“Prophylactic” dialysis was introduced inthe 1960s (179), and the first prospectivestudy was published in 1975 comparing aBUN trigger of 70 mg/dL vs. nearly 150mg/dL (180). Survival was 64% in the“early intervention” group as comparedto 20% in the non-intensive or standardintervention group (p � 0.01). Conven-tional teaching based on this and otherstudies (181, 182) has been to initiateRRT before a BUN exceeds 100 mg/dL.Unfortunately, not only is the “ideal”BUN not established but also BUN per seis an imperfect reference value because itis widely influenced by nonrenal factors.

More recently, a review of the datafrom the PICARD study demonstrated anincreased risk of death associated withinitiation of RRT with a BUN �76 mg/dL

in comparison to �76 mg/dL (183). Animportant limitation of this study is thatpatients who were conservatively man-aged (did not receive RRT) are “invisible”in this analysis, thereby limiting the va-lidity of the findings regarding impact onmortality. In the only randomized studyof timing of CRRT initiation (n � 106),there was no effect on mortality (184).“Early” dialysis was initiated after 6 hrs ofoliguria. Of the 36 patients included inthe “late” arm of this study, six patientsdid not receive RRT, of whom four sur-vived, a fact that likely influenced theresults of this study. Results from a largeprospective multi-centered observationalstudy of �1200 patients were internallyinconsistent and dependent on the defi-nition of “early” or “late” initiation ofRRT (185). In this study, “late” initiationof RRT was associated with worse out-comes (higher crude mortality, longerduration of RRT, increased hospitallength of stay, and greater dialysis depen-dence) when “late” was defined relative to

date of ICU admission. However, therewas no difference in crude mortality if thetiming was defined by serum urea. Fi-nally, there was a lower crude mortality iftiming of RRT initiation was defined bySCr at initiation (higher SCr associatedwith a lower mortality) (185). Unfortu-nately, the question of timing remainsunanswered and controversial (185, 186).There is clearly a need for a large RCT,with a clear definition of “early,” to helpguide the clinician in determining theappropriate timing for initiation of RRTfor AKI in the ICU.

Choosing a Renal Replacement Dose.Six prospective RCT have been publishedaddressing the question of dose of RRT incritically ill adults (37, 184, 187–190; Ta-ble 4. Three of these studies suggest thata higher dose of dialysis translates intoimproved outcomes, specifically de-creased mortality (37, 187, 188). Ronco etal (187) published the first RCT in 2000addressing this question. These investiga-tors compared 20, 35, and 45 mL/kg per

Table 4. Summary of randomized controlled trials of dosing strategies for renal replacement therapy for acute kidney injury in the intensive care unit

Author N Design RRT Modality RRT Doses Prescribed/Delivered Survival

RCT With MortalityDifference

Ronco 2000 (187) 425 Single center CVVH (post-filter dilution) (P) 20 mL/kg/hr 15-day: 41%(P) 35 mL/kg/hr 57%(P) 45 mL/kg/hr 58%(D) �85%

(P) in 100% patientsSchiffl 2002 (37) 160 Single center Intermittent HD: daily vs.

alternate dayDaily HD Kt/V

(P) 1.19(D) 0.92

28 day: 72%

Saudan 2006 (188) 206 Single center CVVH vs. CVVHDF(pre-filter dilution)

Alternate day HD Kt/V(P) 1.21(D) 0.94

54%

(D) mean: 25 mL/kg/hr87% of (P)

28-day: 39%

(D) Mean: 42 mL/kg/hr83% of (P)

59%

Includes mean 24 mL/kg/hrreplacement and 18 mL/kg/hrdialysate

RCT Without MortalityDifference

Bouman 2002 (184) 106 Two centers CVVH (post-filter dilution):early high-volume vs.early low-volume vs. latelow-volume

(D) Mean: 48 ml/kg/hr (early) 28-day: 74%

(D) Mean: 20 ml/kg/hr (early) 69%

(D) Mean: 19 ml/kg/hr (late) 75%

Tolwani 2008 (189) 200 Single center CVVHDF(pre-filter dilution)

(P) 20 mL/kg/hr ICU d/c or30-day: 56%(D) 17 mL/kg/hr

(P) 35 mL/kg/hr 49%(D) 29 mL/kg/hr

Palevsky 2008 (190) 1124 Multicenter Intensive vs. less intensiveRRT (CVVHDF or SLEDor HD)

(P) 21 mL/kg/hr or SLED or HD 3�/wk 60-day: 44%(D) 22 mL/kg/hr or Kt/V 1.3 3�/wk 49%(P) 36 mL/kg/hr or SLED or HD 6�/wk(D) 35 mL/kg/hr or Kt/V 1.3 6�/wk

P, prescribed; D, delivered; CVVH, continuous veno-venous hemofiltration; HD, hemodialysis; CVVHDF, continuous veno-venous hemodiafiltration;SLED, slow low-efficiency dialysis.


hr dosing strategies. There was a highmortality in all groups but a statisticallylower mortality in the two groups withhigher dose of ultrafiltration (35 and 45mL/kg per hr) without any difference incomplication rates between groups (187).In 2002, Schiffl et al (37) found dailydialysis to be superior to alternate daydialysis in a prospective randomizedstudy. There were significantly fewer hy-potensive episodes in the daily dialysisgroup (5% vs. 25%). In an intention-to-treat analysis, mortality was 28% for dailydialysis and 46% for alternate-day dialysis(p � 0.01) (37). An important limitationof this study is that the delivered dose wassignificantly less than the prescribeddose; therefore, the daily dialysis groupreceived only “adequate” therapy asjudged by contemporary standards. Itmay be said, therefore, that it was a com-parison between adequate and inadequatedialysis. In 2006, Saudan et al demon-strated that addition of dialysate (1–1.5L/hr) to CVVH (1–2.5 L/hr; continuousveno-venous hemodiafiltration improved28- and 90-day survival when comparedto CVVH alone in 206 critically ill adults;39% vs. 59%; p � 0.03 and 34% vs. 59%;p � 0.0005, respectively), suggesting thatsmall solute clearance is important (188).

In contrast, three prospective RCThave demonstrated no difference in mor-tality (184, 189, 190; (Table 4). Boumanet al (184), in 2002, showed no differencein 28-day mortality when comparingearly high-volume hemofiltration, earlylow-volume hemofiltration vs. late low-volume hemofiltration with the mediandose (mL/kg per hr) of 48, 20, and 19,respectively. More recently, Tolwani et al(189) compared two different doses, 20mL/kg per hr and 35 mL/kg per hr, ofcontinuous veno-venous hemodiafiltra-tion (pre-filter) and found no differencein 30-day mortality (44% vs. 51%, p �0.32). Of note, the delivered dose in thesetwo groups were 17 mL/kg per hr and 29mL/kg per hr, respectively (189). The larg-est and only multi-centered trial designedto address the question of dose of RRT incritically ill adults is the acute tubular ne-crosis study published in 2008 (190). Thiswas a two-arm study comparing intensiveto standard RRT. The intensive therapygroup underwent daily dialysis, continu-ous veno-venous hemodiafiltration, orsustained low-efficiency dialysis at a doseof 35 mL/kg per hr, whereas the standardtherapy group had alternate day dialysis(three times per wk), continuous veno-venous hemodiafiltration, or sustained

low-efficiency dialysis at 20 mL/kg per hr.Notably, patients were able to move fromintermittent to continuous modalitiesbased on hemodynamic stability but theystayed within their assigned intensive orstandard treatment therapy groups.There was no difference in the primaryoutcome, death from any cause (190).The RENAL study, comparing continuousveno-venous hemodiafiltration 25 mL/kgper hr to 40 mL/kg per hr, has completedenrollment but results have not yet beenpublished.

An important factor in considering theresults of the currently available data arethe difference between study populations,use of solely convective or combinationconvective and diffusive modalities, andthe potential gap between prescribed anddelivered doses. Findings from these neg-ative trials should not be interpreted tomean that dose is not important. On thecontrary, it is likely that dose is impor-tant and, above a minimal dose, furtherescalation may not provide additionalbenefit. Based on currently available data,it is our recommendation that to ensurean actual delivered dose of 20 mL/kg perhr for continuous modalities one mustprescribe a higher dose (e.g., 25 mL/kgper hr) to account for filter clotting, timeoff the machine for interventions, or ra-diographic studies, etc. For intermittentRRT, one should target a Kt/V of 1.2 to1.4 per treatment for alternate day (threetimes per wk) hemodialysis. Further-more, in addition to an appropriate targetdose, there must be close attention givento the actual delivered dose. In summary,one dose does not fit all; RRT dose mustbe weight-adjusted.

Choosing a Renal Replacement Mo-dality. Continuous RRT modalities moreclosely approximate normal physiologywith slow correction of metabolic de-rangements and removal of fluid. There-fore, CRRT is commonly thought to bebetter-tolerated in the critically ill andhemodynamically unstable patient. Thequestion of superiority remains given theabsence of clear evidence that these ap-parent physiologic advantages translateinto a decrease in ICU or hospital mor-tality (191–196).

Since 2000 there have been seven pro-spective RCT designed to address the im-portant clinical question regarding opti-mal RRT modality (192, 193, 195, 197–200); of these, only three were multi-centered studies (193, 198, 200). Of note,many of these trials, although publishedafter 2000, enrolled patients in the 1990s.

In six of the trials, mortality was theprimary outcome. There have been sev-eral meta-analyses and systematic re-views comparing outcomes of intermit-tent vs. continuous renal replacementmodalities with conflicting results (191,201–204). A recent meta-analysis (ninerandomized trials) comparing intermit-tent to continuous renal replacementtherapy (intermittent RRT vs. CRRT) inAKI demonstrated no difference in mor-tality or renal recovery (defined as inde-pendence from RRT) (202). Of note, mor-tality was the primary outcome in eightof the nine included trials. Mortality,however, may not be the only clinicallysignificant outcome. Two studies haveshown that CRRT is associated with bet-ter long-term kidney recovery when com-pared to intermittent RRT (205, 206). Incontrast, four RCT that included renalrecovery as a primary outcome showedno difference in need for chronic RRT(193, 195, 198, 200). In the absence ofdefinitive data in support of a particularmodality (191, 201), the choice of RRTmodality is currently influenced by mul-tiple factors, including individual siteavailability, expertise, resources, cost,and likely clinician bias.

Hybrid therapies include sustainedlow-efficiency dialysis and extended dailydialysis. These modalities utilize standardintermittent hemodialysis machines butprovide a slower solute and fluid removalsimilar to CRRT technologies. Althoughthere have been no prospective random-ized trials evaluating outcomes, hybridtherapies have been shown to be safe andeffective alternatives to treating AKI incritically ill patients (207, 208).

The question of optimal modality hasnot yet been definitively answered. It isimportant to note that although the datastrongly suggest that there is no differ-ence in outcome between intermittentand continuous modalities, several keypatient populations have been excluded.Namely, hemodynamically unstable pa-tients, brain-injured patients, and thosewith fulminant hepatic failure were ex-cluded and are widely believed to requirecontinuous modalities. Furthermore, acritical limitation of all of the studies isthe absence of a standardized dose (bothwithin and between modalities) (202).RRT, like other medical treatments, mustbe considered in terms of dose adequacyto appropriately draw conclusions regard-ing clinical outcomes. Large randomizedtrials may be necessary to identify other


potential subsets of patients who mightbenefit from continuous modalities.

Anticoagulation is frequently requiredto prevent clotting in extracorporeal cir-cuits. There are no large RCT available toguide the choice of anticoagulation: hep-arin (unfractionated or low-molecular-weight heparin) or citrate-based proto-cols. Bleeding complications remain theprimary concern with anticoagulation.Three small RCT, however, have demon-strated both similar or prolonged filterlife and less bleeding and transfusionwith citrate protocols when compared touse of heparins (209–211). In a recentlarger, randomized, non-blinded trial com-paring citrate to nadroparin, circuit sur-vival was similar in both groups, but thecitrate group had a lower mortality rate(212). Currently available data support theuse of citrate for anticoagulation; however,this requires local expertise.

In summary, whereas RRT remainsthe cornerstone of treatment of AKI inthe ICU, many key questions remain con-troversial. This is a rapidly evolving fieldand requires early consultation for appro-priate expertise in the management ofRRT for the critically ill patient with AKI.

On the Horizon

The identification of novel candidatebiomarkers of early AKI provides hope forthe success of future clinical early inter-vention trials. Advances in treatment ofAKI have been limited by the inability todiagnose AKI early. Previously failed in-terventions may portend different out-comes if implemented earlier in thecourse of AKI. Novel pharmacologicagents on the horizon include erythro-poietic agents and natriuretic peptides.Novel interventions include the use ofstem cell therapy, renal tubule assist de-vice, and high-flux hemofiltration forsepsis.

Candidate Biomarkers. Biomarkers ofAKI in the ICU have three primary poten-tial roles: early detection of AKI, differen-tial diagnosis (e.g., hepatorenal syndromevs. acute tubular necrosis), and prognosis(e.g., need for RRT or mortality). Theideal biomarker for AKI would be sensi-tive, specific, inexpensive, available non-invasively as a point-of-care test, and pro-vide a real-time assessment of GFR. Apanel of biomarkers or kidney functiontests may be needed to address the com-plexity and heterogeneity of AKI in theICU (213). Early identification of AKIwith rapid and reproducible biomarkers

is a critical first step toward improvingoutcomes in AKI.

According to several studies in criti-cally ill patients, serum cystatin C is bet-ter than SCr for early detection of AKI(214, 215) and as a more sensitive markerof small changes in GFR (216 –218).However, in one smaller study there wasno correlation between cystatin C andSCr (219). In a recent study, urinary cys-tatin C but not plasma cystatin C wassuperior to conventional plasma markersin the early identification of AKI aftercardiac surgery (220). Whereas rapid au-tomated assays for cystatin C are cur-rently available, more information on theuse of cystatin C in the ICU setting and inspecific patient populations (e.g., post-cardiothoracic surgery, sepsis, andtrauma) is necessary before implementa-tion in clinical practice.

Several studies support neutrophil ge-latinase-associated lipocalin (221–227),kidney injury molecule-1 (228, 229), andIL-18 (222, 230, 231) as promising can-didate biomarkers for the early detectionof AKI. Point-of-care tests for urinaryIL-18 and neutrophil gelatinase-associ-ated lipocalin will likely be available forclinical use soon (213, 231–234). Urinaryexcretion of enzymes (alkaline phospha-tase, gamma glutamyl transaminase, N-acetyl-beta-d-glucosamide) (235), trans-porters (sodium-hydrogen exchangerisoform 3) (236), cytokines (IL-6, IL-8,and IL-18), and protein-like substances(fetuin A) (237) are presumably “shed”into the urine with AKI; therefore, theymay have a role in the early identificationof AKI (232, 233).

In addition to emerging biomarkers,promising real-time imaging for use inearly detection of AKI is on the horizon(238, 239). Ongoing discovery using uri-nary proteomic analyses or analysis ofgenetic polymorphisms may identify sus-ceptibility to AKI (240–244). Overall, bi-omarkers in AKI, although rapidly evolv-ing, are a field still in its relative infancy.Their role in the diagnosis and manage-ment of AKI in the ICU, although prom-ising, remains unproven. Furthermore,judging novel biomarkers against an im-perfect “gold-standard” biomarker (SCr)may have its limitations.

Erythropoietic Agents. The endothe-lium plays a central role in the initiationand maintenance phases of AKI. Animalmodels demonstrate a renal-protective ef-fect of erythropoietin on endotoxin-related kidney injury (245). Decreased se-verity of AKI is proposed to occur

through tubular regeneration from thedirect effects of erythropoietin on tubularepithelial cells (246). These findings sup-port the ongoing trials exploring the roleof erythropoietic agents in the preventionor early intervention for AKI using earlybiomarkers (personal communicationand clinicaltrials.gov NCT00476619).

Atrial Natriuretic Peptide. Recombi-nant human atrial natriuretic peptide de-creased the need for dialysis (21% vs.47%) and improved dialysis-free survivalat 21 days (57% vs. 28%) in a RCT of 61complicated post-cardiopulmonary by-pass patients without preexisting CKD(247). Previously, however, in two multi-centered, prospective, randomized trialsin patients with acute tubular necrosis(248) or late oliguric AKI (249), atrialnatriuretic peptide had no effect on needfor dialysis or overall mortality. Furthertrials are needed before the use of atrialnatriuretic peptide can be recommendedfor routine clinical use in cardiac surgerypatients.

Renal Tubule Assist Device. Resultsfrom a recent RCT of the renal tubuleassist device, in which the renal tubuleassist device added to conventional CRRTwas compared to CRRT alone, are prom-ising with respect to both safety and effi-cacy. There was a non-statistically signif-icant decrease in mortality at 28 days anda statistically significant difference at 180days (secondary outcome) (250).

Hemofiltration for Sepsis. Payen et al(251) recently published the findingsfrom the largest RCT of hemofiltrationfor severe sepsis and septic shock. At in-terim analysis, standard CVVH was foundto be deleterious, with increased organfailures in the CVVH group compared tostandard therapy. The study was stoppedat interim analysis and consequently en-rollment was insufficient to detect a dif-ference in mortality with sufficientpower. These findings contrast with thoseof Honore et al (252) in 2000, suggestinga beneficial role for hemofiltration in re-fractory septic shock. An important dif-ference between these two studies wasthe delivered dose. In the first study, thedose, on average, was approximately 2L/hr, whereas in the second study thedose was, on average, 8.7 L/hr for 4 hrs.

Stem Cells and the Kidney. Progenitorcell therapies represent an exciting futureopportunity for treatment of AKI in thecritically ill. Phase 1 trials of mesenchymalstem cells for treatment of patients at highrisk for cardiac surgery-associated AKI areunderway. A phase 2 RCT will be conducted


if safety is demonstrated in phase 1 (clini-caltrials.gov ID: NCT00733876).

CONCLUSIONS

Many unanswered questions remainwith respect to early identification, pre-vention, optimal timing, dose, and mo-dality of RRT for AKI in the ICU. Withrespect to AKI in the ICU, the fundamen-tal principal that guides all medical ther-apy—do no harm—is especially perti-nent. AKI in the ICU most commonlyresults from multiple insults. Therefore,appropriate and early identification of pa-tients at risk for AKI provides an oppor-tunity to prevent subsequent renal in-sults and ultimately impact overall ICUmorbidity and mortality. Strategies toprevent AKI in these patients are of piv-otal importance. Key components of op-timal prevention and management of thecritically ill patient with AKI includemaintenance of renal perfusion andavoidance of nephrotoxins. Whereasmanagement of AKI remains limited pri-marily to supportive care, there are manypotential therapies and interventions onthe horizon.

Although it is widely accepted thatearly intervention therapies have beenlimited by the lack of tools for early de-tection, there are several promising can-didate biomarkers in the pipeline. Fur-thermore, through the establishment ofAKIN, an international and interdiscipli-nary collaborative network with the over-arching objective to address AKI in theICU, there has been tremendous progressin establishing a uniform definition(AKIN criteria) that is valuable for classi-fication, clinical research study design,and prognosis.

A greater appreciation for the role ofAKI in the ICU as an active contributor tomorbidity and mortality is essential tofurthering our knowledge and under-standing of the influence of AKI in thecritically ill patient. Early detection willfacilitate early intervention. Early inter-vention designed to target the deleterioussystemic effects of AKI will likely improveoverall morbidity and mortality. For now,recognition of risk factors, excellent sup-portive care, and avoidance of clinicalconditions known to cause or worsen AKIremain the cornerstone of managementof AKI in the ICU.

REFERENCES

1. Bagshaw SM, George C, Bellomo R:Changes in the incidence and outcome for

early acute kidney injury in a cohort ofAustralian intensive care units. Crit Care2007; 11:R68

2. Collins AJ, Foley R, Herzog C, et al: Ex-cerpts from the United States Renal DataSystem 2007 annual data report. Am J Kid-ney Dis 2008; 51:S1–S320

3. Waikar SS, Wald R, Chertow GM, et al:Validity of International Classification ofDiseases, Ninth Revision, Clinical Modifica-tion Codes for Acute Renal Failure. J AmSoc Nephrol 2006; 17:1688–1694

4. Xue JL, Daniels F, Star RA, et al: Incidenceand mortality of acute renal failure in Medi-care beneficiaries, 1992 to 2001. J Am SocNephrol 2006; 17:1135–1142

5. Chertow GM, Soroko SH, Paganini EP, et al:Mortality after acute renal failure: modelsfor prognostic stratification and risk adjust-ment. Kidney international 2006; 70:1120–1126

6. Hoste EA, Clermont G, Kersten A, et al:RIFLE criteria for acute kidney injury areassociated with hospital mortality in criti-cally ill patients: a cohort analysis. Crit Care2006; 10:R73

7. Nash K, Hafeez A, Hou S: Hospital-acquiredrenal insufficiency. Am J Kidney Dis 2002;39:930–936

8. Mehta RL, Pascual MT, Soroko S, et al:Spectrum of acute renal failure in the in-tensive care unit: the PICARD experience.Kidney international 2004; 66:1613–1621

9. Uchino S, Kellum JA, Bellomo R, et al:Acute renal failure in critically ill patients: amultinational, multicenter study. Jama2005; 294:813–818

10. Uchino S, Bellomo R, Goldsmith D, et al: Anassessment of the RIFLE criteria for acuterenal failure in hospitalized patients. Criti-cal care medicine 2006; 34:1913–1917

11. Ostermann M, Chang RW: Acute kidney in-jury in the intensive care unit according toRIFLE. Critical care medicine 2007; 35:1837–1843; quiz 1852

12. Lin CY, Chen YC, Tsai FC, et al: RIFLEclassification is predictive of short-termprognosis in critically ill patients with acuterenal failure supported by extracorporealmembrane oxygenation. Nephrol DialTransplant 2006; 21:2867–2873

13. Lopes JA, Jorge S, Resina C, et al: Prognos-tic utility of RIFLE for acute renal failure inpatients with sepsis. Crit Care 2007; 11:408

14. Metnitz PG, Krenn CG, Steltzer H, et al:Effect of acute renal failure requiring renalreplacement therapy on outcome in criti-cally ill patients. Crit Care Med 2002; 30:2051–2058

15. Liano F, Felipe C, Tenorio MT, et al: Long-term outcome of acute tubular necrosis: acontribution to its natural history. Kidney-Int 2007; 71:679–686

16. Bagshaw SM, Laupland KB, Doig CJ, et al:Prognosis for long-term survival and renalrecovery in critically ill patients with severeacute renal failure: a population-basedstudy. Crit Care 2005; 9:R700–R709

17. Chertow GM, Levy EM, HammermeisterKE, et al: Independent association betweenacute renal failure and mortality followingcardiac surgery. Am J Med 1998; 104:343–348

18. Chertow GM, Burdick E, Honour M, et al:Acute kidney injury, mortality, length ofstay, and costs in hospitalized patients.J Am Soc Nephrol 2005; 16:3365–3370

19. Lassnigg A, Schmidlin D, Mouhieddine M,et al: Minimal changes of serum creatininepredict prognosis in patients after cardio-thoracic surgery: a prospective cohortstudy. J Am Soc Nephrol 2004; 15:1597–1605

20. Waikar SS, Liu KD, Chertow GM: The inci-dence and prognostic significance of acutekidney injury. Curr Opin Nephrol Hyper-tens 2007; 16:227–236

21. Coca SG, Peixoto AJ, Garg AX, et al: Theprognostic importance of a small acute dec-rement in kidney function in hospitalizedpatients: a systematic review and meta-analysis. Am J Kidney Dis 2007; 50:712–720

22. Yegenaga I, Hoste E, Van Biesen W, et al:Clinical characteristics of patients develop-ing ARF due to sepsis/systemic inflamma-tory response syndrome: results of a pro-spective study. Am J Kidney Dis 2004; 43:817–824

23. Bagshaw SM, Uchino S, Bellomo R, et al:Septic acute kidney injury in critically illpatients: clinical characteristics and out-comes. Clin J Am Soc Nephrol 2007;2:431–439

24. Bernieh B, Al Hakim M, Boobes Y, et al:Outcome and predictive factors of acute re-nal failure in the intensive care unit. Trans-plant Proc 2004; 36:1784–1787

25. Vincent JL, de Mendonca A, Cantraine F, etal: Use of the SOFA score to assess theincidence of organ dysfunction/failure in in-tensive care units: results of a multicenter,prospective study. Working group on “sep-sis-related problems” of the European Soci-ety of Intensive Care Medicine. Crit CareMed 1998; 26:1793–1800

26. Harbrecht BG, Rosengart MR, Zenati MS, etal: Defining the contribution of renal dys-function to outcome after traumatic injury.Am Surg 2007; 73:836–840

27. Radovic M, Ostric V, Djukanovic L: Validityof prediction scores in acute renal failuredue to polytrauma. Ren Fail 1996; 18:615–620

28. Kuitunen A, Vento A, Suojaranta-Ylinen R,et al: Acute renal failure after cardiac sur-gery: evaluation of the RIFLE classification.Ann Thorac Surg 2006; 81:542–546

29. Thakar CV, Worley S, Arrigain S, et al: In-fluence of renal dysfunction on mortalityafter cardiac surgery: modifying effect ofpreoperative renal function. Kidney inter-national 2005; 67:1112–1119

30. Bove T, Calabro MG, Landoni G, et al: Theincidence and risk of acute renal failureafter cardiac surgery. J Cardiothorac VascAnesth 2004; 18:442–445


31. Holm C, Horbrand F, von DonnersmarckGH, et al: Acute renal failure in severelyburned patients. Burns 1999; 25:171–178

32. Morgera S, Kraft AK, Siebert G, et al: Long-term outcomes in acute renal failure pa-tients treated with continuous renal re-placement therapies. Am J Kidney Dis 2002;40:275–279

33. Prescott GJ, Metcalfe W, Baharani J, et al: Aprospective national study of acute renalfailure treated with RRT: incidence, aetiol-ogy and outcomes. Nephrol Dial Transplant2007; 22:2513–2519

34. Leacche M, Rawn JD, Mihaljevic T, et al:Outcomes in patients with normal serumcreatinine and with artificial renal supportfor acute renal failure developing after cor-onary artery bypass grafting. Am J Cardiol2004; 93:353–356

35. Korkeila M, Ruokonen E, Takala J: Costs ofcare, long-term prognosis and quality of lifein patients requiring renal replacementtherapy during intensive care. IntensiveCare Med 2000; 26:1824–1831

36. Basile DP: Novel approaches in the investi-gation of acute kidney injury. J Am SocNephrol 2007; 18:7–9

37. Schiffl H, Lang SM, Fischer R: Daily hemo-dialysis and the outcome of acute renal fail-ure. N Engl J Med 2002; 346:305–310

38. Macedo E, Bouchard J, Mehta RL: Renalrecovery following acute kidney injury.Curr Opin Crit Care 2008; 14:660–665

39. Mehta RL, Chertow GM: Acute renal failuredefinitions and classification: time forchange? J Am Soc Nephrol 2003; 14:2178–2187

40. Bellomo R, Ronco C, Kellum JA, et al: Acuterenal failure— definition, outcome mea-sures, animal models, fluid therapy and in-formation technology needs: the Second In-ternational Consensus Conference of theAcute Dialysis Quality Initiative (ADQI)Group. Crit Care 2004; 8:R204–R212

41. Dasta JF, Kane-Gill SL, Durtschi AJ, et al:Costs and outcomes of acute kidney injury(AKI) following cardiac surgery. NephrolDial Transplant 2008

42. Mehta RL, Kellum JA, Shah SV, et al: AcuteKidney Injury Network: report of an initia-tive to improve outcomes in acute kidneyinjury. Crit Care 2007; 11:R31

43. Bagshaw SM, George C, Bellomo R: A com-parison of the RIFLE and AKIN criteria foracute kidney injury in critically ill patients.Nephrol Dial Transplant 2008; 23:1569–1574

44. Lopes JA, Fernandes P, Jorge S, et al: Acutekidney injury in intensive care unit pa-tients: a comparison between the RIFLEand the Acute Kidney Injury Network clas-sifications. Crit Care 2008; 12:R110

45. Guerin C, Girard R, Selli JM, et al: Initialversus delayed acute renal failure in theintensive care unit. A multicenter prospec-tive epidemiological study. Rhone-AlpesArea Study Group on Acute Renal Failure.

Am J Respir Crit Care Med 2000; 161:872–879

46. Moran SM, Myers BD: Course of acute renalfailure studied by a model of creatinine ki-netics. Kidney international 1985; 27:928–937

47. Tsai JJ, Yeun JY, Kumar VA, et al: Compar-ison and interpretation of urinalysis per-formed by a nephrologist versus a hospital-based clinical laboratory. Am J Kidney Dis2005; 46:820–829

48. Perazella MA, Coca SG, Kanbay M, et al:Diagnostic value of urine microscopy fordifferential diagnosis of acute kidney injuryin hospitalized patients. Clin J Am SocNephrol 2008; 3:1615–1619

49. da Silva Magro MC, de Fatima FernandesVattimo M: Does urinalysis predict acuterenal failure after heart surgery? Ren Fail2004; 26:385–392

50. Lee VWS, Harris DCH, Anderson RJ, et al:Acute Renal Failure. In: Diseases of the Kid-ney & Urinary Tract. 8th ed. Schrier RW(Ed). Philadelphia, Wolters Kluwer Health/Lippincott Williams & Wilkins, 2007

51. Miller TR, Anderson RJ, Linas SL, et al:Urinary diagnostic indices in acute renalfailure: a prospective study. Ann Intern Med1978; 89:47–50

52. Payen D, de Pont AC, Sakr Y, et al: A posi-tive fluid balance is associated with a worseoutcome in patients with acute renal fail-ure. Crit Care 2008; 12:R74

53. Wiedemann HP, Wheeler AP, Bernard GR,et al: Comparison of two fluid-managementstrategies in acute lung injury. N EnglJ Med 2006; 354:2564–2575

54. Ali T, Khan I, Simpson W, et al: Incidenceand outcomes in acute kidney injury: acomprehensive population-based study.J Am Soc Nephrol 2007; 18:1292–1298

55. Vivino G, Antonelli M, Moro ML, et al: Riskfactors for acute renal failure in traumapatients. Intensive Care Med 1998; 24:808–814

56. De laet I, Malbrain ML, Jadoul JL, et al:Renal implications of increased intra-abdominal pressure: are the kidneys the ca-nary for abdominal hypertension? Acta ClinBelg Suppl 2007:119–130

57. Sugrue M, Jones F, Deane SA, et al: Intra-abdominal hypertension is an independentcause of postoperative renal impairment.Arch Surg 1999; 134:1082–1085

58. Cheatham ML, Malbrain ML, Kirkpatrick A,et al: Results from the International Con-ference of Experts on Intra-abdominal Hy-pertension and Abdominal CompartmentSyndrome. II. Recommendations. IntensiveCare Med 2007; 33:951–962

59. Sharp LS, Rozycki GS, Feliciano DV: Rhab-domyolysis and secondary renal failure incritically ill surgical patients. Am J Surg2004; 188:801–806

60. Obialo CI, Okonofua EC, Tayade AS, et al:Epidemiology of de novo acute renal failurein hospitalized African Americans: compar-ing community-acquired vs hospital-ac-

quired disease. Arch Intern Med 2000; 160:1309–1313

61. Wang Y, Cui Z, Fan M: Hospital-acquiredand community-acquired acute renal fail-ure in hospitalized Chinese: a ten-year re-view. Ren Fail 2007; 29:163–168

62. Kelleher SP, Robinette JB, Conger JD: Sym-pathetic nervous system in the loss of au-toregulation in acute renal failure. Am JPhysiol 1984; 246:F379–F386

63. Schlichtig R, Kramer DJ, Boston JR, et al:Renal O2 consumption during progressivehemorrhage. J Appl Physiol 1991; 70:1957–1962

64. Bourgoin A, Leone M, Delmas A, et al: In-creasing mean arterial pressure in patientswith septic shock: effects on oxygen vari-ables and renal function. Crit Care Med2005; 33:780–786

65. LeDoux D, Astiz ME, Carpati CM, et al:Effects of perfusion pressure on tissue per-fusion in septic shock. Crit Care Med 2000;28:2729–2732

66. Friedrich JO, Adhikari N, Herridge MS, etal: Meta-analysis: low-dose dopamine in-creases urine output but does not preventrenal dysfunction or death. Ann Intern Med2005; 142:510–524

67. Kellum JA, J MD: Use of dopamine in acuterenal failure: a meta-analysis. Crit Care Med2001; 29:1526–1531

68. Marik PE: Low-dose dopamine: a systematicreview. Intensive Care Med 2002; 28:877–883

69. Holmes CL, Walley KR: Bad medicine: low-dose dopamine in the ICU. Chest 2003; 123:1266–1275

70. Dunning J, Khasati N, Barnard J: Low dose(renal dose) dopamine in the critically illpatient. Interact Cardiovasc Thorac Surg2004; 3:114–117

71. Bellomo R, Chapman M, Finfer S, et al:Low-dose dopamine in patients with earlyrenal dysfunction: a placebo-controlled ran-domised trial. Australian and New ZealandIntensive Care Society (ANZICS) ClinicalTrials Group. Lancet 2000; 356:2139–2143

72. Lauschke A, Teichgraber UK, Frei U, et al:‘Low-dose’ dopamine worsens renal perfu-sion in patients with acute renal failure.Kidney Int 2006; 69:1669–1674

73. Argalious M, Motta P, Khandwala F, et al:“Renal dose” dopamine is associated withthe risk of new-onset atrial fibrillation aftercardiac surgery. Crit Care Med 2005; 33:1327–1332

74. Devins SS, Miller A, Herndon BL, et al:Effects of dopamine on T-lymphocyte pro-liferative responses and serum prolactinconcentrations in critically ill patients. CritCare Med 1992; 20:1644–1649

75. Murphy MB, Murray C, Shorten GD:Fenoldopam: a selective peripheral dopam-ine-receptor agonist for the treatment ofsevere hypertension. N Engl J Med 2001;345:1548–1557

76. Samuels J, Finkel K, Gubert M, et al: Effectof fenoldopam mesylate in critically ill pa-


tients at risk for acute renal failure is dosedependent. Ren Fail 2005; 27:101–105

77. Brienza N, Malcangi V, Dalfino L, et al: Acomparison between fenoldopam and low-dose dopamine in early renal dysfunction ofcritically ill patients. Crit Care Med 2006;34:707–714

78. Tumlin JA, Finkel KW, Murray PT, et al:Fenoldopam mesylate in early acute tubularnecrosis: a randomized, double-blind, pla-cebo-controlled clinical trial. Am J KidneyDis 2005; 46:26–34

79. Morelli A, Ricci Z, Bellomo R, et al: Prophy-lactic fenoldopam for renal protection insepsis: a randomized, double-blind, place-bo-controlled pilot trial. Crit Care Med2005; 33:2451–2456

80. Landoni G, Biondi-Zoccai GG, Tumlin JA, etal: Beneficial impact of fenoldopam in criti-cally ill patients with or at risk for acute renalfailure: a meta-analysis of randomized clinicaltrials. Am J Kidney Dis 2007; 49:56–68

81. Schierhout G, Roberts I: Fluid resuscitationwith colloid or crystalloid solutions in crit-ically ill patients: a systematic review ofrandomised trials. BMJ 1998; 316:961–964

82. Alderson P, Schierhout G, Roberts I, et al:Colloids versus crystalloids for fluid resus-citation in critically ill patients. CochraneDatabase Syst Rev 2000:CD000567

83. Alderson P, Bunn F, Lefebvre C, et al: Hu-man albumin solution for resuscitation andvolume expansion in critically ill patients.Cochrane Database Syst Rev 2002:CD001208

84. Alderson P, Bunn F, Lefebvre C, et al: Hu-man albumin solution for resuscitation andvolume expansion in critically ill patients.Cochrane Database Syst Rev 2004:CD001208

85. Choi PT, Yip G, Quinonez LG, et al: Crys-talloids vs. colloids in fluid resuscitation: asystematic review. Critical care medicine1999; 27:200–210

86. Finfer S, Bellomo R, Boyce N, et al: A com-parison of albumin and saline for fluid re-suscitation in the intensive care unit.N Engl J Med 2004; 350:2247–2256

87. Myburgh J, Cooper DJ, Finfer S, et al: Salineor albumin for fluid resuscitation in pa-tients with traumatic brain injury. N EnglJ Med 2007; 357:874–884

88. Schortgen F, Lacherade JC, Bruneel F, et al.Effects of hydroxyethylstarch and gelatin onrenal function in severe sepsis: a multicen-tre randomised study. Lancet 2001; 357:911–916

89. Brunkhorst FM, Engel C, Bloos F, et al:Intensive insulin therapy and pentastarchresuscitation in severe sepsis. N Engl J Med2008; 358:125–139

90. Cittanova ML, Leblanc I, Legendre C, et al:Effect of hydroxyethylstarch in brain-deadkidney donors on renal function in kidney-transplant recipients. Lancet 1996; 348:1620–1622

91. Wiedermann CJ: Systematic review of ran-domized clinical trials on the use of hy-

droxyethyl starch for fluid management insepsis. BMC Emerg Med 2008; 8:1

92. Vincent JL, Sakr Y, Sprung CL, et al: Sepsisin European intensive care units: results ofthe SOAP study. Crit Care Med 2006; 34:344–353

93. Sakr Y, Payen D, Reinhart K, et al: Effects ofhydroxyethyl starch administration on re-nal function in critically ill patients. Br JAnaesth 2007; 98:216–224

94. Finfer S, Chittock DR, Su SY, et al: Inten-sive versus conventional glucose control incritically ill patients. N Engl J Med 2009;360:1283–1297

95. Van den Berghe G, Wilmer A, Hermans G, etal: Intensive insulin therapy in the medicalICU. N Engl J Med 2006; 354:449–461

96. van den Berghe G, Wouters P, Weekers F, etal: Intensive insulin therapy in the criticallyill patients. N Engl J Med 2001; 345:1359–1367

97. Schetz M, Vanhorebeek I, Wouters PJ, et al:Tight blood glucose control is renoprotec-tive in critically ill patients. J Am Soc Neph-rol 2008; 19:571–578

98. Lecomte P, Van Vlem B, Coddens J, et al:Tight perioperative glucose control is asso-ciated with a reduction in renal impairmentand renal failure in non-diabetic cardiacsurgical patients. Crit Care 2008; 12:R154

99. Cheung NW, Napier B, Zaccaria C, et al:Hyperglycemia is associated with adverseoutcomes in patients receiving total paren-teral nutrition. Diabetes Care 2005; 28:2367–2371

100. Rybak MJ, Abate BJ, Kang SL, et al: Pro-spective evaluation of the effect of an ami-noglycoside dosing regimen on rates of ob-served nephrotoxicity and ototoxicity.Antimicrob Agents Chemother 1999; 43:1549–1555

101. Prins JM, Buller HR, Kuijper EJ, et al: Onceversus thrice daily gentamicin in patientswith serious infections. Lancet 1993; 341:335–339

102. Hatala R, Dinh T, Cook DJ: Once-daily ami-noglycoside dosing in immunocompetentadults: a meta-analysis. Ann Intern Med1996; 124:717–725

103. Barza M, Ioannidis JP, Cappelleri JC, et al:Single or multiple daily doses of aminogly-cosides: a meta-analysis. BMJ 1996; 312:338–345

104. Munckhof WJ, Grayson ML, Turnidge JD: Ameta-analysis of studies on the safety andefficacy of aminoglycosides given eitheronce daily or as divided doses. J AntimicrobChemother 1996; 37:645–663

105. Galloe AM, Graudal N, Christensen HR, etal: Aminoglycosides: single or multiple dailydosing? A meta-analysis on efficacy andsafety. Eur J Clin Pharmacol 1995; 48:39–43

106. Ferriols-Lisart R, Alos-Alminana M: Effec-tiveness and safety of once-daily aminogly-cosides: a meta-analysis. Am J Health SystPharm 1996; 53:1141–1150

107. Harbarth S, Pestotnik SL, Lloyd JF, et al:

The epidemiology of nephrotoxicity associ-ated with conventional amphotericin Btherapy. Am J Med 2001; 111:528–534

108. Walsh TJ, Finberg RW, Arndt C, et al: Lipo-somal amphotericin B for empirical therapyin patients with persistent fever and neu-tropenia. National Institute of Allergy andInfectious Diseases Mycoses Study Group.N Engl J Med 1999; 340:764–771

109. Wingard JR, Wood CA, Sullivan E, et al:Caspofungin versus amphotericin B for can-didemia: a pharmacoeconomic analysis.Clin Ther 2005; 27:960–969

110. Mangano DT, Tudor IC, Dietzel C: The riskassociated with aprotinin in cardiac sur-gery. N Engl J Med 2006; 354:353–365

111. Huber W, Eckel F, Hennig M, et al: Prophy-laxis of contrast material-induced nephrop-athy in patients in intensive care: acetylcys-teine, theophylline, or both? A randomizedstudy. Radiology 2006; 239:793–804

112. Joannidis M, Schmid M, Wiedermann CJ:Prevention of contrast media-induced ne-phropathy by isotonic sodium bicarbonate:a meta-analysis. Wien Klin Wochenschr2008; 120:742–748

113. Ho KM, Morgan DJ: Use of isotonic sodiumbicarbonate to prevent radiocontrast ne-phropathy in patients with mild pre-existing renal impairment: a meta-analysis.Anaesth Intensive Care 2008; 36:646–653

114. Hogan SE, L’Allier P, Chetcuti S, et al:Current role of sodium bicarbonate-basedpreprocedural hydration for the preventionof contrast-induced acute kidney injury: ameta-analysis. Am Heart J 2008; 156:414–421

115. Meier P, Ko DT, Tamura A, et al: Sodiumbicarbonate-based hydration prevents con-trast-induced nephropathy: a meta-analysis.BMC Med 2009; 7:23

116. Navaneethan SD, Singh S, Appasamy S, etal: Sodium bicarbonate therapy for preven-tion of contrast-induced nephropathy: asystematic review and meta-analysis. Am JKidney Dis 2009; 53:617–627

117. Brar SS, Shen AY, Jorgensen MB, et al:Sodium bicarbonate vs sodium chloride forthe prevention of contrast medium-inducednephropathy in patients undergoing coro-nary angiography: a randomized trial. JAMA2008; 300:1038–1046

118. Tepel M, van der Giet M, Schwarzfeld C, etal: Prevention of radiographic-contrast-agent-induced reductions in renal functionby acetylcysteine. N Engl J Med 2000; 343:180–184

119. Kelly AM, Dwamena B, Cronin P, et al:Meta-analysis: effectiveness of drugs forpreventing contrast-induced nephropathy.Ann Intern Med 2008; 148:284–294

120. Gonzales DA, Norsworthy KJ, Kern SJ, et al:A meta-analysis of N-acetylcysteine in con-trast-induced nephrotoxicity: unsupervisedclustering to resolve heterogeneity. BMCMed 2007; 5:32

121. Hoffmann U, Fischereder M, Kruger B, et al:The value of N-acetylcysteine in the preven-


tion of radiocontrast agent-induced ne-phropathy seems questionable. J Am SocNephrol 2004; 15:407–410

122. Haase M, Haase-Fielitz A, Ratnaike S, et al:N-Acetylcysteine does not artifactuallylower plasma creatinine concentration.Nephrol Dial Transplant 2008

123. McCullough PA, Bertrand ME, Brinker JA,et al: A meta-analysis of the renal safety ofisosmolar iodixanol compared with low-osmolar contrast media. J Am Coll Cardiol2006; 48:692–699

124. Heinrich MC, Haberle L, Muller V, et al:Nephrotoxicity of iso-osmolar iodixanol com-pared with nonionic low-osmolar contrastmedia: meta-analysis of randomized con-trolled trials. Radiology 2009; 250:68–86

125. Holscher B, Heitmeyer C, Fobker M, et al:Predictors for contrast media-induced ne-phropathy and long-term survival: prospec-tively assessed data from the randomizedcontrolled Dialysis-Versus-Diuresis (DVD)trial. Can J Cardiol 2008; 24:845–850

126. Cirit M, Toprak O, Yesil M, et al: Angioten-sin-converting enzyme inhibitors as a riskfactor for contrast-induced nephropathy.Nephron Clin Pract 2006; 104:c20–c27

127. Onuigbo MA, Onuigbo NT: Does renin-angiotensin aldosterone system blockadeexacerbate contrast-induced nephropathyin patients with chronic kidney disease? Aprospective 50-month Mayo Clinic study.Ren Fail 2008; 30:67–72

128. Rosenstock JL, Bruno R, Kim JK, et al: Theeffect of withdrawal of ACE inhibitors orangiotensin receptor blockers prior to cor-onary angiography on the incidence of con-trast-induced nephropathy. Int Urol Neph-rol 2008; 40:749–755

129. Marenzi G, Marana I, Lauri G, et al: Theprevention of radiocontrast-agent-inducednephropathy by hemofiltration. N EnglJ Med 2003; 349:1333–1340

130. Marenzi G, Lauri G, Campodonico J, et al:Comparison of two hemofiltration protocolsfor prevention of contrast-induced ne-phropathy in high-risk patients. Am J Med2006; 119:155–162

131. Cruz DN, Perazella MA, Bellomo R, et al:Extracorporeal blood purification therapiesfor prevention of radiocontrast-induced ne-phropathy: a systematic review. Am J Kid-ney Dis 2006; 48:361–371

132. Lee PT, Chou KJ, Liu CP, et al: Renal pro-tection for coronary angiography in ad-vanced renal failure patients by prophylac-tic hemodialysis. A randomized controlledtrial. J Am Coll Cardiol 2007; 50:1015–1020

133. Tumlin JA, Wang A, Murray PT, et al:Fenoldopam mesylate blocks reductions inrenal plasma flow after radiocontrast dyeinfusion: a pilot trial in the prevention ofcontrast nephropathy. Am Heart J 2002;143:894–903

134. Stone GW, McCullough PA, Tumlin JA, etal: Fenoldopam mesylate for the preventionof contrast-induced nephropathy: a ran-

domized controlled trial. JAMA 2003; 290:2284–2291

135. Allaqaband S, Tumuluri R, Malik AM, et al:Prospective randomized study of N-acetyl-cysteine, fenoldopam, and saline for preven-tion of radiocontrast-induced nephropathy.Catheter Cardiovasc Interv 2002; 57:279–283

136. Briguori C, Colombo A, Airoldi F, et al:N-Acetylcysteine versus fenoldopam mesy-late to prevent contrast agent-associatednephrotoxicity. J Am Coll Cardiol 2004; 44:762–765

137. Bagshaw SM, Ghali WA: Theophylline forprevention of contrast-induced nephropa-thy: a systematic review and meta-analysis.Arch Intern Med 2005; 165:1087–1093

138. Ix JH, McCulloch CE, Chertow GM: The-ophylline for the prevention of radiocon-trast nephropathy: a meta-analysis. NephrolDial Transplant 2004; 19:2747–2753

139. Akriviadis E, Botla R, Briggs W, et al: Pentoxi-fylline improves short-term survival in severeacute alcoholic hepatitis: a double-blind, pla-cebo-controlled trial. Gastroenterology 2000;119:1637–1648

140. Sort P, Navasa M, Arroyo V, et al: Effect ofintravenous albumin on renal impairmentand mortality in patients with cirrhosis andspontaneous bacterial peritonitis. N EnglJ Med 1999; 341:403–409

141. Gines P, Tito L, Arroyo V, et al: Randomizedcomparative study of therapeutic paracen-tesis with and without intravenous albuminin cirrhosis. Gastroenterology 1988; 94:1493–1502

142. Gluud LL, Kjaer MS, Christensen E: Terlip-ressin for hepatorenal syndrome. CochraneDatabase Syst Rev 2006:CD005162

143. Ortega R, Gines P, Uriz J, et al: Terlipressintherapy with and without albumin for pa-tients with hepatorenal syndrome: resultsof a prospective, nonrandomized study.Hepatology 2002; 36:941–948

144. Solanki P, Chawla A, Garg R, et al: Benefi-cial effects of terlipressin in hepatorenalsyndrome: a prospective, randomized place-bo-controlled clinical trial. J GastroenterolHepatol 2003; 18:152–156

145. Sanyal AJ, Boyer T, Garcia-Tsao G, et al: Arandomized, prospective, double-blind, pla-cebo-controlled trial of terlipressin for type1 hepatorenal syndrome. Gastroenterology2008; 134:1360–1368

146. Martin-Llahi M, Pepin MN, Guevara M, et al:Terlipressin and albumin vs albumin in patientswith cirrhosis and hepatorenal syndrome: a ran-domized study. Gastroenterology 2008; 134:1352–1359

147. Alessandria C, Ottobrelli A, Debernardi-Venon W, et al: Noradrenalin vs terlipressinin patients with hepatorenal syndrome: aprospective, randomized, unblinded, pilotstudy. J Hepatol 2007; 47:499–505

148. Sharma P, Kumar A, Shrama BC, et al. Anopen label, pilot, randomized controlledtrial of noradrenaline versus terlipressin inthe treatment of type 1 hepatorenal syn-

drome and predictors of response. Am JGastroenterol 2008; 103:1689–1697

149. Kiser TH, Fish DN, Obritsch MD, et al:Vasopressin, not octreotide, may be benefi-cial in the treatment of hepatorenal syn-drome: a retrospective study. Nephrol DialTransplant 2005; 20:1813–1820

150. Angeli P, Volpin R, Gerunda G, et al: Rever-sal of type 1 hepatorenal syndrome with theadministration of midodrine and octreotide.Hepatology 1999; 29:1690–1697

151. Esrailian E, Pantangco ER, Kyulo NL, et al:Octreotide/Midodrine therapy significantlyimproves renal function and 30-day survivalin patients with type 1 hepatorenal syn-drome. Dig Dis Sci 2007; 52:742–748

152. Skagen C, Einstein M, Lucey MR, et al:Combination treatment with octreotide,midodrine, and albumin improves survivalin patients with type 1 and type 2 hepato-renal syndrome. J Clin Gastroenterol 2009

153. Cantarovich F, Rangoonwala B, Lorenz H,et al: High-dose furosemide for establishedARF: a prospective, randomized, double-blind, placebo-controlled, multicenter trial.Am J Kidney Dis 2004; 44:402–409

154. Sampath S, Moran JL, Graham PL, et al:The efficacy of loop diuretics in acute renalfailure: assessment using Bayesian evidencesynthesis techniques. Crit Care Med 2007;35:2516–2524

155. Ho KM, Sheridan DJ: Meta-analysis offrusemide to prevent or treat acute renalfailure. BMJ 2006; 333:420

156. Bagshaw SM, Delaney A, Haase M, et al:Loop diuretics in the management of acuterenal failure: a systematic review and meta-analysis. Crit Care Resusc 2007; 9:60–68

157. Mehta RL, Pascual MT, Soroko S, et al:Diuretics, mortality, and nonrecovery of re-nal function in acute renal failure. JAMA2002; 288:2547–2553

158. Uchino S, Doig GS, Bellomo R, et al: Di-uretics and mortality in acute renal failure.Crit Care Med 2004; 32:1669–1677

159. Chertow GM, Lazarus JM, Paganini EP, etal: Predictors of mortality and the provisionof dialysis in patients with acute tubularnecrosis. The Auriculin Anaritide Acute Re-nal Failure Study Group. J Am Soc Nephrol1998; 9:692–698

160. Venkataram R, Kellum JA: The role of di-uretic agents in the management of acuterenal failure. Contrib Nephrol 2001:158–170

161. van der Voort PH, Boerma EC, KoopmansM, et al: Furosemide does not improve renalrecovery after hemofiltration for acute renalfailure in critically ill patients: a doubleblind randomized controlled trial. Criticalcare medicine 2009; 37:533–538

162. Bagshaw SM, Delaney A, Jones D, et al:Diuretics in the management of acute kid-ney injury: a multinational survey. ContribNephrol 2007; 156:236–249

163. McWhirter JP, Pennington CR: Incidenceand recognition of malnutrition in hospital.BMJ 1994; 308:945–948


164. Goldstein-Fuchs D: Assessment of nutri-tional status in renal diseases. LippincottWilliams & Wilkins, 2002

165. Druml W: Nutritional management of acuterenal failure. J Ren Nutr 2005; 15:63–70

166. Frankenfield DC, Reynolds HN: Nutri-tional effect of continuous hemodiafiltra-tion. Nutrition 1995; 11:388 –393

167. Wooley JA, Btaiche IF, Good KL: Metabolicand nutritional aspects of acute renal fail-ure in critically ill patients requiring con-tinuous renal replacement therapy. NutrClin Pract 2005; 20:176–191

168. Cano N, Fiaccadori E, Tesinsky P, et al:ESPEN Guidelines on Enteral Nutrition:Adult renal failure. Clin Nutr 2006; 25:295–310

169. Macias WL, Alaka KJ, Murphy MH, et al:Impact of the nutritional regimen on pro-tein catabolism and nitrogen balance in pa-tients with acute renal failure. JPEN J Par-enter Enteral Nutr 1996; 20:56–62

170. Bellomo R, Tan HK, Bhonagiri S, et al: Highprotein intake during continuous hemodi-afiltration: impact on amino acids and ni-trogen balance. Int J Artif Organs 2002;25:261–268

171. Scheinkestel CD, Kar L, Marshall K, et al:Prospective randomized trial to assess ca-loric and protein needs of critically Ill, an-uric, ventilated patients requiring continu-ous renal replacement therapy. Nutrition2003; 19:909–916

172. Fiaccadori E, Maggiore U, Rotelli C, et al:Effects of different energy intakes on nitro-gen balance in patients with acute renalfailure: a pilot study. Nephrol Dial Trans-plant 2005; 20:1976–1980

173. Ricci Z, Picardo S, Ronco C: Results frominternational questionnaires. ContribNephrol 2007; 156:297–303

174. Ricci Z, Ronco C, D’Amico G, et al: Practicepatterns in the management of acute renalfailure in the critically ill patient: an inter-national survey. Nephrol Dial Transplant2006; 21:690–696

175. Ronco C, Ricci Z, Bellomo R: Currentworldwide practice of dialysis dose prescrip-tion in acute renal failure. Current opinionin critical care 2006; 12:551–556

176. Renal replacement therapy for acute kidneyinjury in Australian and New Zealand inten-sive care units: a practice survey. Crit CareResusc 2008; 10:225–230

177. Hoke TS, Douglas IS, Klein CL, et al: Acuterenal failure after bilateral nephrectomy isassociated with cytokine-mediated pulmonaryinjury. J Am Soc Nephrol 2007; 18:155–164

178. Mehta RL: Indications for dialysis in theICU: renal replacement vs. renal support.Blood Purif 2001; 19:227–232

179. Teschan PE, Baxter CR, O’Brien TF, et al:Prophylactic hemodialysis in the treatmentof acute renal failure. Ann Intern Med 1960;53:992–1016

180. Conger JD: A controlled evaluation of pro-phylactic dialysis in post-traumatic acuterenal failure. J Trauma 1975; 15:1056–1063

181. Gillum DM, Dixon BS, Yanover MJ, et al:The role of intensive dialysis in acute renalfailure. Clin Nephrol 1986; 25:249–255

182. Gettings LG, Reynolds HN, Scalea T: Out-come in post-traumatic acute renal failurewhen continuous renal replacement ther-apy is applied early vs. late. Intensive CareMed 1999; 25:805–813

183. Liu KD, Himmelfarb J, Paganini E, et al:Timing of initiation of dialysis in criticallyill patients with acute kidney injury. ClinJ Am Soc Nephrol 2006; 1:915–919

184. Bouman CS, Oudemans-Van Straaten HM,Tijssen JG, et al: Effects of early high-volume continuous venovenous hemofiltra-tion on survival and recovery of renal func-tion in intensive care patients with acuterenal failure: a prospective, randomizedtrial. Crit Care Med 2002; 30:2205–2211

185. Bagshaw SM, Uchino S, Bellomo R, et al:Timing of renal replacement therapy andclinical outcomes in critically ill patientswith severe acute kidney injury. J Crit Care2009; 24:129–140

186. Bouman CS, Oudemans-van Straaten HM:Timing of renal replacement therapy incritically ill patients with acute kidney in-jury. Curr Opin Crit Care 2007; 13:656–661

187. Ronco C, Bellomo R, Homel P, et al: Effectsof different doses in continuous veno-venous haemofiltration on outcomes ofacute renal failure: a prospective random-ised trial. Lancet 2000; 356:26–30

188. Saudan P, Niederberger M, De Seigneux S,et al: Adding a dialysis dose to continuoushemofiltration increases survival in patientswith acute renal failure. Kidney Int 2006;70:1312–1317

189. Tolwani AJ, Campbell RC, Stofan BS, et al:Standard versus high-dose CVVHDF forICU-related acute renal failure. J Am SocNephrol 2008; 19:1233–1238

190. Palevsky PM, Zhang JH, O’Connor TZ, et al:Intensity of renal support in critically illpatients with acute kidney injury. N EnglJ Med 2008; 359:7–20

191. Tonelli M, Manns B, Feller-Kopman D:Acute renal failure in the intensive careunit: a systematic review of the impact ofdialytic modality on mortality and renal re-covery. Am J Kidney Dis 2002; 40:875–885

192. Uehlinger DE, Jakob SM, Ferrari P, et al:Comparison of continuous and intermittentrenal replacement therapy for acute renalfailure. Nephrol Dial Transplant 2005; 20:1630–1637

193. Vinsonneau C, Camus C, Combes A, et al:Continuous venovenous haemodiafiltrationversus intermittent haemodialysis for acuterenal failure in patients with multiple-organdysfunction syndrome: a multicentre ran-domised trial. Lancet 2006; 368:379–385

194. Bagshaw SM, Bellomo R: Fluid resuscita-tion and the septic kidney. Curr Opin CritCare 2006; 12:527–530

195. Augustine JJ, Sandy D, Seifert TH, et al: Arandomized controlled trial comparing in-termittent with continuous dialysis in pa-

tients with ARF. Am J Kidney Dis 2004;44:1000–1007

196. Rabindranath K, Adams J, Macleod AM, etal: Intermittent versus continuous renal re-placement therapy for acute renal failure inadults. Cochrane Database Syst Rev 2007:CD003773

197. John S, Griesbach D, Baumgartel M, et al:Effects of continuous haemofiltration vs in-termittent haemodialysis on systemic hae-modynamics and splanchnic regional perfu-sion in septic shock patients: a prospective,randomized clinical trial. Nephrol DialTransplant 2001; 16:320–327

198. Mehta RL, McDonald B, Gabbai FB, et al: Arandomized clinical trial of continuous ver-sus intermittent dialysis for acute renal fail-ure. Kidney Int 2001; 60:1154–1163

199. Gasparovic V, Filipovic-Grcic I, Merkler M, etal: Continuous renal replacement therapy(CRRT) or intermittent hemodialysis(IHD)–what is the procedure of choice in crit-ically ill patients? Ren Fail 2003; 25:855–862

200. Lins RL, Elseviers MM, Van der Niepen P, etal: Intermittent versus continuous renal re-placement therapy for acute kidney injurypatients admitted to the intensive care unit:results of a randomized clinical trial. Neph-rol Dial Transplant 2009; 24:512–518

201. Kellum JA, Angus DC, Johnson JP, et al:Continuous versus intermittent renal re-placement therapy: a meta-analysis. Inten-sive Care Med 2002; 28:29–37

202. Bagshaw SM, Berthiaume LR, Delaney A, etal: Continuous versus intermittent renal re-placement therapy for critically ill patientswith acute kidney injury: a meta-analysis.Critical care medicine 2008; 36:610–617

203. Ghahramani N, Shadrou S, Hollenbeak C: Asystematic review of continuous renal re-placement therapy and intermittent haemo-dialysis in management of patients withacute renal failure. Nephrology (Carlton)2008; 13:570–578

204. Pannu N, Klarenbach S, Wiebe N, et al:Renal replacement therapy in patients withacute renal failure: a systematic review.JAMA 2008; 299:793–805

205. Uchino S, Bellomo R, Kellum JA, et al: Patientand kidney survival by dialysis modality incritically ill patients with acute kidney injury.Int J Artif Organs 2007; 30:281–292

206. Bell M, Granath F, Schon S, et al: Continuousrenal replacement therapy is associated withless chronic renal failure than intermittenthaemodialysis after acute renal failure. Inten-sive Care Med 2007; 33:773–780

207. Kumar VA, Yeun JY, Depner TA, et al: Ex-tended daily dialysis vs. continuous hemo-dialysis for ICU patients with acute renalfailure: a two-year single center report. Int JArtif Organs 2004; 27:371–379

208. Kielstein JT, Kretschmer U, Ernst T, et al:Efficacy and cardiovascular tolerability ofextended dialysis in critically ill patients: arandomized controlled study. Am J KidneyDis 2004; 43:342–349

209. Monchi M, Berghmans D, Ledoux D, et al:


Citrate vs. heparin for anticoagulation incontinuous venovenous hemofiltration: aprospective randomized study. IntensiveCare Med 2004; 30:260–265

210. Kutsogiannis DJ, Gibney RT, Stollery D, etal: Regional citrate versus systemic heparinanticoagulation for continuous renal re-placement in critically ill patients. KidneyInt 2005; 67:2361–2367

211. Betjes MG, van Oosterom D, van Agteren M,et al: Regional citrate versus heparin anti-coagulation during venovenous hemofiltra-tion in patients at low risk for bleeding:similar hemofilter survival but significantlyless bleeding. J Nephrol 2007; 20:602–608

212. Oudemans-van Straaten HM, Bosman RJ,Koopmans M, et al: Citrate anticoagulationfor continuous venovenous hemofiltration.Crit Care Med 2009; 37:545–552

213. Dennen P, Parikh CR: Biomarkers of acutekidney injury: can we replace serum creat-inine? Clin Nephrol 2007; 68:269–278

214. Ahlstrom A, Tallgren M, Peltonen S, et al:Evolution and predictive power of serumcystatin C in acute renal failure. Clin Neph-rol 2004; 62:344–350

215. Herget-Rosenthal S, Marggraf G, Husing J,et al: Early detection of acute renal failureby serum cystatin C. Kidney international2004; 66:1115–1122

216. Le Bricon T, Leblanc I, Benlakehal M, et al:Evaluation of renal function in intensivecare: plasma cystatin C vs. creatinine andderived glomerular filtration rate estimates.Clin Chem Lab Med 2005; 43:953–957

217. Delanaye P, Lambermont B, Chapelle JP, et al:Plasmatic cystatin C for the estimation ofglomerular filtration rate in intensive careunits. Intensive Care Med 2004; 30:980–983

218. Villa P, Jimenez M, Soriano MC, et al: Se-rum cystatin C concentration as a marker ofacute renal dysfunction in critically ill pa-tients. Crit Care 2005; 9:R139–R143

219. Mazul-Sunko B, Zarkovic N, Vrkic N, et al:Proatrial natriuretic peptide (1–98), but notcystatin C, is predictive for occurrence ofacute renal insufficiency in critically ill septicpatients. Nephron Clin Pract 2004; 97:c103–c107

220. Koyner JL, Bennett MR, Worcester EM, etal: Urinary cystatin C as an early biomarkerof acute kidney injury following adult car-diothoracic surgery. Kidney Int 2008; 74:1059–1069

221. Mishra J, Dent C, Tarabishi R, et al: Neu-trophil gelatinase-associated lipocalin(NGAL) as a biomarker for acute renal in-jury after cardiac surgery. Lancet 2005;365:1231–1238

222. Parikh CR, Mishra J, Thiessen-Philbrook H,et al: Urinary IL-18 is an early predictivebiomarker of acute kidney injury after car-diac surgery. Kidney Int 2006; 70:199–203

223. Wagener G, Jan M, Kim M, et al: Associationbetween increases in urinary neutrophilgelatinase-associated lipocalin and acute

renal dysfunction after adult cardiac sur-gery. Anesthesiology 2006; 105:485– 491

224. Mishra J, Ma Q, Kelly C, et al: Kidney NGALis a novel early marker of acute injury fol-lowing transplantation. Pediatr Nephrol2006; 21:856–863

225. Parikh CR, Jani A, Mishra J, et al: UrineNGAL and IL-18 are predictive biomarkersfor delayed graft function following kidneytransplantation. Am J Transplant 2006;6:1639–1645

226. Zappitelli M, Washburn KK, Arikan AA, etal: Urine neutrophil gelatinase-associatedlipocalin is an early marker of acute kidneyinjury in critically ill children: a prospectivecohort study. Crit Care 2007; 11:R84

227. Nickolas TL, O’Rourke MJ, Yang J, et al:Sensitivity and specificity of a single emer-gency department measurement of urinaryneutrophil gelatinase-associated lipocalinfor diagnosing acute kidney injury. Ann In-tern Med 2008; 148:810–819

228. Han WK, Bailly V, Abichandani R, et al:Kidney Injury Molecule-1 (KIM-1): a novelbiomarker for human renal proximal tubuleinjury. Kidney Int 2002; 62:237–244

229. Han WK, Waikar SS, Johnson A, et al: Uri-nary biomarkers in the early diagnosis ofacute kidney injury. Kidney Int 2008; 73:863–869

230. Parikh CR, Jani A, Melnikov VY, et al: Uri-nary interleukin-18 is a marker of humanacute tubular necrosis. Am J Kidney Dis2004; 43:405–414

231. Parikh CR, Abraham E, Ancukiewicz M, etal: Urine IL-18 is an early diagnostic markerfor acute kidney injury and predicts mortal-ity in the intensive care unit. J Am SocNephrol 2005; 16:3046–3052

232. Trof RJ, Di Maggio F, Leemreis J, et al:Biomarkers of acute renal injury and renalfailure. Shock 2006; 26:245–253

233. Waikar SS, Bonventre JV: Biomarkers forthe diagnosis of acute kidney injury. CurrOpin Nephrol Hypertens 2007; 16:557–564

234. Bonventre JV: Diagnosis of acute kidneyinjury: from classic parameters to new bi-omarkers. Contrib Nephrol 2007; 156:213–219

235. Westhuyzen J, Endre ZH, Reece G, et al:Measurement of tubular enzymuria facili-tates early detection of acute renal impair-ment in the intensive care unit. NephrolDial Transplant 2003; 18:543–551

236. du Cheyron D, Daubin C, Poggioli J, et al:Urinary measurement of Na�/H� ex-changer isoform 3 (NHE3) protein as newmarker of tubule injury in critically ill pa-tients with ARF. Am J Kidney Dis 2003;42:497–506

237. Zhou H, Pisitkun T, Aponte A, et al: Exoso-mal Fetuin-A identified by proteomics: anovel urinary biomarker for detecting acutekidney injury. Kidney Int 2006; 70:1847–1857

238. Yu W, Sandoval RM, Molitoris BA: Rapid

determination of renal filtration functionusing an optical ratiometric imaging ap-proach. Am J Physiol Renal Physiol 2007;292:F1873–R1880

239. Rabito CA, Panico F, Rubin R, et al: Nonin-vasive, real-time monitoring of renal func-tion during critical care. J Am Soc Nephrol1994; 4:1421–1428

240. Treszl A, Kaposi A, Hajdu J, et al: The extentto which genotype information may add tothe prediction of disturbed perinatal adap-tation: none, minor, or major? Pediatr Res2007; 62:610–614

241. Haase-Fielitz A, Haase M, Bellomo R, et al:Genetic polymorphisms in sepsis- and cardio-pulmonary bypass-associated acute kidney in-jury. Contrib Nephrol 2007; 156:75–91

242. Jaber BL, Pereira BJ, Bonventre JV, et al:Polymorphism of host response genes: im-plications in the pathogenesis and treat-ment of acute renal failure. Kidney Int2005; 67:14–33

243. Liangos O, Balakrishnan VS, Pereira BJ, etal: Cytokine single nucleotide polymor-phism. Role in acute renal failure. ContribNephrol 2004; 144:63–75

244. Nguyen MT, Ross GF, Dent CL, et al: Earlyprediction of acute renal injury using urinaryproteomics. Am J Nephrol 2005; 25:318–326

245. Mitra A, Bansal S, Wang W, et al: Erythro-poietin ameliorates renal dysfunction dur-ing endotoxaemia. Nephrol Dial Transplant2007; 22:2349–2353

246. Sharples EJ, Yaqoob MM: Erythropoietin inexperimental acute renal failure. NephronExp Nephrol 2006; 104:e83–e88

247. Sward K, Valsson F, Odencrants P, et al:Recombinant human atrial natriuretic pep-tide in ischemic acute renal failure: a ran-domized placebo-controlled trial. Crit CareMed 2004; 32:1310–1315

248. Allgren RL, Marbury TC, Rahman SN, et al:Anaritide in acute tubular necrosis. Auricu-lin Anaritide Acute Renal Failure StudyGroup. N Engl J Med 1997; 336:828–834

249. Lewis J, Salem MM, Chertow GM, et al: Atrialnatriuretic factor in oliguric acute renal fail-ure. Anaritide Acute Renal Failure StudyGroup. Am J Kidney Dis 2000; 36:767–774

250. Tumlin J, Wali R, Williams W, et al: Efficacyand safety of renal tubule cell therapy foracute renal failure. J Am Soc Nephrol 2008;19:1034–1040

251. Payen D, Mateo J, Cavaillon JM, et al: Im-pact of continuous venovenous hemofiltra-tion on organ failure during the early phaseof severe sepsis: a randomized controlledtrial. Crit Care Med 2009; 37:803–810

252. Honore PM, Jamez J, Wauthier M, et al:Prospective evaluation of short-term, high-volume isovolemic hemofiltration on thehemodynamic course and outcome in pa-tients with intractable circulatory failureresulting from septic shock. Crit Care Med2000; 28:3581–3587


REVIEWARTICLES

Recent developments in the perioperative management of adultpatients with chronic kidney disease

R. G. Craig* and J. M. Hunter

Division of Clinical Science (Anaesthesia), University Clinical Department, University of Liverpool,

The Duncan Building, Daulby Street, Liverpool L69 3GA, UK

*Corresponding author. E-mail: [email protected]

The complications of chronic kidney disease (CKD) present the anaesthetist with a number of

clinical challenges related in part to altered drug handling and to difficulties with vascular access

and fluid balance. Safe anaesthetic management requires an understanding of CKD pathophysiology

to prevent aggravation of pre-existing disease. This review will consider some recent changes in

the management of adult patients with CKD as they affect the anaesthetist. It will consider medical

problems associated with CKD together with new developments in perioperative management.

Br J Anaesth 2008; 101: 296–310

Keywords: anaesthesia, general; blood, coagulation; cardiovascular system, effects;

kidney, failure

Chronic kidney disease (CKD) is defined as either a

glomerular filtration rate (GFR) of ,60 ml min21 1.73 m22

for 3 months or more, irrespective of cause, or kidney damage

leading to a decrease in GFR, present for 3 months or more.73

The damage may manifest as abnormalities in the composition

of blood or urine, on radiological imaging, or in histology. It

is classified into five stages depending on GFR (Table 1).63

Prevalence

The 2006 UK Renal Registry Report documented the UK

annual incidence of new patients accepted for renal replace-

ment therapy (RRT) as 108 per million population.3 The

prevalence of UK adult patients alive on RRT at the end of

2005 was 694 per million. From 2001 to 2005, there was a

7.3% rise in the number accepted for RRT, due to an

ageing population and an increase in type 2 diabetes melli-

tus.89 In the USA in 2005, 43.8% of all incident cases of

established renal failure were due to diabetes mellitus.137

The number of patients with Stages 1–4 CKD (Table 1) is

likely to exceed the number with established renal failure

by as much as 50 times.89 Most individuals with CKD do

not develop established renal failure, in part due to an

increased mortality secondary to cardiovascular disease, the

advanced age at onset of many renal diseases, and the slow

rate of decline of renal function, especially if treated.73

Assessment of renal function

Estimation of the GFR is used to assess, define, and classify

renal function in CKD (Table 1). A quantitative assessment of

GFR involves measurement of the plasma clearance of an

exogenous marker, such as inulin,124 I-iothalamate, or 48

Cr-EDTA. This is time-consuming and often impracticable in

the clinical setting. Creatinine clearance may be used clini-

cally to estimate GFR, but this usually requires a 24 h urine

collection, which is inconvenient for the patient and prone to

collection errors. A shorter time period for urine collection,

e.g. 2 h, may be used in catheterized patients.129 In general,

creatinine clearance does not improve the estimate of GFR

over that provided by creatinine-based prediction equations

(see below), but it may still be useful in individuals with

exceptional dietary intake (patients on creatinine sup-

plements) or low muscle mass (amputees).80 The serum crea-

tinine concentration is influenced by factors such as age, sex,

muscle mass, and diet; it should not be used alone to assess

kidney function. It is insensitive to mild–moderate decreases

in GFR, which may be reduced by as much as 50% with

serum creatinine still in the normal range.126 To overcome

these limitations, a number of creatinine-based prediction

equations have been developed, and provide a more accurate

assessment of renal function. The most widely used are the

four-variable Modification of Diet in Renal Disease (MDRD)

equation and the Cockcroft–Gault formula (Table 2). The

MDRD equation estimates the GFR with surface area adjust-

ment whereas the Cockcroft–Gault formula estimates the

unadjusted creatinine clearance. These formulae have been

validated in patients with CKD.53 Both formulae are impre-

cise at high values of GFR, and in patients with a grossly

abnormal muscle mass, patients with a very low BMI, preg-

nant patients, and where renal function is changing rapidly.

# The Board of Management and Trustees of the British Journal of Anaesthesia 2008. All rights reserved. For Permissions, please e-mail: [email protected]

British Journal of Anaesthesia 101 (3): 296–310 (2008)

doi:10.1093/bja/aen203 Advance Access publication July 10, 2008

GFR may also be estimated using cystatin C-based equations,

for example, the Filler equation (Table 2). Cystatin C is a

protein (cysteine protease inhibitor) produced at a constant

rate by all nucleated cells. It is freely filtered by the kidney

and catabolized in the proximal tubule. The serum cystatin C

concentration was thought to be independent of body compo-

sition, but it has now been shown to be affected by lean

mass.79 A recent study of renal transplant recipients compared

the GFR measured using radio-labelled diethylenetriamine

pentaacetic acid (99mTc-DTPA) to estimated GFR using

cystatin C-based and creatinine-based equations.143 The Filler

equation correctly classified the stage of CKD in 76% of

patients, compared with only 65% with the four-variable

MDRD equation and 69% with the Cockcroft–Gault formula.

Patients on dialysis may have a degree of residual renal

function. The presence of any residual renal function is

associated with a lower mortality risk, reduced intradialytic

weight gain, and improved solute clearance in haemodialy-

sis (HD) patients.127 Residual renal function, defined as a

24 h urine volume .100 ml, is associated with an adjusted

odds ratio for death of 0.35 (95% CI: 0.18–0.68).127

Aetiology

Table 3 indicates the primary cause of established renal

failure as reported in the 2006 UK Renal Registry Report.3

CKD as a risk factor for perioperativecomplications

CKD is a risk factor for serious postoperative complications,

such as acute renal failure and cardiovascular complications

which are associated with an increased morbidity and mor-

tality.58 59 70 98 102 In a systematic review of preoperative risk

factors for postoperative renal failure,102 preoperative renal

dysfunction was repeatedly found to be predictor of post-

operative renal failure. Howell and colleagues58 conducted a

case–control study of risk factors for cardiovascular death

after elective surgery under general anaesthesia. Patients

were diagnosed as having renal failure if their serum creati-

nine concentration on admission was .150 mmol litre21.

Conditional logistic regression was used to calculate

adjusted odds ratios. The adjusted odds ratio associated with

renal failure was 3.56 with a 95% confidence interval of

1.04–12.10. A similar study by the same authors examined

risk factors for cardiovascular death after urgent or emer-

gency surgery and found that an association between renal

impairment and cardiovascular mortality was evident on

univariate analysis, but not after adjustment for confounding

factors.59 This may have been due to a lack of statistical

power. Lee and colleagues70 conducted a prospective cohort

study to develop an index able to identify patients at higher

risk of cardiac complications after non-cardiac surgery. A

preoperative serum creatinine .176.8 mmol litre21 (2 mg

dl21) was found to be an independent predictor of cardiac

complications, and was associated with major cardiac

complications in 9% of cases. The Euroscore is a scoring

system for the prediction of early mortality in cardiac sur-

gical patients in Europe. A serum creatinine .200 mmol

litre21 is considered an objective risk factor, and contrib-

utes an added 2% to the predicted mortality.98

Pathophysiology

CKD is associated with pathophysiological changes in

many systems, which have implications for the safe

conduct of anaesthesia (Table 4).

Cardiovascular system

CKD is associated with an increased risk of cardiovascular

disease. Myocardial infarction, heart failure, and stroke are

Table 2 Equations to predict GFR using creatinine and cystatin C23 63 143

The four variable or abbreviated MDRD equation

GFR ðml min�1 1:73 m�2Þ ¼ 186� ½s-creatinineðmmol litre�1Þ=88:4��1:154 � age ðyrÞ�0:203 � 0:742 if female� 1:21 if African American

Cockcroft–Gault equation

Creatinine clearance ðml min�1Þ ¼ ð140� ageÞ � weight ðkgÞ=½72� creatinine ðmg dl�1Þ� � 0:85 if female

serum creatinine levels in mmol litre21 can be converted to mg dl21 by dividing by 88.4

Filler equation

Log ðGFR in ml min�11:73 m�2Þ ¼ 1:962þ ½1:123� logð1=cystatin C in mg litre�1Þ�

Table 1 Classification of CKD.63 Adapted from: Joint Specialty Committee

on Renal Medicine of the Royal College of Physicians and the Renal

Association, and the Royal College of General Practitioners.63 Copyright #

2006 Royal College of Physicians. Adapted with permission

Stage 1: Normal GFR; GFR �90 ml min21 1.73 m22 with other evidence of

chronic kidney damage*

Stage 2: Mild impairment; GFR 60–89 ml min21 1.73 m22 with other

evidence chronic kidney damage*

Stage 3: Moderate impairment; GFR 30–59 ml min21 1.73 m22

Stage 4: Severe impairment; GFR 15–29 ml min21 1.73 m22

Stage 5: Established renal failure: GFR ,15 ml min21 1.73 m22 or on

dialysis

*The ‘other evidence of chronic kidney damage’ may include: persistent

microalbuminuria; persistent proteinuria; persistent haematuria, after

exclusion of other causes, e.g. urological disease; structural abnormalities of

the kidneys demonstrated on ultrasound scanning or other radiological tests,

e.g. polycystic kidney disease, reflux nephropathy; biopsy-proven chronic

glomerulonephritis

Recent developments in the perioperative management

297

the leading cause of death in patients with established

renal failure.77

Left ventricular hypertrophy (LVH) occurs due to a

combination of pressure and volume overload. Volume

overload may be due to sodium and water retention, the

presence of an atrioventricular (AV) fistula, or chronic

anaemia with increased stroke volume and heart rate.77

Pressure overload is related to hypertension and arterio-

sclerosis. LVH is associated with myocardial fibrosis and

abnormalities of myocardial relaxation both of which

contribute to diastolic dysfunction and arrhythmias.77

Reduced LV compliance may result in increased sensi-

tivity to volume changes with a small increase in LV

volume precipitating pulmonary oedema.

Accelerated atherosclerosis is a feature of CKD

(Table 4).119 This may be explained by impaired

endothelial function, low grade inflammation, and dyslipi-

daemia. Lipoprotein metabolism is altered in CKD, in par-

ticular high-density lipoprotein levels fall and intermediate

density lipoprotein accumulates.119 Activation of the

renin–angiotensin system (RAS) may also contribute.

Angiotensin II, acting on the AT1 receptor, stimulates the

production of reactive oxygen species. This contributes to

endothelial dysfunction and vascular remodelling.119

Vascular calcification with calcified, stenotic athero-

sclerotic lesions, and valvular heart disease is another

cardiovascular complication of CKD and calcium sup-

plements may enhance this process.76 Calciphylaxis is a

specific dialysis-related type of vascular calcification

characterized by diffuse calcification of the media of small

to medium arteries and arterioles with intimal proliferation

and thrombosis. It results in skin ulcers and can lead to

life-threatening skin necrosis or acral gangrene.

Conduction abnormalities: Myocardial fibrosis and cal-

cification involving the conduction system results in an

increase in the prevalence of second- and third-degree AV

block in patients with CKD and an increase in the inci-

dence of permanent pacing in dialysis patients.71 Hyperka-

laemia and calcium channel blockers may contribute to

the development of complete AV block, as may epidural

anaesthesia, with its attendant reduction in sympathetic

activity.41

Hypertension may be the primary cause of kidney

disease or the result of renal parenchymal or renovascular

disease. Arterial pressure control and block of the RAS are

important in preserving residual function in patients with

both diabetic and non-diabetic renal disease.

The threshold for initiating treatment with antihyperten-

sive medication in patients with renal impairment is an

SBP �140 mm Hg, DBP �90 mm Hg, or both, unless the

urine protein:creatinine ratio is .100 mg mmol21, when a

threshold of 130/80 mm Hg is advocated. An arterial

pressure of ,130/80 mm Hg is required for optimal

control and an arterial pressure of ,125/75 mm Hg may

produce additional benefits in patients with proteinuria of

�1 g/24 h.63 147

Despite the fact that angiotensin-converting enzyme

inhibitors (ACEI) and angiotensin II receptor blockers

(ARBs) may precipitate a deterioration in renal function,

for instance, in the presence of bilateral atherosclerotic

renal artery stenosis, the majority of patients with CKD and

hypertension will benefit from such treatment. Blocking the

RAS may have renoprotective benefits beyond that of arter-

ial pressure control alone.47 ACEI and ARBs have a major

prognostic benefit in proteinuric renal disease.63 There is

Table 4 Complications of CKD

Cardiovascular system

Salt and water retention, hypertension, and LVH

Cardiomyopathy, congestive cardiac failure, and subclinical pulmonary

oedema

Accelerated atherosclerosis and stiffening of large capacitative arteries

Altered lipoprotein metabolism

Complications of AVF/shunts, e.g. heart failure, limb ischaemia, steal

syndrome, pulmonary atheroembolism

Uraemic pericarditis

Cardiovascular autonomic neuropathy with reduced baroreceptor sensitivity,

sympathetic hyperactivity, and parasympathetic dysfunction

Calciphylaxis and vascular calcification resulting in valvular heart disease

and calcified atherosclerotic lesions

Anaemia

Haemostasis and coagulation

Uraemic thrombocytopathy

Prothrombotic tendency/hypercoagulation and reduced fibrinolysis

Vascular access thrombosis

Metabolic acidosis

Bone resorption

Negative nitrogen balance, muscle wasting, growth retardation

Musculoskeletal system

Renal osteodystrophy

Rhabdomyolysis after major surgery

Endocrine system

Secondary and tertiary hyperparathyroidism, vitamin D deficiency

Diabetes mellitus

Gastrointestinal system

Delayed gastric emptying

Anorexia, vomiting, reduced protein intake, malnutrition

Reduced calcium absorption

Immune system

Immunosuppression due to uraemia or drugs

Fluid and electrolyte homeostasis

Hyperkalaemia

Volume overload

Dehydration

Table 3 Aetiology of established renal failure in the UK.3 The data reported

here have been reproduced with permission from the UK Renal Registry of

the Renal Association (http://www.renalreg.com/reports.renal-registry-reports/

2006/). The interpretation and reporting of these data are the responsibility of

the authors and in no way should be seen as an official policy or interpretation

of the UK Renal Registry or the Renal Association

Aetiology % total Male:female ratio

Diabetes mellitus 19.8 1.6

Glomerulonephritis 10.3 1.9

Pyelonephritis 8.2 1.7

Renovascular disease 7.6 1.8

Polycystic kidneys 6.1 1.1

Hypertension 4.8 2.4

Uncertain aetiology/glomerulonephritis unproven 28.0 1.6

Other 15.2 1.4

Craig and Hunter

298

also evidence that a combination of ARB and ACEI may be

more effective than ARB or ACEI alone in protecting renal

survival.96 Other antihypertensives that are commonly used

include thiazide and loop diuretics, the dihydropyridine

calcium channel blockers, a-adrenoceptor blockers, hydra-

lazine, moxonidine, and minoxidil.

Erythropoietin increases blood viscosity and vascular

resistance. This may aggravate hypertension.112

Calcineurin inhibitors such as ciclosporin and corticoster-

oids may also induce hypertension in renal transplant

recipients.

Haemostasis and coagulation

Patients with CKD are often considered to have a bleeding

tendency characterized by platelet dysfunction (uraemic

thrombocytopathy or thrombasthenia). There is, however,

evidence indicating a prothrombotic state in these

patients.109 116 135 136 The evidence of platelet dysfunction

is based upon laboratory findings and a prolonged bleeding

time. A number of abnormalities of platelet function have

been demonstrated including defective interaction of von

Willebrand factor with platelet glycoprotein IIb–IIIa

receptors, reduced platelet ADP content, and reduced

thromboxane A2.111 Platelet count is generally normal, but

anaemia may contribute to the bleeding tendency. A low

haematocrit has been associated with prolonged bleeding

time: patients subjected to a red cell transfusion pro-

gramme showed a normalized bleeding time once haema-

tocrit was .26%.40 Despite the haemostatic defect, there

is also a tendency towards hypercoagulation.

Thromboelastographic indices in patients with CKD

show that all aspects of coagulation are increased, including

initial fibrin formation, fibrin–platelet interaction, and

qualitative platelet function.109 There is also a reduction in

fibrinolysis. Vascular access thrombosis is of particular

importance in patients with Stage 5 CKD on HD as it is

associated with an increased mortality.1 Numerous studies

have attempted to identify factors associated with vascular

access failure46 52 56 61 81 92 130 (Table, see Supplementary

material available at British Journal of Anaesthesia online).

Neuraxial block

The combination of platelet dysfunction and the residual

effects of heparin given during dialysis have raised con-

cerns of an increased risk of epidural haematoma for-

mation. Despite this, there are a number of studies

reporting the use of epidural anaesthesia in patients with

various stages of CKD. The use of hypotensive epidural

anaesthesia in 50 patients, with CKD Stage 3 or more

undergoing total hip replacement,125 did not result in any

acute deterioration of renal function or other complications

from epidural anaesthesia. Beneficial effects in terms of

respiratory function and quality of analgesia were reported

in 13 patients who received combined epidural and

general anaesthesia for renal transplantation, when

compared with a control group who received general

anaesthesia and systemic analgesics.31 This study of only

25 patients was not adequately powered to detect differ-

ences related to haemodynamic stability, graft function, or

other safety issues.

There may be an association between HD and spon-

taneous epidural haematoma formation.124 HD is associ-

ated with a rise in intracranial pressure that may play a

role in its pathogenesis. Epidural anaesthesia in poorly

controlled hypertensive patients may result in haemo-

dynamic instability that could potentially compromise

renal perfusion and increase the likelihood of acute kidney

injury. Although there may be patients with CKD for

whom the benefits of epidural anaesthesia outweigh the

risks, a careful analysis of the individual case is required.

Metabolic acidosis

A reduction in ammonia synthesis and the ability to

excrete hydrogen ions results in metabolic acidosis in

patients with CKD. The potential for sodium bicarbonate

to exacerbate hypertension and volume overload has

caused concern. However, there was no evidence of this in

a recent systematic review.114 The effect of metabolic

acidosis on the perioperative management of CKD patients

relates to their reduced ability to compensate for respirat-

ory acidosis, and altered drug distribution and efficacy.

Preoperative assessment should include measurement of

plasma bicarbonate.

Autonomic neuropathy

Autonomic neuropathy is common in patients with CKD

and may have significant effect on arterial pressure peri-

operatively. A prevalence of 65% in non-diabetic predialy-

sis CKD patients has been noted, whereas studies of

patients with CKD on HD have revealed a prevalence

between 38% and 87.5%.15 117 118

The aetiology may be multifactorial, with uraemia, dia-

betes mellitus, and hyperparathyroidism82 contributing to

the pathogenesis. A significant association between the

radiological signs of osteodystrophy and the presence of

autonomic neuropathy132 has been shown, but not a link

between the biochemical measures of secondary hyperpar-

athyroidism and autonomic neuropathy. Symptoms of per-

ipheral sensory and motor neuropathy correlate with

cardiovascular autonomic neuropathy.118 Delayed gastric

emptying may be present in up to 69% of patients.2 The

autonomic dysfunction associated with CKD is character-

ized by reduced baroreceptor sensitivity, sympathetic over-

activity, and parasympathetic dysfunction,10 66 and may

predispose to the development of arrhythmias periopera-

tively.62 Elevated levels of angiotensin II and deafferenta-

tion of the baroreceptors may be responsible for the

increase in sympathetic tone. Treatment with ACE

inhibitors corrects this sympathetic overactivity.74 The

parasympathetic dysfunction results in reduced heart rate


299

variability and a reduced heart rate response to atropine.10

Of the numerous tests described to assess the cardiac auto-

nomic reflexes, the heart rate variation during deep breath-

ing and arterial pressure response to hand grip exercise

have the best positive predictive value and are of use in pre-

operative assessment.117 The ECG may provide evidence of

autonomic neuropathy in the form of reduced heart rate

variability (reduced R-R interval variation). A simple ECG

rhythm strip may not detect this with great sensitivity.

However, heart rate variation is easy to quantify by record-

ing the ECG during deep breathing at 6 bpm (5 s inspi-

ration:5 s expiration) for 1 min. The mean of the difference

between maximum and minimum heart rates for each of the

six measured cycles is calculated from the R-R interval. A

value of 15 beats min21 or greater is considered normal.37

Fluid and electrolytes

Traditional teaching is that extracellular fluid volume and

electrolyte composition remain normal until the develop-

ment of dialysis-dependent renal failure.140 However, there

is evidence that patients with CKD develop fluid overload

early and this may be a stimulus for inflammation and

accelerated progression of renal disease.106 It is possible

that oedema is associated with altered gut permeability

and an associated inflammatory response.106

Patients with CKD are unable to adapt to large vari-

ations in salt intake and have an impaired ability to con-

centrate and dilute urine. Maximum sodium excretion is a

function of GFR.104 The impaired ability to excrete a

sodium load predisposes these patients to volume overload,

especially when large volumes of saline solutions are

administered. This propensity becomes more marked as

CKD progresses. Infusion of large volumes of saline will

also result in hyperchloraemic metabolic acidosis. The

deleterious effects of metabolic acidosis include depres-

sion of myocardial contractility, reduced cardiac output,

and reduced renal blood flow.105 Furthermore, hyperchlor-

aemia can reduce renal blood flow and GFR.146 If access

to free water is restricted in the perioperative period, the

inability to concentrate urine will result in hypernatraemia

and hypertonicity (Table, see Supplementary material

available at British Journal of Anaesthesia online).

In managing patients on dialysis, the anaesthetist should

establish the patient’s dry weight and compare it with their

weight immediately before coming to theatre. Failure to

achieve dry weight with dialysis is a common problem,

particularly with short duration dialysis prescriptions.

Plasma potassium concentration usually remains normal

until the onset of Stage 5 CKD. This is due to an increase

in the excretion of potassium per functioning nephron and

increased output in the stool.103 Nevertheless, patients

with CKD are at risk of developing hyperkalaemia if chal-

lenged with excessive exogenous potassium or transcellu-

lar potassium shifts. In this respect, acidaemia, insulin

deficiency, hypertonicity, and acute beta-adrenergic receptor

block should be avoided.104 The American College of

Cardiology/American Heart Association 2007 guidelines

on perioperative cardiovascular evaluation includes renal

insufficiency, defined as a serum creatinine .200 mmol

litre21, as a clinical risk factor.43 The guidelines rec-

ommend that the presence of one or more clinical risk

factors, in patients having vascular- or intermediate-risk

surgery, should prompt the anaesthetist to consider beta-

blocker therapy. However, there is a risk of hyperkalaemia

associated with the i.v. administration of beta-blockers.54

This is likely to be of greatest importance for patients with

Stage 5 CKD. Table 5 gives a list of drugs that may con-

tribute to the development of hyperkalaemia. I.V. fluids

containing hydroxyethyl starch have adverse effects on

renal function in renal transplant recipients and in criti-

cally ill patients with severe sepsis or septic shock.13 22 120

Vascular access

Maintenance of vascular access patency is of vital import-

ance in patients with Stage 5 CKD on HD. Vascular

access may be either permanent or temporary. Options for

permanent access include native arteriovenous fistulae

(AVF), arteriovenous grafts (AVG), and long-term

Table 5 Drugs associated with hyperkalaemia

Drug Comment

Succinylcholine Transient increase in serum potassium

concentration of 0.5–1 mmol litre21 under

halothane anaesthesia

May be used in patients with advanced renal

disease provided that preoperative potassium

level is normal. Avoid repeated

administration133

Non-steroidal

anti-inflammatory drugs

Inhibit aldosterone synthesis. Renal

prostaglandins stimulate renin synthesis and

increase the number of open high-conductance

potassium channels in the distal tubular

principal cells107

Beta-adrenergic receptor

blockers

Reduced cellular uptake of potassium and

inhibition of aldosterone secretion86

Heparin Inhibits aldosterone synthesis.107 Occurs within

a few days of the initiation of therapy. Monitor

potassium concentration if receiving heparin for

more than 3 days

ACE inhibitors and ARBs Inhibit aldosterone synthesis107

Digoxin Inhibition of the NAþ–Kþ-ATPase in the

basolateral membrane of the principal cells107

Spironolactone Potassium-sparing diuretic

Blocks the intracellular mineralocorticoid

receptor107

Amiloride and triamterene Potassium-sparing diuretics

Inhibit sodium transport channels in the luminal

membrane of the principal cells of the distal

tubule107

Ciclosporin Inhibition of the NAþ–Kþ-ATPase and apical

secretory potassium channel activity in

principal cells. Reduced aldosterone sectretion.

Increased potassium efflux from cells107

Tacrolimus Inhibition of the NAþ–Kþ-ATPase and

steroid-mediated sodium transport in the distal

tubule115

Craig and Hunter

300

catheters. Temporary vascular access includes: acute short-

term non-cuffed catheters which may or may not be tun-

nelled; long-term tunnelled cuffed catheters; and s.c. port

catheter systems.

Native AVF

A fistula-first approach is advocated as native fistulae have

the best long-term patency rates, require fewer interven-

tions, and are associated with fewer infective compli-

cations than catheters and grafts.100 The mortality risk is

also reduced in patients dialysing through fistulae.110

A period of maturation, characterized by an increase in

blood flow and an increase in the size of the vein, is

required before a newly created AVF can be used. This

takes 1–4 months. The anaesthetist should protect poten-

tial fistula construction sites, especially the cephalic vein,

perioperatively.

Arteriovenous grafts

AVG provide useful permanent access when superficial

veins are not suitable or have been exhausted. AVG may

be synthetic (e.g. polytetrafluoroethylene) or biological

(e.g. bovine mesenteric vein).

Vascular catheters for HD

Factors requiring consideration before the placement of a

central venous catheter for HD include: the duration for

which the catheter is required, the insertion site, the ideal

position of the tip of the catheter, and the method of

insertion.

Duration

The NKF K/DOQI Guidelines state that acute short-term

non-cuffed catheters should be used for ,1 week because

of the risk of infection.100 If a non-cuffed catheter is

required for longer, it should be converted to a tunnelled

cuffed catheter using the same site provided that there is

no evidence of infection.39 With long-term use, cuffed

central catheters are associated with a higher relative risk

of death due to infection than AVFs.32

Insertion site

The right internal jugular vein is the preferred site as the

risk of complications is lower.100 In particular, it is the

risk of stenosis of the vein that is reduced when using this

route.20 The left internal jugular site is associated with a

poorer blood flow rate and a greater rate of stenosis and

thrombosis.100 The subclavian route should be avoided as

the risk of stenosis after catheterization is unacceptably

high, with 40–50% of patients demonstrating some degree

of stricture on venography.20 Subclavian vein stenosis can

result in fistula dysfunction with elevated venous dialysis

pressures and painful arm oedema.121 In patients who are

candidates for renal transplantation, the femoral route

should be avoided to prevent stenosis of the external iliac

vein, as the transplanted kidney is anastomosed to it.100

The femoral route is also associated with the greatest risk

of infection.

Complications

Problems relating to vascular access are a leading cause of

hospitalization, morbidity and the need for anaesthesia in

patients with Stage 5 CKD.100 These include infection,

stenosis, thrombosis, aneurysm, limb ischaemia, limb

oedema, heart failure, pulmonary atheroembolism, steal

syndrome, and recirculation.

Pharmacology

In patients with CKD, the effect of altered clearance, the

production and accumulation of active metabolites, and

the risk of aggravating pre-existing kidney disease on drug

administration must be considered. Dose adjustment is not

usually necessary until the GFR is ,50 ml min21 1.73 m22.

CKD may influence both the pharmacokinetics and the

pharmacodynamics of a drug.

Pharmacokinetic changes

Absorption

Drug absorption may be altered by: gastroparesis causing

delayed gastric emptying, raised gastric pH caused by

gastric urease, converting urea to ammonia, and gut

oedema.101 Reduced intestinal P-glycoprotein, a transpor-

ter found on the apical cell surface of small and large

intestine mucosal cells which protects the body against

toxic compounds by transporting them into the intestinal

lumen, activity in CKD may lead to increased intestinal

absorption and bioavailability of certain compounds.101

Distribution

Volume of distribution may be increased or decreased by

alterations in body composition, especially total body

water, plasma protein binding, and tissue binding.101 Time

since the last dialysis session influences the volume of dis-

tribution of certain drugs, for example, remifentanil

(Table 6).30 A reduced volume of distribution after a dialy-

sis session may result in increased steady-state drug con-

centrations for drugs administered by continuous infusion.

Plasma protein binding of acidic drugs, which bind

mainly to albumin, is altered by the accumulation of

organic acids, such as uric acid and lactic acid, which

compete for binding sites on albumin. The albumin con-

centration is reduced in CKD and there is a change in its

conformational binding site.101 Basic drugs bind mainly to

a1-acid glycoprotein (AAG), an acute phase protein that is

often elevated in CKD. A reduction in plasma protein

binding with an increase in the free fraction of a drug may

result in an increased volume of distribution and clearance

with no significant change in drug exposure.101


301

Table 6 The influence of Stage 5 CKD on drug disposition

Drug T1/2b Systemic clearance Volume of distribution at steady

state

Unbound fraction (%) Reference

Controls CKD Controls CKD Controls CKD Controls CKD

Opioids

Morphine 186 min 185 min 21.3 ml min21 kg21 17.1 ml min21 kg21 3.7 litre kg21 2.8 litre kg21 — — Chauvin and colleagues18

P,0.05

286 min 290 min 741 ml min21 533 ml min21 241 litre 141 litre — — Sear and colleagues123

P¼0.002

Fentanyl 405 min 594 min 14.8 ml min21 kg21 11.8 ml min21 kg21 7.7 litre kg21 9.5 litre kg21 — — Duthie,33 reproduced from Sear122

P,0.05

Alfentanil 83 min 58 min 6.5 ml min21 kg21 5.3 ml min21 kg21 0.46 litre kg21 0.304 litre kg21 — — Van Peer and colleagues138

P,0.05 P,0.05

90 min 107 min 3.1 ml min21 kg21 3.1 ml min21 kg21 0.281 litre kg21 0.405 litre kg21 11% 19% Chauvin and colleagues17

P,0.02 P,0.05

120.2 min 142.4 min 211.8 ml min21 341.9 ml min21 27.6 litre 40.5 litre 10.3% 12.4% Bower and Sear11

P,0.05 P,0.05

Remifentanil 16.3 min 18.86 min 48.7 ml min21 kg21 29.9 ml min21 kg21 0.566 litre kg21 0.358 litre kg21 — — Dahaba and colleagues30

P¼0.045 P¼0.017 P¼0.002

Oxycodone 138 min 234 min 1100 ml min21 840 ml min21 2.39 litre kg21 3.99 litre kg21 — — Kirvela and colleagues64

P¼0.028 P¼0.034

NMBAs

Atracurium 20.6 min 23.7 min 6.1 ml min21 kg21 6.7 ml min21 kg21 0.182 litre kg21 0.224 litre kg21 — — Fahey and colleagues38

19.3 min 20.1 min 5.5 ml min21 kg21 5.8 ml min21 kg21 0.153 litre kg21 0.141 litre kg21 — — Ward and colleagues141

Cisatracurium 30 min 34.2 min 293 ml min21 254 ml min21 — — – — Eastwood and colleagues34

P,0.05

Mivacurium

Cis–cis 68 min 80 min 3.8 ml min21 kg21 2.4 ml min21 kg21 0.227 litre kg21 0.224 litre kg21 — — Head-Rapson and colleagues57

P,0.01

Cis–trans 2 min 4.3 min 106 ml min21 kg21 80 ml min21 kg21 0.278 litre kg21 0.475 litre kg21 – – Head-Rapson 199557

Trans–trans 2.3 min 4.2 min 57 ml min21 kg21 47 ml min21 kg21 0.211 litre kg21 0.270 litre kg21 – – Head-Rapson 199557

Pancuronium 100 min 489 min 74 ml min21 20 ml min21 0.148 litre kg21 0.236 litre kg21 — — McLeod and colleagues87

P,0.05 P,0.005 P,0.05

Vecuronium 52.6 min 83.1 min 5.29 ml min21 kg21 3.08 ml min21 kg21 0.199 litre kg21 0.24 litre kg21 — — Lynam and colleagues78

P,0.05 P,0.05

Rocuronium 57 min 70 min 4.5 ml min21 kg21 2.7 ml min21 kg21 0.194 litre kg21 0.22 litre kg21 — — Robertson and colleagues113

P,0.0001

97.2 min 104.4 min 3.7 ml min21 kg21 2.5 ml min21 kg21 0.207 litre kg21 0.212 litre kg21 — — Cooper and colleagues25

P,0.05

Induction agents

Propofol 27.7 min 23.8 min 33.75 ml min21 kg21 30.66 ml min21 kg21 5.79 litre kg21 11.25 litre kg21 — — Ickx and colleagues60

0.98% 1.11% before HD Costela and colleagues27

0.87% after HD

Thiopental 611 min 583 min 3.2 ml min21 kg21 4.5 ml min21 kg21 1.9 litre kg21 3.0 litre kg21 15.7% 28% Burch and Stanski14

P,0.05 P,0.05 P,0.05

588 min 1069 min 2.7 ml min21 kg21 3.9 ml min21 kg21 1.4 litre kg21 3.2 litre kg21 11% 17.8% Christensen and colleagues19

P,0.05 P,0.05

Craig

and

Hunter

30

2

Local anaesthetics have two plasma protein binding

sites: a high affinity, low capacity site on AAG, and a low

affinity, high capacity site on albumin.88 The albumin

binding site becomes increasingly important as the plasma

concentration of the local anaesthetic increases. Metabolic

acidosis increases the percentage of unbound drug, and

this effect is more pronounced with bupivacaine.97 The

effect of these changes on the toxicity of local anaesthetics

is unclear.

Elimination

The kidneys contribute up to 18% of total cytochrome

P450 (CYP) activated drug metabolism and are also

involved in conjugation reactions.36 Interestingly, non-

renal clearance of many drugs is reduced in patients with

kidney disease.101 Hepatic clearance may vary with

changes in hepatic blood flow, the free fraction of a drug,

and the metabolic capacity of the liver enzymes.36 Animal

studies of experimental CKD have demonstrated a 40%

reduction in total microsomal CYP activity.69 Severe

kidney disease differentially affects hepatic CYP activity;

in particular, the activity of CYP 3A4 and CYP 2C9 is

reduced.101 In contrast, elevated plasma urea concentration

induces CYP 2E1 activity.101 The mechanism of altered

enzyme activity appears to relate in part to altered gene

transcription, although depletion of cofactors may also be

involved.36 101

Potent inhalation agents

Historically, methoxyflurane anaesthesia resulted in ele-

vated serum inorganic fluoride levels and polyuric renal

failure.28 Serum fluoride levels .50 mmol litre21 were

associated with an increased risk of renal damage.

Sevoflurane metabolism also results in elevated fluoride

levels, with peak levels .50 mmol litre21. Sevoflurane

reacts with strong bases in CO2 absorbents to produce

Compound A, a dose-related nephrotoxin in rats.48 A retro-

spective evaluation of pooled renal laboratory data from

22 clinical trials that compared sevoflurane with isoflurane,

enflurane, or propofol found that the incidence of raised

serum creatinine and blood urea nitrogen concentrations

was similar after sevoflurane or the control agents.84 There

was no specific trend with respect to postoperative serum

creatinine and fresh gas flow, type of CO2 absorbent, or

effect of concurrent treatment with nephrotoxic antibiotics

when sevoflurane was used. The authors concluded that

exposure to ,4 MAC h of sevoflurane was not associated

with an increased risk of renal toxicity. Serum fluoride

kinetics after sevoflurane anaesthesia have been studied in

patients with CKD and compared with a group with

normal renal function.99 In CKD patients, the serum inor-

ganic fluoride level and its rate of elimination did not

differ significantly from controls. Indices of renal tubular

function, such as b2-microglobulin and N-acetyl-b-D-

glucosaminidase, do not significantly change after

anaesthesia with sevoflurane in either patients with CKD

or controls.99 Despite initial concerns, sevoflurane is a

suitable choice of potent inhalation anaesthetic agent for

patients with CKD (Table, see Supplementary material

available at British Journal of Anaesthesia online).

Enflurane undergoes greater biotransformation to inor-

ganic fluoride than either isoflurane or desflurane. It is

known to cause vasopressin-resistant polyuria in rats, and

in volunteers prolonged enflurane anaesthesia resulted in a

25% reduction in urine concentrating ability and a transi-

ent reduction in creatinine clearance of 35%.4 85 Case

reports of renal failure after enflurane anaesthesia in

patients with renal dysfunction suggest that it is best

avoided in this group, although studies of the effects of

enflurane in patients with stable renal insufficiency found

no deterioration in function.24 35 144

Desflurane and isoflurane are not associated with renal

toxicity and appear safe to use in patients with CKD.75 148

I.V. anaesthetic agents

Propofol pharmacokinetics are unaltered by established

renal failure (Table 6). The induction dose of propofol

associated with a bispectral index value of 50 is signifi-

cantly higher in patients with established renal failure

compared with controls: 2.03 vs 1.39 mg kg21; P,0.05.49

The time interval between cessation of a propofol infusion

and eye opening is significantly shorter in renal failure

patients than controls (474 vs 714 s; P,0.05), although

blood propofol concentrations are not significantly differ-

ent on waking.60

Thiopental has an increased volume of distribution and

reduced plasma protein binding in renal failure (Table 6).

The brain is exposed to a higher free drug concentration.

The rate of administration should be reduced.122

Neuromuscular blocking and reversal agents

When selecting a neuromuscular blocking agent (NMBA)

for use in patients with CKD, consider the impact of renal

impairment on the elimination of the drug, the potential

for drug accumulation with incremental doses, and the

production of active metabolites. Other factors include the

effect of acidaemia and drug interactions on the intensity

and duration of block. In general, the initial dose required

to produce neuromuscular block (3�ED95, which is the

effective dose to produce 95% twitch depression) is larger

in patients with CKD than in normal subjects. But, with

the exception of atracurium and cisatracurium, the dose

required to maintain block is reduced.50 108 To prevent

postoperative residual curarization (PORC), the anaesthe-

tist should avoid using long-acting NMBA, or agents

which are excreted in part in the urine, and make routine

use of neuromuscular monitoring.

The problems of prolonged neuromuscular block and

PORC were particularly pertinent to the use of

D-tubocurarine and pancuronium, as both agents have a


303

reduced clearance and prolonged half-life in the presence

of CKD. Furthermore, pancuronium has an active metab-

olite, 3-hydroxypancuronium, with half the neuromuscular

blocking potency of the parent compound.

Atracurium is not dependent upon renal or hepatic func-

tion for its elimination, as it undergoes spontaneous break-

down at body temperature and pH, a process known as

Hofmann degradation, and metabolism by non-specific

esterases. One of the products of Hofmann degradation, lau-

danosine, has been shown to be epileptogenic in animals.16

The pharmacodynamics and pharmacokinetics of atracur-

ium are not altered by CKD (Table 6).38 Atracurium is less

potent, results in greater histamine release, and has a

shorter duration of action than cisatracurium.12

Cisatracurium, the 1R cis-10R cis isomer of atracurium,

is subject to Hofmann degradation and ester hydrolysis,

albeit to differing degrees. As it is more potent, the

plasma concentration of laudanosine after cisatracurium is

lower than after an equipotent dose of atracurium. Although

the laudanosine levels are significantly higher in patients

with CKD who have received cisatracurium compared with

healthy controls, these levels are still approximately 1/10th

of those seen after atracurium.34 Renal failure alters the

pharmacokinetics, but has little impact on the pharmacody-

namics of cisatracurium. The clearance is reduced by 13%

and the terminal elimination half-life prolonged by 4.2 min

(Table 6).34 The only difference in onset or recovery

variables is a longer mean time to 90% depression of the

first twitch of the train-of-four response (T1/T0); 3.7 vs 2.4

min, probably due to a poorer cardiac output and slower

delivery to the neuromuscular junction.12

Mivacurium consists of three isomers: cis–trans (37%),

trans–trans (57%), and cis–cis (6%). Clearance of the

cis–cis isomer, the least potent, is significantly reduced

in patients with renal failure and it may accumulate

(Table 6).57 In renal failure, spontaneous recovery is

slower and lower infusion rates are required.108 There is an

acquired decrease in plasma cholinesterase activity in

CKD, and a negative correlation between cholinesterase

activity and time to recovery after mivacurium has been

demonstrated.108

Vecuronium undergoes predominantly biliary excretion,

although up to 30% may be excreted by the kidney.6 It is

also metabolized in the liver to 3-hydroxyvecuronium

which is active at the neuromuscular junction. Renal

failure results in a reduced clearance, increased terminal

elimination half-life, and prolonged duration of action

(Table 6).78 Accumulation occurs with repeat boluses or

constant infusions resulting in prolonged neuromuscular

block.8 72

Rocuronium is excreted primarily in the bile, although

up to 33% may be excreted in the urine within 24 h.145 A

small fraction is metabolized in the liver producing a

metabolite with very low neuromuscular blocking

activity.91 Renal failure reduces the clearance of rocuro-

nium by 39% (Table 6).113 The duration of action (25%

recovery of T1/T0) and time to recovery of the TOF ratio

to 0.7 are significantly prolonged in patients with renal

failure compared with controls: 49 vs 32 min and 88 vs 55

min, respectively.113 Importantly, inter-patient variability

is increased in patients with renal failure.25

Neostigmine clearance is reduced and its half-life is pro-

longed in CKD. This may result in a parasympathomimetic

response, including bradycardia and AV block, especially

when used in combination with atropine rather than the

longer-acting glycopyrronium.142

Sugammadex may prove useful in preventing PORC

when patients have received an aminosteroid NMBA. It is

a modified g-cyclodextrin that selectively encapsulates

steroid-based non-depolarizing NMBAs. The resulting

guest–host complex is water soluble and exists in equili-

brium but with a very low dissociation rate.95

Sugammadex is biologically inactive and does not bind

to plasma proteins. Furthermore, it appears to have rela-

tively few side-effects, although hypotension has been

documented.139 In individuals with normal renal function,

sugammadex is excreted unchanged in the urine and it also

enhances the renal excretion of rocuronium. Sorgenfrei and

colleagues131 showed that 59–77% of sugammadex is

excreted unchanged in the urine within 16 h. But, its effi-

cacy as a reversal agent does not appear to rely on renal

excretion of the cyclodextrin-relaxant complex.95

Analgesic agents

In administering analgesic agents, the anaesthetist needs to

consider: the impact of renal impairment on the distri-

bution and elimination of the parent compound and hence

the need for adjusting the dose or dose interval (Table 6);

the formation of active metabolites; and the risk of com-

promising residual renal function.

Acetaminophen

Oral acetaminophen 40 mg kg21 day21 for 3 days in

normal subjects and patients with CKD produced no

demonstrable change in glomerular or tubular function in

either group.7 Prolonged use of acetaminophen is associ-

ated with analgesic nephropathy, but occasional or mod-

erate use is safe.67 Analgesic nephropathy is mainly

associated with prolonged use of compound analgesics

containing two antipyretic agents with caffeine or codeine.

The use of acetaminophen in the perioperative period is

safe and does not require dose adjustment.

Non-steroidal anti-inflammatory agents

The adverse effects of the non-steroidal anti-inflammatory

drugs (NSAIDs) are likely to outweigh any potential

benefit in the perioperative period. They exacerbate hyper-

tension and precipitate oedema, hyponatraemia, and hyper-

kalaemia. There is an increased risk of gastrointestinal

bleeding, which may be aggravated by the combined

Craig and Hunter

304

effects of uraemic thrombasthenia and drug-induced plate-

let inhibition. Their use is associated with an increased

risk of cardiovascular complications in this at risk popu-

lation.93 They are nephrotoxic agents that precipitate an

acute decrease in GFR and may also cause acute intersti-

tial nephritis as part of an idiosyncratic reaction. The renal

effects of the COX-2 inhibitors are similar to those of the

non-selective NSAIDs.68

Opioids

Opioids have no direct toxic effects on the kidney. They

do, however, have an antidiuretic effect, and they may

cause urinary retention. Very rarely, their use has resulted

in rhabdomyolysis.9

Morphine

Morphine is metabolized in the liver to a number of

metabolites of which morphine-3-glucuronide is the

major one, accounting for 70% of the dose. Morphine-3-

glucuronide antagonizes the analgesic effect of morphine,

and is associated with irritability and a lower seizure

threshold.94 Approximately 5% of a dose of morphine is

metabolized to morphine-6-glucuronide (M6G), which has

potent analgesic properties and may result in delayed

onset of sedation and respiratory depression. The elimin-

ation of M6G is dependent on renal function, and in

patients with renal failure, its half-life is prolonged from 2

to 27 h.94 The metabolite load from an equi-analgesic

dose of morphine given by the oral route is greater than

that from the parenteral route, due to extensive first-pass

metabolism. In renal patients, the dose of morphine should

be reduced and the patient carefully monitored for signs of

delayed onset respiratory depression postoperatively

(Table 6). A significant fraction of morphine will be

removed by HD.5

Fentanyl

Fentanyl undergoes extensive hepatic metabolism with

no active metabolites. Approximately 7% is excreted

unchanged in the urine.68 Clearance is reduced in CKD,

with a strong negative correlation between clearance and

urea concentration.65 HD has little impact on fentanyl

plasma concentration.5

Alfentanil

Elimination half-life and plasma clearance are not altered

in renal failure, although protein binding is reduced with

an increase in the free fraction of alfentanil (Table 6).17

There are no active metabolites. The dose required is

reduced, but the dose interval remains unchanged.94

Remifentanil

Remifentanil is not dependent on renal function for elim-

ination. It undergoes ester hydrolysis and its main metab-

olite is minimally active with 1/300–1/1000 the potency

of the parent compound.30 In patients on HD, remifentanil

had a reduced clearance and prolonged elimination half-

life (Table 6).30 A lower infusion rate is required, but

recovery is not significantly prolonged.29

Oxycodone

Oxycodone is metabolized in the liver to noroxycodone

and oxymorphone. Oxymorphone has analgesic activity

and both the parent compound and the metabolites accum-

ulate in renal failure.64 The elimination half-life of oxyco-

done is prolonged, from 2.3 h in controls to 3.9 h in

established renal failure (Table 6).64 The dose should be

reduced and dose interval increased.

Tramadol

Thirty per cent of tramadol is excreted unchanged in the

urine. O-Demethyl tramadol is an active metabolite which

is excreted by the kidneys. Uraemia is associated with a

lowered seizure threshold, and tramadol may be epilepto-

genic in these circumstances.44 Tramadol is removed by HD.

Other opioids

Meperidine is metabolized to normeperidine which is depen-

dent on renal function for elimination. The use of meperidine

in patients with CKD has been associated with seizures,

myoclonus, and altered mental state.55 Codeine and dihydro-

codeine are also best avoided as their elimination half-life

is significantly prolonged, and conventional doses have

resulted in central nervous system depression.51 83

Immunosupression therapy

Patients with Stage 5 CKD who have undergone renal

transplantation require immunosuppression. These drugs

are usually given by the oral route. If enteral adminis-

tration is inappropriate, then i.v. administration with dose

adjustment will be required. Traditional regimens include

some form of triple therapy, consisting of a calcineurin

inhibitor (ciclosporin or tacrolimus), an antiproliferative

agent (azathioprine or mycophenolate mofetil), and a cor-

ticosteroid.149 Newer regimens attempt to spare or elimin-

ate corticosteroids and calcineurin inhibitors. Polyconal

and monoclonal antibodies also form part of the

armamentarium.

Calcineurin inhibitors

Ciclosporin and tacrolimus are calcineurin inhibitors

which form the mainstay of immunosuppression therapy.

They prevent cytokine mediated T-cell activation and pro-

liferation by blocking the calcineurin phosphatase-

dependent pathway involved in the transcription of several

cytokines, interleukin-2 being the most important.45 Both

ciclosporin and tacrolimus are metabolized by CYP 3A4:

drugs that either induce or inhibit this enzyme will alter

the plasma concentration of these immunosuppressants.

The oral bioavailability of ciclosporin is variable and is


305

influenced by the formulation used. The oral bioavailabil-

ity of the unmodified formulation is 30%. The i.v. dose

should be reduced accordingly and levels monitored.149

The solvent in i.v. ciclosporin, cremophor EL, has been

associated with anaphylactic reactions. Furthermore, i.v.

ciclosporin may cause vasoconstriction and hyperkalae-

mia.134 It should therefore be administered slowly as a

continuous infusion.45 Adverse effects include hyperten-

sion, hyperlipidaemia, hyperkalaemia, gum hypertrophy,

and nephrotoxicity with renal fibrosis. Ciclosporin has

been shown to increase the hypnotic effect of barbiturates

and the analgesic effect of fentanyl in mice.21 A retrospec-

tive study in humans found no evidence of clinically sig-

nificant prolongation of anaesthetic effect.90 Ciclosporin

potentiates neuromuscular block. Its intraoperative use is

associated with an increased risk of postoperative respirat-

ory failure.128

Tacrolimus is similar to ciclosporin, but may have a

better cardiovascular risk profile, with less hypertension

and hyperlipidaemia, and although it is also nephrotoxic,

it is associated with improved long-term post-transplant

renal function.42 Important adverse effects include

disturbed glucose metabolism and diabetes mellitus,

tremor, diarrhoea, neurotoxicity, and nephrotoxicity. Dose

is adjusted according to monitored levels and an i.v. prep-

aration is available.

Antiproliferative agents

Mycophenolate mofetil has replaced azathioprine. It acts

by inhibiting de novo purine synthesis in lymphocytes. It

is dosed empirically and adjusted when side-effects

occur.149 These include leucopaenia, diarrhoea, and infec-

tion. It is not nephrotoxic. The oral bioavailability is 94%.

Azathioprine causes transient antagonism of neuromuscu-

lar block which is unlikely to be of clinical importance.50

Mammalian target of rapamycin inhibitors

Sirolimus and everolimus are used in some newer regi-

mens to replace calcineurin inhibitors as they are less

nephrotoxic. They act by blocking signal transduction in

the interleukin-2 pathway. They are metabolized by CYP

3A4, with enzyme inducers and inhibitors affecting the

level of immunsuppression. Side-effects include hyperlipi-

daemia, leucopaenia, thrombocytopaenia, pneumonitis,

and, rarely, angioedema.149

Antibodies

Examples include antilymphocyte and antithymocyte anti-

bodies (polyclonal), OKT3 which is a monoclonal mouse

antibody directed against the CD3 protein complex, and

anti-interleukin-2 receptor monoclonal antibody. OKT3

has been associated with non-cardiogenic pulmonary

oedema, particularly in combination with pre-existing

increased intravascular volume.26 A biphasic response may

follow administration with fever, hypertension, and tachy-

cardia followed by hypotension and hypoxia.

Conclusions

In managing patients with CKD, the anaesthetist aims to

minimize the risk of perioperative complications. This

requires careful patient assessment and efforts to modify

identified risk factors, for example, hyperkalaemia, to

improve patient outcome. Recent developments in this

regard include: a refined appreciation of the association

between CKD and cardiovascular disease, knowledge of

the importance of blocking the RAS to delay progression

of the condition, and new insights into the complex pro-

thrombotic and haemostatic abnormalities involved. It is

clear that temporary vascular access for HD is to be

avoided and subclavian HD catheters are associated with

an unacceptably high risk of subclavian vein stenosis. The

pharmacokinetic and pharmacodynamic changes must be

taken into consideration: many drugs having reduced renal

and non-renal clearance in CKD. PORC remains a risk.

Supplementary material

Supplementary material is available at British Journal of

Anaesthesia online.

References1 Akman B, Afsar B, Atac FB, et al. Predictors of vascular access

thrombosis among patients on the cadaveric renal transplan-tation waiting list. Transplant Proc 2006; 38: 413–5

2 Alimchandani A, Pai-dhungat JV. A study of gastric emptying inchronic renal failure. J Assoc Physicians India 1997; 45: 835–8

3 Ansell D, Feest TG, Tomson C, Williams AJ, Warwick G. UK

Renal Registry Report 2006. Bristol, UK: UK Renal Registry, 20074 Barr GA, Cousins MJ, Mazze RI, Hitt BA, Kosek JC. A compari-

son of the renal effects and metabolism of enflurane and meth-oxyflurane in Fischer 344 rats. J Pharmacol Exp Ther 1974; 188:257–64

5 Bastani B, Jamal JA. Removal of morphine but not fentanylduring haemodialysis. Nephrol Dial Transplant 1997; 12: 2802–4

6 Bencini AF, Scaf AH, Sohn YJ, et al. Disposition and urinaryexcretion of vecuronium bromide in anesthetized patients withnormal renal function or renal failure. Anesth Analg 1986; 65:

245–517 Berg KJ, Djoseland O, Gjellan A, et al. Acute effects of paraceta-

mol on prostaglandin synthesis and renal function in normalman and in patients with renal failure. Clin Nephrol 1990; 34:

255–628 Bevan DR, Donati F, Gyasi H, Williams A. Vecuronium in renal

failure. Can Anaesth Soc J 1984; 31: 491–69 Blain PG, Lane RJ, Bateman DN, Rawlins MD. Opiate-induced

rhabdomyolysis. Hum Toxicol 1985; 4: 71–4

10 Bondia A, Tabernero JM, Macias JF, Martin-Luengo C. Auto-nomic nervous system in haemodialysis. Nephrol Dial Transplant1988; 3: 174–80

11 Bower S, Sear JW. Disposition of alfentanil in patients receivinga renal transplant. J Pharm Pharmacol 1989; 41: 654–7

12 Boyd AH, Eastwood NB, Parker CJ, Hunter JM. Pharmacody-namics of the 1R cis-10R cis isomer of atracurium (51W89) inhealth and chronic renal failure. Br J Anaesth 1995; 74: 400–4

Craig and Hunter

306

13 Brunkhorst FM, Engel C, Bloos F, et al. Intensive insulin therapyand pentastarch resuscitation in severe sepsis. N Engl J Med2008; 358: 125–39

14 Burch PG, Stanski DR. Decreased protein binding and thiopen-

tal kinetics. Clin Pharmacol Ther 1982; 32: 212–715 Calvo C, Maule S, Mecca F, Quadri R, Martina G, Cavallo PP.

The influence of autonomic neuropathy on hypotension duringhemodialysis. Clin Auton Res 2002; 12: 84–7

16 Chapple DJ, Miller AA, Ward JB, Wheatley PL. Cardiovascularand neurological effects of laudanosine. Studies in mice and rats,and in conscious and anaesthetized dogs. Br J Anaesth 1987; 59:218–25

17 Chauvin M, Lebrault C, Levron JC, Duvaldestin P. Pharmacoki-

netics of alfentanil in chronic renal failure. Anesth Analg 1987;66: 53–6

18 Chauvin M, Sandouk P, Scherrmann JM, Farinotti R, Strumza P,Duvaldestin P. Morphine pharmacokinetics in renal failure.Anesthesiology 1987; 66: 327–31

19 Christensen JH, Andreasen F, Jansen J. Pharmacokinetics andpharmacodynamics of thiopental in patients undergoing renaltransplantation. Acta Anaesthesiol Scand 1983; 27: 513–8

20 Cimochowski GE, Worley E, Rutherford WE, Sartain J, BlondinJ, Harter H. Superiority of the internal jugular over the subcla-

vian access for temporary dialysis. Nephron 1990; 54: 154–6121 Cirella VN, Pantuck CB, Lee YJ, Pantuck EJ. Effects of cyclospor-

ine on anesthetic action. Anesth Analg 1987; 66: 703–622 Cittanova ML, Leblanc I, Legendre C, Mouquet C, Riou B,

Coriat P. Effect of hydroxyethylstarch in brain-dead kidneydonors on renal function in kidney-transplant recipients. Lancet1996; 348: 1620–2

23 Cockcroft DW, Gault MH. Prediction of creatinine clearancefrom serum creatinine. Nephron 1976; 16: 31–41

24 Conzen PF, Nuscheler M, Melotte A, et al. Renal function andserum fluoride concentrations in patients with stable renal

insufficiency after anesthesia with sevoflurane or enflurane.Anesth Analg 1995; 81: 569–75

25 Cooper RA, Maddineni VR, Mirakhur RK, Wierda JM, Brady M,Fitzpatrick KT. Time course of neuromuscular effects and phar-macokinetics of rocuronium bromide (Org 9426) during isoflur-

ane anaesthesia in patients with and without renal failure. Br JAnaesth 1993; 71: 222–6

26 Cosimi AB, Jenkins RL, Rohrer RJ, Delmonico FL, Hoffman M,Monaco AP. A randomized clinical trial of prophylactic OKT3monoclonal antibody in liver allograft recipients. Arch Surg 1990;

125: 781–427 Costela JL, Jimenez R, Calvo R, Suarez E, Carlos R. Serum

protein binding of propofol in patients with renal failure orhepatic cirrhosis. Acta Anaesthesiol Scand 1996; 40: 741–5

28 Cousins MJ, Mazze RI, Kosek JC, Hitt BA, Love FV. The etiologyof methoxyflurane nephrotoxicity. J Pharmacol Exp Ther 1974;

190: 530–4129 Dahaba AA, Von Klobucar F, Rehak PH, List WF. Total intrave-

nous anesthesia with remifentanil, propofol and cisatracurium inend-stage renal failure. Can J Anaesth 1999; 46: 696–700

30 Dahaba AA, Oettl K, Von Klobucar F, Reibnegger G, List WF.

End-stage renal failure reduces central clearance and prolongsthe elimination half life of remifentanil. Can J Anaesth 2002; 49:369–74

31 Dauri M, Costa F, Servetti S, Sidiropoulou T, Fabbi E, Sabato AF.Combined general and epidural anesthesia with ropivacaine for

renal transplantation. Minerva Anestesiol 2003; 69: 873–8432 Dhingra RK, Young EW, Hulbert-Shearon TE, Leavey SF, Port

FK. Type of vascular access and mortality in U.S. hemodialysispatients. Kidney Int 2001; 60: 1443–51

33 Duthie DJR. Renal failure, surgery and fentanyl pharmacokinetics.Abstracts of VII European Congress of Anaesthesiology, volume I.Beitrage zur Anaesthesiologie und Intensivmedizin 1986; 16: 187

34 Eastwood NB, Boyd AH, Parker CJ, Hunter JM. Pharmacoki-

netics of 1R-cis 1’R-cis atracurium besylate (51W89) and plasmalaudanosine concentrations in health and chronic renal failure.Br J Anaesth 1995; 75: 431–5

35 Eichhorn JH, Hedley-Whyte J, Steinman TI, Kaufmann JM,

Laasbert LH. Renal failure following enflurane anesthesia.Anesthesiology 1976; 45: 557–60

36 Elston AC, Bayliss MK, Park GR. Effect of renal failure on drugmetabolism by the liver. Br J Anaesth 1993; 71: 282–90

37 Ewing DJ, Clarke BF. Diagnosis and management of diabetic

autonomic neuropathy. Br Med J (Clin Res Ed) 1982; 285: 916–838 Fahey MR, Rupp SM, Fisher DM, et al. The pharmacokinetics

and pharmacodynamics of atracurium in patients with andwithout renal failure. Anesthesiology 1984; 61: 699–702

39 Falk A, Prabhuram N, Parthasarathy S. Conversion of temporaryhemodialysis catheters to permanent hemodialysis catheters: aretrospective study of catheter exchange versus classic de novo

placement. Semin Dial 2005; 18: 425–3040 Fernandez F, Goudable C, Sie P, et al. Low haematocrit and pro-

longed bleeding time in uraemic patients: effect of red cell trans-fusions. Br J Haematol 1985; 59: 139–48

41 Ferrari F, Nascimento P Jr, Vianna PT. Complete atrioventricularblock during renal transplantation in a patient with Alport’s syn-drome: case report. Sao Paulo Med J 2001; 119: 184–6

42 First MR. Improving long-term renal transplant outcomes withtacrolimus: speculation vs evidence. Nephrol Dial Transplant2004; 19: vi17–22

43 Fleisher LA, Beckman JA, Brown KA, et al. ACC/AHA 2007

guidelines on perioperative cardiovascular evaluation and carefor noncardiac surgery: a report of the American College ofCardiology/American Heart Association Task Force on PracticeGuidelines (Writing Committee to Revise the 2002 Guidelineson Perioperative Cardiovascular Evaluation for Noncardiac

Surgery): developed in collaboration with the American Societyof Echocardiography, American Society of Nuclear Cardiology,Heart Rhythm Society, Society of Cardiovascular Anesthesiolo-gists, Society for Cardiovascular Angiography and Interventions,Society for Vascular Medicine and Biology, and Society for Vas-

cular Surgery. Circulation 2007; 116: e418–9944 Gardner JS, Blough D, Drinkard CR, et al. Tramadol and sei-

zures: a surveillance study in a managed care population.Pharmacotherapy 2000; 20: 1423–31

45 Gaston RS. Maintenance immunosuppression in the renal trans-

plant recipient: an overview. Am J Kidney Dis 2001; 38: S25–3546 George J, Aron A, Levy Y, et al. Anti-cardiolipin, anti-

endothelial-cell and anti-malondialdehyde-LDL antibodies inuremic patients undergoing hemodialysis: relationship with vas-

cular access thrombosis and thromboembolic events. HumAntibodies 1999; 9: 125–31

47 Giatras I, Lau J, Levey AS. Effect of angiotensin-convertingenzyme inhibitors on the progression of nondiabetic renaldisease: a meta-analysis of randomized trials. Angiotensin-

Converting-Enzyme Inhibition and Progressive Renal DiseaseStudy Group. Ann Intern Med 1997; 127: 337–45

48 Gonsowski CT, Laster MJ, Eger EI, Ferrell LD, Kerschmann RL.Toxicity of compound A in rats. Effect of increasing duration ofadministration. Anesthesiology 1994; 80: 566–73

49 Goyal P, Puri GD, Pandey CK, Srivastva S. Evaluation of induc-tion doses of propofol: comparison between endstage renaldisease and normal renal function patients. Anaesth IntensiveCare 2002; 30: 584–7


307

50 Gramstad L. Atracurium, vecuronium and pancuronium in end-stage renal failure. Dose–response properties and interactionswith azathioprine. Br J Anaesth 1987; 59: 995–1003

51 Guay DR, Awni WM, Findlay JW, et al. Pharmacokinetics and

pharmacodynamics of codeine in end-stage renal disease. ClinPharmacol Ther 1988; 43: 63–71

52 Gultekin F, Alagozlu H, Candan F, Nadir I, Bakici MZ, Sezer H. Therelationship between anticardiolipin antibodies and vascular access

occlusion in patients on hemodialysis. ASAIO J 2005; 51: 162–453 Hallan S, Asberg A, Lindberg M, Johnsen H. Validation of the

modification of diet in renal disease formula for estimating GFRwith special emphasis on calibration of the serum creatinineassay. Am J Kidney Dis 2004; 44: 84–93

54 Hamad A, Salameh M, Zihlif M, Feinfeld DA, Carvounis CP.Life-threatening hyperkalemia after intravenous labetolol injec-tion for hypertensive emergency in a hemodialysis patient. Am JNephrol 2001; 21: 241–4

55 Hassan H, Bastani B, Gellens M. Successful treatment of norme-peridine neurotoxicity by hemodialysis. Am J Kidney Dis 2000;35: 146–9

56 Haviv YS. Association of anticardiolipin antibodies with vascularaccess occlusion in hemodialysis patients: cause or effect?Nephron 2000; 86: 447–54

57 Head-Rapson AG, Devlin JC, Parker CJ, Hunter JM.Pharmacokinetics and pharmacodynamics of the three isomers

of mivacurium in health, in end-stage renal failure and in patientswith impaired renal function. Br J Anaesth 1995; 75: 31–6

58 Howell SJ, Sear YM, Yeates D, Goldacre M, Sear JW, Foex P.Risk factors for cardiovascular death after elective surgery

under general anaesthesia. Br J Anaesth 1998; 80: 14–959 Howell SJ, Sear JW, Sear YM, Yeates D, Goldacre M, Foex P.

Risk factors for cardiovascular death within 30 days after anaes-thesia and urgent or emergency surgery: a nested case–controlstudy. Br J Anaesth 1999; 82: 679–84

60 Ickx B, Cockshott ID, Barvais L, et al. Propofol infusion forinduction and maintenance of anaesthesia in patients with end-stage renal disease. Br J Anaesth 1998; 81: 854–60

61 Isbir CS, Akgun S, Yilmaz H, et al. Is there a role of angiotensin-converting enzyme gene polymorphism in the failure of arterio-

venous femoral shunts for hemodialysis? Ann Vasc Surg 2001; 15:443–6

62 Jassal SV, Coulshed SJ, Douglas JF, Stout RW. Autonomic neuro-pathy predisposing to arrhythmias in hemodialysis patients. Am JKidney Dis 1997; 30: 219–23

63 Joint Specialty Committee on Renal Medicine of the RoyalCollege of Physicians of London and the Renal Association, andthe Royal College of General Practitioners. Chronic KidneyDisease in Adults: UK Guidelines for Identification, Management and

Referral. London: Royal College of Physicians, 200664 Kirvela M, Lindgren L, Seppala T, Olkkola KT. The pharmacoki-

netics of oxycodone in uremic patients undergoing renal trans-plantation. J Clin Anesth 1996; 8: 13–8

65 Koehntop DE, Rodman JH. Fentanyl pharmacokinetics in

patients undergoing renal transplantation. Pharmacotherapy 1997;17: 746–52

66 Kurata C, Uehara A, Sugi T, et al. Cardiac autonomic neuropathyin patients with chronic renal failure on hemodialysis. Nephron2000; 84: 312–9

67 Kurth T, Glynn RJ, Walker AM, et al. Analgesic use and changein kidney function in apparently healthy men. Am J Kidney Dis2003; 42: 234–44

68 Launay-Vacher V, Karie S, Fau JB, Izzedine H, Deray G.Treatment of pain in patients with renal insufficiency: the World

Health Organization three-step ladder adapted. J Pain 2005; 6:137–48

69 Leblond FA, Giroux L, Villeneuve JP, Pichette V. Decreased invivo metabolism of drugs in chronic renal failure. Drug Metab

Dispos 2000; 28: 1317–2070 Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and

prospective validation of a simple index for prediction ofcardiac risk of major noncardiac surgery. Circulation 1999; 100:

1043–971 Leman RB, Kratz JM, Gazes PC. Permanent cardiac pacing in

patients on chronic renal dialysis. Am Heart J 1985; 110:1242–4

72 Lepage JY, Malinge M, Cozian A, Pinaud M, Blanloeil Y,Souron R. Vecuronium and atracurium in patients with end-stagerenal failure. A comparative study. Br J Anaesth 1987; 59:

1004–1073 Levey AS, Eckardt KU, Tsukamoto Y, et al. Definition and classi-

fication of chronic kidney disease: a position statement fromKidney Disease: Improving Global Outcomes (KDIGO). KidneyInt 2005; 67: 2089–100

74 Ligtenberg G, Blankestijn PJ, Oey PL, et al. Reduction of sym-pathetic hyperactivity by enalapril in patients with chronic renalfailure. N Engl J Med 1999; 340: 1321–8

75 Litz RJ, Hubler M, Lorenz W, Meier VK, Albrecht DM. Renalresponses to desflurane and isoflurane in patients with renal

insufficiency. Anesthesiology 2002; 97: 1133–676 Locatelli F, Bommer J, London GM, et al. Cardiovascular disease

determinants in chronic renal failure: clinical approach and treat-ment. Nephrol Dial Transplant 2001; 16: 459–68

77 London GM. Cardiovascular disease in chronic renal failure:pathophysiologic aspects. Semin Dial 2003; 16: 85–94

78 Lynam DP, Cronnelly R, Castagnoli KP, et al. The pharmacody-namics and pharmacokinetics of vecuronium in patients anesthe-tized with isoflurane with normal renal function or with renalfailure. Anesthesiology 1988; 69: 227–31

79 Macdonald J, Marcora S, Jibani M, et al. GFR estimation using

cystatin C is not independent of body composition. Am J KidneyDis 2006; 48: 712–9

80 Manjunath G, Sarnak MJ, Levey AS. Estimating the glomerular fil-tration rate. Dos and don’ts for assessing kidney function.Postgrad Med 2001; 110: 55–62

81 Manns BJ, Burgess ED, Parsons HG, Schaefer JP, Hyndman ME,Scott-Douglas NW. Hyperhomocysteinemia, anticardiolipin anti-body status, and risk for vascular access thrombosis in hemodia-lysis patients. Kidney Int 1999; 55: 315–20

82 Massry SG. Is parathyroid hormone a uremic toxin? Nephron

1977; 19: 125–3083 Matzke GR, Chan GL, Abraham PA. Codeine dosage in renal

failure. Clin Pharm 1986; 5: 15–684 Mazze RI, Callan CM, Galvez ST, Delgado-Herrera L, Mayer DB.

The effects of sevoflurane on serum creatinine and blood ureanitrogen concentrations: a retrospective, twenty-two-center,comparative evaluation of renal function in adult surgicalpatients. Anesth Analg 2000; 90: 683–8

85 Mazze RI, Calverley RK, Smith NT. Inorganic fluoride nephro-

toxicity: prolonged enflurane and halothane anesthesia in volun-teers. Anesthesiology 1977; 46: 265–71

86 McCauley J, Murray J, Jordan M, Scantlebury V, Vivas C,Shapiro R. Labetalol-induced hyperkalemia in renal transplantrecipients. Am J Nephrol 2002; 22: 347–51

87 McLeod K, Watson MJ, Rawlins MD. Pharmacokinetics of pan-curonium in patients with normal and impaired renal function.Br J Anaesth 1976; 48: 341–5

Craig and Hunter

308

88 McNamara PJ, Slaughter RL, Pieper JA, Wyman MG, Lalka D.Factors influencing serum protein binding of lidocaine inhumans. Anesth Analg 1981; 60: 395–400

89 Meguid El Nahas A, Bello AK. Chronic kidney disease: the

global challenge. Lancet 2005; 365: 331–4090 Melendez JA, Delphin E, Lamb J, Rose E. Noncardiac surgery in

heart transplant recipients in the cyclosporine era. J CardiothoracVasc Anesth 1991; 5: 218–20

91 Mirakhur RK. Safety aspects of non-depolarizing neuromuscularblocking agents with special reference to rocuronium bromide.

Eur J Anaesthesiol Suppl 1994; 9: 133–4092 Molino D, De Lucia D, Marotta R, et al. In uremia, plasma levels

of anti-protein C and anti-protein S antibodies are associatedwith thrombosis. Kidney Int 2005; 68: 1223–9

93 Mukherjee D, Nissen SE, Topol EJ. Risk of cardiovascular events

associated with selective COX-2 inhibitors. J Am Med Assoc2001; 286: 954–9

94 Murphy EJ. Acute pain management pharmacology for thepatient with concurrent renal or hepatic disease. AnaesthIntensive Care 2005; 33: 311–22

95 Naguib M. Sugammadex: another milestone in clinical neuro-muscular pharmacology. Anesth Analg 2007; 104: 575–81

96 Nakao N, Yoshimura A, Morita H, Takada M, Kayano T, IdeuraT. Combination treatment of angiotensin-II receptor blocker andangiotensin-converting-enzyme inhibitor in non-diabetic renal

disease (COOPERATE): a randomised controlled trial. Lancet2003; 361: 117–24

97 Nancarrow C, Runciman WB, Mather LE, Upton RN, PlummerJL. The influence of acidosis on the distribution of lidocaine and

bupivacaine into the myocardium and brain of the sheep. AnesthAnalg 1987; 66: 925–35

98 Nashef SA, Roques F, Michel P, Gauducheau E, Lemeshow S,Salamon R. European system for cardiac operative risk evalu-ation (EuroSCORE). Eur J Cardiothorac Surg 1999; 16: 9–13

99 Nishiyama T, Aibiki M, Hanaoka K. Inorganic fluoride kineticsand renal tubular function after sevoflurane anesthesia inchronic renal failure patients receiving hemodialysis. AnesthAnalg 1996; 83: 574–7

100 NKF K/DOQI Guidelines, Clinical Practice Guidelines For Vascular

Access. Available from http://www.kidney.org/professionals/kdoqi/guideline_upHD_PD_VA/va_intro.htm

101 Nolin TD, Frye RF, Matzke GR. Hepatic drug metabolism andtransport in patients with kidney disease. Am J Kidney Dis 2003;42: 906–25

102 Novis BK, Roizen MF, Aronson S, Thisted RA. Association ofpreoperative risk factors with postoperative acute renal failure.

Anesth Analg 1994; 78: 143–9103 Obrador GT, Pereira BJ. Systemic complications of chronic

kidney disease. Pinpointing clinical manifestations and best man-agement. Postgrad Med 2002; 111: 115–22

104 Palevsky PM. Perioperative management of patients withchronic kidney disease or ESRD. Best Pract Res Clin Anaesthesiol2004; 18: 129–44

105 Parekh N. Hyperchloremic acidosis. Anesth Analg 2002; 95:1821–2

106 Pecoits-Filho R, Goncalves S, Barberato SH, et al. Impact ofresidual renal function on volume status in chronic renal failure.Blood Purif 2004; 22: 285–92

107 Perazella MA. Drug-induced hyperkalemia: old culprits and newoffenders. Am J Med 2000; 109: 307–14

108 Phillips BJ, Hunter JM. Use of mivacurium chloride byconstant infusion in the anephric patient. Br J Anaesth 1992; 68:492–8

109 Pivalizza EG, Abramson DC, Harvey A. Perioperative hypercoa-gulability in uremic patients: a viscoelastic study. J Clin Anesth1997; 9: 442–5

110 Polkinghorne KR, McDonald SP, Atkins RC, Kerr PG. Vascularaccess and all-cause mortality: a propensity score analysis. J AmSoc Nephrol 2004; 15: 477–86

111 Rabelink TJ, Zwaginga JJ, Koomans HA, Sixma JJ. Thrombosisand hemostasis in renal disease. Kidney Int 1994; 46: 287–96

112 Raine AE. Hypertension, blood viscosity, and cardiovascularmorbidity in renal failure: implications of erythropoietin therapy.

Lancet 1988; 1: 97–100113 Robertson EN, Driessen JJ, Booij LH. Pharmacokinetics and

pharmacodynamics of rocuronium in patients with and withoutrenal failure. Eur J Anaesthesiol 2005; 22: 4–10

114 Roderick P, Willis NS, Blakeley S, Jones C, Tomson C.

Correction of chronic metabolic acidosis for chronic kidneydisease patients. Cochrane Database Syst Rev 2007; CD001890

115 Rokaw MD, West ME, Palevsky PM, Johnson JP. FK-506 andrapamycin but not cyclosporin inhibit aldosterone-stimulatedsodium transport in A6 cells. Am J Physiol 1996; 271:

C194–202116 Sagripanti A, Cupisti A, Baicchi U, Ferdeghini M, Morelli E,

Barsotti G. Plasma parameters of the prothrombotic state inchronic uremia. Nephron 1993; 63: 273–8

117 Sahin M, Kayatas M, Urun Y, Sennaroglu E, Akdur S. Performingonly one cardiovascular reflex test has a high positive predictivevalue for diagnosing autonomic neuropathy in patients with

chronic renal failure on hemodialysis. Ren Fail 2006; 28: 383–7118 Sanya EO, Ogunniyi A. Cardiovascular autonomic neuropathy in

non-diabetic Nigerian patients with chronic renal failure. WestAfr J Med 2004; 23: 15–20

119 Schiffrin EL, Lipman ML, Mann JF. Chronic kidney disease: effectson the cardiovascular system. Circulation 2007; 116: 85–97

120 Schortgen F, Lacherade JC, Bruneel F, et al. Effects of hydro-xyethylstarch and gelatin on renal function in severe sepsis: amulticentre randomised study. Lancet 2001; 357: 911–6

121 Schwab SJ, Quarles LD, Middleton JP, Cohan RH, Saeed M,Dennis VW. Hemodialysis-associated subclavian vein stenosis.Kidney Int 1988; 33: 1156–9

122 Sear JW. Drug handling in renal impairment. Curr Anaesth CritCare 1992; 3: 133–9

123 Sear JW, Hand CW, Moore RA, McQuay HJ. Studies on mor-phine disposition: influence of renal failure on the kinetics ofmorphine and its metabolites. Br J Anaesth 1989; 62: 28–32

124 Shahlaie K, Fox A, Butani L, Boggan JE. Spontaneous epiduralhemorrhage in chronic renal failure. A case report and review.

Pediatr Nephrol 2004; 19: 1168–72125 Sharrock NE, Beksac B, Flynn E, Go G, Della Valle AG.

Hypotensive epidural anaesthesia in patients with preoperativerenal dysfunction undergoing total hip replacement. Br J Anaesth

2006; 96: 207–12126 Shemesh O, Golbetz H, Kriss JP, Myers BD. Limitations of crea-

tinine as a filtration marker in glomerulopathic patients. KidneyInt 1985; 28: 830–8

127 Shemin D, Bostom AG, Laliberty P, Dworkin LD. Residual renal

function and mortality risk in hemodialysis patients. Am J KidneyDis 2001; 38: 85–90

128 Sidi A, Kaplan RF, Davis RF. Prolonged neuromuscular blockadeand ventilatory failure after renal transplantation and cyclospor-ine. Can J Anaesth 1990; 37: 543–8

129 Sladen RN, Endo E, Harrison T. Two-hour versus 22-hour crea-tinine clearance in critically ill patients. Anesthesiology 1987; 67:1013–6


309

130 Song IS, Yang WS, Kim SB, Lee JH, Kwon TW, Park JS.Association of plasma fibrinogen concentration with vascularaccess failure in hemodialysis patients. Nephrol Dial Transplant1999; 14: 137–41

131 Sorgenfrei IF, Norrild K, Larsen PB, et al. Reversal ofrocuronium-induced neuromuscular block by the selectiverelaxant binding agent sugammadex: a dose-finding and safetystudy. Anesthesiology 2006; 104: 667–74

132 Spallone V, Mazzaferro S, Tomei E, et al. Autonomic neuropathyand secondary hyperparathyroidism in uraemia. Nephrol DialTransplant 1990; 5: 128–30

133 Thapa S, Brull SJ. Succinylcholine-induced hyperkalemia inpatients with renal failure: an old question revisited. Anesth

Analg 2000; 91: 237–41134 Toivonen HJ. Anaesthesia for patients with a transplanted organ.

Acta Anaesthesiol Scand 2000; 44: 812–33135 Tomura S, Nakamura Y, Deguchi F, Ando R, Chida Y, Marumo F.

Coagulation and fibrinolysis in patients with chronic renal

failure undergoing conservative treatment. Thromb Res 1991; 64:81–90

136 Turney JH, Woods HF, Fewell MR, Weston MJ. Factor VIIIcomplex in uraemia and effects of haemodialysis. Br Med J (ClinRes Ed) 1981; 282: 1653–6

137 US Renal Data System. USRDS 2007 Annual Data Report: Atlasof Chronic Kidney Disease and End-Stage Renal Disease in theUnited States. Bethesda, MD: National Institutes of Health,National Institute of Diabetes and Digestive and Kidney

Diseases, 2007138 Van Peer A, Vercauteren M, Noorduin H, Woestenborghs R,

Heykants J. Alfentanil kinetics in renal insufficiency. Eur J ClinPharmacol 1986; 30: 245–7

139 Vanacker BF, Vermeyen KM, Struys MM, et al. Reversal of

rocuronium-induced neuromuscular block with the noveldrug sugammadex is equally effective under maintenance

anesthesia with propofol or sevoflurane. Anesth Analg 2007;104: 563–8

140 Wallia R, Greenberg A, Piraino B, Mitro R, Puschett JB. Serumelectrolyte patterns in end-stage renal disease. Am J Kidney Dis

1986; 8: 98–104141 Ward S, Boheimer N, Weatherley BC, Simmonds RJ, Dopson

TA. Pharmacokinetics of atracurium and its metabolites inpatients with normal renal function, and in patients in renal

failure. Br J Anaesth 1987; 59: 697–706142 Webb MD. Type I second-degree AV block after neostigmine

administration in a child with renal failure. Anesth Prog 1995; 42:21–2

143 White C, Akbari A, Hussain N, et al. Chronic kidney disease

stage in renal transplantation classification using cystatin C andcreatinine-based equations. Nephrol Dial Transplant 2007; 22:3013–20

144 Wickstrom I. Enflurane anesthesia in living donor renal trans-plantation. Acta Anaesthesiol Scand 1981; 25: 263–9

145 Wierda JM, Kleef UW, Lambalk LM, Kloppenburg WD,Agoston S. The pharmacodynamics and pharmacokinetics ofOrg 9426, a new non-depolarizing neuromuscular blockingagent, in patients anaesthetized with nitrous oxide, halothaneand fentanyl. Can J Anaesth 1991; 38: 430–5

146 Wilcox CS. Regulation of renal blood flow by plasma chloride.J Clin Invest 1983; 71: 726–35

147 Williams B, Poulter NR, Brown MJ, et al. Guidelines for manage-ment of hypertension: report of the fourth working party of

the British Hypertension Society, 2004-BHS IV. J Hum Hypertens2004; 18: 139–85

148 Zaleski L, Abello D, Gold MI. Desflurane versus isoflurane inpatients with chronic hepatic and renal disease. Anesth Analg1993; 76: 353–6

149 Zand MS. Immunosuppression and immune monitoring afterrenal transplantation. Semin Dial 2005; 18: 511–9

Craig and Hunter

310

FCA(SA) Part II

THE COLLEGES OF MEDICINE OF SOUTH AFRICA Incorporated Association not for gain

Reg No 1955/000003/08

Final Examination for the Fellowship of the College of Anaesthetists of South Africa

16 March 2010

Paper I (3 hours) All questions are to be answered. Each question to be answered in a separate book (or books if more than one is required for the one answer) Al die vrae moet beantwoord word. Elke vraag moet in ’n aparte boek (of boeke indien meer as een nodig is vir ’n vraag) geskryf word 1 Write detailed notes on the complete perioperative management of an acute ascending

aorta dissection (including anaesthesia). [100] 1 Skryf gedetailleerde notas oor die volledige perioperatiewe hantering van ’n akute

stygende aorta disseksie (narkose ingesluit). [100] 2 A 45-year-old female with a body mass index (BMI) of 32 presents for an elective

laparoscopic cholecystectomy a) Describe your pre-operative assessment and recommendations. (30) b) Write short notes about the possible intra- and post-operative complications. (30) c) What anaesthetic technique would you employ? (40)

[100]

2 ’n Vyf-en-veertigjarige vrou met ligaamsmassa indeks (LMI) van 32 presenteer vir elektiewe laparoskopiese cholesistektomie a) Beskryf u pre-operatiewe evaluasie en aanbevelings. (30) b) Skryf kort notas oor die moontlike intra- en post-operatiewe komplikasies. (30) c) Watter narkosetegniek sal u gebruik? (40)

[100] 3 Discuss the management of a 6-year-old HIV positive child, on anti-retroviral therapy,

under the following headings a) Potential peri-operative drug interactions. b) The management of pain during the peri-operative period. c) Risks of infection. [100]

3 Bespreek die hantering van ’n 6-jaar-oue MIV positiewe kind, op anti-retrovirale terapie,

onder die volgende hoofde a) Potensiële peri-operatiewe geneesmiddel interaksies.

PTO/ Page 2 Question 3 ii)...

-2-

b) Die hantering van pyn in die peri-operatiewe periode. c) Risiko’s vir infeksie. [100]

4 A 40-year-old man is scheduled to have a living donor renal transplant. He has been on

haemodialysis for 5-years. Discuss briefly a) The more likely possible causes of end-stage renal disease in this patient. (5) b) Common medical problems secondary to end-stage renal disease that may impact

on the anaesthetic management of the patient. (20) c) The anaesthetic management and monitoring of the recipient. (40) d) The anaesthetic management and monitoring of the living donor. (20) e) Post-operative pain management in the recipient and donor. (15)

[100]

4 ’n 40-Jarige man is geskeduleer om ’n lewende donor nieroorplanting te ondergaan. Hy is al vir 5-jare op hemodialise. Bespreek kortliks a) Die meer waarskynlike moontlike oorsake vir eindstadium nierversaking in hierdie

pasiënt. (5) b) Algemene mediese probleme sekondêr tot eindstadium nierversaking wat ’n invloed

mag hê op die narkosehantering van die pasiënt. (20) c) Die narkosehantering en monitering van die ontvanger. (40) d) Die narkosehantering en monitering van die lewende skenker. (20) e) Post-operatiewe pynhantering by die ontvanger en skenker. (15)

[100]

FCA(SA) Part II

THE COLLEGES OF MEDICINE OF SOUTH AFRICA Incorporated Association not for gain

Reg No 1955/000003/08

Final Examination for the Fellowship of the College of Anaesthetists of South Africa

17 March 2010

Paper II (3 hours) All questions are to be answered. Each question to be answered in a separate book (or books if more than one is required for the one answer) Al die vrae moet beantwoord word. Elke vraag moet in ’n aparte boek (of boeke indien meer as een nodig is vir ’n vraag) geskryf word 1 Discuss CO2 under the following headings

a) Pathophysiological effects of an abnormal pCO2. (30) b) Useful applications of CO2 gas in theatre. (20) c) Permissive hypercarbia. (50)

[100] 1 Bespreek CO2 onder die volgende hoofde

a) Die patofisiologiese effekte van ’n abnormale pCO2. (30) b) Bruikbare toepassings van CO2 gas in die teater. (20) c) Toegelate hiperkarbie. (50)

[100] 2 a) What are the risks involved in performing a lumbar epidural on a labouring

parturient, and how can these risks be prevented? (40) b) How may an epidural for labour be converted to anaesthesia for caesarean

section? (20) c) Why does central neuraxial anaesthesia sometimes fail to work, and how can this

failure be prevented? (40) [100]

2 a) Wat is die risiko’s betrokke in die uitvoer van ’n lumbale epiduraal op ’n pasiënt in

kraam, en hoe kan hierdie risiko’s voorkom word? (40) b) Hoe kan ’n epiduraal vir kraam verander word in narkose vir keisersnit? (20) c) Hoekom werk sentrale neuraksiale narkose soms nie, en hoe kan hierdie

mislukking voorkom word? (40) [100]

-2- 3 A patient presents to your hospital with 60% lower limb and torso burns.

Describe your management of the following a) Initial resuscitation. (40) b) Anaesthesia for initial escharotomy . (20) c) Sedation for dressing changes. (20) d) Anaesthesia for skin grafting . (20)

[100] 3 ’n Pasiënt presenteer by jou hospitaal met 60% brandwonde van die onderbene en

torso. Beskryf jou hantering van die volgende a) Aanvanklike resussitasie. (40) b) Narkose vir aanvanklike eskarotomie. (20) c) Sedasie vir verband omruiling. (20) d) Narkose vir veloorplanting. (20)

[100] 4 Patients with rheumatoid arthritis pose several challenges to the anaesthesiologist.

Discuss. [100] 4 Pasiënte met rheumatoiede arthritis bied verskeie uitdagings aan die anestesioloog.

Bespreek. [100]

The Colleges of Medicine of South Africa Incorporated Association not for gain (Reg. No. 1955/000003/08)

Nonprofit Organisation (Reg No 009-874 NPO) 27 Rhodes Ave, PARKTOWN WEST 2193

Private Bag X23, BRAAMFONTEIN 2017 Tel: +27 11 726-7037/8/9

Fax: +27 11 726-4036 General: [email protected]

Academic Registrar: [email protected] Website: http://www.collegemedsa.ac.za

FCA(SA) Part II

DATA INTERPRETATION – PAPER III

Question 1 - 2

18 MARCH 2010

Time: 3 hours

CANDIDATE NUMBER……………………

Question 1 / Vraag 1 A 65-year-old woman is booked for an elective cholecystectomy. On history the patient complains of neck pain that became worse since she had a car accident 4-years ago. On examination she has a limited range of neck movement. The x-ray below (Figure 1) was taken pre-operatively. ’n 65-Jarige vrou is bespreek vir elektiewe cholesistektomie. Sy gee geskiedenis van nekpyn wat erger geword het sedert ’n motorongeluk 4-jaar gelede. By ondersoek het sy ’n beperkte nekbeweging. Die x-straal hieronder (Figuur 1) is pre-operatief geneem. a) Describe the findings on the neck x-ray.

Beskryf die bevindinge op hierdie x-foto. (2)

.............................................................................................................................................. .............................................................................................................................................. .............................................................................................................................................. .............................................................................................................................................. b) What are the special anaesthetic considerations with regard to the findings on this x-

ray? Wat is die special narkoseoorwegings rakende die x-straal bevindinge? (3)

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

.. c) Would you like further investigations of the neck complaint before the procedure? Why?

Sou u verdere ondersoeke rakende die nek klagte wou hê voor die prosedure? Hoekom?(2)

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

d) If you decide to anaesthetise the patient what would be your anaesthetic technique? As u besluit om die pasiënt narkose te gee, wat sou u narkosetegniek wees? (3)

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

[10] Question 2 / Vraag 2 A 70-year-old woman has fractured her hip 5-days ago. She presents to theatre for a total hip replacement. On examination her pulse rate is 95 beats/min. Figure 2 below is the patient’s ECG. ’n 70-Jarige vrou het haar heup 5-dae gelede fraktuur. Sy presenteer in teater vir ’n totale heupvervanging. By ondersoek is haar polstempo 95 slae/minuut. Figuur 2 hieronder is die pasiënt se EKG. a) Interpret the ECG.

Interpreteer die EKG. (6)

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................ b) What would be your next investigation and what do you expect to find?

Wat sou u volgende ondersoek wees en wat verwag u om te vind? (4)

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

............................................................................................................................................

..

[10]






FCA(SA) Part II


Question 3 - 5

18 MARCH 2010

Time: 3 hours

CANDIDATE NUMBER……………………

Question 3 / Vraag 3 A fifty-year-old patient develops sudden hypotension following manipulation of a long bone fracture under general anaesthesia ’n Vyftigjarige pasiënt ontwikkel skielik hipotensie na manipulasie van ’n langbeen fraktuur onder algemene narkose a) What are your differential diagnoses of this hypotension?

Wat is u differensiële diagnose van hierdie hipotensie?

i) ...............................................................................................................................

ii) ...............................................................................................................................

½ mark each (total 1 mark) b) How would your monitoring of this patient help with the diagnosis?

Hoe sou u monitering van hierdie pasiënt help met diagnose?

i) ...............................................................................................................................

ii) ...............................................................................................................................

iii) ...............................................................................................................................

iv) ...............................................................................................................................

v) ...............................................................................................................................

(5) c) Outline principles of management of this patient’s hypotension.

Omlyn beginsels van hantering van hierdie pasiënt se hipotensie

i) ...............................................................................................................................

ii) ...............................................................................................................................

iii) ...............................................................................................................................

iv) ...............................................................................................................................

(4) [10]

Question 4 / Vraag 4 A 25-year-old para 2 gravida 3 with an atrial septal defect measuring 24 mm, severe pulmonary hypertension, and an ejection fraction of 62% is not cyanosed,nor is she in heart failure. She is scheduled for surgical delivery. ’n 25-Jarige para 2 gravida 3 met atriale septale defek van 24 mm deursnee, erge pulmonale hipertensie, en ejeksiefraksie van 62 % is nie sianoties of in hartversaking nie. Sy is geskeduleer vir chirurgiese verlossing.

a) What problems can you expect during anaesthesia? Watter probleme kan u verwag tydens narkose? (4) ........................................................................................................................................

........................................................................................................................................

........................................................................................................................................

........................................................................................................................................

........................................................................................................................................

........................................................................................................................................

........................................................................................................................................

b) Outline your anaesthetic plan for this patient.

Omlyn u narkoseplan vir hierdie pasiënt. (6) ........................................................................................................................................

........................................................................................................................................

........................................................................................................................................

........................................................................................................................................

........................................................................................................................................

........................................................................................................................................

........................................................................................................................................

........................................................................................................................................

........................................................................................................................................

[10] Question 5 / Vraag 5 Outline the electrolyte abnormalities in the following gastro-intestinal tract obstructions Verskaf die elektrolietafwykings in die volgende gastrointestinale traktus obstruksies a) Gastric outlet obstruction.

Maaguitgang obstruksie. i) ...............................................................................................................................

ii) ...............................................................................................................................

iii) ...............................................................................................................................

iv) ...............................................................................................................................

v) ...............................................................................................................................

½ mark each (total 2 ½ )

b) Duodenal atresia. Duodenale atresie.

i) ...............................................................................................................................

ii) ...............................................................................................................................

iii) ...............................................................................................................................

iv) ...............................................................................................................................

v) ...............................................................................................................................

½ mark each (2 ½ marks) c) Intersussception of the small bowel.

Intussusepsie van die dunderm.

i) ...............................................................................................................................

ii) ...............................................................................................................................

iii) ...............................................................................................................................

iv) ...............................................................................................................................

v) ...............................................................................................................................

½ mark each (total 2 ½ marks)

d) Large bowel volvulus. Dikderm volvulus.

i) ...............................................................................................................................

ii) ...............................................................................................................................

iii) ...............................................................................................................................

iv) ...............................................................................................................................

v) ...............................................................................................................................

½ mark each (total 2 ½ marks) [10]






FCA(SA) Part II


Question 6 - 8

18 MARCH 2010

Time: 3 hours CANDIDATE NUMBER……………………

Question 6 / Vraag 6 The trends of heart rate and blood pressure in Figure 3 were taken from a healthy 30-year-old man who underwent a laparoscopic Nissen fundoplication Die opname oor tyd van hart-tempo en bloedruk in Figuur 3 is geneem van ’n 30-jarige gesonde man wat ’n Nissen fundoplikasie ondergaan het

a) Explain the haemodynamic changes seen between the two arrows.

Verduidlik die hemodinaimiese veranderings wat tussen die twee pyltjies plaasvind. (4)

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

b) How would you have responded to these changes?

Hoe sou jy hierdie veranderings behandel het? (6) ..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

[10]

Question 7/ Vraag 7 A 25-year-old female presents with respiratory distress and a temperature of 39oC. She is intubated in the emergency room and the chest x-ray (Figure 4) is taken ’n 25-Jarige vrou presenteer met respiratorise nood en ’n koors van 39oC. Sy is in die nood-eenheid geintubeer en die bors x-straal (Figuur 4) is geneem a) What is the most likely diagnosis?

Wat is die mees waarskynlike diagnose? (2) ..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

b) Describe your initial ventilatory management in the intensive care unit.

Beskryf jou aanvanklike ventilatoriese behandeling in die waakeenheid. (8) ..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

[10]

Question 8 / Vraag 8 a) Describe the feature of note in the flow-time loop below (Figure 5).

Beskryf die kenmerk in die onderstaande vloei-tyd lus (Figuur 5). (4)

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

b) Which three ventilator settings could have been changed to achieve the flow-time loop

below (Figure 6)? Watter drie ventilator stelling mag verander word om die lus onder (figuur 6) te bereik? (6) ..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

[10]






FCA(SA) Part II


Question 9 - 11

18 MARCH 2010


Question 9 / Vraag 9 A 35-year-old man presents with diarrhoea and asthma. Clinical examination reveals facial telangectasia and a systolic murmur loudest in the left second intercostal space. ’n 35-Jarige man presenteer met diaree en asma. Kliniese ondersoek toon telangiektase van die gesig en ’n sistoliese geruis, hardste in die linker interkostale spasie.

a) What syndrome should be suspected?

Watter sindroom moet vermoed word? (1)

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

b) What tests should be performed to confirm this diagnosis?

Watter toetse behoort gedoen te word om hierdie diagnose te bevestig? (3)

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

c) How should this patient be managed in order to prevent the potential adverse effects of

this syndrome intraoperatively? Hoe moet die pasiënt hanteer word ten einde die potensiële newe-effekte van die sindroom intra-operatief te hanteer? (6)

……..................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

..........................................................................................................................................

[10]

Question 10 / Vraag 10 The man shown in this drawing (Figure 7) presented with headaches and visual disturbance. Die man op die skets (Figuur 7) het presenteer met hoofpyn en gesigsteurnis. a) What is his most likely medical condition?

Wat is sy mees waarskynlike mediese toestand? (1)

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

His pre-operative blood tests were normal. Surgery to treat his condition is performed and post-operatively his blood tests are Sy pre-operatiewe bloedtoetse was normaal. Chirurgie vir sy toestand word gedoen en post-operatief is sy bloeduitslae Sodium/natrium 156 mmol/l Potassium/kalium 4.2 mmol/l Urea/ureum 9.8 mmol/l Bicarbonate/bicarbonaat 28 mmol/l Plasma osmolality/plasma osmolaliteit 330/mosmol/kg H2O. b) What post-operative complication do you suspect has occurred and how would you

treat it? Watter post-operatiewe komplikasies het plaasgevind en hoe sal u dit behandel? (9) ………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

[10]

Question 11 / Vraag 11 A 76-year-old lady attends the pre-operative anaesthetic clinic. She has been booked for a total hip replacement in two-weeks time. She does not have any cardiac symptoms although she can only walk 10m due to hip pain. This is her ECG (Figure 8) ’n 76-Jarige dame besoek die pre-operatiewe narkosekliniek. Sy is bespreek vir ’n totale heupvervanging binne 2-weke. Sy het geen kardiale simptome maar kan slegs 10 meter loop weens heuppyn. Hierdie is haar EKG (Figuur 8)

a) What is the abnormality and what are the possible causes?

Wat is die abnormaliteit en wat is die moontlike oorsake? (4)

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

b) How does this affect your peri-operative management of this patient?

Hoe beïnvloed dit u peri-operatiewe hantering van hierdie pasiënt? (6)

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

………………………………………………………………………………………………….....

[10]






FCA(SA) Part II


Question 12 - 14

18 MARCH 2010


Question 12 / Vraag 12 A surgery intern asks your opinion about a serum potassium of 6.0 mmol/l in a patient booked for anesthesia. ’n Intern by chirurgie vra u opinie oor ’n serum kalium van 6.0 mmol/l in ’n pasiënt wat bespreek word vir narkose. a) What other measurement(s) from the blood investigations would you ask about?

Watter ander bloed ondersoek bevinding(e) sou u oor navraag doen? (2)

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

b) List important causes of hyperkalaemia.

Maak ’n lys van die belangrike oorsake van hiperkalemie. (5)

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

c) Describe your immediate treatment of a patient with symptomatic hyperkalaemia.

Beskryf u onmiddellike behandeling van ’n pasiënt met simptomatiese hiperkalemie.(3)

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

[10]

Question 13 / Vraag 13 You administer heparin 7000 IU intravenously to an adult weighing 70kg for aorta- bifemoral bypass. His activated clotting time after three minutes is 110 seconds. His pre-induction ACT was 100 seconds. U dien heparien 7000 IE intraveneus toe aan ’n 70kg volwassene wat aorta-bifemorale omleiding ondergaan. Die geaktiveerde stol tyd drie minute later is 110 sekondes. Dit was voor induksie 100 sekondes. a) List the causes of apparent heparin resistance.

Lys oorsake van skynbare heparien weerstandigheid. (5)

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

b) Briefly discuss your management.

Beskryf u hantering kortliks. (5)

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

[10]

Question 14 / Vraag 14 You ventilate a relaxed, intubated child of 30 kg using your anaesthetic ventilator. You have real-time spirometry on your display (Figure 9). U ventileer ’n geïntubeerde, verslapte kind van 30kg met u narkose ventilator. U narkose monitor het ’n intydse spirometrie vertoon (Figuur 9).

a) Comment on the spirometry curve.

Lewer kommentaar op die spirometrie kurwe. (2)

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

b) What causes would you consider?

Watter oorsake sal u in ag neem? (3)

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

c) What are the risks to the patient?

Wat is die gevare vir die pasïent? (3)

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

d) In the absence of other reversible causes, how would you adjust your ventilator? Hoe sal u die ventilator verstel nadat u ander omkeerbare oorsake uitgeskakel het? (2)

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

…………………………………………………………………………………………………….

[10]






FCA(SA) Part II

DATA INTERPRETATION

Question 15 - 17

18 MARCH 2010


Question 15 / Vraag 15 A 55-year-old woman is brought to the casualty unit by ambulance. She is semi-comatose and ill for several days and has a past history of heart failure. Current medication is digoxin and a thiazide diuretic. An arterial blood gas is obtained whilst breathing room air ’n 55-jarige vrou word per ambulans ingebring na die ongevalle-eenheid. Sy is semi-komateus, siek vir verskeie dae en het ook ’n geskiedenis van hartversaking. Huidige medikasie is digoksien en ’n tiasied-diuretikum. ’n Arteriële bloedgas word gedoen op kamerlug pH = 7.41 PCO2 = 32 mmHg PO2 = 85 mmHg HCO3

- = 19 mmol/l Anion gap = 33 mmol/l S-glucose = 67 mmol/l S-K+ = 2.7 mmol/l a) Give 2 likely causes for the abnormal anion gap.

Gee 2 waarskynlike oorsake vir die abnormale anioongaping. (2)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

b) A pre-existing high S-HCO3

- is likely. Explain why? Calculate the likely pre-existing S-HCO3

-. Show your calculations. ’n Voorafgaande hoë S-HCO3

- is waarskynlik. Verduidelik hoekom? Bereken die waarskynlike voorafgaande S-HCO3

-. Wys u berekeninge. (4)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

c) What is the expected PaCO2 for a chronic S-HCO3

- = 19 mmol/l? Wat is die verwagte PaCO2 vir ’n chroniese S-HCO3

- = 19 mmol/l? (2)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

d) What conclusion can be drawn from your finding in (c)? Watter gevolgtrekking kan gemaak word uit u bevinding in (c)? (2)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

[10] Question 16 / Vraag 16 A nerve stimulator is applied over the ulnar nerve on the forearm of a patient with post-operative apnoea. The motor function of the adductor pollicis is monitored. Draw the motor response of this muscle schematically as found in the following possible scenarios ’n Senuweestimulator word geplaas oor die ulnare senuwee in die voorarm van ’n pasiënt met post-operatiewe apnee. Die motorfunksie van die adduktor pollicis word gemoniteer. Teken skematies die motorrespons van hierdie spier soos gevind met die volgende moontlike scenario’s a) A partial depolarising block when using a tetanic stimulus.

’n Gedeeltelike depolariserende blok wanneer ’n tetaniese stimulus toegedien word. (2)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

……………………………………………………………………………………………………… b) A partial nondepolarising block when using a train-of-four stimulation.

’n Gedeeltelike nie-depolariserende blok wanneer ’n rits-van-vier stimulasies gebruik word. (2)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

c) Testing for posttetanic facilitation in the presence of a complete depolarising block.

Toets vir posttetaniese potensiasie in die aanwesigheid van ’n volledige depolariserende blok. (2)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

d) Testing for posttetanic facilitation in the presence of a partial nondepolarising block. Toets vir posttetaniese potensiasie in die teenwoordigheid van ’n gedeeltelike nie-depolariserende blok. (2)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

e) Applying a tetanic stimulus in the absence of a neuromuscular block in a man with

myasthenic syndrome. Toediening van ’n tetaniese stimulus in die afwesigheid van ’n neuromuskulêre blok by ’n man met miasteniese sindroom. (2)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

[10] Question 17 / Vraag 17 a) A neonate is breathing spontaneously with a fast respiratory rate utilising the Jackson-Rees

system. The fresh gas flow is 5 litres per minute. The capnograph side stream aspiration tube is connected to the paediatric airway filter on to the distal end of the endotracheal tube.

’n Neonaat haal spontaan asem met ’n Jackson-Reessisteem teen ’n vinnige respiratoriese tempo. Die varsgasvloei is 5 liter per minuut. Die kapnograaf-systroomaspirasiepypie is gekonnekteer aan die pediatriese lugwegfilter aan die distale einde van die endotrageale buis.

i) Identify two seemingly obvious abnormalities identifiable from the capnogram tracing in Figure 10. Identifiseer twee oënskynlik duidelike abnormaliteite identifiseerbaar op die kapnogramuitdruk in Figuur 10 (1)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

ii) Describe the interpretational difficulties associated with a capnogram lacking a plateau phase. Beskryf die interpretasieprobleme geassosieer met ’n kapnogram sonder ’n platofase. (2)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

iii) Describe the interpretational difficulties associated with a capnogram at such a fast respiratory rate. Beskryf die interpretasieprobeleme geassosieer met ’n kapnogram waar die respiratoriese tempo so vinnig is. (2)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

b) Volume control ventilation is applied in an adequately paralyzed young female during a

laparotomy for a ruptured ectopic pregnancy. The tidal volume is 400 ml and the ventilatory rate is 12 per minute. Her blood pressure is 65/35 when an arterial bloodgas determination is done. The PaCO2 is 68 mmHg although the ETCO2 recording at the moment of blood sampling read 48 mmHg. Volume-beheerventilasie word toegepas gedurende ’n laparotomie vir ’n geruptuurde ektopiese swangerskap by ’n voldoende verslapte jong vrou. Die getyvolume is 400 ml en die ventilatoriese tempo is 12 per minuut. Haar bloeddruk is 65/35 wanneer ’n arteriële bloedgasanalise gedoen word. Die PaCO2 is 68 mmHg alhoewel die ETCO2-lesing op die oomblik van die bloedmonsterneming 48 mmHg was.

i) What PaCO2-ETCO2 gradient is considered to be acceptable during ideal physiological circumstances? Watter PaCO2-ETCO2-gradiënt word as aanvaarbaar beskou gedurende idiale fisiologies omstandighede? (1)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

ii) Explain the possible cause of the abnormal PaCO2-ETCO2 gradient in the mentioned patient.

Verduidelik die moontlike oorsaak van die abnormale PaCO2-ETCO2-gradiënt in die genoemde geval. (2)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

iii) Under which unique circumstance can the PaCO2-ETCO2 gradient be reversed for a limited period (negative gradient, not necessarily related to the above case)? Onder watter unieke omstandighede kan die PaCO2-ETCO2-gradiënt omgekeerd wees vir ’n beperkte periode (negatiewe gradient, nie noodwendig verwant aan bogenoemde pasiënt nie)? (1)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

iv) Explain briefly how significant hypercarbia can contribute towards tissue hypoxia in this clinical scenario.

Verduidelik kortliks hoe betekenisvolle hiperkarbie in hierdie scenario kan bydra tot weefselhipoksie. (1)

………………………………………………………………………………………………………

………………………………………………………………………………………………………

………………………………………………………………………………………………………

[10]

T h e n e w e ngl a nd j o u r na l o f m e dic i n e

n engl j med 363;19 nejm.org november 4, 2010 1833

review article

Medical Progress

HemodialysisJonathan Himmelfarb, M.D., and T. Alp Ikizler, M.D.

From the Kidney Research Institute, De-partment of Medicine, Division of Nephrol-ogy, University of Washington, Seattle ( J.H.); and the Department of Medicine, Division of Nephrology, Vanderbilt Uni-versity Medical Center, Nashville (T.A.I.). Address reprint requests to Dr. Himmel-farb at the Kidney Research Institute, Box 359606, 325 9th Ave., Seattle, WA 98104, or at [email protected].

N Engl J Med 2010;363:1833-45.Copyright © 2010 Massachusetts Medical Society.

Fifty years ago, Belding Scribner and his colleagues at the Uni-versity of Washington developed a blood-access device using Teflon-coated plastic tubes, which facilitated the use of repeated hemodialysis as a life-sus-

taining treatment for patients with uremia.1,2 The introduction of the Scribner shunt, as it became known, soon led to the development of a variety of surgical techniques for the creation of arteriovenous fistulas and grafts. Consequently, hemodialysis has made survival possible for more than a million people throughout the world who have end-stage renal disease (ESRD) with limited or no kidney function. The expan-sion of dialysis into a form of long-term renal-replacement therapy transformed the field of nephrology and also created a new area of medical science, which has been called the physiology of the artificial kidney. This review describes the medical, social, and economic evolution of hemodialysis therapy.

G oa l s of Hemodi a lysis

Dialysis is defined as the diffusion of molecules in solution across a semipermeable membrane along an electrochemical concentration gradient.3 The primary goal of hemodialysis is to restore the intracellular and extracellular fluid environment that is characteristic of normal kidney function. This is accomplished by the transport of solutes such as urea from the blood into the dialysate and by the transport of solutes such as bicarbonate from the dialysate into the blood (Fig. 1A). Solute con-centration and molecular weight are the primary determinants of diffusion rates. Small molecules, such as urea, diffuse quickly, whereas compartmentalized and larger molecules, such as phosphate, β2-microglobulin, and albumin, and protein-bound solutes, such as p-cresol, diffuse much more slowly (Fig. 1B and 1C). In addi-tion to diffusion, solutes may pass through pores in the membrane by means of a convective process driven by hydrostatic or osmotic pressure gradients — a process called ultrafiltration.4 During ultrafiltration, there is no change in solute concentra-tions; its primary purpose is the removal of excess total body water.

For each dialysis session, the patient’s physiological status should be assessed so that the dialysis prescription can be aligned with the goals for the session. This is accomplished by integrating the separate but related components of the dialysis prescription to achieve the desired rates and total amount of solute and fluid re-moval (Table 1). By replacing kidney excretory function, dialysis is intended to elimi-nate the symptom complex known as the uremic syndrome, although ascribing particular cellular or organ dysfunction to the accumulation of specific solutes in uremia has proved to be difficult.5

Qua n tif y ing the D ose a nd A dequac y of Di a lysis

Measuring the clearance of solutes that accumulate in patients with uremia has be-come the mainstay for calculating the dose of dialysis and determining its adequa-

The New England Journal of Medicine Downloaded from www.nejm.org by RENI ROUNCIVELL on November 7, 2010. For personal use only. No other uses without permission.

Copyright © 2010 Massachusetts Medical Society. All rights reserved.


n engl j med 363;19 nejm.org november 4, 20101834

cy as delivered. Precise standards and goals of di-alysis adequacy are based on the clearance of urea, a byproduct of protein catabolism, which can be readily and accurately measured. The volume of distribution of urea, which is neither lipophilic nor highly protein-bound, reflects total body water; consequently, urea is an attractive molecule for quantifying dialysis adequacy through mathemat-ical modeling based on changing blood concen-trations.3,6 Urea kinetic modeling predicts mor-bidity and mortality better than kinetic modeling of any other known solute. The amount of urea to be removed is usually calculated according to the patient’s body size with the use of the following dimensionless construct, which relates the clear-ance of urea to its volume of distribution in the patient: Kt/Vurea, where K is the urea clearance of the dialyzer, t is the duration of dialysis, and Vurea is the patient’s volume of urea distribution. This construct has been readily adopted by the nephrol-ogy community to calculate the dialysis dose.6 Some investigators have suggested that adjusting the amount of solute clearance according to the volume of distribution rather than according to the patient’s body-surface area may result in un-derdosing in small patients and women.7-9 Al-though alternative means of adjusting clearance for body size have been proposed, none currently constitute the standard of care.

The importance of clearance of middle-molec-ular-weight solutes (500 to 30,000 daltons) with respect to clinical outcomes has long been debat-ed.10 Current high-flux hemodialysis membranes have larger pores than did earlier-generation mem-branes, and they permit the passage of larger uremic toxins. Since the β2-microglobulin concen-tration is easy to measure, it is frequently used as a marker solute for middle-molecular-weight sol-utes. Several retrospective, observational studies have suggested an association between the use of high-flux hemodialysis membranes and reduced mortality.11-14 However, increased clearance of middle-molecular-weight solutes has not been con-clusively shown to be an important factor in a well-powered, prospective, randomized trial.

Tr e atmen t Time

An important component of the dialysis prescrip-tion is treatment time, which can influence the ability to safely remove solutes and accumulated excess fluid. In the 1980s, shortening the treatment

time to cut costs while maintaining an adequate level of urea clearance became common practice in the United States. However, subsequent studies revealed that outcomes were adversely affected by shorter treatment times.15 Advocates for longer treatment times pointed to the better outcomes in Europe and Asia, where treatment times are pro-longed.16,17 Patients who gain more weight with dialysis are at increased risk for death,18 and a longer treatment time is often required for such patients to help maintain fluid balance. However, to date, little effort has been made to evaluate dif-ferent fluid-removal strategies in controlled studies.

Reports from individual centers that have used extended dialysis sessions (8 to 12 hours per treat-ment, often provided overnight) are receiving more attention. Extended treatment times clearly im-prove blood-pressure control and phosphate re-moval while having a modest effect on overall solute clearance.19,20 Excellent outcomes have been reported from these centers, although, again, not in the context of randomized clinical trials.21,22 It is unclear whether extended treatments provided at night are practical and would be accepted by most patients undergoing dialysis.

Fr equenc y of Di a lysis

For more than four decades, the standard sched-ule for hemodialysis has continued to be three sessions a week, largely owing to logistic and cost concerns. Although several centers have treated a small number of patients with more frequent he-modialysis, a systematic study of outcomes after such therapy is only now being undertaken. Most available reports are from case–control studies or uncontrolled interventional studies.23 A major-ity of such studies have shown reductions in blood-pressure levels and in the need for antihypertensive medications, with variable effects on regression of left ventricular hypertrophy, a frequent occurrence among patients receiving long-term hemodialysis. Health-related quality-of-life measures appear to improve with more frequent dialysis treatments, whereas mixed results are reported for measures of anemia control and calcium phosphate metabo-lism.24 A recent randomized, controlled pilot trial compared daily nocturnal hemodialysis with con-ventional thrice-weekly hemodialysis.25 In the pri-mary analysis, there was a significant reduction in left ventricular mass in the group treated with daily dialysis, as compared with the conventionally



Medical Progress


Blood Urea Concentration (%)

100

End

Star

tSt

art

Star

t

Dia

lysi

s

Star

t

Blo

odB

lood

Dia

lysa

teD

ialy

sate

End

A

Blood Phosphate (%)

100

End

Dia

lysi

s

Star

t

Blo

odB

lood

Dia

lysa

teD

ialy

sate

End

B

Blood P-Cresol Sulfate (%)

100

00

0En

d

Dia

lysi

s

Star

t

Blo

odB

lood

Dia

lysa

teD

ialy

sate

End

C

Blo

od c

ells

Phos

phat

eB

lood

cel

lsPr

otei

nP-

cres

ol s

ulfa

teU

rea

Blo

od c

ells

Bic

arbo

nate

Figu

re 1

. Pat

tern

s of

Sol

ute

Rem

oval

wit

h H

emod

ialy

sis.

Low

-mol

ecul

ar-w

eigh

t so

lute

s su

ch a

s ur

ea a

re r

eadi

ly d

ialy

zed

(Pan

el A

). T

he e

ffic

ienc

y of

tot

al u

rea

rem

oval

dec

lines

ove

r th

e co

urse

of t

reat

men

t ow

ing

to d

ecre

asin

g pl

asm

a ur

ea c

once

ntra

tion

s. A

fter

dia

lysi

s, t

here

is a

mod

est

rebo

und

in u

rea

leve

ls (

abou

t 15

%).

Sol

utes

suc

h as

bic

arbo

nate

als

o di

alyz

e re

adily

; how

ever

, the

dir

ecti

on o

f tra

nspo

rt is

fr

om t

he d

ialy

sate

to

the

bloo

d be

caus

e of

the

con

cent

rati

on d

iffe

renc

es t

hat

driv

e th

e di

ffus

ion

proc

ess.

For

com

part

men

taliz

ed s

olut

es, s

uch

as p

hosp

hate

, the

pla

sma

spac

e is

ra

pidl

y cl

eare

d (P

anel

B),

whe

reas

mos

t ph

osph

ate

in c

ells

and

bon

e re

mai

ns t

here

. Aft

er d

ialy

sis,

the

re is

a r

obus

t re

boun

d in

the

pla

sma

phos

phat

e co

ncen

trat

ion

owin

g to

equ

ili-

brat

ion

of t

he b

ody

com

part

men

ts; t

hus,

ove

rall

rem

oval

is in

effi

cien

t. L

onge

r di

alys

is t

imes

pro

vide

mor

e ef

fect

ive

clea

ranc

e. D

ialy

sis

of p

rote

in-b

ound

sol

utes

, suc

h as

p-c

reso

l su

lfate

, is

limit

ed b

ecau

se p

rote

in b

indi

ng li

mit

s th

e fr

ee-s

olut

e co

ncen

trat

ion,

whi

ch is

the

dri

ving

for

ce f

or d

iffu

sion

(Pa

nel C

). S

trat

egie

s de

sign

ed t

o in

crea

se t

he r

emov

al o

f pro

-te

in-b

ound

sol

utes

incl

ude

pres

crib

ing

high

er-t

han

-usu

al d

ialy

sate

flo

ws,

usi

ng d

ialy

zers

wit

h ve

ry la

rge

mem

bran

e po

res,

and

add

ing

sorb

ents

to

the

dial

ysat

e.





treated group. Improvements in blood-pressure control, serum calcium–phosphorus product, and selected quality-of-life measures were also ob-served. The Frequent Hemodialysis Network, spon-sored by the National Institutes of Health, is currently conducting two studies: in one, daily in-center dialysis involving short treatment times is being compared with conventional thrice-weekly dialysis, and in the other, daily nocturnal dialysis involving longer treatment times is being com-

pared with conventional thrice-weekly dialysis.26 Outcomes will include survival, change in left ven-tricular mass, and quality of life.

E volu tion of Hemodi a lysis in the Uni ted S tates

Long-term dialysis was initially available only for patients who were enrolled in a handful of pro-grams (Fig. 2). In 1972, President Richard Nixon

Table 1. Key Components of the Hemodialysis Prescription.

Component Comments

Dialyzer

Configuration Hollow-fiber dialyzers are preferred owing to improved safety.

Membrane biomaterials Synthetic membranes are used more frequently than cellulose membranes owing to fewer blood–membrane interactions.

Membrane permeability High-flux membranes are constructed with larger pores, which allow greater removal of higher-molecular-weight solutes, with similar removal of lower-molecular-weight solutes as compared with low-flux mem-branes.

Treatment time Usual treatment time is about 4 hours.Longer treatment times allow more fluid removal with less risk of intradialytic hypotension, and the removal

of compartmentalized solutes such as phosphate is increased; nevertheless, increased dialysis time has limited effects on removal of many solutes because of decreasing plasma concentrations.

Treatment frequency Usual frequency is 3 times per week.Increasing the frequency of dialysis to >3 times per week improves solute clearance and fluid removal; effects

on clinical outcomes and quality of life are being evaluated in randomized trials.

Blood flow rate Usual prescription is 200 to 400 ml per minute.Achievable blood flow depends on the type and quality of vascular access. Increasing blood flow increases sol-

ute removal; however, increased flow resistance will eventually limit the augmented clearance.

Dialysate flow rate Usual rate is twice the achieved blood flow rate in order to attain near-maximal solute clearance.

Ultrafiltration rate Should be less than 10 ml per kilogram of body weight per hour to reduce the risk of intradialytic hypo tension.

Dialysate composition

Sodium Between 130 and 145 mmol per liter.Higher sodium concentrations decrease the risk of intradialytic hypotension but increase thirst and inter-

dialytic weight gain.

Potassium Generally 2 to 3 mmol per liter.Lower levels of dialysate potassium are associated with sudden cardiac death; intradialytic potassium removal

is highly variable, and plasma potassium levels rebound about 30% after dialysis.

Calcium Generally 1.25 to 1.75 mmol per liter.Only non–protein-bound calcium is removed; higher levels of dialysate calcium increase intradialytic blood

pressure.

Magnesium Generally 0.5 mmol per liter.The optimal level of magnesium is unresolved, and magnesium flux is difficult to predict.

Alkaline buffers Commonly 30 to 40 mmol per liter.Predominantly bicarbonate with a small amount of acetate; bicarbonate concentration can be adjusted to

correct metabolic acidosis.

Chloride Defined by prescribed cations and alkaline buffers in dialysate.

Glucose Commonly 100 to 200 mg per deciliter.Higher levels of glucose promote hypertriglyceridemia.

Intradialytic medications Erythropoietin, iron, vitamin D analogues, antibiotics.

Anticoagulation Heparin or other agents.



Medical Progress


signed legislation authorizing Medicare coverage for the costs of ESRD treatments, including dialysis and kidney transplantation, for all eligible Amer-icans.27 With little public or congressional debate, the passage of this legislation heralded an era of nearly universal entitlement for ESRD care, in marked contrast to other organ failure–related dis-ease states such as end-stage heart disease or liver disease. Legislators approved the law with the understanding that dialysis would provide high-level rehabilitation and social benefit to a rela-

tively small number of people at low cost.28 Since 1972, a geometric increase in the number of pa-tients receiving dialysis has expanded the scope of the Medicare ESRD Program enormously. From an economic and societal perspective, salient changes during the evolution of this program have included steadily increasing aggregate costs to Medicare, diminished per-treatment reimburse-ment in inflation-adjusted dollars, a dramatic in-crease in costs associated with medications given during dialysis, a steady decline in the use of home

1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010

Allows for repeated,long-term dialysis

1960 Scribner shuntinvented

Sets the stagefor eventual

congressionalaction on

funding ESRDcare; projectsa low rate of

dialysis treat-ment, with a

high rehabili-tation rate

1968 Incorporation of National Medical Care,the first for-profit dialysis provider

1967 Gottschalk Committee report

AuthorizesMedicarepayment forESRD treat-ment,includingdialysis andkidneytransplan-tation

1972 Public law 92-603, section 2991

1978 Public law 95-292

1978 Congress authorizes ESRD networks

Paves the wayfor a bundledcomposite rateof paymentfor dialysisservices

Facilitates qualityassurance andcontinuous qualityimprovement

1989 Medicarecoverage of

erythropoietin

1988 Establishmentof U.S. RenalData System

Creates a government-mandated comprehensive

data set on dialysis outcomes

Results in coverage of injectablemedications in dialysis treatment

1991 Institute ofMedicine’s report:

Kidney Failureand the Federal

Government

Initiatespublicreportingof qualitymeasuresand out-comes

1999 Establishmentof ESRD ClinicalPerformanceMeasures Project

Facilitatesdevelopmentof qualitymonitoringin dialysis

Initiates successfulfocused quality-

improvementprogram

2003 Launch of Fistula FirstBreakthrough Initiative by CMS

2005–06 Consolidation ofdialysis industry throughmergers and acquisitions

2008 Passage of MedicareImprovements for Patientsand Providers Act

Creates an oligopolywhereby two companiescontrol more than 60%

of the market

Increases bundlingof payments fordialysis in 2011;establishes reimburse-ment for preventive care and institutes payments for quality

Figure 2. Timeline for the Evolution of Hemodialysis.

The widespread implementation of dialysis therapy over the past 50 years has been accompanied by considerable medical, social, eco-nomic, and regulatory changes in the U.S. End-Stage Renal Disease (ESRD) Program. Fueled by concerns about increasing costs for the program, associated high morbidity and mortality, and incomplete rehabilitation for patients undergoing dialysis, developments in the evolution of the ESRD Program provided an early window into trends that have subsequently affected the U.S. health care system as a whole. The ESRD Program has served as an incubator for efforts to develop robust vehicles to collect data, measure quality and deter-mine performance indicators, and initiate quality-improvement projects. Legislative changes are under way to institute payments for the delivery of high-quality care and to increase the bundling of payments for ESRD services. CMS denotes Centers for Medicare and Medic-aid Services.





dialysis (including peritoneal dialysis), and the rise and consolidation of a for-profit dialysis-provider industry.29,30

The demographics of the dialysis population have also changed dramatically over time. Less stringent selection of patients has led to treatment of an increasing proportion of elderly patients, patients with diabetes, and patients who are frail and have complex coexisting conditions. The ini-tiation of dialysis in patients with higher levels of residual kidney function has occurred concomi-tantly, particularly among patients older than 75 years of age.31 Among elderly nursing-home resi-dents, the initiation of dialysis is associated with a substantial decline in functional status and high mortality.32 The factors driving these clinical prac-tices, and their societal implications, are only beginning to be studied but may well lead to increased consideration of conservative manage-ment and palliative-care options for some pa-tients.30,33-35

The Medicare Improvements for Patients and Providers Act (MIPPA) of 2008 contains provisions that are likely to change the Medicare ESRD Pro-gram substantially. Currently, dialysis facilities receive a bundled payment for each dialysis ses-sion they provide, which includes funds to cover supplies, staffing, and selected ESRD-related lab-oratory tests; the costs of intravenous medications are billed separately. MIPPA mandates an expan-sion of the bundled-payment system to include funds for all ESRD-related medications and labo-ratory tests, beginning in 2011. Including medi-cation reimbursement along with these bundled payments to dialysis providers should remove any possible financial incentive to overprescribe med-ications during treatment, but simultaneously, it may create incentives for providers to underuse medications or choose to treat patients on the basis of characteristics that may translate into re-duced medication expenditures. MIPPA relies on adjustments for case mix to prevent providers from deselecting or cherry-picking patients and also mandates the development of a payment system for quality indicators by 2012.

Me a sur ing a nd Improv ing Qua li t y in Di a lysis C a r e

The ability to evaluate outcomes among patients with ESRD increased dramatically after 1988, when the United States Renal Data System (USRDS) was

established to record and issue reports that would track mortality and morbidity and determine fac-tors affecting clinical outcomes. Perhaps the most robust disease-specific data sets available within the entire Medicare population, these USRDS re-ports have greatly facilitated the development of quality goals and metrics, at the same time as evidence-based clinical practice guidelines, such as those issued by the Kidney Disease Outcomes Quality Initiative and the Kidney Disease: Improv-ing Global Outcomes program, have been devel-oped.36-38 In 2003 the Centers for Medicare and Medicaid Services (CMS) and other key stakehold-ers jointly developed a national quality-improve-ment effort to increase the use of arteriovenous fistulas as the preferred choice for vascular ac-cess — a choice that had historically lagged be-hind other indicators of high-quality care. This collaborative initiative, known as Fistula First, led to a dramatic increase in the use of fistulas.39 In an effort to facilitate patient choice and promote quality improvement, the CMS developed Dialysis Facility Compare, a Web site that allows consum-ers to compare the mandatory reported perfor-mance of dialysis facilities.40

Patien t S a fe t y a nd Technic a l A dva nces

Hemodialysis is now substantially safer than it was initially, and deaths directly related to the di-alysis procedure are rare. Improved dialysate de-livery systems, more reliable monitoring devices, and automated safety mechanisms have reduced the risk of complications. Other technical im-provements include the standard use of the more physiologic bicarbonate-based dialysate, better water-quality standards, volumetric ultrafiltration controls, and computer-controlled sodium and po-tassium modeling.41 Several in-line devices now allow dynamic monitoring of the rate of blood flow through the vascular access,42 changes in the hematocrit (to measure vascular refilling during ultrafiltration), and changes in the electrical con-ductivity of the dialysate (to estimate the amount of solute being removed).43

Thus, dialysis machines with feedback-control systems currently allow for computer-controlled, real-time adjustments in the critical components of dialysis, such as the ultrafiltration rate.44 Au-tomated control of dialysate temperature helps maintain a constant body temperature during di-



Medical Progress


alysis, which may reduce the incidence of intra-dialytic hypotension.45 Although studies in small groups of patients have suggested possible bene-fits from in-line monitoring or feedback-control systems, evidence of improved outcomes in large, rigorously controlled trials is lacking.46

Tr ends in Ou t comes in the Uni ted S tates

The steady improvement in procedure-focused and process-related measures of quality has led to a noticeable improvement in survival over the past two decades.36 Nevertheless, the death rate among U.S. patients undergoing dialysis continues to ex-ceed 20% per year during the first 2 years after maintenance dialysis is begun. Unfortunately, hos-pitalization rates have remained nearly constant, averaging almost 13 hospital days and two admis-sions per patient-year.47 The exclusive reliance on dialysis-focused quality measures (e.g., adequacy of dialysis, presence or absence of anemia, and mineral metabolism) has previously been ques-tioned, since such measures may account for only 15% of the variations in mortality and morbidi-ty.48 It has been suggested that quality measures also include an assessment of risk factors for car-diovascular disease and infection, which consti-tute the major causes of hospitalization and death in the population receiving dialysis. Practice pat-terns also vary greatly, such as variations in the placement and use of fistulas, the rate of coronary revascularization, and the rate of pneumococcal vaccination. Practice-pattern variations in achiev-ing quality-of-care goals, including predialysis care with timely fistula placement, represent a poten-tial area for the improvement of outcomes.

Clinical care of the patient undergoing dialysis is highly complex, given the insidious but pro-tean manifestations of uremia (Table 2).37,38,49-62 Although symptoms of uremia are often nonspe-cific, virtually every organ system in the body is affected by the disruption in metabolic homeo-stasis associated with ESRD. Physicians who treat patients receiving dialysis must be cognizant of the numerous complications that can result from the loss of kidney function and of the complex rela-tionships between uremia and dialysis treatment. For example, uremia-induced alterations in gastro-intestinal tract function can alter nutrient intake and result in poor nutritional status, which in turn increases the risks of cardiovascular disease

and infection, particularly when dialysis involves tunneled catheters. Given the problems associat-ed with ESRD, physicians who care for patients receiving hemodialysis face unique and difficult challenges.63 Caring for such patients is particu-larly difficult because of the lack of high-level evidence in support of target ranges for many of the important components of dialysis care, such as optimal concentrations of parathyroid hormone and low-density lipoprotein (LDL) cholesterol or blood-pressure levels (Table 2).

In ter nationa l Compa r isons

Many investigators have noted that crude mortal-ity rates are consistently higher in the United States than in Europe or Japan. Probably the best avail-able comparative data come from the Dialysis Out-comes and Practice Patterns Study (DOPPS), which uses a prospective design and attempts to har-monize data collection across several countries and continents.64 The DOPPS reported that crude 1-year mortality rates from 1996 to 2002 were 6.6% in Japan, 15.6% in Europe, and 21.7% in the Unit-ed States.65 Although dramatic differences in de-mographic characteristics, clinical factors, com-pleteness of data ascertainment, and access to kidney transplantation can limit the validity of these transnational comparisons, the relative risk of death after adjustments have been made for age and multiple coexisting disorders is still higher in the United States than in Japan or Europe.66 Features of practice patterns in the United States that differ from those in the other two countries may account in part for the observed differences in the risk of death. Such features include shorter treatment times, less frequent use of fistulas, and staffing of dialysis units with patient care techni-cians rather than nurses.67

One trend of concern in the United States is the increasing proportion of patients who begin dialysis with a tunneled catheter rather than with a more permanent type of vascular access. The use of such catheters is strongly associated with in-creases in the rate of hospitalization, the risk of death, and the cost of care, owing in large part to the risk of catheter-related bacteremia.55 Although the preponderance of available data suggests worse outcomes in the United States, not all investiga-tors agree that differences in practice patterns account for the differences in outcomes. Alterna-tively, results of a study that used the World Health





Organization mortality database suggest that much of the international variation in mortality is attributable to differences in the risk of death that are related to cardiovascular disease in the respective general populations.68

Con trolled Tr i a l s of Di a lysis Ther a py

Several randomized, controlled clinical trials with sufficient power to detect changes in mortality or hospitalization rates have evaluated the ade-quacy of dialysis therapy. The National Coopera-tive Dialysis Study (NCDS), performed during the

1970s, was designed to determine whether alter-ing the time-averaged concentration of urea or the treatment time — each of which is consid-ered an important determinant of the adequacy of hemodialysis — would affect hospitalization rates.69 The results of the NCDS indicated that a high urea concentration was significantly associ-ated with increased hospitalizations. On the ba-sis of the NCDS results, a minimum delivered di-alysis dose equivalent to a single-pool Kt/Vurea value of 1.2 was initially established as a standard, which was incorporated into clinical practice guidelines and performance measures. The NCDS was not powered to evaluate mortality as an outcome. Al-

Table 2. Clinical Care of Patients Receiving Hemodialysis.*

Variable Goals and Targets

Dialysis dose Monitor urea kinetic modeling; target single-pool Kt/Vurea >1.4.†

Fluid management and estimated body weight

Carry out individualized management and assessment; interdialytic weight gain should ideally be less than 5% of total body weight.

Dialysate quality Monitor endotoxin and bacteria concentrations in water used for dialysate; the use of ultrapure dialy-sate may reduce inflammation.49

Anemia Try to attain a hemoglobin level of 10 to 12 g per deciliter (although current recommendations may change on the basis of results from clinical trials involving patients with chronic kidney dis-ease50-53); avoid high-dose erythropoietin; evaluate patients with erythropoietin resistance for inflammation and iron deficiency; monitor iron levels and treat iron deficiency; the long-term safety and efficacy of iron administration in patients with high ferritin levels have not been well established.54†‡

Vascular access Implement strategies to increase the placement and use of fistulas and eliminate catheter use when-ever feasible55; monitor to detect possible access dysfunction.†§

Bone and mineral disorders Aim for a serum calcium level of 8.4 to 9.5 mg per deciliter and a serum phosphate level of 3.5 to 5.5 mg per deciliter; monitor serum levels of intact PTH; although the optimal target PTH level has not been well defined, maintain PTH level at >2 times the upper limit of the normal range to minimize risk of low bone turnover; suppress rising PTH levels with vitamin D analogues, calcimi-metics, and phosphate binders.¶

Nutrition Aim for serum albumin level >4.0 g per deciliter; consider enteral supplementation for progressive signs of protein energy wasting; refer patient to dietitian for nutritional counseling; restrict phos-phorus, sodium, and potassium intake, as guided by laboratory studies.†

Blood pressure Optimal targets and management strategies have not been well defined.57

LDL cholesterol Aim for LDL cholesterol level of <100 mg per deciliter; the relationship between LDL cholesterol and cardiovascular risk is confounded by inflammation; statins are without proven benefit.58-60

Diabetes management Balance benefits of tighter glycemic control, which carries an increased risk of hypoglycemia, by means of individualized therapy; glycated hemoglobin targets have not been well defined61; manage other aspects of diabetes, such as peripheral vascular disease, intestinal dysmotility, and eye problems.

Transplantation referral Provide education about transplantation and timely referrals for suitable candidates; monitor status of wait-listed patients.§

Quality-of-life and psychosocial evaluation

The evaluation, conducted by a social worker with the support of a multidisciplinary team, should be aimed at optimizing adjustment to kidney failure and its treatment; the Kidney Disease Quality of Life (KDQOL-36) instrument is often used for the evaluation.62§

* LDL denotes low-density lipoprotein, and PTH parathyroid hormone.† This recommendation is supported by the Kidney Disease Outcomes Quality Initiative Clinical Practice Guidelines.38

‡ This recommendation is compatible with those of the Food and Drug Administration and the European Medicines Agency.§ This recommendation is mandated by the Centers for Medicare and Medicaid Services Conditions of Coverage for end-stage renal disease

facilities.56

¶ This recommendation is supported by the Kidney Disease: Improving Global Outcomes Clinical Practice Guidelines.37



Medical Progress


though the NCDS was an exemplary early ran-domized, controlled trial of the adequacy of di-alysis, the delivered dose in the low-dose treatment group was well below that routinely delivered to-day, and the patient population was not represen-tative of current dialysis recipients.

In the 1990s, several large, observational stud-ies suggested that doses of dialysis that were higher than standard doses and the use of dialy-sis membranes with higher-permeability charac-teristics (or flux) were associated with lower mor-tality.11-13,70-72 The Hemodialysis (HEMO) Study, funded by the National Institutes of Health, sub-sequently compared the effects of a standard di-alysis dose (a single-pool Kt/Vurea value of 1.25) with a higher dose for urea clearance and also compared the effects of high versus low mem-brane flux on morbidity and mortality. A total of 1846 subjects were followed for 7 years, and the study had reasonably high power to detect a re-duction in mortality. The results of the HEMO study showed no significant differences in all-cause mortality or in seven other prespecified outcomes among any of the treatment groups.73 Another well-conducted, randomized trial, the Adequacy of Peritoneal Dialysis in Mexico (ADEMEX) trial, also showed no relationship be-tween dialysis dose and outcomes among pa-tients receiving peritoneal dialysis.74 Taken col-lectively, the results of these important trials suggest that there is a threshold–plateau relation-ship between the dose of dialysis and outcomes and that increasing the dose to greater than the currently recommended target of a single-pool Kt/Vurea value of 1.4 (in order to ensure an achieved dose of at least 1.2) does not improve important outcomes. These studies also illustrate the limi-tations of what is achievable with current dialy-sis practice and underscore the need for more innovative approaches.

Con trolled Tr i a l s t o E va luate C a r diova scul a r R isk

Many randomized, controlled trials have focused on mitigating cardiovascular events and mortal-ity in patients undergoing hemodialysis, but re-sults have been disappointing. Notably, two inves-tigations evaluating the use of atorvastatin and rosuvastatin showed no improvement in major outcomes, despite being well powered because of the high cardiovascular event rate in each treat-ment group.58,59 Trials evaluating homocysteine-

lowering drugs,75 non–calcium-containing phos-phorus binders,76 and erythropoietin at doses that target higher hemoglobin concentrations77 have all supported the null hypothesis or even suggest-ed harm. In interpreting such trial results, it is important to consider that many metabolic and structural contributors to cardiovascular risk among patients undergoing dialysis may differ from those among patients without kidney disease (Fig. 3). Furthermore, in cross-sectional studies, conventional cardiovascular risk factors, such as elevated serum cholesterol levels and blood pres-sure or a high degree of adiposity, have been found to be less predictive of risk among patients receiving dialysis than among persons with pre-served kidney function.

Uremic cardiovascular disease is characterized by a high prevalence of medial vascular calcifica-tion, arterial stiffness, and altered left ventricu-lar geometry.78-80 The development of aggressive intimal hyperplasia is common after either coro-nary angioplasty or the establishment of arterio-venous access.81 Cardiac arrest and congestive heart failure are more prominent causes of car-diovascular death than is acute myocardial in-farction in patients with uremia.82 Metabolically, ESRD is strongly associated with acute inflamma-tion, oxidative stress, endothelial dysfunction, in-sulin resistance, and excess sympathetic tone.83-91 A number of uremic toxins that are highly pro-tein-bound or sequestered within cells or bone, such as p-cresol sulfate, indoxyl sulfate, and phos-phate, may contribute directly to cardiovascular risk and are not sufficiently removed by means of conventional dialysis (Fig. 1).5,10 Further re-search is needed to understand more precisely how uremic toxins contribute to cardiovascular risk and to evaluate novel approaches to reducing this risk. Innovative experimental approaches to dialysis have been advocated, such as wearable artificial kidneys and the use of nanotechnology for more rational membrane design.92-94 However, the large-scale implementation of any of these novel experi-mental approaches is not likely to occur in the near future.95

Conclusions

Over the past half century, the widespread use of dialysis to prolong life for people without kidney function has been a remarkable achievement. As a result of its growth and evolution, the U.S. ESRD Program has often provided an early window into





Figure 3. Pathobiology of Increased Cardiovascular Risk in End-Stage Renal Disease (ESRD).

The factors leading to an increased risk of cardiovascular disease among patients with ESRD are multifaceted and in many cases incom-pletely understood. Although patients who undergo dialysis often have conventional cardiovascular risk factors, the presence of such fac-tors does not appear to fully explain the high level of risk. In patients with uremia, multiple mediators related to the metabolic changes resulting from the loss of kidney function appear to contribute to this increased risk by causing multiple functional and structural chang-es in the heart and blood vessels. These mediators include increased inflammation, greater sympathetic-nerve activity, oxidative stress, disturbed mineral balance, and profound endothelial dysfunction. Concurrent medical problems, such as anemia, hypertension, and hy-pervolemia, also contribute to structural cardiac and vascular alterations, frequently resulting in heart failure. Excessive vascular calcifi-cation, increased myocardial fibrosis, and increased intimal hyperplasia are common findings. In addition to ischemic cardiovascular events, the incidence of sudden death is excessively high in patients receiving hemodialysis. FGF-23 denotes fibroblast growth factor 23. The histologic photomicrographs are courtesy of Dr. Michael Laflamme and the angiographic image is courtesy of Dr. Thomas S. Hatsukami.



Medical Progress


social, political, and economic developments in health care, and these changes have later been reflected throughout the U.S. health care system. Despite such successes, the use of dialysis in the treatment of ESRD is problematic in some respects. The number of patients treated, especially in the United States, has escalated and is far beyond early estimates. Aggregate dialysis-associated costs have increased accordingly, and morbidity and mortality among treated patients remain high despite considerable technical and scientific im-provements. Our knowledge of which uremic tox-ins confer injury and of how they can be optimally removed during dialysis therapy remains incom-plete. The limited number of clinical trials that have attempted to improve outcomes have had disappointing results, so more well-designed and adequately powered clinical trials are needed.

Ongoing studies are assessing whether longer or more frequent dialysis treatments, or both, can improve outcomes and whether these changes would be acceptable to most patients. However, substantive improvements for patients receiving

dialysis will probably require major technologi-cal breakthroughs that will be predicated on an improved understanding of uremic toxins and ure-mic complications.

Dr. Himmelfarb reports serving on the Shire Pharmaceuticals scientific advisory board on Outcomes Studies in Hemodialysis Patients and the CytoPherx (formerly Nephrion) medical advi-sory board and receiving consulting fees from KAI Pharmaceu-ticals, as well as receipt by his former institution, the Maine Medical Center Research Institute, of research support from Fresenius Medical Services of North America (now Fresenius Medical Care North America). All payments for board member-ship and consulting were donated to the Kidney Research Insti-tute at the University of Washington. Dr. Ikizler reports serving on the boards of the American Board of Internal Medicine and Satellite Healthcare, receiving consulting fees from Amgen, Ab-bott Renal Care, Novo Nordisk, Renal Advantage, Eli Lilly, and Fresenius Medical Care North America, royalties from Wolter Kluwer, as well as receipt by his institution, the Vanderbilt Uni-versity Medical Center, of research support from Medical Nutri-tion USA and Fresenius Medical Care North America. Dr. Ikizler reports having been the medical director of the Vanderbilt Dialy-sis Unit, which is owned by Vanderbilt University Medical Center and managed by Fresenius Medical Care North America, until June 2010. No other potential conflict of interest relevant to this article was reported.

Disclosure forms provided by the authors are available with the full text of this article at NEJM.org.

References

1. Scribner BH, Caner JE, Buri R, Quin-ton W. The technique of continuous hemo-dialysis. Trans Am Soc Artif Intern Organs 1960;6:88-103.2. Quinton W, Dillard D, Scribner BH. Cannulation of blood vessels for pro-longed hemodialysis. Trans Am Soc Artif Intern Organs 1960;6:104-13.3. Depner TA. Prescribing hemodialysis: a guide to urea modeling. Boston: Kluwer Academic, 1991.4. Locatelli F, Manzoni C, Di Filippo S. The importance of convective transport. Kidney Int Suppl 2002;115-20.5. Meyer TW, Hostetter TH. Uremia. N Engl J Med 2007;357:1316-25.6. Gotch FA, Sargent JA. A mechanistic analysis of the National Cooperative Di-alysis Study (NCDS). Kidney Int 1985;28: 526-34.7. Daugirdas JT, Depner TA, Greene T, et al. Surface-area-normalized Kt/V: a meth-od of rescaling dialysis dose to body sur-face area — implications for different-size patients by gender. Semin Dial 2008;21: 415-21.8. Port FK, Wolfe RA, Hulbert-Shearon TE, McCullough KP, Ashby VB, Held PJ. High dialysis dose is associated with lower mortality among women but not among men. Am J Kidney Dis 2004;43:1014-23.9. Spalding EM, Chandna SM, Daven-port A, Farrington K. Kt/V underestimates the hemodialysis dose in women and small men. Kidney Int 2008;74:348-55.

10. Vanholder R, Baurmeister U, Brunet P, Cohen G, Glorieux G, Jankowski J. A bench to bedside view of uremic toxins. J Am Soc Nephrol 2008;19:863-70.11. Hornberger JC, Chernew M, Petersen J, Garber AM. A multivariate analysis of mortality and hospital admissions with high-f lux dialysis. J Am Soc Nephrol 1992;3:1227-37.12. Koda Y, Nishi S, Miyazaki S, et al. Switch from conventional to high-f lux membrane reduces the risk of carpal tun-nel syndrome and mortality of hemodialy-sis patients. Kidney Int 1997;52:1096-101.13. Locatelli F, Mastrangelo F, Redaelli B, et al. Effects of different membranes and dialysis technologies on patient treatment tolerance and nutritional parameters. Kid-ney Int 1996;50:1293-302.14. Port FK, Wolfe RA, Hulbert-Shearon TE, et al. Mortality risk by hemodialyzer reuse practice and dialyzer membrane characteristics: results from the USRDS dialysis morbidity and mortality study. Am J Kidney Dis 2001;37:276-86.15. Held PJ, Levin NW, Bovbjerg RR, Pau-ly MV, Diamond LH. Mortality and dura-tion of hemodialysis treatment. JAMA 1991;265:871-5.16. Marshall MR, Byrne BG, Kerr PG, Mc-Donald SP. Associations of hemodialysis dose and session length with mortality risk in Australian and New Zealand pa-tients. Kidney Int 2006;69:1229-36.17. Saran R, Bragg-Gresham JL, Levin

NW, et al. Longer treatment time and slower ultrafiltration in hemodialysis: as-sociations with reduced mortality in the DOPPS. Kidney Int 2006;69:1222-8.18. Kalantar-Zadeh K, Regidor DL, Koves-dy CP, et al. Fluid retention is associated with cardiovascular mortality in patients undergoing long-term hemodialysis. Cir-culation 2009;119:671-9.19. Charra B, Calemard M, Laurent G. Importance of treatment time and blood pressure control in achieving long-term survival on dialysis. Am J Nephrol 1996; 16:35-44.20. Powell JR, Oluwaseun O, Woo YM, et al. Ten years experience of in-center thrice weekly long overnight hemodialysis. Clin J Am Soc Nephrol 2009;4:1097-101.21. Bugeja A, Dacouris N, Thomas A, et al. In-center nocturnal hemodialysis: another option in the management of chronic kid-ney disease. Clin J Am Soc Nephrol 2009;4:778-83.22. Troidle L, Hotchkiss M, Finkelstein F. A thrice weekly in-center nocturnal he-modialysis program. Adv Chronic Kidney Dis 2007;14:244-8.23. Kliger AS. High-frequency hemodial-ysis: rationale for randomized clinical tri-als. Clin J Am Soc Nephrol 2007;2:390-2.24. Suri RS, Nesrallah GE, Mainra R, et al. Daily hemodialysis: a systematic review. Clin J Am Soc Nephrol 2006;1:33-42.25. Culleton BF, Walsh M, Klarenbach SW, et al. Effect of frequent nocturnal he-





modialysis vs conventional hemodialysis on left ventricular mass and quality of life: a randomized controlled trial. JAMA 2007;298:1291-9.26. Suri RS, Garg AX, Chertow GM, et al. Frequent Hemodialysis Network (FHN) randomized trials: study design. Kidney Int 2007;71:349-59.27. Schreiner GE. How end-stage renal disease (ESRD)-Medicare developed. Am J Kidney Dis 2000;35:Suppl 1:S37-S44.28. Rettig RA. The policy debate on pa-tient care financing for victims of end-stage renal disease. Law Contemp Probl 1976;40:196-230.29. Himmelfarb J, Berns A, Szczech L, Wesson D. Cost, quality, and value: the changing political economy of dialysis care. J Am Soc Nephrol 2007;18:2021-7.30. Knauf F, Aronson PS. ESRD as a win-dow into America’s cost crisis in health care. J Am Soc Nephrol 2009;20:2093-7.31. Rosansky SJ, Clark WF, Eggers P, Glassock RJ. Initiation of dialysis at high-er GFRs: is the apparent rising tide of early dialysis harmful or helpful? Kidney Int 2009;76:257-61.32. Kurella Tamura M, Covinsky KE, Chertow GM, Yaffe K, Landefeld CS, Mc-Culloch CE. Functional status of elderly adults before and after initiation of dialy-sis. N Engl J Med 2009;361:1539-47.33. Arnold RM, Zeidel ML. Dialysis in frail elders — a role for palliative care. N Engl J Med 2009;361:1597-8.34. Moss AH. Shared decision-making in dialysis: the new RPA/ASN guideline on appropriate initiation and withdrawal of treatment. Am J Kidney Dis 2001;37:1081-91.35. Murtagh FE, Marsh JE, Donohoe P, Ekbal NJ, Sheerin NS, Harris FE. Dialysis or not? A comparative survival study of patients over 75 years with chronic kidney disease stage 5. Nephrol Dial Transplant 2007;22:1955-62.36. Himmelfarb J, Kliger AS. End-stage renal disease measures of quality. Annu Rev Med 2007;58:387-99.37. Kidney Disease: Improving Global Outcomes (KDIGO) home page. (http://kdigo.org/.)38. National Kidney Foundation Kidney Disease Outcomes Quality Initiative (NKF KDOQI). New York: National Kid-ney Foundation. (http://www.kidney.org/professionals/KDOQI/.)39. National AV fistula rate reaches 55.3% in April 2010. Midlothian, VA: Fistula First Breakthrough Initiative. (http://www .fistulafirst.org.)40. Dialysis Facility Compare (DFC). Bal-timore: Centers for Medicare & Medicaid Services. (http://www.cms.gov/dialysis facilitycompare/.)41. Oliver MJ, Edwards LJ, Churchill DN. Impact of sodium and ultrafiltration pro-filing on hemodialysis-related symptoms. J Am Soc Nephrol 2001;12:151-6.

42. Lopot F, Nejedlý B, Sulková S, Bláha J. Comparison of different techniques of he-modialysis vascular access flow evalua-tion. J Vasc Access 2004;5:25-32.43. Maduell F, Vera M, Arias M, et al. In-fluence of the ionic dialysance monitor on Kt measurement in hemodialysis. Am J Kidney Dis 2008;52:85-92.44. Ward RA, Ronco C. Improvements in technology: a path to safer and more ef-fective hemodialysis. Blood Purif 2009;27: 6-10.45. Maggiore Q, Pizzarelli F, Santoro A, et al. The effects of control of thermal balance on vascular stability in hemodi-alysis patients: results of the European randomized clinical trial. Am J Kidney Dis 2002;40:280-90.46. Locatelli F, Buoncristiani U, Canaud B, Köhler H, Petitclerc T, Zucchelli P. Hae-modialysis with on-line monitoring equip-ment: tools or toys? Nephrol Dial Trans-plant 2005;20:22-33.47. US Renal Data System. USRDS 2009 annual data report: atlas of chronic kid-ney disease and end-stage renal disease in the United States. 2009. (http://www.usrds .org/adr.htm.)48. Lowrie EG, Lew NL. Death risk in he-modialysis patients: the predictive value of commonly measured variables and an evaluation of death rate differences be-tween facilities. Am J Kidney Dis 1990; 15:458-82.49. Ward RA. Ultrapure dialysate. Semin Dial 2004;17:489-97.50. Pfeffer MA, Burdmann EA, Chen CY, et al. A trial of darbepoetin alfa in type 2 diabetes and chronic kidney disease. N Engl J Med 2009;361:2019-32.51. Singh AK, Szczech L, Tang KL, et al. Correction of anemia with epoetin alfa in chronic kidney disease. N Engl J Med 2006;355:2085-98.52. Drüeke TB, Locatelli F, Clyne N, et al. Normalization of hemoglobin level in pa-tients with chronic kidney disease and anemia. N Engl J Med 2006;355:2071-84.53. Unger EF, Thompson AM, Blank MJ, Temple R. Erythropoiesis-stimulating agents — time for a reevaluation. N Engl J Med 2010;362:189-92.54. Spiegel DM, Chertow GM. Lost with-out directions: lessons from the anemia debate and the DRIVE study. Clin J Am Soc Nephrol 2009;4:1009-10.55. Lacson E Jr, Lazarus JM, Himmelfarb J, Ikizler TA, Hakim RM. Balancing Fis-tula First with Catheters Last. Am J Kid-ney Dis 2007;50:379-95.56. Center for Medicaid and State Opera-tions/Survey & Certification Group. End-stage renal disease (ESRD) program in-terpretive guidance, version 1.1. Baltimore: Centers for Medicare & Medicaid Servic-es, 2008. (http://www.cms.gov/Survey CertificationGenInfo/downloads/ SCletter09-01.pdf.)57. Agarwal R. Assessment of blood pres-

sure in hemodialysis patients. Semin Dial 2002;15:299-304.58. Fellström BC, Jardine AG, Schmeider RE, et al. Rosuvastatin and cardiovascular events in patients undergoing hemodialy-sis. N Engl J Med 2009;360:1395-407. [Er-ratum, N Engl J Med 2010;362:1450.]59. Wanner C, Krane V, März W, et al. Atorvastatin in patients with type 2 dia-betes mellitus undergoing hemodialysis. N Engl J Med 2005;353:238-48. [Erratum, N Engl J Med 2005;353:1640.]60. Liu Y, Coresh J, Eustace JA, et al. As-sociation between cholesterol level and mortality in dialysis patients: role of in-flammation and malnutrition. JAMA 2004;291:451-9.61. Williams ME, Lacson E Jr, Teng M, Ofsthun N, Lazarus JM. Hemodialyzed type I and type II diabetic patients in the US: characteristics, glycemic control, and survival. Kidney Int 2006;70:1503-9.62. Lowrie EG, Curtin RB, LePain N, Schatell D. Medical Outcomes Study Short Form-36: a consistent and powerful pre-dictor of morbidity and mortality in dialy-sis patients. Am J Kidney Dis 2003;41: 1286-92.63. Himmelfarb J, Chuang P, Schulman G. Hemodialysis. 8th ed. Philadelphia: W.B. Saunders, 2008.64. Young EW, Goodkin DA, Mapes DL, et al. The Dialysis Outcomes and Practice Patterns Study (DOPPS): an international hemodialysis study. Kidney Int 2000;57: Suppl:S74-S81.65. Goodkin DA, Young EW, Kurokawa K, Prutz KG, Levin NW. Mortality among he-modialysis patients in Europe, Japan, and the United States: case-mix effects. Am J Kidney Dis 2004;44:16-21.66. Goodkin DA, Bragg-Gresham JL, Koenig KG, et al. Association of comorbid conditions and mortality in hemodialysis patients in Europe, Japan, and the United States: the Dialysis Outcomes and Prac-tice Patterns Study (DOPPS). J Am Soc Nephrol 2003;14:3270-7.67. Foley RN, Hakim RM. Why is the mortality of dialysis patients in the Unit-ed States much higher than the rest of the world? J Am Soc Nephrol 2009;20:1432-5.68. Yoshino M, Kuhlmann MK, Kotanko P, et al. International differences in dialysis mortality reflect background general pop-ulation atherosclerotic cardiovascular mor-tality. J Am Soc Nephrol 2006;17:3510-9.69. Lowrie EG, Laird NM, Parker TF, Sar-gent JA. Effect of the hemodialysis pre-scription of patient morbidity: report from the National Cooperative Dialysis Study. N Engl J Med 1981;305:1176-81.70. Collins AJ, Ma JZ, Umen A, Keshaviah P. Urea index and other predictors of he-modialysis patient survival. Am J Kidney Dis 1994;23:272-82.71. Held PJ, Port FK, Wolfe RA, et al. The dose of hemodialysis and patient mortal-ity. Kidney Int 1996;50:550-6.



Medical Progress


72. Owen WF Jr, Lew NL, Liu Y, Lowrie EG, Lazarus JM. The urea reduction ratio and serum albumin concentration as pre-dictors of mortality in patients undergo-ing hemodialysis. N Engl J Med 1993; 329:1001-6.73. Eknoyan G, Beck GJ, Cheung AK, et al. Effect of dialysis dose and membrane flux in maintenance hemodialysis. N Engl J Med 2002;347:2010-9.74. Paniagua R, Amato D, Vonesh E, et al. Effects of increased peritoneal clearances on mortality rates in peritoneal dialysis: ADEMEX, a prospective, randomized, con-trolled trial. J Am Soc Nephrol 2002;13: 1307-20.75. Jamison RL, Hartigan P, Kaufman JS, et al. Effect of homocysteine lowering on mortality and vascular disease in advanced chronic kidney disease and end-stage re-nal disease: a randomized controlled trial. JAMA 2007;298:1163-70. [Erratum, JAMA 2008;300:170.]76. Suki WN, Zabaneh R, Cangiano JL, et al. Effects of sevelamer and calcium-based phosphate binders on mortality in hemodialysis patients. Kidney Int 2007; 72:1130-7.77. Besarab A, Bolton WK, Browne JK, et al. The effects of normal as compared with low hematocrit values in patients with cardiac disease who are receiving hemodialysis and epoetin. N Engl J Med 1998;339:584-90.78. Middleton RJ, Parfrey PS, Foley RN. Left ventricular hypertrophy in the renal patient. J Am Soc Nephrol 2001;12:1079-84.79. Block GA, Raggi P, Bellasi A, Kooien-

ga L, Spiegel DM. Mortality effect of coro-nary calcification and phosphate binder choice in incident hemodialysis patients. Kidney Int 2007;71:438-41.80. Raggi P, Bellasi A, Ferramosca E, Is-lam T, Muntner P, Block GA. Association of pulse wave velocity with vascular and valvular calcification in hemodialysis pa-tients. Kidney Int 2007;71:802-7.81. Li L, Terry CM, Shiu YT, Cheung AK. Neointimal hyperplasia associated with synthetic hemodialysis grafts. Kidney Int 2008;74:1247-61.82. Herzog CA, Mangrum JM, Passman R. Sudden cardiac death and dialysis pa-tients. Semin Dial 2008;21:300-7.83. Converse RL Jr, Jacobsen TN, Toto RD, et al. Sympathetic overactivity in pa-tients with chronic renal failure. N Engl J Med 1992;327:1912-8.84. Annuk M, Zilmer M, Lind L, Linde T, Fellström B. Oxidative stress and endo-thelial function in chronic renal failure. J Am Soc Nephrol 2001;12:2747-52.85. Zoccali C, Mallamaci F, Tripepi G, et al. Adiponectin, metabolic risk factors, and cardiovascular events among patients with end-stage renal disease. J Am Soc Nephrol 2002;13:134-41.86. DeFronzo RA, Alvestrand A, Smith D, Hendler R, Hendler E, Wahren J. Insulin resistance in uremia. J Clin Invest 1981; 67:563-8.87. Himmelfarb J, Stenvinkel P, Ikizler TA, Hakim RM. The elephant in uremia: oxidant stress as a unifying concept of cardiovascular disease in uremia. Kidney Int 2002;62:1524-38.88. Ikizler TA, Wingard RL, Harvell J,

Shyr Y, Hakim RM. Association of mor-bidity with markers of nutrition and in-flammation in chronic hemodialysis pa-tients: a prospective study. Kidney Int 1999;55:1945-51.89. Yilmaz MI, Carrero JJ, Ortiz A, et al. Soluble TWEAK plasma levels as a novel biomarker of endothelial function in pa-tients with chronic kidney disease. Clin J Am Soc Nephrol 2009;4:1716-23.90. Vallance P, Leone A, Calver A, Collier J, Moncada S. Accumulation of an endog-enous inhibitor of nitric oxide synthesis in chronic renal failure. Lancet 1992; 339:572-5.91. Passauer J, Büssemaker E, Range U, Plug M, Gross P. Evidence in vivo showing increase of baseline nitric oxide genera-tion and impairment of endothelium-dependent vasodilation in normotensive patients on chronic hemodialysis. J Am Soc Nephrol 2000;11:1726-34.92. Fissell WH, Roy S. The implantable artificial kidney. Semin Dial 2009;22:665-70.93. Gura V, Macy AS, Beizai M, Ezon C, Golper TA. Technical breakthroughs in the wearable artificial kidney (WAK). Clin J Am Soc Nephrol 2009;4:1441-8.94. Nissenson AR, Ronco C, Pergamit G, Edelstein M, Watts R. Continuously func-tioning artificial nephron system: the promise of nanotechnology. Hemodial Int 2005;9:210-7.95. Rastogi A, Nissenson AR. Techno-logical advances in renal replacement therapy: five years and beyond. Clin J Am Soc Nephrol 2009;4:Suppl 1:S132-S136.Copyright © 2010 Massachusetts Medical Society.

apply for jobs at the nejm careercenter

Physicians registered at the NEJM CareerCenter can apply for jobs electronically. A personal account created when you register allows you to apply for positions,

using your own cover letter and CV, and keep track of your job-application history. Visit NEJMjobs.org for more information.



Anesthesiology Clin

24 (2006) 523–547

Renal Disease: The Anesthesiologist’sPerspective

Gebhard Wagener, MD, Tricia E. Brentjens, MD*Department of Anesthesiology, College of Physicians and Surgeons at Columbia University,

New York Presbyterian Hospital, Ph-5, 633 W. 168th Street, New York, NY 10032, USA

The kidney is a remarkable organ that accomplishes a multitude of tasks.Its main functions include maintaining fluid and electrolyte balance, excret-ing metabolic waste products, controlling vascular tone, and regulatinghematopoiesis and bone metabolism. The kidneys are paired organs locatedin the retroperitoneal space lateral to the vertebral bodies of T12. Eachkidney weighs approximately 125 g, is 8 to 10 cm long and 4 cm wide. Thekidneys receive 20% of cardiac output, whereas they comprise only 0.5%body weight, which makes them the best-perfused organs per gram weight.

Vascular anatomy

A single renal artery supplies each kidney, although accessory renal ar-teries are not uncommon. The renal arteries branch off the aorta, enterthe kidneys at the hilum, and divide to form an anterior and a posteriorbranch. The anterior branch then divides into segmental arteries, whichare ‘‘end arteries’’ because there is no collateral circulation between them.The segmental arteries further divide into interlobular arteries. Afferent ar-terioles arise from smaller branches of the interlobular arteries. Each neph-ron is supplied by an afferent arteriole that delivers blood to the glomerulartuft of capillaries. The afferent arteriolar sphincter regulates blood flow tothe glomerular tuft. Constriction of the sphincter reduces blood flow inresponse to an increase in glomerular filtration rate (GFR), by way of thesecretion of renin and angiotensin.

Blood drains from the glomerular tuft into the efferent arteriole, past theefferent arteriolar sphincter. Efferent arterioles of the cortical nephrons thenbecome peritubular capillaries that drain into the interlobular veins and the

* Corresponding author.

E-mail address: [email protected] (T.E. Brentjens).

0889-8537/06/$ - see front matter � 2006 Elsevier Inc. All rights reserved.

doi:10.1016/j.atc.2006.04.001 anesthesiology.theclinics.com

mailto:[email protected]

524 WAGENER & BRENTJENS

arcuate veins. Efferent arterioles of juxtamedullary nephrons become thedescending vasa recta to supply medullary capillaries. Blood then flowsthrough the ascending vasa recta directly into the arcuate veins, which drainblood into the segmental veins and then finally into the renal veins (Fig. 1).

Glomerular function

Each kidney contains approximately 1 million nephrons, each composedof a glomerulus and a proximal, intermediate and distal tubule. The glomer-ular tuft contains a tortuous convolute of capillaries that cover the glomer-ular basement membrane. The glomerulus is responsible for the productionof an ultrafiltrate of plasma. In the glomerulus, hydrostatic pressure pushesfluid through the capillaries, the glomerular basement membrane, and thevisceral epithelium into Bowman’s space, a blind-ending sack, consistingof visceral and parietal epithelium, that drains filtrate into the proximaltubule.

The capillaries have 50- to 100-nm wide pores, and the glomerular base-ment membrane has openings of 350 nm. The capillaries avert cells from be-ing filtrated, whereas the glomerular basement membrane inhibits the

Fig. 1. Anatomy of the nephron. (FromMariebNM.Human anatomy& physiology, 5th edition.

San Francisco; London: Benjamin Cummings; 2001. p. 1010; with permission.)

525RENAL DISEASE: THE ANESTHESIOLOGIST’S PERSPECTIVE

filtration of proteins. Any molecule that is smaller than 1.8 nm is filtratedfreely, whereas only cations between 1.8 and 3.6 nm are filtrated [1].

The amount of filtrate depends on the hydrostatic and oncotic pressuresof the capillary and the filtrate, as well as the permeability of the barrier.The afferent and efferent arterioles regulate the hydrostatic pressure of thecapillaries. Dilation of the afferent arteriolar sphincter or constriction ofthe efferent arteriolar sphincter results in an increase in hydrostatic pressure,and therefore the GFR. On average, the GFR is about 180 mL/min.

Tubular function

The filtrate enters the proximal tubule, which consists of a convolutedsegment (the pars convolute) and a straight portion (the pars recta). Thepars recta enters the outer medulla to become the descending portion ofthe loop of Henle. The two subsets of nephrons are cortical and juxtamedul-lary. Cortical nephrons, which comprise approximately 40% of the neph-rons, have short loops that extend only to the outer medulla. Theirglomeruli are located mostly in the outer cortex and they do not have an as-cending thick portion of the loop of Henle [2]. Long-looped nephrons orig-inating from the juxtamedullary glomeruli have long loops of Henle thatinclude a thick ascending limb.

The cortical segment of the thick ascending loop reapproaches its glomer-ulus to form the macula densa. The macula densa and the connecting por-tions of the afferent and efferent arterioles form the juxtaglomerularapparatus. Chemoreceptors in the macula densa sense the concentrationof sodium chloride, NaCl. They release renin in response to an increase inchloride concentration in the tubules. The release of renin leads to theformation of angiotensin, which constricts the afferent vascular sphincterand decreases GFR.

Filtrate empties from Bowman’s space into the proximal tubule. Hydro-gen ions and sodium are absorbed actively from the proximal tubule to theapical cells in exchange for bicarbonate. Sodium is then transported to theinterstitial space, using a Na-K-ATPase transport system, resulting in anincreased interstitial osmolality, and leading to the reabsorption ofwater, sodium, and chloride that was filtered into Bowman’s space. Or-ganic anions and cations, and drugs such as bile salts, creatinine, dopa-mine, cimetidine, and penicillin are secreted actively into the tubularlumen at the proximal tubule. These active transport systems can becomesaturated.

The remaining filtrate then enters the loop of Henle, which consists offour segments: the thin descending limb and the thin ascending limb, fol-lowed by the medullary thick ascending limb and the cortical thick ascend-ing limb. The function of the loop of Henle is to allow for dramatic changesin the osmolality of urine during times of volume depletion or volumeoverload.


The descending limb is water permeable, and urine osmolarity equili-brates passively with the medullary interstitium. The thin ascending limbis not water permeable, and contains an active transport system that movesNaCl into the interstitium. Rising interstitial NaCl concentrations increasethe interstitial osmolarity, which results in more water being pulled fromthe descending limb. Consequently, the highest intratubular and interstitialosmolarity can be reached at the 180-degree turn in the inner medulla. Theosmolarity at the papillary tip is increased further by the active transport ofurea from the collecting tubules into the medullary interstitium [3,4]. Thecountercurrent mechanism of the loop of Henle and the collecting ductsproduces an increase in osmolarity from the cortex to the medulla, andallows for the production of concentrated or dilute urine in response tohypo/hypervolemia (Fig. 2).

Response to ischemia

Although the kidneys receive about 20% of the cardiac output, bloodflow distribution within the kidney is very heterogeneous. The cortex re-ceives more than 85% of the blood flow, even though the medulla is highlyactive metabolically. Tissue pO2 is approximately 50 to 100 mm Hg in thecortex, whereas it is as low as 10 to 15mm Hg in the medullary thick ascend-ing limb. This progressive fall in tissue pO2 from the cortex to the medulla issecondary to countercurrent oxygen exchange. Energy is required for thevarious active transport mechanisms, especially in the metabolically activethick ascending limb of the loop of Henle. Therefore, the medulla is moreprone to ischemic injury secondary to hypoperfusion. During periods ofischemia, sympathoadrenal activation redistributes blood preferentially tothe medulla to attenuate medullary ischemia [5,6].

Prolonged ischemia leads to a breakdown of active transport systems inthe thick ascending loop, increasing intratubular NaCl concentration. In themacula densa, the increased NaCl concentration decreases GFR by way ofthe secretion of renin and angiotensin. This feedback mechanism decreasesGFR and results in decreased delivery of solute to the thick ascending limb,which helps to prevent massive fluid losses in the case of tubular dysfunc-tion, and resulting disturbances of tubular water reabsorption. This preven-tion of massive fluid loss after renal ischemia has been phrased ‘‘acute renalsuccess’’ rather than acute renal failure [7].

Regulation of blood flow

Under normal circumstances, renal blood flow and glomerular filtrationrate are maintained by autoregulation (ie, at a constant rate over a widerange of renal perfusion pressures) [8]. The afferent arterioles dilate andthe efferent arterioles constrict in response to a decrease in perfusion


pressure to maintain the transglomerular pressure. Only when the mean ar-terial pressure drops below 70 mm Hg do transglomerular pressure andGFR decrease as well. It is thought that afferent arterioles contain myogenicstretch receptors that cause vasoconstriction in response to stretch, second-ary to an increase in perfusion pressure. However, the exact mechanism ofrenal autoregulation is not completely understood [9,10]. A number of me-diators play a role in the modulation of renal blood flow, including angio-tensin II, dopamine, vasopressin, prostaglandins, atrial natriuretic peptide(ANP), endothelin, nitric oxide, and adenosine.

If renal perfusion pressure decreases because of hypotension or hypovole-mia, specialized cells in the afferent arteriole release renin from secretory

Fig. 2. Tubular electrolytes and fluid transport.


granules. Renin is also secreted by the activation of the sympathetic-adrenalsystem and as a response to increased chloride concentrations in the distaltubule [11].

Renin converts circulating angiotensinogen, produced primarily in theliver, to angiotensin I. Angiotensin I is then converted to angiotensin IIby angiotensin-converting enzyme (ACE). ACE concentrations are highestin the lung but are also present in the vascular endothelium, kidneys, adre-nal glands, and brain [9,10]. At low levels, angiotensin II decreases corticalperfusion and constricts efferent arterioles, allowing the GFR to remain pre-served [12]. At higher concentrations, angiotensin II results in systemic va-soconstriction and water and sodium reabsorption in the proximal tubulethrough the angiotensin II receptors, AT-1 and AT-2, helping to restorenormotension and euvolemia. It constricts the glomerular mesangium anddecreases the effective surface area available for filtration, leading to a reduc-tion in GFR. It also increases the sensitivity of the afferent arteriole to thevasoconstrictor effects of the juxtaglomerular apparatus [13–15].

Excessive decreases of GFR by angiotensin II are prevented in two ways:(1) angiotensin II stimulates prostaglandin release from glomeruli, causingrenal vasodilation; (2) at high concentrations it has a more substantial effecton efferent arterioles than on afferent arterioles, thereby maintaining glo-merular perfusion pressure. Even at very low concentrations, angiotensinII stimulates the release of aldosterone from the adrenal cortex. Aldosteronecauses sodium and water retention as well as Kþ and Hþ secretion in thedistal and collecting tubules (Fig. 3).

The kidney receives sympathetic innervation through autonomic fibers ofthe celiac plexus, but has no parasympathetic innervation. Adrenergic stim-ulation, by way of the neuronal release of circulating catecholamines as a re-sponse to stress or hypovolemia, causes alpha-mediated vasoconstriction.Beta-2–sympathetic receptors are rarely found in the kidney. Low concen-trations of epinephrine and alpha stimulation constrict efferent arteriolesand preserve GFR, whereas higher concentrations constrict afferent arteri-oles and decrease GFR [16]. Adrenergic stimulation also causes the releaseof renin and further vasoconstriction, mediated by way of angiotensin II[17]. Direct neurohormonal adrenergic stimulation of the nephron causes in-creased sodium reabsorption, which results in water retention.

DA-1 dopamine receptors are located in the splanchnic and renal vascu-lature and in the proximal tubules. Activation of DA-1 receptors causes re-nal vasodilation and sodium reabsorption, resulting in increased renal bloodflow, GFR, and natriuresis [18,19]. DA-2 receptors are presynaptic postgan-glionic receptors that inhibit the release of norepinephrine, leading to sys-temic vasodilation without a direct renal effect.

Arginine-vasopressin (AVP) is produced in the anterior hypothalamusand then transported to granules of the posterior pituitary gland. Secretionof AVP is regulated by hypothalamic osmo-receptors. When plasma osmo-lality increases to more than approximately 280 mmol/l, AVP is released to


aid in the maintenance of intravascular volume. There are two types of va-sopressin receptors. Stimulation of renal V2 receptors by AVP in the collect-ing ducts causes increased water reabsorption and urine osmolality,effectively decreasing urine output. At higher concentrations, AVP stimu-lates V1 receptors, located in vascular smooth muscle and the vasa recta,and results in vasoconstriction [20,21].

Prostaglandins help regulate intrarenal blood flow distribution. Phospho-lipase A2 converts phospholipids into arachidonic acid, which is then con-verted by cyclooxygenase 1 to PGG2. PGG2 can be converted into threedifferent prostaglandins: PGD2, PGE2 and PGI2 (prostacyclin). These pros-taglandins cause direct vasodilation and maintain renal cortical blood flowin response to hypotension, catecholamine release, angiotensin II, and AVP.They modulate the effects of these renal vasoconstrictors, acting as

Fig. 3. The Renin-Angiotensin-Aldosterone System. (From Palmer BF. Current Concepts:

Managing hyperkalemia caused by inhibitors of the rennin–angiotensin–aldosterone system.

N Engl J Med 2004;351:588; with permission.)


a negative feedback loop, to prevent excessive renal vasoconstriction. Pros-taglandins also cause natriuresis by reducing sodium reabsorption in thethick ascending loop and the collecting duct.

At baseline, prostaglandin levels are low, and blockade of cyclooxygenasein healthy subjects by nonsteroidal anti-inflammatory drugs (NSAIDs) hasa significant effect on GFR and renal perfusion only in those with increasedrenal vasoconstriction [15,22]. Cyclooxygenase 2 mediates the formation ofPGF2 and TXA2 (thromboxane), which act as vasoconstrictors. They con-strict mesangial cells through TXA2 receptors, thereby reducing the areaavailable for filtration and GFR, and decreasing renal blood flow [23].

ANP is secreted from myocardial cells of the atria in response to stretch,secondary to hypervolemia [24,25]. ANP release causes systemic vasodila-tion by way of the activation of cam. It blocks Naþ reabsorption in theproximal tubule and increases GFR, with no effect on renal blood flow.ANP inhibits renin, aldosterone, and AVP secretion, resulting in increasedurine output and sodium excretion [26–28].

Brain natriuretic peptide was detected initially in the central nervous sys-tem, but is also secreted by cardiac ventricles, as a response to stretch. It hasan effect similar to ANP, but is much less potent [29,30]. C-natriuretic pep-tide, produced by vascular smooth muscle and the kidney, has been identi-fied as well. Its role and significance in renal function remain uncertain(Fig. 4) [26].

Endothelin is released by vascular endothelial cells and acts as a potentvasoconstrictor. This vasoconstrictor effect is attenuated by the endothelin-induced release of prostaglandins. In postischemic states, endothelin maybe released from damaged endothelium and can effectively worsen renal per-fusion and renal failure [27,28].

Nitric oxide is formed by endothelial nitric oxide synthase and acts bymediating an increase of intracellular cyclic GMP, which relaxes smoothmuscle and inhibits inflammation. Nitric oxide seems to modulate the vaso-constrictor effect of endothelin and angiotensin II, and protects against ex-cessive renal vasoconstriction [31,32].

Adenosine appears to be a mediator of juxtaglomerular feedback that de-creases GFR in response to an increase in tubular chloride concentration.Four subtypes of adenosine receptors are identified in the kidney. Depend-ing on the subtype, adenosine causes increases or decreases in renin releaseand results in either vasoconstriction or vasodilation. Stimulation of A1 orA2 adenosine receptors before ischemic injury has been shown to be protec-tive [33–35].

Hematopoiesis

Another function of the kidney is to aid in hematopoiesis. Interstitial re-nal cells and proximal tubular cells release erythropoietin in response to de-creased oxygen delivery. The kidney produces 90% of erythropoietin, and


the liver produces the remaining 10%. Circulating erythropoietin binds toa receptor on the surface of erythrocyte precursor cells, inducing their pro-liferation and maturation to erythrocytes.

A reduction in renal blood flow leads to a compensatory decrease in theactivity of active tubular transport systems, which are metabolically verycostly. This decrease allows renal oxygen content to remain relatively stablein the kidney, even with decreased renal blood flow. Erythropoietin releasedepends on changes in arterial oxygen content and is independent of varia-tions in renal blood flow [36,37].

Evaluation of renal function

No sensitive and specific marker evaluates both tubular and glomerularfunction. Most clinically used tests evaluate the ability of the kidney to clearplasma of a specified substance. The glomerular filtration rate allows one toquantify the glomeruli’s ability to filter fluid. Rather than measuring the

Fig. 4. Physiology of the natriuretic peptides. (FromWilkins MR. The natriuretic peptides fam-

ily. Lancet 1997;349:1308; with permission.)


amount of filtrated fluid directly, one can measure the ability of the kidneyto clear the plasma of a substance easily filtered into the Bowman space. Ide-ally, this substance would undergo neither tubular secretion nor reabsorp-tion, to allow for the measurement of only glomerular function.

Creatinine is a product of skeletal muscle and dietary meat metabolism.Its production is relatively constant, it is filtered freely across the glomerularmembrane, and it is not absorbed or metabolized in the tubules. However,tubular secretion accounts for 15% of urinary creatinine [38]. If one ignoresthe tubular secretion, creatinine clearance should therefore equal GFR:

CLCreatinine ¼ GFR

¼ CLUrine creatinine � Urine flow

CLPlasma creatinine

; where CL is clearance

Normal values for GFR are approximately 95 � 20 mL/min in femalesand 120 � 25 mL/min in males. Urine flow is measured during either2- or 24-hour intervals [39]. A calculated creatinine clearance usually over-estimates GFR, because it does not take into account secreted creatinine.

Two formulas allow for the estimation of creatinine clearance when theserum creatinine concentration is known: the Modification of Diet in RenalDisease [40] and Cockroft-Gault equation [41]. The Cockroft-Gault equa-tion corrects for gender, age, and body mass:

CLCreatinine ðml=minÞ ¼ ð140�AgeÞ�Lean body weightðkgÞCðmg=dLÞ� 72

ð� 0:85 for femalesÞ; whereC is creatinine

The estimation of creatinine clearance is only appropriate in a steadystate, because serum creatinine does not reflect acute changes in GFR.

Inulin is neither secreted nor reabsorbed in the renal tubules and is fil-tered freely across the glomerular membrane. Therefore, inulin clearanceequals GFR, unlike creatinine clearance, which overestimates GFR. Inulinmust be given as an intravenous infusion to achieve steady state beforeplasma and urine levels are measured. Inulin clearance is used rarely for clin-ical use, because the technique is cumbersome.

Tests of tubular function are an attempt to differentiate between prerenalazotemia and intrinsic renal failure. The kidney loses the ability to concen-trate urine appropriately with impaired tubular function. During periods ofhypotension, the proximal and distal tubules reabsorb most of the filteredsodium and water, resulting in urine with low sodium concentration. If tu-bular function deteriorates, as in acute tubular necrosis (ATN), the distal tu-bule is unable to reabsorb sodium appropriately, and urine sodiumincreases. Urinary sodium excretion of greater than 40 mEq/L suggests animpaired ability of the renal tubules to resorb sodium. Concomitant use


of diuretics impairs tubular sodium reabsorption and increases urine sodiumconcentration, even if tubular function is intact.

Functioning kidneys attempt to preserve water and sodium as a responseto hypovolemia. The kidneys are capable of increasing urine osmolality to atleast 1.5 times that of the plasma osmolality in response to hypovolemia. Inthe case of tubular damage, this concentrating ability is impaired and urineosmolality may remain low, despite a high plasma osmolality.

Markers of renal injury

Tests of renal function often detect deterioration in function long afterthe causative insult. In order for therapy to be successful, it must be imple-mented early in the course of renal injury. Early markers of renal injury aretherefore important. Neutrophil gelatinase-associated lipocalin (NGAL) isa protein that is not detectable in the urine of normal subjects or thosewith chronic renal insufficiency. However, shortly after even a small renalinjury, whether it is an ischemic or nephrotoxic insult, urinary NGAL in-creases significantly [42–44]. Additional studies are needed to evaluate ifNGAL will be clinically useful and will allow the early detection and treat-ment of renal injury.

Effect of anesthetics and surgery on renal function

Surgical stress causes the release of catecholamines, renin, angiotensin,AVP, and other vasoactive substances [45]. A small surgical stress may re-sult in an increase in sodium and water reabsorption, whereas a high degreeof surgical stress can lead to a redistribution of renal blood flow, afferentarteriolar vasoconstriction, and a decrease in GFR. The end-result is a de-crease in urine output that can affect even adequately hydrated patients.

Usually, general anesthetics do not impair renal blood flow or function inany considerable way; however, if the anesthetic affects systemic hemody-namics by causing hypotension or a decrease in systemic vascular resistanceor cardiac output, this can lead to a reduction in renal perfusion. Sevoflur-ane undergoes significant metabolism that results in the production of inor-ganic fluoride. Inorganic fluoride is a tubular nephrotoxin, but theadministration of sevoflurane has not been shown to cause any clinically sig-nificant renal injury in humans [45]. Enflurane has been implicated in signif-icant renal injury; it is no longer widely used.

Positive pressure ventilation used during general anesthesia can decreasecardiac output, renal blood flow, and GFR, especially when used in combi-nation with high levels of positive end-expiratory pressure. Decreased car-diac output leads to a release of catecholamines, renin, and angiotensinII, and the activation of the sympathoadrenal system, which results in de-creased renal blood flow. Insufflation of the abdomen during laparoscopic


surgery has a similar effect on renal blood flow and GFR. Additionally, dur-ing laparoscopic surgery, the increased intra-abdominal pressure is transmit-ted directly to the kidney and results in a further reduction of renal bloodflow.

Regional anesthesia that achieves a sympathetic blockade of levels T4 toT10 also blocks the sympathetic innervation of the kidney and thereby sup-presses the neurohormonal effect of surgical stress. If systemic vascular re-sistance and cardiac output are maintained during spinal or epiduralanesthesia, the resulting sympathetic blockade can cause an attenuation ofcatecholamine-induced renal vasoconstriction. Clinically, there is no evi-dence of a clear advantage of regional over general anesthesia with regardto renal function and protection.

Anesthetic management

Many drugs commonly used during anesthesia depend to some degree onrenal excretion for elimination. When devising an anesthetic plan for a pa-tient who has renal insufficiency or renal failure, the anesthesiologist musttake this into consideration. Patients with renal disease display an increasedsensitivity to barbiturates and benzodiazepines as a result of decreased pro-tein binding. Morphine and meperidine should be used judiciously becausethey have active metabolites and may have prolonged activity in the settingof renal insufficiency. Succinylcholine can be used if the serum potassiumconcentration is normal at the time of induction. However, if the potassiumconcentration is not known, it is best to avoid its use, if possible. Cisatracu-rium and atracurium are metabolized by enzymatic ester hydrolysis andHofmann elimination and are unaffected by renal failure. The reversalagents rely on renal excretion; therefore, their effects are prolonged, so thereversal of neuromuscular blockade should not be an issue in the settingof renal failure. Many antimicrobial agents must be dosed to account for re-nal dysfunction. NSAIDs should be avoided because they may aggravate re-nal insufficiency.

Patients with end-stage renal disease presenting for elective surgeryshould be dialyzed ideally the day of, or the day before, surgery to optimizethe patient’s fluid and metabolic status. Intravenous fluids are usually keptto a minimum for small procedures, such as catheters for dialysis access ora hernia repair. Normal saline, 0.9% NaCl, is usually the crystalloidof choice, although lactated ringers may be used safely, as it has only4 mEq/L of potassium [46]. During larger procedures, it may be advisableto place a central line to help guide intraoperative fluid management. Ifblood transfusion is necessary, one should keep in mind both the volumeload and potassium load. Frequent monitoring of electrolytes and bloodgases is often necessary. If a patient becomes hyperkalemic, severely aci-dotic, or volume overloaded, it may be necessary to inform the dialysisunit that the patient may need urgent dialysis in the Post Anesthesia Care


Unit (PACU) after surgery. In the operating room, medical managementdincluding calcium, insulin, and glucose for the treatment of hyperkalemia,and hyperventilation to improve metabolic acidosisdcan be invaluable.

Chronic renal failure

Chronic renal failure results from a progressive and irreversible loss offunctioning nephrons, with a resultant decrease in GFR. The many causesof chronic renal failure include diabetic nephropathy, hypertensive nephro-sclerosis, progressive glomerulonephritis, polycystic kidney disease, andpharmacologic toxicity, to name a few. Patients have decreased renal reservebut remain asymptomatic with at least 40% functioning nephrons. Renal in-sufficiency results when only 10% to 40% of nephrons are functioning,which puts the patient at risk for renal failure with even small stresses, be-cause there is virtually no renal reserve. Loss of more than 90% of function-ing nephrons results in renal failure, necessitating either hemodialysis (HD)or peritoneal dialysis (PD). Patients have moderate renal insufficiency whencreatinine clearance decreases to less than 40 ml/min. Patients exhibit renalfailure with a creatinine clearance of less than 25 ml/min and will be dialysisdependent at creatinine clearances of less than 10 ml/min.

Patients with renal failure can develop a myriad of metabolic abnormal-ities, including hyperkalemia, hyperphosphatemia, hypocalcemia, uremia,and metabolic acidosis, to name a few. Cardiac dysrhythmias often resultbecause of the metabolic disarray. An obvious problem in chronic renal fail-ure is fluid overload, which can lead to congestive heart failure. Other prob-lems include anemia, platelet dysfunction, leukocyte dysfunction, andpruritus.

Acute renal failure

Acute postoperative renal failure is associated with a high morbidity andmortality [47,48]. Ahlstrom and colleagues [49] showed recently that criti-cally ill patients requiring renal replacement therapy have a 28% mortality.Various mechanisms can lead to acute renal failure. Factors such as pre-existing renal insufficiency, advanced age, diabetes, and cardiovascular co-morbidities increase the risk of renal failure significantly. Perioperativedisturbances of renal blood flow secondary to relative fluid depletion, andmild degrees of hypotension that are well tolerated in the healthy patient,can tip the balance toward acute renal failure in the compromised patient(Fig. 5).

The absence of a universally accepted definition of renal failure makes thecomparison of studies very difficult [50]. Commonly used definitions ofacute renal failure include an increase in creatinine of more than 0.5 mg/dlor by more than 50% over baseline; a decrease in GFR by more than


50% or to less than 30 mg/min; and renal failure requiring renal replacementtherapy [51]. The Second International Consensus Conference of the AcuteDialysis Quality Initiative recently suggested a five-step definition of acuterenal failure, the RIFLE criteria, that is based on either GFR or urinary out-put (Fig. 6.) This classification includes the spectrum of renal failure froma mild reduction in renal function to end-stage kidney disease, and may bethe first step toward a universally accepted definition of renal failure [52].

Prerenal injury

Prerenal failure is defined as a decrease in renal perfusion with preservedtubular and glomerular function [53]. Decreases in renal perfusion due tohypovolemia or cardiogenic shock lower the transglomerular pressureand, consequently, GFR. Sodium and water reabsorption in the thick as-cending limb increases, causing increased medullary oxygen demand ata time when oxygen delivery is decreased. Sodium reabsorption is increasedby multiple mechanisms, including increased renin and AVP release, and ac-tivation of the sympathetic system.

The effects of prerenal failure are easily reversible when renal perfusion isreestablished, unless it is prolonged or the kidney suffers additional insults.Concomitant administration of NSAIDs or ACE inhibitors prevents the re-lease of protective prostaglandins and can convert prerenal failure into acuterenal failure [54]. The transformation of prerenal into intrinsic renal failureis a continuum; prolonged prerenal states or additional insults to the kidneymay lead to ischemic renal failure.

Fig. 5. Classification of acute renal failure. (From Lameire N, Van Biesen W, Vanholder R.

Acute renal failure. Lancet 2005;365(9457):418; with permission.)


Intrinsic renal failure

Renal hypoperfusion for an extended period causes depletion of ATPstores, especially in the metabolically active thick ascending limb in the me-dulla. The tubular brush border disappears, tubular cells lose their polarity,and Na-K transport systems move from the basolateral to the tubular sideof the cells [55]. Further ischemia causes tubular cell death, through eithernecrosis or apoptosis. The sloughing of cellular material into the tubularlumen results in cast obstruction and leads to ATN. Backpressure to theglomerulus and activation of the juxtaglomerular apparatus secondary toincreased intratubular NaCl concentration decreases GFR and ultimatelyprevents massive fluid loss after ischemia [38].

Cell injury leads to depletion of energy stores and the release of metabo-lites of ATP, including inosine, adenosine, and hypoxanthine, that releaseactivated oxygen species. Activated oxygen radicals further worsen tubularcell damage and renal failure [56]. Circulating neutrophils adhere to vascularendothelium in the kidney after ischemic injury, through the expression ofendovascular intracellular adhesion molecules. Activated leukocytes migrateinto the extravascular tissue by way of chemotaxis, controlled by comple-ment activation. Neutrophils, macrophages, and T cells release activated

Fig. 6. RIFLE criteria for acute renal failure. (From Bellomo R, Ronco C, Kellum JA, et al.

Acute renal failure - definition, outcome measures, animal models, fluid therapy and informa-

tion technology needs: the Second International Consensus Conference of the Acute Dialysis

Quality Initiative (ADQI) Group. Crit Care 2004;8:R206; with permission.)


oxygen species, proteases, and myeloperoxidases, which cause cell lysis andincrease vascular permeability, further propagating cellular injury [57].

Proximal tubules are able to undergo repair, regeneration, and prolifera-tion after damage. During the reparative phase, growth factors are ex-pressed that stimulate the spread and proliferation of new epithelialtubular cells [58]. These cells then repolarize, create a new brush border,and resume their function [59].

Acute Renal Failure (ARF) occurs in 23% of patients with severe sepsisand in 51% with septic shock and positive blood cultures [60]. The hallmarkof sepsis is vasodilation with a resultant decrease in systemic vascular resis-tance, leading to a compensatory increase in cardiac output. Potent renalvasoconstrictors, including norepinephrine, angiotensin II, and thrombox-ane, are released in sepsis to compensate for these hemodynamic alterations,which results in a decrease in GFR and renal ischemia. Afferent arteriolarvasoconstriction and efferent arteriolar vasodilatation cause a decrease intransglomerular perfusion pressure and GFR [61,62]. Studies were able toshow that renal failure occurs in septic, hyperdynamic shock, even if renalblood flow is maintained [63,64].

Plasma vasopressin levels are low in profound septic shock and theadministration of vasopressin (0.03–0.1 U/kg/hr) can significantly improvevasomotor tone and renal perfusion, and increase GFR. In vasopressin-deficient states, such as sepsis, there is a substantial increase in the vasopressinsensitivity of endovascular cells, and replacement of vasopressin improvesvasomotor tone and renal perfusion dramatically [21,65,66].

Cardiopulmonary bypass (CPB) can cause significant decreases in bloodpressure with nonpulsatile flow. Additionally, CPB results in high levels ofcirculating catecholamines, the stimulation of renin and angiotensin II pro-duction, and the release of other vasoactive substances that generate pro-found renal vasoconstriction and result in the reduction of GFR [67,68].No agent to date has proved to provide adequate protection from renal in-jury during CPB. Early postoperative serum creatinine may be misleading ifused to assess renal function. Serum creatinine often decreases within thefirst 24 hours after cardiac surgery, even if the GFR acutely deterioratesbecause of the hypervolemia and hemodilution that can result after CPB.Acute renal failure after cardiac surgery is surprisingly rare, with an inci-dence of between 1% and 15%, but if it occurs it increases mortality tomore than 60% [69–72].

Patients undergoing open aortic aneurysm repair are at increased risk forATN. Infrarenal aortic cross clamping leads to a reduction in renal bloodflow of up to 40%, secondary to an increase in renal vascular resistance[73], thought, in turn, to be secondary to the decrease in cardiac output dur-ing the cross clamp. During suprarenal cross clamping of the aorta, it is im-portant to keep the ischemic time to the kidneys as brief as possible. It is alsoimperative to try to maintain adequate volume status and hemodynamics, tominimize the ischemic insult.


Hepatorenal syndrome presents with deteriorating renal function and oli-guria in the presence of hepatic cirrhosis and ascites. Splanchnic vasodila-tion is thought to cause a reflex renal vasoconstriction that initially leadsto prerenal azotemia and then progresses to acute renal failure [74]. Patientsusually do not respond to any treatment unless they undergo liver transplan-tation. Recently, the vasopressin analog terlipressin has been used success-fully in some cases to treat hepatorenal syndrome [75]. Concomitantepisodes of sepsis or bleeding further exacerbate renal failure in these patients.

Nephrotoxic renal failure

Agents impairing renal perfusion rarely cause acute renal failure in theabsence of concomitant renal insufficiency, hypovolemia, advanced age, di-abetes, or other additional nephrotoxic insults. Common pharmacologicagents frequently implicated in nephrotoxic renal failure include NSAIDs,ACE inhibitors, calcineurin inhibitors, radiocontrast agents, and antibiotics.

NSAIDs inhibit prostaglandin synthesis by blocking cyclooxygenase 1.Use of NSAIDs in states of renal hypoperfusion leaves angiotensin-II–induced vasoconstriction unopposed and causes medullary renal ischemia[76]. Therefore, NSAIDs, which rarely cause renal failure in a healthy indi-vidual, are nephrotoxic in combination with pre-existing renal insufficiencyor when renal perfusion has already decreased, as in heart failure, liver fail-ure, and hypovolemia [77].

ACE inhibitors block the transformation of angiotensin I into angioten-sin II. In normovolemic patients, ACE inhibition may improve renal func-tion and prevent decreases in GFR caused by high levels of angiotensin II[78]. ACE inhibitors have been shown to be beneficial in the long-term treat-ment of hypertension, because improved cardiac output by way of afterloadreduction in hypertensive heart failure results in improved renal perfusion[78,79]. However, in the face of hypovolemia or pre-existing renal dysfunc-tion, ACE inhibition can result in a diminished vasoconstrictor effect of an-giotensin II on efferent arterioles, thereby decreasing GFR and potentiallyinducing acute renal failure.

Aminoglycosides accumulate in the renal cortex and can lead to ATN.This toxicity is usually reversible if the drug is discontinued. High peaklevels with once daily dosing is less likely to cause damage, comparedwith sustained high levels seen with more frequent dosing [80].

Calcineurin inhibitors, such as tacrolimus, FK-506, and cyclosporine,cause renal vasoconstriction though the release of vasoactive substancessuch as thromboxane and endothelin. Hypertension and renal failure are es-pecially common during the induction phase for immunosuppression aftertransplantation [81]. Most patients treated with cyclosporine after trans-plantation will develop a modest degree of renal insufficiency, and approx-imately 16.5% of all patients receiving nonrenal solid organ transplants willdevelop chronic renal failure [82].


Contrast-induced nephropathy is the third-leading cause of hospital-acquired renal failure [83]. The increase of endovascular surgical procedureswill most likely add to this problem. Radiocontrast agents cause renal vaso-constriction and direct tubular toxicity [84]. Direct toxicity to nephrons isrelated to the osmolality of the contrast; therefore, low osmolar contrastshould be used in at-risk patients, if possible. Risk factors for contrast-induced renal failure include renal insufficiency, volume depletion, diabetes,pre-existing hypertension, periprocedure hypotension, and large volumes ofcontrast. Adequate hydration before exposure to radiocontrast agents sig-nificantly decreases the incidence of renal failure. If possible, nephrotoxicdrugs should be discontinued before a procedure requiring contrast. Multi-ple treatment regimens are used for preventing radiocontrast-induced renalfailure. Of these, bicarbonate infusion and acetylcysteine have shown prom-ising results [85,86].

Rhabdomyolysis is the third-leading cause of acute renal failure in theUS. Trauma, crush injury, malignant hyperthermia, neuroleptic malignantsyndrome, diabetic ketoacidosis, and illicit drugs are some causes of rhabdo-myolysis. Severe injury to skeletal muscle causes muscle necrosis and the re-lease of myoglobin. Myoglobin is filtrated through the glomerular basementmembrane and then converted to ferrihematin in the acidotic filtrate. Ferri-hematin precipitates and obstructs renal tubules, leads to cytokine release,and may result in acute renal failure. Additionally, rhabdomyolysis causesrenal vasoconstriction by increasing circulating levels of renin and activatingthe sympathetic nervous system. Treatment consists of aggressive hydrationto maintain a high urine output to flush out precipitates, and alkalinizationof the urine to increase the solubility of myoglobin [87]. Mannitol may beuseful because it is a renal vasodilator, osmotic diuretic, and free radicalscavenger, and it prevents the precipitation of myoglobin.

Postrenal obstruction

In all cases of oliguria, postrenal obstruction should be ruled out. Mostcommonly, it is caused by prostatic hypertrophy and blockage of the ure-thra, but can also be caused by bladder cancer, calculi, hematoma, or ob-struction of the ureter in patients with a solitary kidney [38]. Occluded orkinked urinary catheters are frequent culprits that cause obstruction of urineflow. Postrenal obstruction may be accompanied by flank or abdominalpain and urgency. Prompt relief of the obstruction usually alleviates ob-structive renal failure quickly, although prolonged obstruction can resultin thinning of the renal cortex and irreversible damage.

Renal protection: prevention and treatment of acute tubular necrosis

The best strategy to prevent ATN is to ensure adequate hydration andhemodynamics to avoid renal ischemia. To date, no agent provides adequate


renal protection in cases of major renal injury. Renal protective strategiesare usually implemented after renal function is already deteriorated. Moststudies evaluating renal protective strategies are underpowered and a signif-icant effect cannot be concluded. Renal failure requiring renal replacementtherapy after renal injury is rare, and requires a large number of subjects if itused as an endpoint. More frequent endpoints, such as changes in serum cre-atinine, commonly have only a weak correlation with outcomes and are of-ten not useful as substitute endpoints.

The definition of acute renal failure needs to be standardized to allow thecomparison of studies and the inclusion of clinical trials into meta-analysis.Further understanding of the molecular mechanisms of renal response to,and recovery after, injury that surpasses mechanistic models such as renalblood flow may allow for the development of new renal protective agentsin the future.

Dopamine increases renal blood flow and induces natriuresis. It has longbeen thought to be renal protective at low doses of 2 to 3 mcg/kg/min.Bellomo and colleagues [87] showed in a large randomized placebo-controlled study of critically ill subjects at risk for renal failure that low-dose dopamine had no protective effects. Other studies have confirmed thisfinding [88,89]. Dopamine may be valuable to increase renal perfusion inlow cardiac output states, but it does not have any intrinsic renal protectiveproperties.

Fenoldopam is a selective D-1 dopamine receptor agonist that causes sys-temic vasodilation and leads to increased renal blood flow and natriuresis. Itis used for the treatment of hypertension and one study has shown a renalprotective effect in abdominal aortic aneurysm surgery [90].

Diuretics increase urinary output, and their theoretic benefit in the treat-ment of ATN is to flush out necrotic cell debris from tubules after an ische-mic insult, which would otherwise obstruct these tubules. Cantarovich andcolleagues [91] found no effect of furosemide on mortality or morbidity ina large randomized trial of subjects with acute renal failure.

Mannitol is an osmotic diuretic that increases intratubular osmolalityand prevents water reabsorption. Its osmotic effect may also decrease tubu-lar cell swelling after an ischemic injury. Its clinical effectiveness has notbeen proven in a large randomized trial.

Acetylcysteine is an antioxidant that has been shown to prevent acute re-nal failure after exposure to radiocontrast agents [92,93]. Others failed toshow an effect [93]. It appears to be more effective if given before the injuryand in patients who are at high risk for renal failure.

Renal replacement therapy

When renal function deteriorates to a degree that the kidneys are unableto maintain orthostasis, fluid, electrolytes, and metabolic products accumu-late and end-stage renal disease ensues. The patient will now require renal


replacement therapy or a renal transplant. The two modes of commonlyused renal replacement therapy are hemodialysis (HD) and peritonealdialysis (PD).

During HD, blood is drawn from a venous access device and flows acrossa semipermeable membrane. On the other side of this membrane, a dialysatefluid with a similar electrolyte concentration as the blood runs in oppositedirection. Electrolytes and waste products transverse the membrane fromthe blood to the dialysate fluid and alkali moves from the dialysate to theblood by way of concentration gradients. Negative pressure on the dialysateside results in fluid removal from the patient. The membrane consists of hol-low fibers with a surface area of 1 to 1.8 m2. Purified blood is returned to thepatient. Vascular access is often achieved with an arteriovenous fistula orgraft that is created in the arm or by way of a venous catheter [94,95].

The duration of HD depends on the flow of the dialysate fluid across themembrane. Unfortunately, high flow rates cause rapid changes in electro-lytes and fluid status that are frequently not well tolerated. HD is usuallyscheduled three times a week for 4 to 6 hours. Hypotension is common dur-ing HD and is not explained entirely by intravascular fluid removal. The ad-equacy of HD is assessed by the removal of urea, and inadequate HD is notuncommon.

Patients receiving intermittent HD have a high mortality, and approxi-mately 5% of these patients die each year primarily because of cardiovascu-lar complications and infections [94]. The short-term mortality may be lowerfor younger patients when continuous or nightly intermittent abdominal PDis used instead of HD. PD is labor intensive and requires a high degree ofcompliance on the patient’s part, but it reduces the effects of sudden fluidshifts during dialysis and eliminates the need to stay in a dialysis facilitythree times a week.

PD utilizes the peritoneum as the semipermeable membrane. Dialysatefluid is inserted through a catheter in the abdominal wall into the peritonealspace and is usually exchanged three times a day, while the fluid dwells in theabdomen at night. Many other intermittent or continuous schedules havebeen developed with none having a clear advantage over the other. Perito-nitis is a common complication of PD.

Acute renal failure in critically ill patients is frequently due to shock andhypotension, and these patients often do not tolerate the rapid fluid shiftsassociated with intermittent HD. Continuous veno-venous hemodialysis(CVVHD) or filtration (CVVHF), allows renal replacement therapy in thehemodynamically unstable patient. Blood is pulled from a venous catheterthat crosses a dialysis membrane and is then returned to the patient usinga different lumen of the same catheter. The indications for CVVHD in acuterenal failure include volume overload, metabolic acidosis, hyperkalemia,uremic encephalopathy and intoxication with a dialyzable substance.

When using hemofiltration, only fluid is removed, whereas CVVHD alsoremoves electrolytes and waste products and correcting metabolic acidosis.


Because fluid removal is continuous and can be adjusted to the intravascularvolume status of the patient, CVVHF/D does not cause any significant he-modynamic derangements and can be used even in the setting of multisys-tem organ failure with high doses of vasopressors [96].

Summary

The kidney is a remarkable organ whose functions include maintainingfluid and electrolyte balance, excreting metabolic waste products, and con-trolling vascular tone. Blood flow within the kidney is very heterogeneous,which places the metabolically active medulla at high risk for ischemic in-jury. A number of mediators play a role in the modulation of renal bloodflow, including angiotensin II, dopamine, vasopressin, prostaglandins, atrialnatriuretic peptide, endothelin, and nitric oxide. Early markers of renal in-jury elicit strong interest, although currently there is no reliable markeravailable.

Surgery causes the release of catecholamines, renin, angiotensin, andAVP that lead to a redistribution of renal blood flow and a decrease inGFR. Additionally, general anesthesia often results in some degree of hypo-tension and depressed cardiac output, which further reduces renal perfusionand potentially jeopardizes renal function. A careful anesthetic plan is im-perative in the patient with renal insufficiency or failure because acute renalfailure in the perioperative period is associated with a high morbidity andmortality. Factors including advanced age, diabetes, underlying renal insuf-ficiency, and heart failure place a patient at high risk for developing acuterenal failure. It is imperative to maintain euvolemia, normotension, and car-diac output, and to avoid nephrotoxic agents to optimize renal blood flowand renal perfusion as the best prevention of renal dysfunction. Furtherstudies are needed to establish if any therapies exist to prevent or treat renaldysfunction effectively.

References

[1] KanwarYS, Liu ZZ,KashiharaN, et al. Current status of the structural and functional basis

of glomerular filtration and proteinuria. Semin Nephrol 1991;11:390–413.

[2] JacobsonHR. Functional segmentation of themammalian nephron. Am J Physiol 1981;241:

F203–18.

[3] Sands JM,Kokko JP. Current concepts of the countercurrent multiplication system.Kidney

Int Suppl 1996;57:S93–9.

[4] Sands JM, Kokko JP. Countercurrent system. Kidney Int 1990;38:695–9.

[5] HollenbergNK,EpsteinM,Rosen SM, et al. Acute oliguric renal failure inman: evidence for

preferential renal cortical ischemia. Medicine (Baltimore) 1968;47:455–74.

[6] Schrier RW. Effects of adrenergic nervous system and catecholamines on systemic and renal

hemodynamics, sodium andwater excretion and renin secretion.Kidney Int 1974;6:291–306.

[7] Thurau K, Boylan JW. Acute renal success. The unexpected logic of oliguria in acute renal

failure. Am J Med 1976;61:308–15.


[8] Navar LG. Renal autoregulation: perspectives from whole kidney and single nephron

studies. Am J Physiol 1978;234:F357–70.

[9] Kifor I, Moore TJ, Fallo F, et al. Potassium-stimulated angiotensin release from superfused

adrenal capsules and enzymatically dispersed cells of the zona glomerulosa. Endocrinology

1991;129:823–31.

[10] Schunkert H, Ingelfinger JR, HirschAT, et al. Evidence for tissue-specific activation of renal

angiotensinogenmRNAexpression in chronic stable experimental heart failure. J Clin Invest

1992;90:1523–9.

[11] Skott O, Jensen BL. Cellular and intrarenal control of renin secretion. Clin Sci (Lond) 1993;

84:1–10.

[12] Ichikawi I, Harris RC. Angiotensin actions in the kidney: renewed insight into the old

hormone. Kidney Int 1991;40:583–96.

[13] Casellas D, Moore LC. Autoregulation and tubuloglomerular feedback in juxtamedullary

glomerular arterioles. Am J Physiol 1990;258:F660–9.

[14] Goodfriend TL, Elliott ME, Catt KJ. Angiotensin receptors and their antagonists. N Engl

J Med 1996;334:1649–54.

[15] Romero JC, Knox FG. Mechanisms underlying pressure-related natriuresis: the role of the

renin-angiotensin and prostaglandin systems. State of the art lecture. Hypertension 1988;11:

724–38.

[16] Myers BD, Deen WM, Brenner BM. Effects of norepinephrine and angiotensin II on the

determinants of glomerular ultrafiltration and proximal tubule fluid reabsorption in the

rat. Circ Res 1975;37:101–10.

[17] Oliver JA, Pinto J, Sciacca RR, et al. Increased renal secretion of norepinephrine and pros-

taglandin E2 during sodium depletion in the dog. J Clin Invest 1980;66:748–56.

[18] Jose PA, Eisner GM, Felder RA. Role of dopamine receptors in the kidney in the regulation

of blood pressure. Curr Opin Nephrol Hypertens 2002;11:87–92.

[19] Steinhausen M, Endlich K, Wiegman DL. Glomerular blood flow. Kidney Int 1990;38:

769–84.

[20] Holmes CL, Landry DW, Granton JT. Science review: Vasopressin and the cardiovascular

system part 1–receptor physiology. Crit Care 2003;7:427–34.

[21] Holmes CL, Landry DW, Granton JT. Science Review: Vasopressin and the cardiovascular

system part 2dclinical physiology. Crit Care 2004;8:15–23.

[22] Breyer MD, Breyer RM. Prostaglandin receptors: their role in regulating renal function.

Curr Opin Nephrol Hypertens 2000;9:23–9.

[23] Welch WJ, Wilcox CS. Potentiation of tubuloglomerular feedback in the rat by thrombox-

ane mimetic. Role of macula densa. J Clin Invest 1992;89:1857–65.

[24] de Bold AJ, Borenstein HB, Veress AT, et al. A rapid and potent natriuretic response to

intravenous injection of atrial myocardial extract in rats. Life Sci 1981;28:89–94.

[25] Webster MW, Sharpe DN, Coxon R, et al. Effect of reducing atrial pressure on atrial natri-

uretic factor and vasoactive hormones in congestive heart failure secondary to ischemic and

nonischemic dilated cardiomyopathy. Am J Cardiol 1989;63:217–21.

[26] Mattingly MT, Brandt RR, Heublein DM, et al. Presence of C-type natriuretic peptide in

human kidney and urine. Kidney Int 1994;46:744–7.

[27] Kohan DE. Endothelins in the normal and diseased kidney. Am J Kidney Dis 1997;29:

2–26.

[28] Chou SY, DahhanA, Porush JG. Renal actions of endothelin: interaction with prostacyclin.

Am J Physiol 1990;259:F645–52.

[29] MukoyamaM, Nakao K, Hosoda K, et al. Brain natriuretic peptide as a novel cardiac hor-

mone in humans. Evidence for an exquisite dual natriuretic peptide system, atrial natriuretic

peptide and brain natriuretic peptide. J Clin Invest 1991;87:1402–12.

[30] Davidson NC, Struthers AD. Brain natriuretic peptide. J Hypertens 1994;12:329–36.

[31] Ito S, Carretero OA, Abe K. Nitric oxide in the regulation of renal blood flow. New Horiz

1995;3:615–23.


[32] Berthold H, Just A, Kirchheim HR, et al. Interaction between nitric oxide and endogenous

vasoconstrictors in control of renal blood flow. Hypertension 1999;34:1254–8.

[33] Weihprecht H, Lorenz JN, Schnermann J, et al. Effect of adenosine1-receptor blockade on

renin release from rabbit isolated perfused juxtaglomerular apparatus. J Clin Invest 1990;85:

1622–8.

[34] Hansen PB, Schnermann J. Vasoconstrictor and vasodilator effects of adenosine in the

kidney. Am J Physiol Renal Physiol 2003;285:F590–9.

[35] Lee HT, XuH, Nasr SH, et al. A1 adenosine receptor knockout mice exhibit increased renal

injury following ischemia and reperfusion. Am J Physiol Renal Physiol 2004;286:F298–306.

[36] Pfeilschifter J, Huwiler A. Erythropoietin is more than just a promoter of erythropoiesis.

J Am Soc Nephrol 2004;15:2240–1.

[37] Fisher JW.Erythropoietin: physiologyandpharmacologyupdate.ExpBiolMed (Maywood)

2003;228:1–14.

[38] Lameire N, Van Biesen W, Vanholder R. Acute renal failure. Lancet 2005;365:417–30.

[39] Sladen RN, Endo E, Harrison T. Two-hour versus 22-hour creatinine clearance in critically

ill patients. Anesthesiology 1987;67:1013–6.

[40] Levey AS, Bosch JP, Lewis JB, et al. A more accurate method to estimate glomerular filtra-

tion rate from serum creatinine: a new prediction equation. Modification of Diet in Renal

Disease Study Group. Ann Intern Med 1999;130:461–70.

[41] Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine.

Nephron 1976;16:31–41.

[42] Mishra J, Dent C, Tarabishi R, et al. Neutrophil gelatinase-associated lipocalin (NGAL) as

a biomarker for acute renal injury after cardiac surgery. Lancet 2005;365:1231–8.

[43] Mishra J, Mori K, Ma Q, et al. Neutrophil gelatinase-associated lipocalin: a novel early

urinary biomarker for cisplatin nephrotoxicity. Am J Nephrol 2004;24:307–15.

[44] Wagener GJM, KimM, Mori M, et al. Association between increases in urinary neutrophil

gelatinase-associated lipocalin and acute renal dysfunction after adult cardiac surgery. An-

esthesiology 2006, in press.

[45] Herd JA. Cardiovascular response to stress. Physiol Rev 1991;71:305–30.

[46] O’Malley CM, Frumento RJ, Bennett-Guerrero E. Intravenous fluid therapy in renal trans-

plant recipients: results of a US survey. Transplant Proc 2002;34:3142–5.

[47] ChertowGM.Dialysis: cost-effective ‘‘SUPPORT’’ for patients with acute renal failure. Am

J Kidney Dis 1998;31:545–8 [discussion: 548–9].

[48] Chertow GM, Lazarus JM, Paganini EP, et al. Predictors of mortality and the provision of

dialysis in patients with acute tubular necrosis. TheAuriculin Anaritide Acute Renal Failure

Study Group. J Am Soc Nephrol 1998;9:692–8.

[49] Ahlstrom A, Tallgren M, Peltonen S, et al. Survival and quality of life of patients requiring

acute renal replacement therapy. Intensive Care Med 2005;31:1222–8.

[50] Palevsky PM,Metnitz PG, Piccinni P, et al. Selection of endpoints for clinical trials of acute

renal failure in critically ill patients. Curr Opin Crit Care 2002;8:515–8.

[51] Lameire N, Hoste E. Reflections on the definition, classification, and diagnostic evaluation

of acute renal failure. Curr Opin Crit Care 2004;10:468–75.

[52] Bellomo R, Ronco C, Kellum JA, et al. Acute renal failure - definition, outcome measures,

animal models, fluid therapy and information technology needs: the Second International

Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care

2004;8:R204–12.

[53] BadrKF, Ichikawa I. Prerenal failure: a deleterious shift from renal compensation to decom-

pensation. N Engl J Med 1988;319:623–9.

[54] Blantz RC. Pathophysiology of pre-renal azotemia. Kidney Int 1998;53:512–23.

[55] Alejandro VS, Nelson WJ, Huie P, et al. Postischemic injury, delayed function and Na þ /

K(þ)-ATPase distribution in the transplanted kidney. Kidney Int 1995;48:1308–15.

[56] Nath KA, Norby SM. Reactive oxygen species and acute renal failure. Am JMed 2000;109:

665–78.


[57] Bonventre JV, Weinberg JM. Recent advances in the pathophysiology of ischemic acute

renal failure. J Am Soc Nephrol 2003;14:2199–210.

[58] LameireNH, Vanholder R. Pathophysiology of ischaemic acute renal failure. Best Pract Res

Clin Anaesthesiol 2004;18:21–36.

[59] Nigam S, Lieberthal W. Acute renal failure. III. The role of growth factors in the process of

renal regeneration and repair. Am J Physiol Renal Physiol 2000;279:F3–11.

[60] Schrier RW, Wang W. Acute renal failure and sepsis. N Engl J Med 2004;351:159–69.

[61] Badr KF, Kelley VE, Rennke HG, et al. Roles for thromboxane A2 and leukotrienes in

endotoxin-induced acute renal failure. Kidney Int 1986;30:474–80.

[62] Ravikant T, Lucas CE. Renal blood flow distribution in septic hyperdynamic pigs. J Surg

Res 1977;22:294–8.

[63] Brenner M, Schaer GL, Mallory DL, et al. Detection of renal blood flow abnormalities in

septic and critically ill patients using a newly designed indwelling thermodilution renal

vein catheter. Chest 1990;98:170–9.

[64] Di Giantomasso D, Morimatsu H, May CN, et al. Intrarenal blood flow distribution in

hyperdynamic septic shock: Effect of norepinephrine. Crit Care Med 2003;31:2509–13.

[65] Landry DW, Levin HR, Gallant EM, et al. Vasopressin pressor hypersensitivity in vasodi-

latory septic shock. Crit Care Med 1997;25:1279–82.

[66] Holmes CL, Patel BM, Russell JA, et al. Physiology of vasopressin relevant to management

of septic shock. Chest 2001;120:989–1002.

[67] Hilberman M, Myers BD, Carrie BJ, et al. Acute renal failure following cardiac surgery.

J Thorac Cardiovasc Surg 1979;77:880–8.

[68] Ip-Yam PC, Murphy S, Baines M, et al. Renal function and proteinuria after cardiopulmo-

nary bypass: the effects of temperature and mannitol. Anesth Analg 1994;78:842–7.

[69] Chertow GM, Levy EM, Hammermeister KE, et al. Independent association between acute

renal failure and mortality following cardiac surgery. Am J Med 1998;104:343–8.

[70] Chertow GM, Lazarus JM, Christiansen CL, et al. Preoperative renal risk stratification.

Circulation 1997;95:878–84.

[71] Conlon PJ, Stafford-Smith M, White WD, et al. Acute renal failure following cardiac

surgery. Nephrol Dial Transplant 1999;14:1158–62.

[72] deMoraes Lobo EM, Burdmann EA,Abdulkader RC. Renal function changes after elective

cardiac surgery with cardiopulmonary bypass. Ren Fail 2000;22:487–97.

[73] Gamulin Z, Forster A, Morel D, et al. Effects of infrarenal aortic cross-clamping on renal

hemodynamics in humans. Anesthesiology 1984;61:394–9.

[74] Gines P, Guevara M, Arroyo V, et al. Hepatorenal syndrome. Lancet 2003;362:1819–27.

[75] Uriz J, Gines P, Cardenas A, et al. Terlipressin plus albumin infusion: an effective and safe

therapy of hepatorenal syndrome. J Hepatol 2000;33:43–8.

[76] Clive DM, Stoff JS. Renal syndromes associated with nonsteroidal antiinflammatory drugs.

N Engl J Med 1984;310:563–72.

[77] Oates JA, FitzGerald GA, Branch RA, et al. Clinical implications of prostaglandin and

thromboxane A2 formation. N Engl J Med 1988;319.

[78] Schmieder RE. Optimizing therapeutic strategies to achieve renal and cardiovascular risk

reduction in diabetic patients with angiotensin receptor blockers. J Hypertens 2005;23:

905–11.

[79] Lewis EJ,HunsickerLG,ClarkeWR, et al. Renoprotective effect of the angiotensin-receptor

antagonist irbesartan in patients with nephropathy due to type 2 diabetes. N Engl J Med

2001;345:851–60.

[80] Zager RA. Endotoxemia, renal hypoperfusion, and fever: interactive risk factors for amino-

glycoside and sepsis-associated acute renal failure. Am J Kidney Dis 1992;20:223–30.

[81] Rodicio JL. Calcium antagonists and renal protection from cyclosporine nephrotoxicity:

long-term trial in renal transplantation patients. J Cardiovasc Pharmacol 2000;35:S7–11.

[82] Ojo AO, Held PJ, Port FK, et al. Chronic renal failure after transplantation of a nonrenal

organ. N Engl J Med 2003;349:931–40.


[83] Nash K, Hafeez A, Hou S. Hospital-acquired renal insufficiency. Am J Kidney Dis 2002;39:

930–6.

[84] Liss P, Nygren A, Erikson U, et al. Injection of low and iso-osmolar contrast medium

decreases oxygen tension in the renal medulla. Kidney Int 1998;53:698–702.

[85] Solomon R. Contrast-medium-induced acute renal failure. Kidney Int 1998;53:230–42.

[86] Itoh Y, Yano T, Sendo T, et al. Clinical and experimental evidence for prevention of acute

renal failure induced by radiographic contrast media. J Pharmacol Sci 2005;97:473–88.

[87] BellomoR, ChapmanM, Finfer S, et al. Low dose dopamine in patients with early renal dys-

function: a placebo-controlled randomized trial. Australian and New Zealand Intensive

Care Society (ANZICS) Clinical Trials Group. Lancet 2000;356:2139–43.

[88] Lassnigg A, Donner E, Grubhofer G, et al. Lack of renoprotective effects of dopamine and

furosemide during cardiac surgery. J Am Soc Nephrol 2000;11:97–104.

[89] Woo EB, Tang AT, el-Gamel A, et al. Dopamine therapy for patients at risk of renal

dysfunction following cardiac surgery: science or fiction? Eur J Cardiothorac Surg 2002;

22:106–11.

[90] HalpennyM,RusheC, Breen P, et al. The effects of fenoldopamon renal function in patients

undergoing elective aortic surgery. Eur J Anaesthesiol 2002;19:32–9.

[91] Cantarovich F, Rangoonwala B, Lorenz H, et al. High-dose furosemide for established

ARF: a prospective, randomized, double-blind, placebo-controlled, multicenter trial. Am

J Kidney Dis 2004;44:402–9.

[92] Kay J, Chow WH, Chan TM, et al. Acetylcysteine for prevention of acute deterioration of

renal function following elective coronary angiography and intervention: a randomized

controlled trial. JAMA 2003;289:553–8.

[93] Shyu KG, Cheng JJ, Kuan P. Acetylcysteine protects against acute renal damage in patients

with abnormal renal function undergoing a coronary procedure. J AmColl Cardiol 2002;40:

1383–8.

[94] Mallick NP, Gokal R. Haemodialysis. Lancet 1999;353:737–42.

[95] Pastan S, Bailey J. Dialysis therapy. N Engl J Med 1998;338:1428–37.

[96] Forni LG, Hilton PJ. Continuous hemofiltration in the treatment of acute renal failure.

N Engl J Med 1997;336:1303–9.

Fax +41 61 306 12 34E-Mail [email protected]

Minireview

Nephron Clin Pract 2009;112:c222–c229 DOI: 10.1159/000224788

Renal Replacement Therapy for Acute Kidney Injury

Heather Fieghen a Ron Wald a, b Bertrand L. Jaber c

a Division of Nephrology, St. Michael’s Hospital and the University of Toronto, and b The Keenan Research Centrein the Li Ka Shing Knowledge Institute of St. Michael’s Hospital, Toronto, Ont. , Canada; c Division of Nephrology,St. Elizabeth’s Medical Center and Tufts University School of Medicine, Boston, Mass. , USA

hospitalized patients. This is to some extent related to the lack of a uniform definition for AKI. Two classification systems have emerged that may serve to unify communi-cation concerning AKI ( table 1 ) [3, 4] . The RIFLE classi-fication system was formulated by the Acute Dialysis Quality Initiative and defines three strata of increasing severity of AKI: Risk, Injury and Failure, based on rela-tive changes in serum creatinine and urine output, and two outcome strata (Loss and End-stage renal failure) [5] . The three-stage Acute Kidney Injury Network classifica-tion system largely overlaps with the three first stages of RIFLE, with the addition of an absolute serum creatinine increase of 0.3 mg/dl to qualify for stage 1, a limit on the time interval for the creatinine rise to 48 h, and categori-zation of patients started on renal replacement therapy (RRT) as stage 3 regardless of creatinine or urine output. In both classification systems, escalating severity of AKI is associated with incrementally worse outcomes [6] .

Although understanding of the epidemiology of AKI has improved, there are no proven therapies that reverse the course of established AKI. RRT is a key component of the supportive care given to patients with severe AKI. Ap-proximately 4% of all critically ill patients will require RRT [7] . Among RRT-requiring patients who survive the critical phase of their illness, the majority will be free of RRT at the time of hospital discharge [7, 8] . This review will discuss novel diagnostic strategies for AKI but will focus on the management of AKI, with an emphasis on renal replacement strategies.

Key Words

Acute kidney injury � Biomarkers � Continuous renal replacement therapy � Intermittent hemodialysis � Sustained low-efficiency dialysis � Dialysis dose

Abstract

The treatment of established acute kidney injury (AKI) is largely supportive in nature. Renal replacement therapy re-mains the cornerstone of management for the minority of patients who have severe AKI. Optimization of renal replace-ment therapy may modulate the high mortality associated with AKI. Recent trials indicated that continuous renal re-placement therapy does not confer a survival advantage as compared to intermittent hemodialysis. Furthermore, there is no evidence to support a more intensive strategy of renal replacement therapy in the setting of AKI. There is compara-tively limited data regarding the ideal timing of renal re-placement therapy initiation and the preferred mode of sol-ute clearance. Copyright © 2009 S. Karger AG, Basel

Introduction

Acute kidney injury (AKI) is a frequent complication of hospitalization that is associated with substantial mor-bidity, mortality and health care expenditures [1, 2] . There is significant variability in the reported prevalence (1–25%) and mortality (20–60%) associated with AKI in

Published online: June 16, 2009

Ron Wald, MDCM, MPH Division of Nephrology, St. Michael’s Hospital 61 Queen Street East, 9th Floor Toronto, Ont. M5C 2T2 (Canada) Tel. +1 416 867 3703, Fax +1 416 867 3709, E-Mail [email protected]

© 2009 S. Karger AG, Basel1660–2110/09/1124–0222$26.00/0

Accessible online at:www.karger.com/nec

http://dx.doi.org/10.1159%2F000224788


Nephron Clin Pract 2009;112:c222–c229 c223

New Biomarkers in AKI

Current diagnostic paradigms for AKI are limited by reliance on serum creatinine, which is affected by age, gender and muscle mass. In addition, elevations in serum creatinine may occur several days after the actual injury. The search for AKI biomarkers has focused on identify-ing alternatives to serum creatinine. Urinary neutrophil gelatinase-associated lipocalin (NGAL) and interleukin-18 may provide insights into the cause of AKI [9, 10] . Sim-ilarly, urinary and serum NGAL, serum cystatin C and urinary kidney injury molecule-1 (KIM-1) may facilitate the early diagnosis of AKI. KIM-1 also shows promise in predicting adverse events in patients with established AKI [11] . Ongoing studies are clarifying the role of these biomarkers in larger patient cohorts. Ultimately, it will need to be shown that integration of novel biomarkers into clinical decision-making impacts on patient-rele-vant outcomes.

Pharmacological Treatment of AKI

Pharmacological interventions in AKI have targeted the prevention of renal ischemia or modulation of the en-suing inflammatory or hormonal milieus. Low-dose do-pamine, historically thought to improve renal perfusion and thus prevent AKI, has recently been shown in a meta-analysis to have no effect on mortality and RRT require-ment [12] . Similarly, atrial natriuretic peptide (ANP), a vasoactive endogenous hormone that increases glomeru-lar filtration by dilating afferent and constricting efferent

arterioles, was felt to be a promising therapeutic option. Anaritide, an ANP analogue, was initially shown to have no effect on dialysis-free survival and caused increased rates of hypotension in a randomized controlled trial (RCT) of AKI of variable etiology when given at a dose of 200 ng/kg/min [13] . However, in a small randomized study of 59 postcardiac surgery patients with AKI, anarit-ide at a lower dose (50 ng/kg/min) was found to signifi-cantly increase the rate of 21-day dialysis-free survival [14] . A recent meta-analysis of 19 different RCTs showed a trend towards reduction of RRT requirement when ANP was used in the prevention of AKI, however an overall increase in mortality was observed when ANP was used in the treatment of established AKI (particu-larly with higher doses) [15] .

Recombinant human insulin-like growth factor 1 showed encouraging results at reducing renal tubular apoptosis and inflammation in mice when administered immediately after renal ischemia [16] . In a small RCT of 72 critically ill patients with AKI, insulin-like growth factor 1 failed to show any benefit [17] . More recently, N-acetylcysteine and sodium bicarbonate have received a great deal of attention in the setting of contrast ne-phropathy and postcardiac surgery AKI prevention [18, 19] . The clinical benefits of these interventions remain controversial.

There are a number of other agents in preclinical stud-ies that show promise in the prevention or early treatment of AKI. However, the efficacy of these therapies in clin-ical practice may depend on the early identification of AKI [20] .

Table 1. The Acute Dialysis Quality Initiative and Acute Kidney Injury Network classification systems for AKI

Stage Serum creatinine change Urine output criteria

Acute Dialysis Quality Initiative RIFLE criteria [4, 5]Risk increase in serum creatinine of 1.5–2 times baseline less than 0.5 ml/kg/h for more than 6 hInjury increase in serum creatinine of 2–3 times baseline less than 0.5 ml/kg/h for more than 12 hFailure increase in serum creatinine of >3 times baseline OR if baseline creatinine

>4 mg/dl, any 0.5 mg/dl increase qualifiesless than 0.3 ml/kg/h for more than 24 h OR anuria for >12 h

Loss persistent need for RRT for >4 weeksEnd-stage persistent need for RRT for >3 months

Acute Kidney Injury Network criteria1 increase in serum creatinine of 1.5–2 times baseline OR 0.3 mg/dl

increase from baselineless than 0.5 ml/kg/h for more than 6 h

2 increase in serum creatinine of 2–3 times baseline less than 0.5 ml/kg/h for more than 12 h3 increase in serum creatinine of >3 times baseline OR if baseline creatinine

>4 mg/dl, any 0.5 mg/dl increase qualifies OR if any RRT givenless than 0.3 ml/kg/h for more than 24 h OR anuria for >12 h

Fieghen /Wald /Jaber


RRT Modality Choice

Intermittent hemodialysis (IHD), continuous renal re-placement therapies (CRRT) and sustained low-efficien-cy dialysis (SLED) are the principal RRT modalities that are used in the acute setting. Although institutional pol-icies may determine the local availability of these mo-dalities, CRRT and SLED tend to be used in patients with greater hemodynamic instability. There is likely substan-tial intercenter variability with respect to how each of these forms of RRT is utilized and prescribed.

IHD is typically administered with conventional di-alysis machinery that is used in the chronic dialysis pop-ulation with session length ranging from 3 to 5 h. CRRT is applied with an intended treatment time of 24 h and generally requires dedicated machines that operate at comparatively lower blood and dialysate pump speeds. CRRT may be administered as hemodialysis (continu-ous venovenous hemodialysis, CVVHD), hemofiltration (continuous venovenous hemofiltration, CVVH) or a combination of these (continuous venovenous hemodia-filtration, CVVHDF). SLED, sometimes referred to as extended dialysis, is considered a ‘hybrid’ of IHD and

CRRT. SLED is administered using conventional dialysis technology but typical sessions run for 8–12 h using blood and dialysis flows that are intermediate to those pre-scribed in IHD and CRRT. Table 2 summarizes the dif-ferences between these three modalities.

CRRT versus IHD Most trials that addressed the question of optimal

RRT modality compared CRRT and IHD ( table 3 ). A tri-al by Mehta et al. [21] revealed higher ICU mortality in patients treated with CRRT as compared to IHD (59.5 vs. 41.5%). This finding was tempered by apparent baseline imbalances between the groups, whereby patients ran-domized to CRRT had a greater severity of illness. Renal recovery did not differ between the groups.

In a larger multicenter RCT conducted in France com-paring CRRT and IHD in critically ill patients with AKI [22] , 60-day mortality was not reduced by CRRT, and there was no significant difference in the duration of RRT dependence. Three additional trials have shown no sur-vival benefit with CRRT [23–25] .

The presumption of greater hemodynamic stability with CRRT also remains controversial. Three RCTs sug-

Table 2. Practical comparison of acute RRT modalities

Intermittenthemodialysis

Sustained low-efficiency dialysis

Continuous renalreplacement therapy

Session duration, h 3–5 8–12 24Blood flow, ml/min 300–400 200–300 100–200Dialysate flow, ml/min 500–800 200–350 25–40Anticoagulation requirement heparin or none heparin or none heparin or regional citrate

Table 3. Results of individual RCTs comparing CRRT to IHD

Study Sample size Primary endpoint Mortality, % Persistent dialysisrequirementa, %

CRRT IHD CRRT IHD CRRT IHD

Mehta et al. [21] 84 82 ICU mortality 59.5 41.5 14.0 7.0Augustine et al. [23] 40 40 in-hospital mortality 67.5 70.0 61.5 66.7Uehlinger et al. [24] 70 55 in-hospital mortality 47.0 51.0 2.7 3.7Vinsonneau et al. [22] 175 184 60-day mortality 67.4 68.5 1.8 0.0Lins et al. [25] 172 144 in-hospital mortality 58.1 62.5 16.9b 25.5b

a Defined as dialysis dependence at the time of hospital discharge among survivors.b Glomerular filtration rate <15 ml/min at the time of hospital discharge.



gested no advantage with CRRT as compared to IHD [22, 24, 26] , while others demonstrated more favorable hemo-dynamics with CRRT [23, 27] . Even if CRRT confers a hemodynamic benefit, it is unclear whether this trans-lates into improvements in the patient-relevant outcomes of survival and renal recovery.

CRRT versus SLED In addition to the absence of a survival advantage,

CRRT is more costly than IHD and is associated with a number of obstacles such as continuous patient immo-bilization, the requirement for anticoagulation and the need for specialized machines and premixed commercial solutions [28] . This has stimulated a search for a strategy that incorporates the putative hemodynamic benefits of CRRT without the associated logistic and resource con-straints. SLED meets many of these criteria.

Observational data from single centers suggest that SLED is a feasible way of providing RRT that is adequate, hemodynamically well tolerated, potentially anticoagu-lation-free and possibly cost-effective [29, 30] . However, only two small RCTs have compared SLED and CRRT [31, 32] . Kielstein et al. [31] randomized 39 critically ill patients with AKI to receive either 24 h of CVVH or 12 h of SLED. Using invasive monitoring, these authors found no significant differences in all measured hemo-dynamic parameters (mean arterial pressure, systemic vascular resistance, cardiac output) with comparable re-moval of creatinine and urea. A smaller study random-ized 16 patients to receive three sessions with either CVVH or SLED (with an added hemofiltration compo-nent), and showed that fluid removal and hemodynamic parameters were similar in both groups [32] . Although these preliminary data suggest that SLED may supplant CRRT as the modality of choice for hemodynamically unstable patients with AKI, further studies that utilize patient-relevant outcomes are required to define the pre-cise role of SLED.

A strategy has recently been described whereby CRRT machinery was applied over a contracted treatment time of 9 h, using increased blood and effluent flow rates. Termed accelerated venovenous hemofiltration, this mo-dality retains many of the feasibility advantages of SLED, but dedicated commercial solutions were still required. A retrospective case series demonstrated adequate solute removal, acceptable hemodynamic tolerability and the ability to avoid systemic anticoagulation [33] .

Intensity of RRT

The impact of RRT intensity or dose on patient out-comes has been a matter of controversy. Most definitions of acute RRT dose are based on small molecule removal, as exemplified by urea, while ignoring other crucial as-pects of RRT adequacy in AKI, such as volume and elec-trolyte control. Furthermore, assumptions pertaining to urea kinetic modeling that are applied to end-stage renal disease are inappropriate in the acute setting. Finally, many of the benefits derived from acute RRT may relate to large-molecular-weight solute clearance, which is poorly quantified.

Schiffl et al. [34] compared daily to thrice-weekly IHD in 146 critically ill patients with AKI. Patients receiving daily hemodialysis had significantly lower mortality as compared to those receiving alternate-day dialysis (28 vs. 46%) and a shorter time to renal recovery, as defined by independence from dialysis (mean of 9 vs. 16 days). The study was designed to provide a minimum Kt/V urea of 1.2 for each treatment session in both arms of the study. However, the authors demonstrated that the actual de-livered dose per session was significantly lower than in-tended, albeit similar in both arms of the study (mean Kt/V urea of 0.92 vs. 0.94). It is therefore unclear if the sur-vival benefit demonstrated in this study was mediated by the mere provision of an adequate dialysis dose in the daily dialysis arm or by the true merits of more intensive therapy.

Ronco et al. [35] randomized 425 critically ill patients with AKI to continuous hemofiltration rates of 20, 35 or 45 ml/kg/h. They showed enhanced survival at 15 days following discontinuation of RRT in the two high-dose groups when compared to the low-dose group. The ben-efits of high-intensity CRRT were also seen in a trial by Saudan et al. [36] who compared CVVH (mean effluent flow of 25 ml/kg/h) with CVVHDF (mean effluent flow of 42 ml/kg/h) in 206 critically ill patients. Ninety-day survival was significantly higher among patients treated with CVVHDF as compared with CVVH (59 vs. 34%). These results contrasted with two other trials demon-strating no benefit with intensified therapy [37, 38] .

In the recently completed Veterans Administration/National Institutes of Health Acute Renal Failure Trial Network (ATN) study, over 1,100 critically ill patients with AKI requiring RRT were randomized to an inten-sive or a less intensive treatment strategy [39] . In the in-tensive arm, hemodynamically unstable patients received CRRT at 35 ml/kg/h or 6 sessions per week of SLED; if patients were hemodynamically stable, IHD was admin-



istered 6 times per week. In the less intensive arm, hemo-dynamically unstable patients received CRRT at 20 ml/kg/h or thrice-weekly SLED; stable patients received thrice-weekly IHD. Patients were able to shift between RRT modalities depending on their evolving hemody-namic status, thereby reflecting usual clinical practice. The trial was well executed with minimal crossover and excellent follow-up. Sixty-day mortality was 53.6 and 51.8% in the intensive and less intensive arms, respec-tively (NS). There were no appreciable differences in renal recovery.

Two important limitations of the ATN study should be considered. In both treatment arms, CRRT was delivered as an equal mix of hemodialysis and hemofiltration. The benefits of more intensive RRT may be conferred by the latter component, and it is notable that hemofiltration was administered exclusively in the positive trial by Ron-co et al. [35] . The generalizability of the ATN study is also limited by the exclusion of patients with significant pre-morbid chronic kidney disease, a population that com-prises a large proportion of AKI patients seen in clinical practice [7, 40] . The Randomised Evaluation of Normal vs. Augmented Level (RENAL) Replacement Therapy study (ClinicalTrials.gov identifier NCT00221013) en-rolled 1,500 critically ill patients in Australia and New Zealand with AKI at multiple sites. Preliminary study re-sults have been reported (although not published to date) and indicate that CVVHDF at a dose of 40 ml/kg/h did not confer improved 90-day survival as compared to a dose of 25 ml/kg/h.

Timing of RRT Initiation

Conventional indications for RRT initiation in AKI include refractory hyperkalemia, severe metabolic acido-sis, hypervolemia, and uremic end-organ complications. However, one or more of these indications may emerge well after the time of the kidney insult. The resulting de-lay in initiating RRT may prolong patient exposure to the uncorrected metabolic effects of kidney failure, which may in turn contribute to adverse outcomes.

There is a very limited evidence base to guide the tim-ing of RRT initiation in AKI. A recent meta-analysis identified 23 relevant studies and found a wide array of study designs and definitions for ‘early’ and ‘late’ RRT initiation [41] . In the 5 RCTs that addressed this issue, early initiation of RRT was associated with a trend to-wards mortality reduction [risk ratio (RR) 0.64, 95% con-fidence interval (CI) 0.40–1.05] and in the 18 observa-

tional studies, a mortality risk reduction of 28% (RR 0.72, 95% CI 0.64–0.82) was observed. It is noteworthy that all of the RCTs cited in the meta-analysis had small sample sizes and were generally of poor quality.

Defining the optimal timing of RRT initiation in AKI remains a research priority. Prior to embarking on a de-finitive RCT, substantial preliminary work will be re-quired to determine the prevailing standard of care in the timing of RRT initiation. This should set the parameters for the control arm in such a trial. An ‘early start’ strat-egy could then be derived relative to the standard of care.

Convection and Removal of Large-Molecular-Weight

Solutes

Convection, as provided through hemofiltration, and diffusion, as provided through hemodialysis, are the principal modes of solute removal in AKI. Convection utilizes hydrostatic pressure to effect the translocation of water across the semi-permeable membrane while con-comitantly dragging solutes with molecular weights that are below the pore size of the membrane. Diffusion fuels the movement of solutes across a semi-permeable mem-brane down concentration gradients for the respective solutes. Both modes provide reliable clearance of low-molecular-weight molecules such as urea, whereas larger-molecular-weight solutes are more effectively removed by convection. Large molecules of particular interest in-clude inflammatory cytokines such as tumor necrosis factor- � and interleukin-6 that may serve as key media-tors in the pathogenesis of AKI. Despite the conceptual appeal of convective clearance, there is a paucity of pa-tient-level data on the optimal mode of clearance in AKI resulting in a great deal of practice variability. A small RCT evaluated outcomes of 20 patients who received CVVH versus CVVHD. There were no significant differ-ences in mortality, renal recovery or duration of ICU stay [42] .

Removal of large-molecular-weight solutes may also be enhanced by the use of hemodiafilters with larger pore sizes. These so-called high cutoff hemodiafilters have pores of 50–60 kDa (as compared to 20–30 kDa for con-ventional hemodiafilters), thereby enabling the translo-cation of larger-sized inflammatory mediators. Haase et al. [43] recently completed a double-blind crossover RCT of 10 septic patients with AKI who received 4-hour ses-sions of hemodialysis with a high cutoff dialyzer and a standard high-flux dialyzer, respectively, in random se-



quence. Interleukin-6, interleukin-8 and interleukin-10 concentrations were significantly reduced following ther-apy with the high cutoff dialyzers with little change in these markers following treatment with the high-flux di-alyzer. Similar small solute control was achieved with both dialyzers. Although the clinical implications of these findings need further clarification, other work done by this group suggested improved hemodynamic control in patients treated with a high cutoff hemodiafilter [44] .

Renal Tubule Cell Therapy

Contemporary RRT devices are designed to replace the filtration component of kidney function but tubular function is not addressed. Humes et al. [45] have devel-oped a bioartificial kidney known as a renal assist device (RAD), composed of sheets of human tubular cells that are housed within a high-flux hemodiafilter. Preliminary data indicated that these cells successfully performed many of the transport and endocrinologic functions nor-mally attributed to the renal tubular epithelium [46] . A phase II RCT of 58 critically ill patients with AKI com-pared 72 h of continuous treatment with the RAD and a conventional filter (in series with each other) versus con-ventional CRRT [47] . Treatment with the RAD was found to be safe with a trend towards reduced mortality at 28 days (33 vs. 61% in patients treated with conventional CRRT). The RR of death in the RAD group, adjusted for comorbid illness, was 0.48 (95% CI = 0.23–0.99). In addi-tion, higher rates of renal recovery were seen in patients treated with the RAD. It should be noted that only 10 of the 40 patients treated with RAD completed the planned 72 h of therapy.

Conclusions

New classification systems for AKI may enhance stan-dardization around diagnosis and staging of this clinical syndrome. Novel biomarkers for the early diagnosis of AKI may represent a breakthrough for clinicians if they are accurate, reproducible and applicable in different set-tings. There are currently no specific therapeutic inter-ventions for patients with established AKI.

It is possible that optimization of the RRT prescription will impact on patient outcomes. Recently completed RCTs have provided clinicians with guidance on some aspects of the RRT prescription but several areas remain unclear. Despite the conceptual advantages of CRRT, multiple RCTs have shown no evidence of improved pa-tient outcomes with this modality, as compared to con-ventional IHD. The logistic challenges associated with CRRT and the relatively high costs of this modality may stimulate the increased use of SLED. However, well-de-signed RCTs are still needed to better characterize the reported benefits of SLED prior to its widespread adop-tion. Recent data have also clarified important questions regarding RRT intensity in AKI. The current balance of evidence suggests that among hemodynamically unsta-ble patients, CRRT need not be administered at doses higher than 20 ml/kg/h, and in more stable patients, al-ternate-day IHD is acceptable.

There are several areas in the AKI arena where high-quality evidence is lacking. The optimal timing of RRT initiation in AKI remains a critical area of uncertainty. A well-powered trial to examine this issue, in which an ear-ly intervention may be guided by novel AKI biomarkers, is needed. Further trials to address the clinical benefits conferred by convective clearance, high cutoff filters and the bioartificial kidney are also required.

References

1 Nash K, Hafeez A, Hou S: Hospital-acquired renal insufficiency. Am J Kidney Dis 2002; 39: 930–936.

2 Liangos O, Wald R, O’Bell JW, Price L, Pereira BJ, Jaber BL: Epidemiology and out-comes of acute renal failure in hospitalized patients: a national survey. Clin J Am Soc Nephrol 2006; 1: 43–51.

3 Mehta RL, Kellum JA, Shah SV, Molitoris BA, Ronco C, Warnock DG, Levin A: Acute kidney injury network: report of an initiative to improve outcomes in acute kidney injury. Crit Care 2007; 11:R31.

4 Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P: Acute renal failure – Definition,

outcome measures, animal models, f luid therapy and information technology needs: the Second International Consensus Confer-ence of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 2004; 8:R204–R212.

5 Ostermann M, Chang RW: Acute kidney in-jury in the intensive care unit according to RIFLE. Crit Care Med 2007; 35: 1837–1843;quiz 1852.

6 Bagshaw SM, George C, Bellomo R: A com-parison of the RIFLE and AKIN criteria for acute kidney injury in critically ill patients. Nephrol Dial Transplant 2008; 23: 1569–1574.

7 Uchino S, Kellum JA, Bellomo R, Doig GS, Morimatsu H, Morgera S, Schetz M, Tan I, Bouman C, Macedo E, Gibney N, Tolwani A, Ronco C: Acute renal failure in critically ill patients: a multinational, multicenter study. JAMA 2005; 294: 813–818.

8 Bagshaw SM, Laupland KB, Doig CJ, Mortis G, Fick GH, Mucenski M, Godinez-Luna T, Svenson LW, Rosenal T: Prognosis for long-term survival and renal recovery in critically ill patients with severe acute renal failure: a population-based study. Crit Care 2005; 9:R700–R709.



9 Nickolas TL, O’Rourke MJ, Yang J, Sise ME, Canetta PA, Barasch N, Buchen C, Khan F, Mori K, Giglio J, Devarajan P, Barasch J: Sen-sitivity and specificity of a single emergency department measurement of urinary neu-trophil gelatinase-associated lipocalin for diagnosing acute kidney injury. Ann Intern Med 2008; 148: 810–819.

10 Parikh CR, Jani A, Melnikov VY, Faubel S, Edelstein CL: Urinary interleukin-18 is a marker of human acute tubular necrosis. Am J Kidney Dis 2004; 43: 405–414.

11 Liangos O, Perianayagam MC, Vaidya VS, Han WK, Wald R, Tighiouart H, MacKin-non RW, Li L, Balakrishnan VS, Pereira BJ, Bonventre JV, Jaber BL: Urinary N-acetyl-beta-(D)-glucosaminidase activity and kid-ney injury molecule-1 level are associated with adverse outcomes in acute renal failure. J Am Soc Nephrol 2007; 18: 904–912.

12 Friedrich JO, Adhikari N, Herridge MS, Beyene J: Meta-analysis: low-dose dopamine increases urine output but does not prevent renal dysfunction or death. Ann Intern Med 2005; 142: 510–524.

13 Allgren RL, Marbury TC, Rahman SN, Weisberg LS, Fenves AZ, Lafayette RA, Sweet RM, Genter FC, Kurnik BR, Conger JD, Sayegh MH: Anaritide in acute tubular ne-crosis. Auriculin Anaritide Acute Renal Failure Study Group. N Engl J Med 1997; 336: 828–834.

14 Sward K, Valsson F, Odencrants P, Samuels-son O, Ricksten SE: Recombinant human atrial natriuretic peptide in ischemic acute renal failure: a randomized placebo-con-trolled trial. Crit Care Med 2004; 32: 1310–1315.

15 Nigwekar SU, Navaneethan SD, Parikh CR, Hix JK: Atrial natriuretic peptide for man-agement of acute kidney injury: a systematic review and meta-analysis. Clin J Am Soc Nephrol 2009; 4: 261–272.

16 Goes N, Urmson J, Vincent D, Ramassar V, Halloran PF: Effect of recombinant human insulin-like growth factor-1 on the inflam-matory response to acute renal injury. J Am Soc Nephrol 1996; 7: 710–720.

17 Hirschberg R, Kopple J, Lipsett P, Benjamin E, Minei J, Albertson T, Munger M, Metzler M, Zaloga G, Murray M, Lowry S, Conger J, McKeown W, O’Shea M, Baughman R, Wood K, Haupt M, Kaiser R, Simms H, Warnock D, Summer W, Hintz R, Myers B, Haenftling K, Capra W, et al: Multicenter clinical trial of recombinant human insulin-like growth factor I in patients with acute renal failure. Kidney Int 1999; 55: 2423–2432.

18 Birck R, Krzossok S, Markowetz F, Schnulle P, van der Woude FJ, Braun C: Acetylcysteine for prevention of contrast nephropathy: meta-analysis. Lancet 2003; 362: 598–603.

19 Kelly AM, Dwamena B, Cronin P, Bernstein SJ, Carlos RC: Meta-analysis: effectiveness of drugs for preventing contrast-induced ne-phropathy. Ann Intern Med 2008; 148: 284–294.

20 Jo SK, Rosner MH, Okusa MD: Pharmaco-logic treatment of acute kidney injury: why drugs haven’t worked and what is on the ho-rizon. Clin J Am Soc Nephrol 2007; 2: 356–365.

21 Mehta RL, McDonald B, Gabbai FB, Pahl M, Pascual MT, Farkas A, Kaplan RM: A ran-domized clinical trial of continuous versus intermittent dialysis for acute renal failure. Kidney Int 2001; 60: 1154–1163.

22 Vinsonneau C, Camus C, Combes A, Costa de Beauregard MA, Klouche K, Boulain T, Pallot JL, Chiche JD, Taupin P, Landais P, Dhainaut JF: Continuous venovenous hae-modiafiltration versus intermittent haemo-dialysis for acute renal failure in patients with multiple-organ dysfunction syndrome: a multicentre randomised trial. Lancet 2006; 368: 379–385.

23 Augustine JJ, Sandy D, Seifert TH, Paganini EP: A randomized controlled trial compar-ing intermittent with continuous dialysis in patients with ARF. Am J Kidney Dis 2004; 44: 1000–1007.

24 Uehlinger DE, Jakob SM, Ferrari P, Eichel-berger M, Huynh-Do U, Marti HP, Mohaupt MG, Vogt B, Rothen HU, Regli B, Takala J, Frey FJ: Comparison of continuous and in-termittent renal replacement therapy for acute renal failure. Nephrol Dial Transplant 2005; 20: 1630–1637.

25 Lins RL, Elseviers MM, Van der Niepen P, Hoste E, Malbrain ML, Damas P, Devriendt J: Intermittent versus continuous renal re-placement therapy for acute kidney injury patients admitted to the intensive care unit: results of a randomized clinical trial. Nephrol Dial Transplant 2009; 24: 512–518.

26 Misset B, Timsit JF, Chevret S, Renaud B, Ta-mion F, Carlet J: A randomized cross-over comparison of the hemodynamic response to intermittent hemodialysis and continuous hemofiltration in ICU patients with acute re-nal failure. Intensive Care Med 1996; 22: 742–746.

27 John S, Griesbach D, Baumgartel M, Weih-precht H, Schmieder RE, Geiger H: Effects of continuous haemofiltration vs intermittent haemodialysis on systemic haemodynamics and splanchnic regional perfusion in septic shock patients: a prospective, randomized clinical trial. Nephrol Dial Transplant 2001; 16: 320–327.

28 Tonelli M, Manns B, Wiebe N, Shrive F, Pan-nu N, Doig C, Klarenbach S: Continuous Re-nal Replacement Therapy in Adult Patients with Acute Renal Failure: Systematic Review and Economic Evaluation. Ottawa, Canadi-an Agency for Drugs and Technologies in Health, 2007.

29 Kumar VA, Craig M, Depner TA, Yeun JY: Extended daily dialysis: a new approach to renal replacement for acute renal failure in the intensive care unit. Am J Kidney Dis 2000; 36: 294–300.

30 Berbece AN, Richardson RM: Sustained low-efficiency dialysis in the ICU: cost, anti-coagulation, and solute removal. Kidney Int 2006; 70: 963–968.

31 Kielstein JT, Kretschmer U, Ernst T, Hafer C, Bahr MJ, Haller H, Fliser D: Efficacy and cardiovascular tolerability of extended dial-ysis in critically ill patients: a randomized controlled study. Am J Kidney Dis 2004; 43: 342–349.

32 Baldwin I, Bellomo R, Naka T, Koch B, Fealy N: A pilot randomized controlled compari-son of extended daily dialysis with filtration and continuous veno-venous hemofiltra-tion: f luid removal and hemodynamics. IntJ Artif Organs 2007; 30: 1083–1089.

33 Gashti CN, Salcedo S, Robinson V, Rodby RA: Accelerated venovenous hemofiltration: early technical and clinical experience. Am J Kidney Dis 2008; 51: 804–810.

34 Schiffl H, Lang SM, Fischer R: Daily hemo-dialysis and the outcome of acute renal fail-ure. N Engl J Med 2002; 346: 305–310.

35 Ronco C, Bellomo R, Homel P, Brendolan A, Dan M, Piccinni P, La Greca G: Effects of dif-ferent doses in continuous veno-venous hae-mofiltration on outcomes of acute renal fail-ure: a prospective randomised trial. Lancet 2000; 356: 26–30.

36 Saudan P, Niederberger M, De Seigneux S, Romand J, Pugin J, Perneger T, Martin PY: Adding a dialysis dose to continuous hemo-filtration increases survival in patients with acute renal failure. Kidney Int 2006; 70: 1312–1317.

37 Bouman CS, Oudemans-Van Straaten HM, Tijssen JG, Zandstra DF, Kesecioglu J: Ef-fects of early high-volume continuous veno-venous hemofiltration on survival and re-covery of renal function in intensive care patients with acute renal failure: a prospec-tive, randomized trial. Crit Care Med 2002; 30: 2205–2211.

38 Tolwani AJ, Campbell RC, Stofan BS, Lai KR, Oster RA, Wille KM: Standard versus high-dose CVVHDF for ICU-related acute renal failure. J Am Soc Nephrol 2008; 19: 1233–1238.

39 Palevsky PM, Zhang JH, O’Connor TZ, Chertow GM, Crowley ST, Choudhury D, Finkel K, Kellum JA, Paganini E, Schein RM, Smith MW, Swanson KM, Thompson BT, Vijayan A, Watnick S, Star RA, Peduzzi P: Intensity of renal support in critically ill pa-tients with acute kidney injury. N Engl J Med 2008; 359: 7–20.

40 Cho KC, Himmelfarb J, Paganini E, Ikizler TA, Soroko SH, Mehta RL, Chertow GM: Survival by dialysis modality in critically ill patients with acute kidney injury. J Am Soc Nephrol 2006; 17: 3132–3138.

41 Seabra VF, Balk EM, Liangos O, Sosa MA, Cendoroglo M, Jaber BL: Timing of renal re-placement therapy initiation in acute renal failure: a meta-analysis. Am J Kidney Dis 2008; 52: 272–284.



42 Daud KM, Leong G, Visvanathan R: Acute dialytic support for the critically ill: continu-ous venovenous haemodialysis versus con-tinuous venovenous haemofiltration. Int Med J 2006; 13: 37–42.

43 Haase M, Bellomo R, Baldwin I, Haase-Fielitz A, Fealy N, Davenport P, Morgera S, Goehl H, Storr M, Boyce N, Neumayer HH: Hemodialysis membrane with a high-molec-ular-weight cutoff and cytokine levels in sep-sis complicated by acute renal failure: a phase 1 randomized trial. Am J Kidney Dis 2007; 50: 296–304.

44 Morgera S, Haase M, Kuss T, Vargas-Hein O, Zuckermann-Becker H, Melzer C, Krieg H, Wegner B, Bellomo R, Neumayer HH: Pilot study on the effects of high cutoff hemofil-tration on the need for norepinephrine in septic patients with acute renal failure. Crit Care Med 2006; 34: 2099–2104.

45 Humes HD, Fissell WH, Weitzel WF: The bioartificial kidney in the treatment of acute renal failure. Kidney Int Suppl 2002: 121–125.

46 Humes HD, Fissell WH, Weitzel WF, Buff-ington DA, Westover AJ, MacKay SM, Guti-errez JM: Metabolic replacement of kidney function in uremic animals with a bioartifi-cial kidney containing human cells. Am J Kidney Dis 2002; 39: 1078–1087.

47 Tumlin J, Wali R, Williams W, Murray P, Tolwani AJ, Vinnikova AK, Szerlip HM, Ye J, Paganini EP, Dworkin L, Finkel KW, Kraus MA, Humes HD: Efficacy and safety of renal tubule cell therapy for acute renal failure. J Am Soc Nephrol 2008; 19: 1034–1040.

This excellent review by Fieghen, Wald and Jaber from Toronto and Boston highlights recent advances in the classification of AKI whilst drawing attention to the lim-itations of clinical trials in the field. It also brings to the reader’s awareness the recent interest in the predictive value of a number of biomarkers including NGAL, IL-18 and KIM-1. These biomarkers have been put forward as early diagnostic and prognostic indicators in AKI.

More recently, Jaber’s group in Boston has reported on the influence of patients’ genotype on outcomes in AKI. They noted a relationship to adverse outcome of a non-synonymous polymorphism in the coding region of the HIF-1 � gene where a C to T substitution occurs at posi-tion +85 in exon 12 [Kolyada et al., 2009]. This associa-tion warrants further investigation.

In their minireview, the authors thoroughly and criti-cally review a number of clinical trials of pharmacologi-cal interventions as well as those based on replacement therapy modalities or intensity in AKI. Most of these in-

Editorial Comment

M. El Nahas, Sheffield

tervention trials are negative as they fail to show signifi-cant advantage. This may be due in part to the fact that clinical trials in AKI are very difficult to conduct due to the heterogeneity of the patients involved and associated comorbidities.

Recent interest has also emerged in the impact AKI may have on the incidence of chronic kidney disease as well as on the acceleration of the progression of estab-lished chronic kidney disease to end-stage renal disease. This is likely to be particularly relevant in the elderly whose recovery from AKI is often limited and may pre-cipitate end-stage renal disease.

Reference

Kolyada AY, Tighiouart H, Perianayagam MC, Liangos O, Madias NE, Jaber BL: A genetic variant of hypoxia-inducible factor-1alpha is associ-ated with adverse outcomes in acute kidney injury. Kidney Int 2009; 75:1322–1329.

Indications and timing of renal replacement therapy in acutekidney injury

Paul M. Palevsky, MD

Conventional indications forinitiating renal replacementtherapy (RRT) in acute kidneyinjury (AKI) include volume

overload, hyperkalemia, metabolic acidosis,and overt uremic manifestation such as en-cephalopathy and pericarditis (Table 1).Acute dialysis also is indicated when AKIoccurs in the setting of acute intoxicationwith a dialyzable drug or toxin. While theseindications are well accepted, they also aresubject to interpretation. How severe a de-gree of volume overload, hyperkalemia, ormetabolic acidosis? What, if any, medicaltherapies should be tried before initiatingRRT? If diuretic therapy is initiated, whatdose constitutes diuretic resistance? Inmany patients, RRT is initiated in the ab-sence of specific indications in response topersistent oliguria unresponsive to volumeadministration or progressive azotemiawithout uremic manifestations. Observa-tion practice patterns within and across in-stitutions suggest that there is no uniformstandard of care and that wide variations inclinical practice prevail (1).

In the past, the paradigm for manage-ment of severe AKI was that patients diedwith, but did not die of, their renal failure,so long as acute uremic complications wereprevented. The corollary of this view wasthat management of RRT merely needed toassure that patients did not succumb tohyperkalemia, metabolic acidosis, or vol-ume overload; and that overt uremic com-plications, such as pericarditis and enceph-alopathy, were prevented. During the pastdecade, this paradigm has been challengedby data demonstrating that AKI is an inde-pendent risk factor for mortality (2–6). Animplication of these data is that the specificmanagement of RRT may impact the out-comes of AKI and that optimization of renalsupport may reduce its high mortality (7–9). Although multiple recent clinical trialshave prospectively evaluated the impact ofdose (10–13) and modality (14–17) of RRT,the literature on timing of initiation of RRTin AKI is far less robust. In the remainder ofthis review, we will summarize the currentdata on timing of initiation of RRT in AKIthat guide current clinical practices (Table2) and discuss issues that need to be ad-dressed in future clinical trials.

1950S to 1970S: RetrospectiveCase Series and ObservationalStudies

The concept of prophylactic hemodi-alysis in AKI was introduced by Dr. Te-schan and colleagues (18, 19) almost 50yrs ago, within the first decade after theintroduction of hemodialysis into routineclinical practice for the management of

severe acute renal failure. In their land-mark report, Dr. Teschan and colleaguesdescribed their experience using “prophy-lactic” hemodialysis in 15 patients witholiguric acute renal failure treated at theRenal Center of the U.S. Army SurgicalResearch Unit (18). Hemodialysis was ini-tiated before the blood nonprotein nitro-gen reached 200 mg/dL and obvious ure-mic symptoms appeared. Although nocontrol group was included in this report,the authors stated that the results con-trasted dramatically with their own pastexperience in patients in whom dialysiswas not initiated until “conventional” in-dications were present, with a mortalityrate of only 33% and with patients expe-riencing a “stable, convalescent clinicalcourse . . . free from uremic symptoms orchemical imbalances . . .” (18).

Subsequently, multiple retrospectivecase series and observational studies inthe 1960s and early 1970s compared“early” initiation of hemodialysis (as de-fined by blood urea nitrogen [BUN] con-centrations ranging from �93 mg/dL tolevels of approximately 150 mg/dL) to“late” initiation of therapy (as defined byBUN levels of 163 mg/dL to �200 mg/dL)(20–22). All of these studies (Table 2)demonstrated improved survival withearlier initiation of hemodialysis.

1970S to 1980S: SmallProspective Clinical Trials

The first prospective evaluation of“prophylactic” dialysis in acute renal fail-

From the Renal Section, Veterans Affairs PittsburghHealthcare System, Pittsburgh, PA, and Renal-ElectrolyteDivision, Department of Medicine, University of PittsburghSchool of Medicine, Pittsburgh, PA (PMP).

The author has not disclosed any potential con-flicts of interest.

For information regarding this article, E-mail:[email protected]

Copyright © 2008 by the Society of Critical CareMedicine and Lippincott Williams & Wilkins

DOI: 10.1097/CCM.0b013e318168e3fb

The optimal timing for initiation of renal replacement therapyin patients with acute kidney injury remains uncertain. Conven-tionally accepted indications include volume overload, hyperka-lemia, metabolic acidosis, overt uremia, and even progressiveazotemia in the absence of specific symptoms; however, precisedefinitions for these indications are lacking. Data from recentobservational trials have suggested that early initiation of renalreplacement therapy may be associated with decreased mortality;

however, the results of these studies are inconclusive. Existingdata on timing of initiation of renal replacement therapy in acutekidney injury that guide current clinical practice are summarizedand issues that need to be addressed in future clinical trials arediscussed. (Crit Care Med 2008; 36[Suppl.]:S224–S228)

KEY WORDS: renal replacement therapy; acute kidney injury;volume overload; hyperkalemia; metabolic acidosis

S224 Crit Care Med 2008 Vol. 36, No. 4 (Suppl.)

ure was reported by Dr. Conger in 1975(23). In this study, conducted on the U.S.Naval Hospital Ship USS Sanctuary be-tween April and October 1970, 18 pa-tients with post-traumatic AKI were as-signed alternately to an intensivehemodialysis regimen that maintainedpredialysis BUN �70 mg/dL and serumcreatinine �5 mg/dL, or to a noninten-sive regimen in which dialysis was notcarried out until BUN and serum creati-nine approached 150 mg/dL and 10 mg/dL, respectively, or clinical indicationsfor therapy (e.g., hyperkalemia, volume

overload, or uremic encephalopathy) su-pervened. Survival was 64% (five of eightpatients) in the intensive treatmentgroup as compared with 20% (two of tenpatients) with the nonintensive dialysisstrategy (p � .14). In addition, majorcomplications, including hemorrhageand Gram-negative sepsis, were less fre-quent in the intensive hemodialysis arm.

Approximately a decade later, Dr. Con-ger and associates (24) re-examined theintensity of hemodialysis in 34 patientstreated at the University of ColoradoHealth Sciences Center Hospitals. Pairsof patients were assigned randomly whenserum creatinine reached 8 mg/dL to ei-ther an intensive regimen, designed tomaintain the predialysis BUN �60 mg/dLand serum creatinine �5 mg/dL or to aless intensive regimen in which BUN andserum creatinine were permitted to reach100 mg/dL and 9 mg/dL, respectively.While the treatment strategies in thisstudy were not designed to strictly eval-uate timing of imitation of therapy, theaverage time from onset of AKI to initia-tion of dialysis was 5 � 2 days in the intensivetherapy arm as compared with 7 � 3 days inthe nonintensive regimen. Mortality washigher in the more intensively dialyzedgroup; however, given the small sample

size, this difference was not statisticallysignificant (p � .73).

On the basis of these data, the conven-tional teaching became that in the ab-sence of specific symptoms, hemodialysisshould be initiated when BUN reaches alevel of approximately 100 mg/dL, butthat no additional benefit was associatedwith earlier initiation or more intensivetherapy. It should be recognized, how-ever, that the studies on which this argu-ment is based had inadequate statisticalpower to draw definitive conclusions.

The Past Decade

No additional studies examining thetiming of initiation of hemodialysis orother modalities of renal replacementtherapy were published until the end ofthe 1990s. During the past decade, how-ever, multiple studies have re-examinedthis issue, focusing almost exclusively onthe timing of initiation of continuous re-nal replacement therapy (CRRT). In 1999,Dr. Gettings and colleagues (25) pub-lished a retrospective analysis of the tim-ing of initiation of CRRT on outcomes in100 consecutive patients with post-traumatic AKI. Patients were classified asearly or late initiation of therapy based on

Table 1. Conventional indications for renal re-placement therapy in acute kidney injury

Intravascular volume overload unresponsive todiuretic therapy

Hyperkalemia refractory to medicalmanagement

Metabolic acidosis refractory to medicalmanagement

Concomitant intoxication with dialyzable drugor toxin

Overt uremic symptomsEncephalopathyPericarditisUremic bleeding diathesis

Progressive azotemia in the absence of specificsymptoms

Table 2. Summary of studies evaluating the timing of initiation of renal replacement therapy (RRT)

Study YrMode

of RRT Study Design No.

Criteria for Initiation of RRT Survival (%)

Early Late Early Late

Parsons et al (20) 1961 IHD Retrospective 33 BUN 120–150 mg/dL BUN �200 mg/dL 75 12Fischer et al (21) 1966 IHD Retrospective 162 BUN �150 mg/dL BUN �200 mg/dL 43 26Kleinknecht et al (22) 1972 IHD Retrospective 500 BUN �93 mg/dL BUN �163 mg/dL 73 58Conger (23) 1975 IHD RCT 18 BUN �70 mg/dL or

SCr �5 mg/dLBUN �150 mg/dL,

SCr �10 mg/dL, orclinical indications

64 20

Gillum et al (24) 1986 IHD RCT 34 SCr 8 mg/dLTreatment goal:BUN �60 mg/dL,SCr �5 mg/dL

BUN �100 mg/dL orSCr �9 mg/dL

41 53

Gettings et al (25) 1999 CRRT Retrospective 100 BUN �60 mg/dL BUN �60 mg/dL 39 20Bouman et al (12) 2002 CRRT RCT 106 �12 hrs after

meeting AKIdefinition

BUN �112 mg/dL,SK �6.5 mmol/L, orpulmonaryedema

LV: 69HV: 74

LV: 75

Demirkilic et al (26) 2004 CRRT Retrospective 61 UOP �100 mL/8 hr SCr �5.0 mg/dL orSK �5.5 mmol/L

77 45

Elahi et al (27) 2004 CRRT Retrospective 64 UOP �100 mL/8 hr BUN �4 mg/dL,SCr �2.8 mg/dL,or SK �6 mmol/L

78 57

Piccinni et al (28) 2006 CRRT Retrospective 80 �12 hrs after ICUadmission

“Conventional”indications

55 28

Liu et al (29) 2006 IHD & CRRT Observational 243 BUN �76 mg/dL BUN �76 mg/dL 65 59

IHD, intermittent hemodialysis; CRRT, continuous renal replacement therapy; RCT, randomized controlled trial; BUN, blood urea nitrogen; SCr, serumcreatinine; AKI, acute kidney injury; UOP, urine output; ICU, intensive care unit; SK, serum potassium; LV, low-volume hemofiltration; HV, high-volumehemofiltration.

S225Crit Care Med 2008 Vol. 36, No. 4 (Suppl.)

BUN at initiation of therapy, using avalue of 60 mg/dL to separate the twogroups. In the early group, CRRT wasinitiated on hospital day 10 � 15, with amean BUN of 43 � 13 mg/dL, as comparedwith initiation of therapy after 19 � 27 dayswith a mean BUN of 94 � 28 mg/dL inthe late group. Survival was 39% in theearly initiation group, as compared with20% in the late group (p � .041). Al-though baseline demographic character-istics and severity of illness scores ofpatients in the two groups were compa-rable, a greater percentage of patients inthe late cohort had multisystem organfailure or sepsis. In addition, the rationalefor earlier as opposed to later initiation oftherapy in individual patients was notprovided. Although the timing of therapymay merely have reflected random varia-tion in physician behavior, it is morelikely that individual patient characteris-tics contributed to nonrandom decisionsregarding the timing of therapy. For ex-ample, 56% of patients in the early initi-ation arm were oliguric on the first day ofCRRT, as compared with only 39% in thelate initiation group. Thus, the observeddifferences in outcome may well havebeen confounded by unreported differ-ences underlying the decision for early asopposed to late initiation of therapy.

Similar results have been reported intwo retrospective analyses of timing ofCRRT in patients following cardiac sur-gery (26, 27). Dr. Demirkiliç and col-leagues (26) reported on a series of 61patients treated at a single center in Tur-key between March 1992 and September2001 who required postoperative contin-uous venovenous hemodiafiltration(CVVHDF) after undergoing cardiac sur-gery. In the 27 patients treated beforeJune 1996, CVVHDF was not initiated un-til the serum creatinine level exceeded 5mg/dL or the serum potassium level ex-ceeded 5.5 molEq/L despite medical ther-apy, independent of urine output (group1), while in the remaining 34 patients,treated after June 1996, CVVHDF was ini-tiated as soon as the urine volume was�100 mL for 8 hrs, despite administra-tion of furosemide (group 2). Treatmentwas initiated 2.6 � 1.7 days after surgeryin group 1 as compared with 0.9 � 0.3days after surgery in group 2. Early initi-ation was associated with lower intensivecare (17.6% vs. 48.1%; p � .05) and hos-pital mortality (23.5% vs. 55.5%; p � .05)and decreased duration of both mechan-ical ventilation and intensive care lengthof stay. Similarly, Dr. Elahi and col-

leagues (27) identified 64 consecutive pa-tients who underwent cardiac surgery be-tween January 2002 and January 2003 ina single center in the United Kingdomand who received postoperative CVVHDF.In 28 patients, the CVVHDF was not ini-tiated until BUN was �84 mg/dL, thecreatinine was �2.8 mg/dL, or the serumpotassium was �6 molEq/L despite med-ical therapy, regardless of urine output(group 1), while in the remaining 36 pa-tients CVVHDF was initiated when theurine volume was �100 mL for 8 hrsdespite furosemide infusion (group 2). Asin the prior study, the reported demo-graphic and baseline clinical characteris-tics of the two groups were similar. Theinterval between surgery and initiation ofrenal support was 2.6 � 2.2 days in group1 as compared with 0.8 � 0.2 days ingroup 2. Hospital mortality was 43% ingroup 1 and 22% in group 2 (p � .05).

Dr. Piccinni and colleagues (28) havedescribed the use of “early isovolemic he-mofiltration” in oliguric patients withsepsis. In their report, they compare out-comes in 40 patients with sepsis and oli-guria in whom high volume (45 mL/kg/hr) hemofiltration with no net fluidremoval was initiated within 12 hrs ofintensive care admission as comparedwith 40 historical controls, in whom re-nal replacement therapy was initiated forconventional indications. In the patientstreated with isovolemic hemofiltration,28-day survival was 55% as comparedwith 27.5% in the historical control co-hort. Although the authors describe earlyisovolemic hemofiltration in terms ofearly initiation of therapy, comparison ofthe interval between onset of renal dys-function and initiation of therapy is notprovided and biochemical indexes of re-nal function at initiation of renal supportare similar, with a trend toward higherBUN and serum creatinine in the earlyisovolemic hemofiltration cohort (110 �38 mg/dL vs. 120 � 30 mg/dL and 1.7 �2 vs. 1.8 � 2, respectively), suggestingthat the actual timing of treatment wasnot significantly different betweengroups.

Dr. Liu and colleagues (29) analyzeddata on the timing of initiation of renalreplacement therapy (both intermittentand CRRT) from the Program to ImproveCare in Acute Renal Disease, a multi-center observational study of AKI. Theystratified the 243 patients in the databasewho received RRT into early and late ini-tiation groups based on the median BUN(76 mg/dL) at initiation of therapy. Al-

though patients in the late (BUN, �76mg/dL) group had a reduced burden oforgan failure, the survival rates at 14 daysand 28 days in this group (75% and 59%,respectively) were slightly lower than inthe early (BUN, �76 mg/dL) group (8%and 65%, respectively). After adjustmentfor age and clinical factors and stratifica-tion by site and initial modality of RRT ina multivariate analysis, the relative risk ofdeath associated with dialysis initiationwith more severe azotemia (using theearly initiation group as the comparator)was 1.85 (95% confidence interval, 1.16–2.96). Using a propensity analysis to ad-just for factors predicting initiation oftherapy at a higher as compared with alower BUN, they also found an increasedrelative risk for death in the high BUNgroup (2.07; 95% confidence interval,1.30–3.29).

There are several important limita-tions that must be considered in analyz-ing the results of these retrospectivestudies. First, the use of BUN as a surro-gate measure for duration of AKI is prob-lematic. Urea generation is not constant,with wide variation in generation ratesfrom patient to patient and even withinan individual patient as the degree ofcatabolic stress waxes and wanes. Simi-larly, the volume of distribution of ureain critically ill patients is highly variableand inconstant. Thus, the rate of increasein BUN will vary across patients, and maynot even be constant in an individualpatient over time. Second is the issue ofbias by indication. Renal support was ini-tiated for oliguria in the early groups andfor azotemia or hyperkalemia in the lategroups in both of the postcardiac surgerystudies (26, 27). Although the reasons forearly and late initiation of treatment inthe studies by Dr. Gettings and col-leagues (25) and Dr. Liu and colleagues(29) were not specified, it is likely earlierinitiation was prompted by volume over-load and electrolyte disturbances,whereas late initiation of therapy wasmore likely to be prompted by progres-sive azotemia. Whether there is a rela-tionship between indication for therapyand outcome is not known. Most impor-tantly, the design of all four of these stud-ies limited analysis to patients who re-ceived renal replacement therapy,ignoring the subset of patients with AKIwho recover or die without RRT.

Only one recent study has attemptedto address the timing of CRRT prospec-tively (12). Dr. Bouman and colleaguesrandomized 106 critically ill patients


with AKI at two centers to three groups:early high-volume CVVHDF (n � 35),early low-volume CVVHDF (n � 35),and late low-volume CVVHDF (n � 36).Treatment was initiated in the two earlygroups within 12 hrs of meeting studyinclusion criteria (the presence of oligu-ria for �6 hrs despite hemodynamic op-timization or a measured creatinineclearance of �20 mL/min on a 3-hr timedurine collection), while in the late grouprenal support was not initiated until BUNwas �112 mg/dL, potassium was �6.5mEq/L, or pulmonary edema was present.No significant differences in survivalwere observed among the three groups.Of note, however, the overall 28-daymortality for subjects in this study wasonly 27%, substantially lower thanmortality rates reported in most otherstudies of critically ill patients withAKI, suggesting a lower disease burdenin this cohort. In addition, as a result ofthe small sample size, the statisticalpower of the study was low.

Future Directions

From the above discussion, it shouldbe readily apparent that current data re-

main inadequate to answer the questionof appropriate indications and timing ofinitiation of renal support in AKI.Whether earlier initiation of RRT or pro-vision of therapy in patients currentlymanaged conservatively improves sur-vival remains an open question. Althoughobservational data suggests improvedoutcomes with early initiation of therapy,this may merely reflect inclusion of pa-tients with a lesser degree of organ in-jury, whose outcomes would be betterregardless of treatment strategy. All ofthe observational studies have been basedon identification of patients with AKI whoultimately were treated with RRT. How-ever, the vast majority of patients withAKI are never treated with RRT (30).Thus, the paradigm for answering thequestion of timing of initiation of RRTneeds to change (Fig. 1). To be able toanswer the question, a study must in-clude all patients for whom early initi-ation of RRT is a consideration. Whileobservational data that include patientswho never receive RRT may help informthe question, observational data cannotprovide a definitive answer because ofthe issue of bias by indication.

Ultimately, a definitive answer will re-quire prospective randomized trials.However, the design of such trials posessignificant challenges, most critically theneed for early identification of patientswith persistent and severe renal injury.Unfortunately, current clinical and bio-chemical parameters are inadequate toprospectively identify the appropriatestudy cohort. In the study by Dr. Boumanand colleagues (12), six of the 36 patients(16.7%) in the late therapy group neverreceived RRT, two patients because theydied before meeting criteria for RRT andfour patients because they recovered re-nal function. Although the RIFLE criteria(a mnemonic for the progression of anacronym for staging of AKI as risk ofrenal dysfunction, injury to the kidney,failure of kidney function, loss of kidneyfunction, and end-stage kidney disease)have been proposed to help stratify sever-ity of renal injury (31), they are inade-quate for identification of the necessarystudy cohort. In a retrospective analysisof �5000 critically ill patients at a singleinstitution, Dr. Hoste and colleagues (30)observed that only 14.2% of patientsmeeting the RIFLE-Loss criteria (corre-sponding to the Acute Kidney Injury Net-work criteria for stage 3 AKI) receivedRRT. Without reliable markers to identifythis population, a substantial number of

patients who would not otherwise receiveRRT will need to be randomized into anearly therapy arm and subjected to therisks of RRT, risks that include morbidityand mortality associated with vascularcatheters and the possibility that expo-sure to hemodialysis may in and of itselfdelay recovery of renal function and ad-versely impact patient survival (32).Thus, robust biomarkers and/or clinicalpredictors of the course of AKI are neededbefore such a study can be ethically un-dertaken.

CONCLUSIONS

The optimal timing for initiation ofRRT in patients with AKI remains uncer-tain. Conventionally accepted indicationsinclude volume overload, hyperkalemia,metabolic acidosis, overt uremia, andeven progressive azotemia in the absenceof specific symptoms; however, precisedefinitions for these indications are lack-ing. In addition, a number of observa-tional and retrospective analyses havesuggested improved survival with earlierinitiation of renal support; however, theabsence of inclusion of patients with AKIwho meet criteria for early initiation ofRRT but who never receive therapy limitsthe validity of these analyses. A definitiveanswer to this important clinical issueultimately will require data from prospec-tive randomized controlled trials; how-ever, the conduct of such trials requiresmore robust biomarkers and/or clinicalpredictors of the course of AKI than arecurrently available.

REFERENCES

1. Overberger P, Pesacreta M, Palevsky PM:Management of renal replacement therapy inacute kidney injury: A survey of practitionerprescribing practices. Clin J Am Soc Nephrol2007; 2:623–630

2. Levy EM, Viscoli CM, Horwitz RI: The effectof acute renal failure on mortality. A cohortanalysis. JAMA 1996; 275:1489–1494

3. Bates DW, Su L, Yu DT, et al: Correlates ofacute renal failure in patients receiving par-enteral amphotericin B. Kidney Int 2001;60:1452–1459

4. Chertow GM, Levy EM, Hammermeister KE,et al: Independent association between acuterenal failure and mortality following cardiacsurgery. Am J Med 1998; 104:343–348

5. Metnitz PG, Krenn CG, Steltzer H, et al:Effect of acute renal failure requiring renalreplacement therapy on outcome in criticallyill patients. Crit Care Med 2002; 30:2051–2058

6. Chertow GM, Burdick E, Honour M, et al:

Figure 1. Design of clinical trials of timing ofinitiation of renal replacement therapy (RRT) inacute kidney injury (AKI). A, design of publishedobservational studies. Study cohorts have con-sisted of patients treated with RRT and have notincluded patients with AKI who never receiveRRT. B, required study design to adequately an-swer clinical question of timing of RRT in AKI.Not all patients included in the late initiationstrategy actually will receive RRT; conversely,some patients included in the early initiationstrategy would not be managed with RRT if in-cluded in the late initiation strategy.

S227Crit Care Med 2008 Vol. 36, No. 4 (Suppl.)

Acute kidney injury, mortality, length ofstay, and costs in hospitalized patients. J AmSoc Nephrol 2005; 16:3365–3370

7. Liano F, Junco E, Pascual J, et al: The spec-trum of acute renal failure in the intensivecare unit compared with that seen in othersettings. The Madrid Acute Renal FailureStudy Group. Kidney Int 1998; 66(Suppl):S16–S24

8. Liano F, Pascual J: Epidemiology of acuterenal failure: A prospective, multicenter,community-based study. Madrid Acute RenalFailure Study Group. Kidney Int 1996; 50:811–818

9. Uchino S, Kellum JA, Bellomo R, et al: Acuterenal failure in critically ill patients: A mul-tinational, multicenter study. JAMA 2005;294:813–818

10. Ronco C, Bellomo R, Homel P, et al: Effectsof different doses in continuous veno-venoushaemofiltration on outcomes of acute renalfailure: A prospective randomised trial. Lan-cet 2000; 356:26–30

11. Schiffl H, Lang SM, Fischer R: Daily hemo-dialysis and the outcome of acute renal fail-ure. N Engl J Med 2002; 346:305–310

12. Bouman CS, Oudemans-Van Straaten HM,Tijssen JG, et al: Effects of early high-volumecontinuous venovenous hemofiltration onsurvival and recovery of renal function inintensive care patients with acute renal fail-ure: A prospective, randomized trial. CritCare Med 2002; 30:2205–2211

13. Saudan P, Niederberger M, De Seigneux S, etal: Adding a dialysis dose to continuous he-mofiltration increases survival in patientswith acute renal failure. Kidney Int 2006;70:1312–1317

14. Mehta RL, McDonald B, Gabbai FB, et al: Arandomized clinical trial of continuous ver-

sus intermittent dialysis for acute renal fail-ure. Kidney Int 2001; 60:1154–1163

15. Augustine JJ, Sandy D, Seifert TH, et al: Arandomized controlled trial comparing inter-mittent with continuous dialysis in patientswith ARF. Am J Kidney Dis 2004; 44:1000–1007

16. Uehlinger DE, Jakob SM, Ferrari P, et al:Comparison of continuous and intermittentrenal replacement therapy for acute renalfailure. Nephrol Dial Transplant 2005; 20:1630–1637

17. Vinsonneau C, Camus C, Combes A, et al:Continuous venovenous haemodiafiltrationversus intermittent haemodialysis for acuterenal failure in patients with multiple-organdysfunction syndrome: A multicentre ran-domised trial. Lancet 2006; 368:379–385

18. Teschan P, Baxter C, O’Brian T, et al: Pro-phylactic hemodialysis in the treatment ofacute renal failure. Ann Intern Med 1960;53:992–1016

19. Teschan PE: Acute renal failure during theKorean War. Ren Fail 1992; 14:237–239

20. Parsons FM, Hobson SM, Blagg CR, et al:Optimum time for dialysis in acute reversiblerenal failure. Description and value of animproved dialyser with large surface area.Lancet 1961; 1:129–134

21. Fischer RP, Griffen WO Jr, Reiser M, et al:Early dialysis in the treatment of acute renalfailure. Surg Gynecol Obstet 1966; 123:1019–1023

22. Kleinknecht D, Jungers P, Chanard J, et al:Uremic and non-uremic complications inacute renal failure: Evaluation of early andfrequent dialysis on prognosis. Kidney Int1972; 1:190–196

23. Conger JD: A controlled evaluation of pro-phylactic dialysis in post-traumatic acute

renal failure. J Trauma 1975; 15:1056 –1063

24. Gillum DM, Dixon BS, Yanover MJ, et al: Therole of intensive dialysis in acute renal fail-ure. Clin Nephrol 1986; 25:249–255

25. Gettings LG, Reynolds HN, Scalea T: Out-come in post-traumatic acute renal failurewhen continuous renal replacement therapyis applied early vs late. Intensive Care Med1999; 25:805–813

26. Demirkiliç U, Kuralay E, Yenicesu M, et al:Timing of replacement therapy for acute re-nal failure after cardiac surgery. J Card Surg2004; 19:17–20

27. Elahi MM, Lim MY, Joseph RN, et al: Earlyhemofiltration improves survival in post-cardiotomy patients with acute renal failure.Eur J Cardiothorac Surg 2004; 26:1027–1031

28. Piccinni P, Dan M, Barbacini S, et al: Earlyisovolaemic haemofiltration in oliguric pa-tients with septic shock. Intensive Care Med2006; 32:80–86

29. Liu KD, Himmelfarb J, Paganini E, et al:Timing of initiation of dialysis in critically illpatients with acute kidney injury. Clin J AmSoc Nephrol 2006; 1:915–919

30. Hoste EA, Clermont G, Kersten A, et al:RIFLE criteria for acute kidney injury areassociated with hospital mortality in criti-cally ill patients: A cohort analysis. Crit Care2006; 10:R73

31. Kellum JA, Levin N, Bouman C, et al: Devel-oping a consensus classification system foracute renal failure. Curr Opin Crit Care2002; 8:509–514

32. Conger J: Does hemodialysis delay recoveryfrom acute renal failure? Semin Dial 1990;3:146–148


Claudio RoncoZaccaria Ricci

Renal replacement therapies:physiological review

Received: 21 April 2008Accepted: 23 July 2008Published online: 13 September 2008� Springer-Verlag 2008

C. Ronco ())Department of Nephrology,Dialysis and Transplantation,S.Bortolo Hospital, Viale Rodolfi,36100 Vicenza, Italye-mail: [email protected].: ?39-0444-993869Fax: ?39-0444-993949

Z. RicciDepartment of Pediatric Cardiosurgery,Bambino Gesu Hospital,Piazza S. Onofrio, 4, 00100 Rome, Italye-mail: [email protected].: ?39-06-68592449Fax: ?39-06-68592552

Abstract Introduction: A physio-logical review on renal replacementtherapies (RRT) is a challenging task:there is nothing ‘‘physiologic’’ aboutRRT, since the most accurate, safeand perfectly delivered extracorporealtherapy would still be far from‘‘physiologically’’ replacing thefunction of the native kidney.Methods: This review will addressthe issues of physiology of fluid andsolute removal, acid base control andimpact on mortality during intermit-tent and continuous therapies:different RRT modalities and relativeprescriptions will provide different‘‘physiological clinical effects’’ tocritically ill patients with acutekidney injury (AKI), with the aim ofrestoring lost ‘‘renal homeostasis’’.On the other side, however, the

‘‘pathophysiology’’ of RRT, consistswith unwanted clinical effects causedby the same treatments, generallyunder-recognized by current literaturebut often encountered in clinicalpractice. Physiology and pathophysi-ology of different RRT modalitieshave been reviewed. Conclu-sion: Physiology andpathophysiology of RRT often coex-ist during dialysis sessions.Improvement in renal recovery andsurvival from AKI will be achievedfrom optimization of therapy andincreased awareness of potentialbenefits and dangers deriving fromdifferent RRT modalities.

Keywords Renal failure �Dialysis and hemofiltration

Physiology of RRT

Fluid removal

All critically ill patients need a high daily amount ofvolume infusions: blood and fresh frozen plasma, vaso-pressors and other continuous infusions, parenteral andenteral nutrition, which should be delivered withoutrestriction or interruption. It is not uncommon for patientswith acute kidney injury (AKI) and associated septicshock, to receive large amounts of fluid resuscitation,leading to fluid overload. The consequent positive fluidbalance and tendency to interstitial edema causes thenecessity for water removal and, possibly, the achieve-ment of a negative daily fluid balance. The process of

ultrafiltration occurs when plasma water is driven by ahydrostatic force through an extracorporeal circuit acrossa semi-permeable membrane and then removed frompatient total body water. Extracorporeal renal replacementtherapies (RRT) are typically utilized for ultrafiltration.Ultrafiltered water has a similar osmolarity to plasmawater; for this reason the process of ‘‘isolated ultrafiltra-tion’’ substantially corresponds to blood dehydration, withpossible increase of hematocrit values and without mod-ification of (small) solutes concentration.

Continuous renal replacement therapy (CRRT) slowlyand continuously removes patient’s plasma water, mim-icking urine output, whereas thrice weekly intermittenthemodialysis (IHD) must extract, in few hours, the

Intensive Care Med (2008) 34:2139–2146DOI 10.1007/s00134-008-1258-6 REVIEW

equivalent of 2 days of administered fluids plus excessbody water which may be present in the anuric patient.The intravascular volume depletion associated withexcessive ultrafiltration rate is due to both the high rate offluid removal required and the trans-cellular and inter-stitial fluid shifts caused by the rapid dialytic loss ofsolute [1]. The major consequence of rapid fluid removalis hemodynamic instability. In the case of a septic patientwith AKI who is receiving a high amount of vasopressorsbecause of hemodynamic instability and is needingappropriate fluid resuscitation, supplementation of nutri-tion and blood product administration. The renalreplacement modality of choice seems to be the one thatwarrants slow fluid removal, prolonged it for many hoursper day, in order to easily meet the higly variable requireddaily fluid balance. In particular, when volemic and ure-mic control are not a problem, an aggressive, protein-richnutritional policy (1.5–2.5 g/day) can be implemented inthe care of AKI patients receiving CRRT, resulting in amarked improvement in daily nitrogen balance withpossible favorable effects on immune function and overalloutcome [2–5]. Safe prescription of fluid loss during RRTrequires intimate knowledge of the patient’s underlyingcondition, understanding of the process of ultrafiltrationand close monitoring of the patient’s cardiovascularresponse to fluid removal. In order to preserve tissueperfusion in patients with AKI, it is important to optimizefluid balance by removing patient’s excess water withoutcompromising the effective circulating fluid volume. It isstill a matter of controversy which clinical parameter(actual patient weight/patient dry weight, mean arterialpressure, central venous pressure, wedge pressure, sys-temic saturation, mixed venous saturation, bioimpedance,etc.) or currently available monitorization (central venouscatheter, swan ganz catheter, transesophageal echocardi-ography, etc.) should be utilized in order to uniformlydefine the concept of ‘‘volume overload’’. In patients whoare clinically fluid overloaded, however, it is extremelyimportant to accurately evaluate the amount of fluid toremove [6]. In these kinds of patients, one of the mainfeatures of slow and constant ultrafiltration is the possi-bility for interstitial fluid to slowly and constantly refillthe ‘‘dehydrated’’ bloodstream. This phenomenon is dri-ven by hydrostatic and osmotic forces and allows for theelimination of high plasma water volumes per day with areduced risk of hypovolemia and hypotension. In criti-cally ill children the correction of water overload isconsidered a priority. It has been show that restoringadequate water content in small children is the mainindependent variable for outcome prediction [7, 8]. Sim-ilar results have been recently found in a large cohort ofadult critically ill patients with AKI [9].

All patients who are at risk of, or who already have,increased intracranial pressure (neurosurgical patients,patients with encephalitis or meningo-encephalitis, trau-matic brain injury or acute liver failure) CRRT in

preference to IHD is strongly indicated in case of AKI. Inan evaluation of brain density by repeated computerizedtomography (CT), La Greca and co authors found that afterintermittent dialysis densitometric values fell significantlyfrom normal values (27–40 Hounsfield units), especially inthe basal ganglia; this might point to a build-up of fluid inthe parenchyma. No significant variations was observed inrepeated examinations of three normal individuals. Inter-estingly, the authors found no significant relation betweenbiochemical data and the CT pattern, hypothesizing thatdecreased density of brain tissue and post-dialysis edemamight be a consequence of the local production of ‘‘hy-drogenic osmoles’’, due to a reduction in the pH factor ofCSF and cerebral tissue [10]. CRRT has been shown toprevent intracranial pressure increase associated withintermittent renal replacement therapies [11].

Solute removal

Physiology of solute removal is one of the main issues ofdialysis and it is partially responsible for safety, tolera-bility and outcome of extracorporeal RRTs. Soluteremoval is a very broad concept that is generally descri-bed by the elimination of a marker solute. This markersolute should be reasonably representative of all solutesthat are normally removed from blood by the kidney.

Unfortunately, a reference solute that represents all thesolutes accumulating during AKI is not currently avail-able, because kinetics and volume of distribution aredifferent for each molecule. Then, ‘‘single-solute control’’during RRT represents only a rough estimate of treatmentefficiency. With these specifications, urea is generallyutilized as ‘‘imperfect’’ marker molecule, because of itsaccumulation in all patients with AKI and the ease ofserum level measurement. Furthermore, in spite of itsmoderate toxicity, urea is the final product of proteinmetabolism; its accumulation describes the need for dial-ysis and its removal describes treatment efficiency. It is asmall molecule and its volume of distribution is similar tototal body water. It is not bound to protein and it freelypasses through tissues and cell membranes. Creatinine hassimilar characteristics and is another commonly usedmarker solute. One of the measures utilized to quantifyurea/creatinine removal is dialysis dose. An in depthanalysis of the concept of dialysis dose goes beyond theaim of this review and can be found elsewhere [12]. One ofthe main aspects of dose to be understood, however, is theconcept of clearance (K); K is the volume of blood clearedfrom a given solute over a given time. K does not reflectthe overall solute removal rate (mass transfer) but rather itsvalue normalized by the serum concentration. Even whenK remains stable over time, the removal rate will vary ifthe blood levels of the reference molecule change. Kdepends on solute molecular size, intercompartmentaltransmittance (Kc), transport modality (diffusion or

2140

convection), and circuit operational characteristics (bloodflow rate, dialysate flow rate, ultrafiltration rate, hemodi-alyzer type and size). In its original conceivement, K isutilized to evaluate renal function among disparate indi-viduals whose kidneys, however, are operating 24 h perday and urea/creatinine blood levels are at steady state. Forthis reason, the concept of K is easily applicable to con-tinuous treatments and its utilization to describeintermittent therapies efficiency is a sort of ‘‘adaptation’’.For this reason, since K represents only the instantaneousefficiency of the system, during treatments with differenttime schedules the information about the time span duringwhich K is delivered is fundamental to compare the dif-ferent RRT doses. For example, K is typically higher inIHD than in continuous renal replacement therapy (CRRT)and sustained low efficiency daily dialysis (SLEDD).However, mass removal may be greater during SLEDD orCRRT because the K is applied for 12/24 h (Table 1). Inany case, from a physiological point of view, even if acontinuous and an intermittent therapy were prescribed inorder to provide exactly the same marker solute removal,still they could not be comparable: during continuoustreatments, where a relatively low K is applied, a slow butprolonged removal of solutes approaches a pseudo-steadystate slope (Fig. 1). In highly intermittent therapies, theintensive K, limited to 4–6 h per day, thrice a week, causesthe saw tooth slope in solute removal and the eventualrebound during the time span without treatment. Thesepeaks and valleys of solutes, bicarbonate, electrolytes,plasma osmolarity and volemia are not physiologic andmight have a detrimental impact on patients’ hemody-namics, electrolyte, acid base and other ‘‘osmoles’’balance. Furthermore, in the case of IHD, the Kc, i.e., thevariable tendency of different tissues to ‘‘release’’ a soluteinto the bloodstream, is much more relevant than duringslow efficiency treatments. As a matter of fact, finally,solute control is optimized during CRRT: it has beencalculated that if the solute target was, for example, in a70-kg patient, a mean blood urea nitrogen level of 60 mg/dL, this would be easily obtainable with a ‘‘standard’’CVVH dose, but it might be very difficult to be reached by

even intensive IHD regimens [13]. Some authors haverecently suggested to express CRRT daily dose as Kindexed to patient body weight. It is now recommended toadminister a CRRT dose of at least 35 mL/(h kg) 9 24 h[14–16]. Simplifying for low molecular weight solutes(see [15]), K equals replacement solution and/or dialysateflow, and CRRT ‘‘standard’’ dose may be expressed in our70-kg patient as about 2,500 mL/h (35 mL/h 9 70 kg) per24 h or 60 L/day (2,500 mL/h 9 24 h) of replacementsolution during continuous venovenous hemofiltration(CVVH) or of dialysate during continuous venovenoushemodialysis (CVVHD) [12]. It is expected that this rec-ommended dose to be modified in the next years, after theproduction of new evidence in this field (see below).

Acid–base control

Few studies have been done to detect which modality isbetter in terms of acid–base control [17]. Oligo anuricpatients often have mild acidemia secondary to increasedunmeasured anions (strong ion gap: SIG 12.3 mEq/L),hyperphosphatemia and hyperlactatemia. This acidosis isattenuated by the alkalizing effect of hypoalbuminemia.Uchino and co-workers [18] compared the effect on acid–base balance of IHD and CVVHDF. Before treatment,metabolic acidosis was common in both groups (63.2%for IHD and 54.3% for CVVHDF). Both IHD andCVVHDF corrected metabolic acidosis. However, therate and degree of correction differed significantly.

Table 1 Different simultaneous clearances (K) may lead to almostequivalent final daily clearance (K 9 time). Nonetheless, continu-ous therapies (CRRT) seem to achieve a final better urea control(urea removed) with resepct to intermittent hemodialysis (IHD) dueto the different ‘‘physiology’’ of solute removal (slow, steady,prolonged and independent from tissues intercompartmentaltransmittance)

CRRT IHD

K (mL/min) 35 200Urea start (mg/dL) 110 110Urea end (mg/dL) 90 30Treatment time (min) 1440 240Total K (K 9 time) 50.5 48Urea removed (g) 25 18

Fig. 1 Typical solute removal patterns during different renalreplacement therapies. The slow and steady clearance of continuoustreatments allows lower average serum urea levels than duringintermittent therapies, also avoiding potentially dangerous peaks ofsolute increase. Furthermore, since clearance does not reflect theoverall solute removal rate (mass transfer) but rather its valuenormalized by the solute serum concentration, when soluteconcentration rapidly decreases (intermittent dialysis) the finalresult will be a lower mass transfer than when solute levels aresteady (continuous treatments)

2141

CVVHDF normalized metabolic acidosis more rapidlyand more effectively during the first 24 h than did IHD(P \ 0.01). IHD was also associated with a higher inci-dence of metabolic acidosis than was CVVHDF duringthe subsequent 2 weeks treatment period. Accordingly,CVVHDF can be considered physiologically superior toIHD in the correction of metabolic acidosis. In a com-parison between CVVH and peritoneal dialysis, allpatients randomized to CVVH achieved correction ofacidosis by 50 h of treatment, compared with only 15% ofthose treated by peritoneal dialysis (P \ 0.001) [19].

Rocktaschel showed that once CVVH is commenced,acidemia is corrected within 24 h. This change is asso-ciated with a decreased SIG, and decreased phosphate andchloride concentrations. After 3 days of CVVH, patientsdevelop alkalemia secondary to metabolic alkalosis due toa further decrease in SIG and in serum phosphate con-centration in the setting of persistent hypoalbuminemia[20].

Mortality

It has been hypothesized that safety features and ‘‘phys-iology’’ of different RRT modalities delivered for AKI incritically ill patients might partially give explanation fordifferent clinical outcomes such as mortality, recovery ofrenal function and dialysis independence. Theoreticallyspeaking, CRRT seems to represent the ideal therapy inorder to ‘‘physiologically’’ restore renal homeostasis,steadily achieve adequate fluid balance, maintain hemo-dynamic stability and effectively control metabolicderangements of AKI and septic syndrome. However,four recently published randomized clinical trials and onemulticenter observational study contested that outcomeswith CRRT are superior to those of IHD [21–25]. None ofthese studies showed a superior outcome for CRRTcompared with IHD. The results of these studies in somecases have been criticized for methodology and groupsrandomization [26]. Nevertheless, they do not support thebelief that CRRT provides better outcomes than IHD.Three meta-analyses published in the last year agreed thatCRRT does not differ from IHD with respect to in-hos-pital mortality, ICU mortality, number of survivingpatients not requiring RRT and hemodynamic instability[27–29]. One of the common key points of these findingscan be that IHD has become safer and more efficaciouswith contemporary dialytic techniques. Furthermore, aliberal and extended use of CRRT might have becomeless safe and/or efficacious than previously considered orexpected. The concept that CRRT can provide morehemodynamic stability, more-effective volume homeo-stasis and better blood pressure support than IHD hasbeen challenged by technical advances in the delivery ofIHD, that have dramatically decreased the propensity ofIHD to cause intradialytic hypotension. These advances

include the introduction of volume-controlled dialysismachines, the routine use of biocompatible syntheticdialysis membranes, the use of bicarbonate based dialy-sate and the delivery of higher doses of dialysis, theutilization of ‘‘hybrid techniques’’ such as extended dailydialysis [30, 31]. In an important study, Schortgenet al.32] demonstrated that there was a lower rate ofhemodynamic instability and better outcomes afterimplementation of a clinical practice algorithm designedto improve hemodynamic tolerance to IHD. Recommen-dations included priming the dialysis circuit with isotonicsaline, setting dialysate sodium concentration at above145 mmol/L, discontinuing vasodilator therapy and set-ting dialysate temperature to below 37�C. Thus, theoriginal rationale for the widely held assumption thatCRRT is a superior therapy may have dissipated overtime [33].

There is also urgent need for prospective high-qualityand suitably powered trials to adequately address theimpact of ‘‘dialysis dose’’ on mortality. New high level ofevidence is however coming from two very recent trials.A (small) randomized controlled trial on 200 critically illpatients with AKI concluded that patient survival or renalrecovery was not different between patients receivinghigh-dosage (35 mL/(kg h) or standard-dosage (20 mL/(kg h) CVVHDF [34]. In the second study, under thesponsorship of The Veterans Affairs/National Institutes ofHealth (VA/NIH) Acute Renal Failure Trial Network,1,124 critically ill patients with AKI and failure of at leastone nonrenal organ or sepsis were randomly assigned toreceive intensive or less-intensive RRT [35]. The studywas a multicenter, prospective, randomized, parallel-group trial conducted between November 2003 and July2007 at 27 VA and university-affiliated medical centers.In both groups only hemodynamically stable patientsunderwent IHD, whereas hemodynamically unstablepatients underwent CVVHDF or SLED. Patients receiv-ing the intensive treatment strategy underwent IHD andSLED six times per week and CVVHDF at 35 mL/h perkilogram of body weight; for patients receiving the less-intensive treatment strategy, the corresponding treatmentswere provided thrice weekly and at 20 mL/h per kilo-gram. The rate of death from any cause by day 60 was53.6% with intensive therapy and 51.5% with less-inten-sive therapy. There was no significant difference betweenthe two groups in the duration of RRT or in the rate ofrecovery of kidney function or nonrenal organ failure.This is the first multicenter clinical trial with adequatestatistical power on different RRT strategies, so far. Thefindings of this study contrast with other single centertrials [18, 19] and are similar to smaller studies byTolwani and Bouman [34, 36]. However, these results addhigh level of evidence to the debate on dialysis dose andare going to be discussed for a long time. Nonetheless,many concerns about external validity of the study haverisen. First of all, patients were allowed to be transitioned

2142

from one dialysis method to another. In this condition, asdiscussed above, the dialysis dose is impossible to com-pare and unlikely to be equivalent. Finally it is not knownif the study findings can be generalized to different healthcare systems from United States and to different RRTapproaches (for example the sole use of continuousvenovenous hemofiltration of many European and Aus-tralian centers). In any case, it is possible that otherstrategies than only increasing RRT dose might help AKIpatients. As the accompanying editorial correctly pointsout, current approaches to dialysis are probably adequateto replace critical functions such as regulation of volumeand electrolyte and acid–base homeostasis. Still lackingare methods that efficiently down-regulate the inflam-matory response, which might play a major role in thepathophysiology of AKI [37].

Pathophysiology of RRT

Although considerable attention has focused on the per-ceived benefits of CRRT, there has been less emphasis onthe possibility that CRRT might confer increased risk[38]. As a continuous extracorporeal therapy, CRRTrequires continuous contact of patient’s blood with for-eign surfaces. This event activates the coagulation andcomplement cascade, leukocytes and platelets [39].Activated leukocytes release inflammatory mediators andinduce oxidative stress, transforming lipids and proteinsand contributing to endothelial injury. Activated plateletsaggregate and stimulate thrombin generation. Thus, bio-incompatibility of RRT materials potentially enhancescoagulation and inflammation pathways that are alreadytriggered in the critically ill patients and that RRT iscalled to treat.

Continuous anticoagulation does increase bleedingrisk. Conversely, clotting of the extracorporeal circuit alsooccurs frequently with CRRT, which might contribute toblood loss and could exacerbate anemia in critically illpatients. Interestingly, it has been shown that citrateanticoagulation abolishes polymorphonuclear and plateletdegranulation in the filter in chronic hemodialysis. Fur-thermore, lower levels of oxidized low-density lipoproteinwere found, indicating less lipid peroxidation [40]; citratemay thus have beneficial effects beyond the reduction ofbleeding.

RRT has important metabolic consequences because itis associated with large nonselective solute shifts. In thenormal kidney, tubular modification of the glomerularfiltrate includes re-absorption of beneficial substancessuch as amino acids, water-soluble vitamins and traceelements. During RRT these substances are lost, therebyreducing antioxidant defense. Amino acids are lostthrough the filter and it has been estimated that theyrepresent at most approximately 10% of overall amino

acid supplementation [41]. A negative balance of water-soluble vitamins, glutamine, carnitine, selenium andcopper has been shown [42]. Zinc is also lost, but totalbalance appeared to be positive, because the replacementsolution contained zinc. These micronutrients are crucialfor antioxidant defense [43]. Consequently, patients onRRT need a sufficient intake of protein and micronutrientsto compensate for increased losses. Moreover, CRRTcorrects metabolic acidosis by removing metabolic acidsand replacing buffer. In addition to citrate, lactate andbicarbonate are the most frequently used buffers. If liverfunction and tissue perfusion are not severely disturbed,and CRRT dose is sufficient, lactate buffering is generallysafe and adequate. However, the generation of bufferfrom lactate requires three molecules of oxygen for eachmolecule of bicarbonate. Furthermore, with the exclusiveuse of lactate-based fluids patients develop hyperlactat-emia [44]. Even if it is generally postulated thathyperlactatemia is not harmful, it indicates that theinfused amount of lactate (about 150–300 g a daydepending on CRRT dose) overcomes metabolic capacity[45]. If exogenous lactate supply exceeds the capacity ofthe citric acid cycle, a higher proportion of lactate is usedfor gluconeogenesis, which is not energy-effective. Inpatients with multiple organ failure, lactate buffering(compared with bicarbonate buffering) increased glucoseintolerance [45]. Furthermore, RRT may alter actualserum lactate level and impair the utilization of thismolecule as a marker of metabolic acidosis or systemichypoperfusion.

The majority of critically ill patients receive someform of antimicrobial treatment during their stay in theintensive care unit (ICU). Clinical data on the removal ofantimicrobial drugs by CVVH are scarce and in mostcases give general informations, whereas antibioticremoval rate is highly variable [46, 47]. Clearance ofantibiotics is influenced by prescribed RRT dose andmodality, protein binding, drug–membrane interaction,size and charge of molecules, characteristics and lifespanof the hemofilter. However, although interpatient vari-ability and the potential need of monitoring serum levelsof each molecule, it has been shown that for clinicalpractice dose adjustment according to predicted CVVHremoval provides a reliable estimate compared to effec-tive CVVH removal for many antibiotics (amoxicillin,ceftazidime, ciprofloxacin, flucloxacillin, fluconazole andmetronidazole) [48]. Nevertheless, plasma levels shouldbe strictly monitored when CVVH is performed for thoseantibiotics that are eliminated predominantly by the kid-ney, and that have a low therapeutic threshold (e.g.vancomycin). In this case, especially when high volumeRRT is used, the risk of therapy underdosing exists [49].

During CRRT heat is lost as a result of blood leavingthe body, cooling by the dialysate and/or the infusion oflarge amounts of replacement fluids. Although short-termadverse effects of cooling on gastric mucosal acidosis

2143

(pHi) were not seen [50], long-term cooling has hugemetabolic consequences. Among others, loss of heatimplies loss of energy, shivering, which increases oxygendemands, vasoconstriction, inhibition of leukocyte func-tion and coagulation. In an experimental septic-shockhemofiltration model, the nonheating of replacement flu-ids caused early death [51] even if there are no currentclinical trials in humans. The heater of several moderndevices for CRRT is inadequate, especially if highervolumes are exchanged. Monitoring of body temperature,prevention of hypothermia and timely provision ofexternal heat are recommended.

Also of concern are recent reports that technicalproblems with the delivery of CRRT, including machinemalfunction, medication errors and compounding errors,might contribute to increased patient morbidity andmortality. Detection of safety problems and/or adverseevents is particularly difficult when there are high rates ofexpected morbidity and mortality in the populationundergoing a procedure, as is the case with CRRT incritically ill patients with AKI. Currently, few availablestudies provide substantive information on the safety oradverse effects of CRRT or IHD. The possibility ofmaking fluid balance errors during CRRT was easier withthe early technology and the advent of automatedmachines has partially overcome this problem. Further-more, a new generation of dedicated CRRT machines hasbeen recently released with strict safety features [52, 53].Nevertheless, there are conditions and operation modes inwhich the potential for fluid balance errors is still present.For example, the simultaneous use of other extracorporealtherapies continuously tests CRRT machines safety andaccuracy features. Dr. Shaheen and colleagues [54]recently presented their experience with two differentsubgroups of children: one group that required hemofil-tration alone and the second group that required

hemofiltration and extracorporeal membrane oxygenation(ECMO). Not surprisingly, the authors identified a highermortality rate in those patients requiring CVVH andECMO compared with those patients requiring hemofil-tration alone. The authors promoted the concept thatcertain therapies should be reserved to experienced teams.Accuracy and safety features of modern CRRT machinesallow effective and harmless net ultrafiltration delivery inhigh-risk pediatric patients, when coupled with constantclinical monitorization and deep knowledge of theextracorporeal circuits. We recently showed that netultrafiltration error ranges between -1% and about 2% ofprescribed rate [55]. We currently have no information onhow ‘‘acceptable’’ is a machine-provoked error and netultrafiltration unbalance cannot be reduced to zero,because ‘‘external’’ factors must be considered, such asuncorrect opening of dialysis solution bags, delay in bagsubstitution, troubleshooting. Finally, the CRRT deliveryin parallel with ECMO circuit might have an importantrole in net ultrafiltration derangements [55].

Conclusions

Physiology and physiopathology of RRT are veryimportant and often coexistent features during dialysissessions. An in depth knowledge of both may helpoperators to approach the ideal treatment and help totailor RRT to critically ill patients. It is possible that afurther improvement in renal recovery and survival fromAKI will depend from increased awareness of potentialbenefits and dangers that different RRT modalities cancarry to different patients.

References

1. Lauer A, Alvis R, Avram M (1988)Hemodynamic consequences ofcontinuous arteriovenoushemofıltration. Am J Kidney Dis12:110–115

2. Bouffard Y, Viale JP, Annat G,Delafosse B, Guillaume C, Motin J(1987) Energy expenditure in the acuterenal failure patient mechanicallyventilated. Intensive Care Med 13:401–404

3. Frankenfield DC, Reynolds HN (1995)Nutritional effect of continuoushemodiafiltration. Nutrition 11:388–393

4. Kierdorf HP (1995) The nutritionalmanagement of acute renal failure in theintensive care unit. New Horiz3:699–707

5. Bellomo R, Tan K, Bhonagiri S, GopalI, Seacombe J, Daskalakis M, Boyce N(2002) High protein intake duringcontinuous hemodialfiltration: impacton aminoacids and nitrogen balance. IntJ Artif Organs 25:261–268

6. Gibney N, Cerda J, Davenport A,Ramirez J, Singbartl K, Leblanc M,Ronco C (2008) Volume managementby renal replacement therapy in acutekidney injury. Int J Artif Organs31:145–155

7. Goldstein SL, Currier H, Graf C, CosioCC, Brewer ED, Sachdeva R (2001)Outcome in children receivingcontinuous veno-venous hemofiltration.Pediatrics 107:1309–1312

8. Foland JA, Fortenberry JD, WarshawBL, Pettignano R, Merritt RK, HeardML, Rogers K, Reid C, Tanner AJ,Easley KA (2004) Fluid overload beforecontinuous hemofiltration and survivalin critically ill children: a retrospectiveanalysis. Crit Care Med 32:1771–1776

9. Payen D, de Pont AC, Sakr Y, Spies C,Reinhart K, Vincent JL, The SepsisOccurrence in Acutely Ill Patients(SOAP) Investigators (2008) A positivefluid balance is associated with a worseoutcome in patients with acute renalfailure. Crit Care 12:R74

10. La Greca G, Dettori P, Biasioli S,Fabris A, Feriani M, Pinna V, Pisani E,Ronco C (1980) Brain density studiesduring dialysis. Lancet 13:580

2144

11. Davenport A, Will EJ, Davison AM(1993) Effect of renal replacementtherapy on patients with combinedacute renal failure and fulminanthepatic failure. Kidney Int 43:S245–S251

12. Ricci Z, Ronco C (2008) Dose andefficiency of renal replacement therapy:continuous renal replacement therapyversus intermittent hemodialysis versusslow extended daily dialysis. Crit CareMed 36:S229–S237

13. Clark WR, Mueller BA, Alaka KJ,Macias VL (1994) A comparison ofmetabolic control by continuous andintermittent therapies in acute renallailure. J Am Soc Nephrol 4:1113–1120

14. Dellinger RP, Levy MM, Carlet JM,Bion J, Parker MM, Jaeschke R,Reinhart K, Angus DC, Brun-BuissonC, Beale R, Calandra T, Dhainaut JF,Gerlach H, Harvey M, Marini JJ,Marshall J, Ranieri M, Ramsay G,Sevransky J, Thompson BT, TownsendS, Vender JS, Zimmerman JL, VincentJL (2004) Surviving sepsis campaignguidelines for management of severesepsis and septic shock. Crit Care Med32:858–873

15. Ronco C, Bellomo R, Homel P,Brendolan A, Dan M, Piccinni P, LaGreca G (2000) Effects of differentdoses in continuous veno-venoushaemofiltration on outcomes of acuterenal failure: a prospective randomisedtrial. Lancet 356:26–30

16. Saudan P, Niederberger M, DeSeigneux S, Romand J, Pugin J,Perneger T, Martin PY (2006) Adding adialysis dose to continuoushemofiltration increases survival inpatients with acute renal failure. KidneyInt 70:1312–1317

17. Naka T, Bellomo R (2004) Bench-to-bedside review: treating acid–baseabnormalities in the intensive careunit—the role of renal replacementtherapy. Crit Care 8:108–114

18. Uchino S, Bellomo R, Ronco C (2001)Intermittent versus continous renalreplacement therapy in the ICU: impacton electrolyte and acid–base balance.Intensive Care Med 27:1037–1043

19. Phu NH, Hien TT, Mai NT, Chau TT,Chuong LV, Loc PP, Winearls C, FarrarJ, White N, Day N (2002)Hemofitlration and peritoneal dialysisin infection-associated acute renalfailure in Vietnam. N Engl J Med347:895–902

20. Rocktaschel J, Morimatsu H, Uchino S,Ronco C, Bellomo R (2003) Impact ofcontinuous veno-venous hemofiltrationon acidbase balance. Int J Artif Organs26:19–25

21. Mehta RL, McDonald B, Gabbai FB,Pahl M, Pascual MT, Farkas A, KaplanRM, Collaborative Group for Treatmentof ARF in the ICU (2001) Arandomized clinical trial of continuousversus intermittent dialysis for acuterenal failure. Kidney Int 60:1154–1163

22. Uehlinger DE, Jakob SM, Ferrari P,Eichelberger M, Huynh-Do U, MartiHP, Mohaupt MG, Vogt B, Rothen HU,Regli B, Takala J, Frey FJ (2005)Comparison of continuous andintermittent renal replacement therapyfor acute renal failure. Nephrol DialTransplant 20:1630–1637

23. Augustine JJ, Sandy D, Seifert TH,Paganini EP (2004) A randomizedcontrolled trial comparing intermittentwith continuous dialysis in patients withARF. Am J Kidney Dis 44:1000–1007

24. Vinsonneau C, Camus C, Combes A,Costa de Beauregard MA, Klouche K,Boulain T, Pallot JL, Chiche JD, TaupinP, Landais P, Dhainaut JF, For theHemodiafe Study Group (2006)Continuous venovenous haemodiafiltration versus intermittenthaemodialysis for acute renal failure inpatients with multiple-organdysfunction syndrome: a multicentrerandomised trial. Lancet 368:379–385

25. Cho KC, Himmelfarb J, Paganini E,Ikizler TA, Soroko SH, Mehta RL,Chertow GM (2006) Survival bydialysis modality in critically illpatients with acute kidney injury. J AmSoc Nephrol 17:3132–3138

26. Kellum J, Palevsky P (2006) Renalsupport in acute kidney injury. Lancet368:344–345

27. Pannu N, Klarenbach S, Wiebe N,Manns B, Tonelli M, Alberta KidneyDisease Network (2008) Renalreplacement therapy in patients withacute renal failure: a systematic review.JAMA 299:793–805

28. Bagshaw SM, Berthiaume LR, DelaneyA, Bellomo R (2008) Continuous versusintermittent renal replacement therapyfor critically ill patients with acutekidney injury: a meta-analysis. CritCare Med 36:610–617

29. Rabindranath K, Adams J, MacleodAM, Muirhead N (2007) Intermittentversus continuous renal replacementtherapy for acute renal failure in adults.Cochrane Database Syst Rev18:CD003773

30. Kielstein JT, Kretschmer U, Ernst T,Hafer C, Bahr MJ, Haller H, Fliser D(2004) Efficacy and cardiovasculartolerability of extended dialysis incritically ill patients: a randomizedcontrolled study. Am J Kidney Dis43:342–349

31. Baldwin I, Naka T, Koch B, Fealy N,Bellomo R (2007) A pilot randomisedcontrolled comparison of continuousveno-venous haemofiltration andextended daily dialysis with filtration:effect on small solutes and acid–basebalance. Intensive Care Med 33:830–835

32. Schortgen F, Soubrier N, Delclaux C,Thuong M, Girou E, Brun-Buisson C,Lemaire F, Brochard L (2000)Hemodynamic tolerance of intermittenthemodialysis in critically ill patients:usefulness of practice guidelines. Am JRespir Crit Care Med 162:197–202

33. Himmelfarb J (2007) Continuousdialysis is not superior to intermittentdialysis in acute kidney injury of thecritically ill patient. Nat Clin PractNephrol 3:120–121

34. Tolwani AJ, Campbell RC, Stofan BS,Lai KR, Oster RA, Wille KM (2008)Standard versus high-dose CVVHDFfor ICU-related acute renal failure. JAm Soc Nephrol 19:1233–1238

35. The VA/NIH Acute Renal Failure TrialNetwork (2008) Intensity of renalsupport in critically ill patients withacute kidney injury. N Engl J Med359:7–20

36. Bouman CS, Oudemans-Van StraatenHM, Tijssen JG, Zandstra DF,Kesecioglu J (2002) Effects of earlyhigh-volume continuous venovenoushemofiltration on survival and recoveryof renal function in intensive carepatients with acute renal failure: aprospective, randomized trial. Crit CareMed 30:2205–2211

37. Bonventre JV (2008) Dialysis in acutekidney injury—more is not better. NEngl J Med 359:82–84

38. Oudemans van Straaten HM (2007)Primum non nocere, safety ofcontinuous renal replacement therapy.Curr Opin Crit Care 13:635–637

39. Gorbet MB, Sefton MV (2004)Biomaterial-associated thrombosis:roles of coagulation factors,complement, platelets and leukocytes.Biomaterials 25:5681–5703

40. Gritters M, Grooteman MP, Schoorl M,Schoorl M, Bartels PC, Scheffer PG,Teerlink T, Schalkwijk CG,Spreeuwenberg M, Nube MJ (2006)Citrate anticoagulation abolishesdegranulation of polymorphonuclearcells and platelets and reduces oxidativestress during haemodialysis. NephrolDial Transplant 21:153–159

41. Wooley JA, Btaiche IF, Good KL(2005) Metabolic and nutritionalaspects of acute renal failure incritically ill Patients RequiringContinuous Renal ReplacementTherapy. Nutr in Clin Pract 20:176–191

2145

42. Story DA, Ronco C, Bellomo R (1999)Trace element and vitaminconcentrations and losses in critically illpatients treated with continuousvenovenous hemofiltration. Crit CareMed 27:220–223

43. Heyland DK, Dhaliwal R, Suchner U,Berger MM (2005) Antioxidantnutrients: a systematic review of traceelements and vitamins in the criticallyill patient. Intensive Care Med 31:327–337

44. Cole L, Bellomo R, Baldwin I, HayhoeM, Ronco C (2003) The impact oflactate buffered high volumehemofiltraiton on acid base balance.Intensive Care Med 29:1113–1120

45. Bollmann MD, Revelly JP, Tappy L,Berger MM, Schaller MD, Cayeux MC,Martinez A, Chiolero RL (2004) Effectof bicarbonate and lactate buffer onglucose and lactate metabolism duringhemodiafiltration in patients withmultiple organ failure. Intensive CareMed 30:1103–1110

46. Vidal L, Shavit M, Fraser A, Paul M,Leibovici L (2005) Systematiccomparison of four sources of druginformation regarding adjustment ofdose for renal function. BMJ 331:263

47. Trotman RL, Williamson JC,Shoemaker DM, Salzer WL (2005)Antibiotic dosing in critically ill adultpatients receiving continuous renalreplacement therapy. Clin Infect Dis41:1159–1166

48. Bouman CS, van Kan HJ, KoopmansRP, Korevaar JC, Schultz MJ, VroomMB (2006) Discrepancies betweenobserved and predicted continuous venovenous hemofiltration removal ofantimicrobial agents in critically illpatients and the effect of dosing.Intensive Care Med 32:2013–2019

49. Uchino S, Cole L, Morimatsu H,Goldsmith D, Bellomo R (2002)Clearance of vancomycin during highvolume hemofiltration: impact of pre-dilution. Intensive Care Med 28:1664–1667

50. John S, Griesbach D, Baumgartel M,Weihprecht H, Schmieder RE, Geiger H(2001) Effects of continuoushaemofiltration vs intermittenthaemodialysis on systemichaemodynamics and splanchnicregional perfusion in septic shockpatients: a prospective, randomizedclinical trial. Nephrol Dial Transplant16:320–327

51. Rogiers P, Sun Q, Dimopoulos G, Tu Z,Pauwels D, Manhaeghe C, Su F,Vincent JL (2006) Bloodwarmingduring hemofiltration can improvehemodynamics and outcome in ovineseptic shock. Anesthesiology104:1216–1222

52. Ricci Z, Bonello M, Salvatori G,Ratanarat R, Brendolan A, Dan M,Ronco C (2004) Continuous renalreplacement technology: from adaptivetechnology and early dedicatedmachines towards flexible multipurposemachine platforms. Blood Purif22:269–276

53. Ronco C, Ricci Z, Bellomo R, BaldwinI, Kellum J (2005) Management of fluidbalance in CRRT: a technical approach.Int J Artif Organs 28:765–776

54. Shaheen IS, Harvey B, Watson AR,Pandya HC, Mayer A, Thomas D(2007) Continuous venovenoushemofiltration with or withoutextracorporeal membrane oxygenationin children. Pediatr Crit Care Med8:362–365

55. Ricci Z, Morelli S, Vitale V, Di ChiaraL, Cruz D, Picardo S (2007)Management of fluid balance incontinuous renal replacement therapy:technical evaluation in the pediatricsetting. Int J Artif Organs 30:896–901

2146

Acute Kidney Injury inthe Intensive Care Unit
Susan T. Crowley, MDa,b,*, Aldo J. Peixoto, MDa,b
KEYWORDS� Acute kidney injury � Critical care � Dialysis� Renal replacement therapy � Acute tubular necrosis

Acute kidney injury (AKI) is common in hospitalizedpatients. The incidence of AKI in the intensive careunit (ICU) varies from 1.5% to 24%,1 where it is as-sociated with mortality rates as high as 78% (inpatients who require dialysis), and as many asone third of survivors may remain on chronic dialy-sis.2 In this article, we review recent advancesrelated to AKI in critically ill patients. Our goal is toprovide an update on the topic to the initiatedreader, focusing on recent developments in thedefinition of AKI, its diagnosis, and therapeuticoptions, particularly in the realm of renal replace-ment therapy (RRT).

THE SPECTRUMOFACUTE KIDNEY INJURYIN THE ICU

Recent multinational databases have providedgood detail on the incidence, spectrum, and out-comes of AKI in ICU patients. Evaluating datafrom 54 centers in 23 countries, Uchino and col-leagues1 reported on 1738 cases of severe AKIcomplicating 22,269 ICU admissions in patientsaged 12 and older. The criteria for enrollmentincluded a blood urea nitrogen higher than84 mg/dL and/or oliguria defined as urine outputless than 200 mL for 12 hours. Medical patientscontributed a larger fraction of AKI cases (59%)than surgical patients (41%), and the overall in-hospital mortality was 60%. A multivariable modelwas used to analyze the impact of different factorson mortality. It revealed significant associationsbetween mortality and older age (odds ratio [OR]1.02 [per year]), delayed fulfillment of inclusioncriteria (OR 1.02 [per day between admission

a Section of Nephrology, Yale University School of Medicb VA Connecticut Healthcare System, 950 Campbell Aven* Corresponding author. Renal Section (111F), VA ConnecHaven, CT 06516.E-mail address: [email protected] (S. T. Crowley).

Clin Chest Med 30 (2009) 29–43doi:10.1016/j.ccm.2008.09.0020272-5231/08/$ – see front matter. Published by Elsevier I

and inclusion in study]), Simplified Acute Physio-logic Score II (OR 1.02 [per point]), mechanicalventilation (OR 2.11), use of vasopressors and/orinotropes (OR 1.95), a hematological medical diag-nosis as cause for admission to the ICU (OR 2.7),sepsis (OR 1.36), cardiogenic shock (OR 1.41),hepatorenal syndrome (OR 1.87), admission toa ‘‘specific’’ ICU compared with a general ICU(OR 1.64), and a lower number of ICU beds (OR0.57 for ICUs with fewer than 10 beds comparedwith those with more than 30 beds).1

The timing of development of AKI in relationshipto the admission to the ICU may have relevant prog-nostic value. An important view of the problem wasprovided by a large multicenter study in France thatanalyzed 1086 AKI cases in the ICU setting.3 The in-vestigators categorized subjects according to thetiming of occurrence of AKI: admission throughsecond ICU day (736 cases), third through sixthICU day (202 cases), or after the seventh day (148cases). Mortality was lower in those with AKI onadmission (61% versus 71% versus 81%), and sowas the need for dialysis (51% versus 58% versus64%), mostly a reflection of the larger number ofpatients with pre–renal azotemia in the first group.3

In addition, the burden of comorbid conditionsshared by critically ill patients with AKI is remark-able. A multicenter study of 618 cases of AKI infive ICUs in the United States outlined the highprevalence of coexiting chronic diseases (30%chronic kidney disease, 37% coronary arterydisease, 29% diabetes, 21% chronic liver disease)and the severity of the acute illness (average 2.9failed organ systems).4 This comorbidity is likelyto account for the lack of improvements in overall

ine, FMP 107, 333 Cedar Street, New Haven, CT, USAue, West Haven, CT 06516, USAticut Healthcare System, 950 Campbell Avenue, West

nc. ches

tmed

.thec

lini

cs.c

om

mailto:[email protected]

http://chestmed.theclinics.com

Crowley & Peixoto30

mortality in AKI over time, despite significantadvances in renal and critical care support.

NEWDEFINITIONS OFACUTE KIDNEY INJURY:RISK, INJURY, FAILURE, LOSS, AND END-STAGERENAL DISEASE ANDACUTE KIDNEYINJURY NETWORK

There is wide variability in the definition of AKI inclinical studies. To foster uniformity in both re-search and clinical practice, an expert group(Acute Dialysis Quality Initiative [ADQI]) developeda new classification of AKI that has been increas-ingly used (risk, injury, failure, loss, and end-stagerenal disease [RIFLE] classification, www.adqi.net). The RIFLE classification (Table 1) differedfrom previous approaches by including both bio-chemical measures of renal function and urine out-put as components of the definition. Although theuse of urine output can be challenged on thegrounds that it may decrease in the absence ofa change in glomerular filtration rate, we view itas an important measure because the persistenceof oliguria following appropriate volume manage-ment is a marker of parenchymal injury. The otherinteresting aspect of this classification is the attri-bution of time parameters to define ‘‘persistentloss of function’’ (4 weeks or longer) and ‘‘end-stage kidney disease’’ (3 months). These lead togreater uniformity in outcomes assessment in clin-ical trials, and guide clinicians to two important ele-ments of prognosis: dialysis-requiring AKI that lastsmore than 4 weeks is severe and likelihood of re-covery is decreased, and its persistence for more

Table1The RIFLE and Acute Kidney Injury Network classification

GFR Criteria

Risk Increased SCreat �1.5 or GFRdecrease >25%

Injury Increased SCreat �2 or GFR decrease >50%

Failure Increased SCreat �3, GFR decrease>75% or SCreat >4 mg/dL (acute rise>0.5 mg/dL)

Loss Persistent AKI: complete loss of kidneyfunction >4 weeks

ESKD End-stage kidney disease: complete lossof kidney function >3 months

AKIN 1 Increased SCreat by 1.5–2� above baselinor by 0.3 mg/dL

AKIN 2 Increased Screat by 2–3� above baseline

AKIN 3 Increased SCreat by >3� above baselineor by R0.3 mg/dL in patients withbaseline SCreat >4 mg/dL

Abbreviations: AKI, acute renal failure; GFR, glomerular filtrat

than 3 months is rarely associated with enough re-covery to remove the patient from dialysis.

The more recent Acute Kidney Injury Network(AKIN) classification is based on the RIFLE system,but has introduced a few relevant modifications(see Table 1).5 First, it uses smaller increments inserum creatinine for the diagnosis of AKI in re-sponse to growing data in the literature (see nextsection). Second, it introduces a time element(48 hours) to diagnosis. This modification has thegoal of focusing on truly acute changes in renalfunction, a factor that is of great relevance to theperformance of clinical trials. Last, it eliminatesthe ‘‘LOSS’’ and ‘‘FAILURE’’ categories. This isan acknowledgment that these categories repre-sent outcomes, and thus should not be listed aspart of diagnosis.

These classifications have undergone evalua-tion of their prognostic value. In a meta-analysisthat included patient-level data from more than71,000 patients in 13 studies, RIFLE criteria dis-played a graded association with adverse out-comes.6 The overall mortality in the study was6.9% in the group without AKI and 31.2% in theAKI group. RIFLE criteria were associated with in-creased risk of death and decreased likelihood ofrenal recovery. Compared with patients withRIFLE R, patients with RIFLE I had 2.2-fold greaterodds of ICU mortality. These odds for RIFLE Fwere 4.9-fold. Renal recovery was also less com-mon among RIFLE I and F patients. Despite wideconfidence intervals, all point estimates were sta-tistically significant, and held up in several ICUand non-ICU settings. Overall, we believe they

s of acute kidney injury

Urine Output Criteria

UO <0.5 mL/kg/h � 6 h

UO <0.5 mL/kg/h � 12 h

UO <0.3 mL/kg/h � 24 h or anuria � 12 h

e UO <0.5 mL/kg/h � 6 h

UO <0.5 mL/kg/h � 12 h

UO <0.3 mL/kg/h � 24 h or anuria for 12 h

ion rate; SCreat, serum creatinine; UO, urine output.

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

http://www.adqi.net

Acute Kidney Injury in the Intensive Care Unit 31

substantiate the prognostic relevance of this clas-sification, although most of these data are basedon the creatinine-based component of RIFLE, andthere is evidence that the urine output–based crite-ria are not associated with mortality.7 As for theAKIN criteria, Barrantes and colleagues recentlypresented data on 471 patients admitted to a med-ical ICU over a 1-year period.8 Patients with a rise inserum creatinine of 0.3 mg/dL or more (or a 50% in-crease frombaseline)within48hours,orwithaurineoutput less than 0.5 mL/kg/h for a least 6 hours(AKIN 1) had greater hospital mortality (45.8% ver-sus 16.4%, adjusted odds ratio 3.7, P < .01) andhospital length of stay (14 versus 7 days, adjustedodds ratio 3.0, P < .01) when compared with pa-tients without AKI. The urine output component ofthe criteria did not hold a significant associationwith mortality. Unfortunately, this small observa-tional study did not address whether a graded rela-tionship exists according to AKIN class.

THRESHOLD FOR DEFINITIONOFACUTE KIDNEY INJURYAND PROGNOSIS

The issue of what thresholds to use in the definitionof AKI remains a work in progress. Recent evi-dence indicates that even minor declines in glo-merular filtration rate are associated withincreased mortality in different populations of hos-pitalized patients. A meta-analysis of eight studiesobserved a graded relationship between theamount of elevation of serum creatinine and mor-tality in AKI.9 Creatinine elevations of 10% to24% above baseline resulted in a relative risk of1.8 (1.3–2.5) for short-term mortality (30 days orless); patients with a rise of 25% to 49% had a rel-ative risk of 3.0 (1.6–5.8), and those with greaterthan 50% increase had a risk of 6.9 (2.0–24.5).9

Results were similar regardless of clinical scenarioor ICU type. These data, along with those of Bar-rantes and colleagues,8 justify the ‘‘tighter’’ criteriaproposed in the AKIN classification.

LONG-TERMOUTCOMES OFACUTEKIDNEY INJURY

The comprehensive impact of AKI on outcomesrequires an assessment of not only short-term out-comes but long-term effects on survival, renal re-covery, and quality of life as well. Recentexamination of the long-term outcome of renalfunction and patient survival following AKI sus-tained during an episode of critical illness hasyielded disquieting results. In addition to experi-encing uniformly high hospital mortality rates, sur-vivors of critical illness complicated by AKI havebeen reported to sustain excess mortality in the

post-hospitalization period as well. Absolute mor-tality differences between hospital discharge and1-year post AKI have ranged up to 18% suggest-ing the possibility of residual mortality risk attribut-able to the preceding AKI episode.10,11 The impactof severe AKI on health-related quality of life(HRQOL) has also been recently examined. Whilenot all, most patients surviving AKI who requireRRT12 report impaired QOL, specifically in the do-mains of physical strength and daily activities formonths to years after hospital discharge com-pared with the general age- and gender-matchedpopulation.13,14

These findings suggest that the true burden ofdisease associated with AKI may be seriouslyunderestimated. The 1-year patient survival, renalfunctional recovery, RRT re-initiation rates, andHRQOL of survivors from the Acute Renal FailureTrial Network (ATN) Study interventional cohortwill be forthcoming and should substantially illumi-nate these underappreciated outcomes. In addi-tion, to further examine the long-term renalconsequences of AKI, the National Institutes ofHealth initiated a U01 grant program in 2007 forancillary studies in the natural history of AKI.

THE DEVELOPMENT OF BIOMARKERS FOR THEEARLY IDENTIFICATION OFACUTE KIDNEY INJURY

An exciting recent development is the use of se-rum and urine biomarkers to identify early renal in-jury (see Coca and colleagues15 for a detailedsystematic review). The main value of thesemarkers would be to allow for earlier intervention.As of now, knowledge of the likelihood of develop-ing AKI still does not translate into interventionsthat may abort its development and alter the clini-cal course (see later in this article). However, thedevelopment of novel therapies directed at spe-cific injury pathways may change this in the rela-tively near future. Previously, late diagnosis mayhave been a factor hampering the success oftested interventions.

Neutrophil gelatinase–associated lipocalin(NGAL) is a protein expressed in multiple tissuesthat is up-regulated in proximal tubular cells imme-diately following ischemic injury.16 Its value in theevaluation of risk of developing AKI was firstshown in 71 children undergoing cardiac surgeryfor congenital heart disease.17 In the 20 childrenwho developed clinical AKI, urinary NGAL levelsrose within 2 hours of bypass, and preceded anychanges in serum creatinine by 34 hours. NGALlevels were unchanged in the 51 children withoutAKI. Since this groundbreaking study, serumand/or urine NGAL has been shown to identifyAKI early in several groups, including adult

Crowley & Peixoto32

patients undergoing coronary bypass surgery, AKIin critical care settings, contrast nephropathy, anddelayed graft function following renaltransplantation.16

Interleukin-18 (IL-18) is a proinflammatory cyto-kine that is activated in proximal renal tubules fol-lowing injury.16 In a nested case-control study of138 patients developing clinical AKI in the ARDSNetwork study, Parikh and colleagues18 demon-strated that a rise in urinary IL-18 levels was 73%predictive of the diagnosis of AKI 24 hours beforethe elevation of serum creatinine.

Cystatin C is a cysteine protease inhibitor that issynthesized and secreted by all nucleated cells; itis freely filtered by the glomerulus and metabolizedby proximal tubule cells.16 It is a better marker ofGFR than serum creatinine, and could identifythe development of AKI 1 to 2 days before eleva-tion in serum creatinine in one study,19 althoughthis finding was not corroborated by anotherstudy.20 It is possible that cystatin C levels will al-low us to make better estimates of GFR in earlyAKI, but its value is still uncertain at this time.

Kidney injury molecule 1 (KIM-1) is a transmem-brane protein that is overexpressed in dedifferen-tiated proximal tubular cells following injury.16

Urine KIM-1 can effectively identify tubular injuryfollowing cardiac surgery in both adults andchildren.21,22

Some of the biomarkers listed above have as-says that are already commercially available forclinical use. Overall, it appears that several bio-markers, used either alone or in combination, willbe useful tools for the early diagnosis of AKI inthe near future.

THE DIAGNOSTIC APPROACH TOACUTE KIDNEYINJURY

The basic diagnostic approach to patients withAKI includes a detailed history and examination,urinalysis, selected urine chemistries, and imagingof the urinary tree. The history should focus on thetempo of loss of function (if known), associatedsystemic diseases, and symptoms related to theurinary tract (especially those that suggest ob-struction). A concise review of systemic diseasesassociated with renal dysfunction has been re-cently published.23 We strongly believe that a re-view of the medications looking for potentiallytoxic drugs is essential. Table 2 presents a list ofsome medications associated with AKI and illus-trates the wide spectrum of possible patterns of in-jury. The physical examination is directed towardthe identification of findings of a systemic diseaseand a detailed assessment of hemodynamic sta-tus. This latter goal often requires invasive

monitoring, especially in the oliguric patient withconflicting clinical findings, where the physicalexamination has limited accuracy.24

Excluding urinary tract obstruction is necessaryin all cases. This can be achieved initially witha bladder scan or placement of a bladder catheterto rule out bladder outlet obstruction. It is also pos-sible that a bladder catheter already in place can beobstructed, and this possibility should not be over-looked. In most cases, more detailed evaluation isindicated through radiologic evaluation of theurinary tree, most commonly with an ultrasound.

Microscopic evaluation of the urine sedimentfrom a freshly voided urine sample is essential tothe evaluation of AKI. Dysmorphic red blood cellsand red blood cell casts suggest glomerulonephri-tis or vasculitis. Sterile pyuria or white blood cellcasts should raise the possibility of interstitial ne-phritis. ‘‘Muddy brown’’ casts (heavily pigmentedcasts resulting from tubular cell debris) and/or tu-bular epithelial cell casts are typically seen in pa-tients with acute tubular necrosis (ATN). Theirpresence is an important tool in the distinction be-tween ATN and prerenal azotemia, which is char-acterized by a normal sediment, or by occasionalhyaline casts. Eosinophiluria, which is often usedto screen for interstitial nephritis, has limited spec-ificity and positive predictive value, since it can beseen in other conditions associated with AKI suchas acute glomerulonephritis and atheroembolic re-nal disease, as well as other common diseases inacutely ill patients, such as pyelonephritis, andprostatitis.25,26

Distinguishing between the two most commoncauses of AKI (prerenal azotemia and ATN) is oftendifficult, especially because the clinical examina-tion is often misleading in the setting of mild vol-ume depletion or overload.24 Urinary chemistriesare routinely used to help in this distinction. Themost discerning test is the fractional excretion ofsodium (FENa 5 U/P Na: U/P creatinine � 100).Values of less than 1% are seen in prerenal azote-mia, whereas values greater than 3% are seen inATN (1%–3% is the diagnostic ‘‘gray zone’’ forthis test). A frequent problem in the interpretationof the FENa is the concomitant use of diuretics,which may result in a high FENa despite a low ef-fective circulating volume. To address this prob-lem, investigators have tested the value of thefractional excretion of urea (FEurea 5 U/P urea ni-trogen: U/P creatinine � 100). A prospective studyindicated that the FEurea identifies patients whohave prerenal azotemia despite the use of di-uretics,27 confirming previous retrospective obser-vations.28 Cutoff values for FEurea are less than35% for prerenal azotemia and greater than 50%for ATN. A recent study was unable to corroborate

Table 2The spectrum of acute kidney injury induced by drugs used in the acute care setting

Type of Renal Injury Drugs

Induction of volume depletion Diuretics

Changes in intra-renal hemodynamicsthat amplify renal sensitivity tohypoperfusion

NSAIDs (including selective COX-2inhibitors), ACE inhibitors,Angiotensin-2 receptor blockers,renin inhibitors, cyclosporine/tacrolimus, vasopressor agents

Glomerular injury

Glomerulonephritis NSAIDs, zoledronate, pamidronate

Microangiopathy Ticlopidine, clopidogrel, cyclosporine,gemcitabine

Tubular injury Iodinated contrast, aminoglycosides,amphotericin B, pentamidine,foscarnet, cisplatinum,acetaminophen, cidofovir, adefovir,tenofovir, melphalan, IVIG,hetastarch, mannitol

Interstitial nephritis NSAIDs, beta-lactams, quinolones,sulfonamides, phenytoin, allopurinol,diuretics (thiazides, loop), indinavir,proton pump inhibitors

Obstruction of the urinary tract

Crystal deposition (intra-renalobstruction)

Indinavir, sulfadiazine,sulfamethoxazole, methotrexate,high-dose acyclovir

Retroperitoneal fibrosis (ureteralobstruction)

Ergotamine, sotalol, propranolol,bromocriptine (all rare)

Abbreviations: ACE, angiotensin-converting enzyme; COX, cyclooxygenase; IVIG, intravenous immunoglobulin; NSAID,nonsteroidal anti-inflammatory drug.


these observations;29 however, this latter studyused a definition of ‘‘transient AKI’’ (return of se-rum creatinine to baseline over a 7-day period)as the equivalent of prerenal azotemia, and only37% of these patients had reached baseline bythe third day following the diagnosis of AKI. There-fore, one could certainly argue that many of thesepatients had some degree of tubular injury thatwent beyond prerenal azotemia. Therefore, whilethese data appear to indicate that FEurea is nota good test to distinguish between ‘‘transient’’and ‘‘permanent’’ AKI, our current interpretation,which is corroborated by our clinical experience,is that the FEurea is valuable to distinguish prere-nal azotemia and ATN, especially in patientsreceiving diuretics.

Biomarkers of tubular injury, some of them dis-cussed in the section above, have been exploredto refine this distinction by allowing the detection

of lesser degrees of tubular injury.30 NGAL, IL-18,and KIM-1 may have value in distinguishing ATNfrom other etiologies of AKI. An important recentpaper analyzed the etiology of renal dysfunctionin 635 patients seen in an emergency depart-ment.31 It demonstrated that NGAL successfullyidentified patients with AKI (area under the curvefor diagnosis 5 0.948). NGAL, alpha-1 microglo-bulin, N-acetyl glycosamine, and FENa success-fully distinguished between ATN and prerenalazotemia, with NGAL having the best discriminat-ing ability. With the exception of the FENa, thesemarkers could also distinguish between ATN andchronic kidney disease. On the other hand, onlythe FENa could distinguish between prerenal azo-temia and chronic kidney disease.31 Therefore, itappears that the use of new biomarkers may alsoadd to our ability to distinguish between differentetiologies of AKI. Ongoing studies are analyzing

Crowley & Peixoto34

this issue in greater detail, also addressing thecombined use of several proteins to improvespecificity.

RISKS ASSOCIATEDWITH GADOLINIUMEXPOSURE IN PATIENTSWITH ACUTEKIDNEY INJURY

An issue of great concern to clinical nephrologistsis the association between gadolinium exposureand nephrogenic systemic fibrosis, a devastatingfibrotic disease, affecting primarily the skin.32

While this disease was initially described amongpatients with end-stage kidney disease on dialysis,approximately 10% of cases occur in patients withchronic kidney disease not on dialysis.33 Of rele-vance, this disease has been reported in patientswith severe AKI.32,33 Until these issues are re-solved, we recommend that patients with AKI donot receive gadolinium. In cases when a magneticresonance study demands contrast, one shouldavoid the potentially more toxic agents (gadodia-mide [Omniscan] and gadopentetate dimeglumine[Magnevist]),34 and the patient or surrogate shouldsign an informed consent form. For patients ondialysis, hemodialysis should be performed imme-diately following exposure to gadolinium althoughit should be noted that dialysis has not beenshown to diminish the risk of development ofnephrogenic systemic fibrosis.32

RENAL REPLACEMENT THERAPY IN ACUTEKIDNEY INJURY

The spectrum of choices of RRT for the manage-ment of critically ill patients with AKI has growncontinuously since the commencement of dialysisfor clinical use. Contemporary renal support is nowprovided via a panoply of intermittent to continu-ous methods using a host of biocompatible mem-branes and an increasing spectrum ofanticoagulation methodologies, as detailed in anexcellent recent systematic review.35 What willbe discussed here are the results of several recentrandomized controlled trials (RCTs) specificallyexamining the effect of ‘‘dose’’ and modality ofRRT on outcomes in AKI. The genesis of these tri-als arose from observations gleaned from severaldescriptive studies conducted over the past 2decades.

Dialysis Modality

Ten years ago, the National Kidney Foundation36

surveyed American nephrologists about their pref-erences and practice in the management of AKI. Itwas identified that intermittent hemodialysis (IHD)was the preferred modality for renal support,

used in more than 75% of cases by most nephrol-ogists, while continuous renal replacement ther-apy (CRRT) and peritoneal dialysis (PD) wereused in a minority of treatments (<10%) accordingmost practitioners. More recently, a survey of in-tensivists and nephrologists practicing at partici-pating sites of the multicenter ATN Study,suggests that IHD remains the most frequentlyused modality of renal support, estimated to beused in 57% of treatments. However, CRRT isgaining in popularity in the United States and isnow reported to account for over a third (36%) ofprescribed RRT treatments.37 The Program toImprove Care in Acute Renal Disease (PICARD)observational study mirrored the new survey re-sults. Among the five participating US tertiarycare centers in the PICARD study, IHD was themodality of RRT used in the greatest percentageof patients, but CRRT use had become such com-monplace practice that 60% of dialyzed patientshad received CRRT for some or all of their renalsupport.4 Notably, there was substantial intersitevariability in modality preference.

In contrast to the reported US experience, arecent survey of an international multidisciplinarycohort of renal practitioners showed that continu-ous therapies had become the norm for AKI sup-port outside of the United States.38 Confirmingthe international survey results, the multinationalprospective epidemiologic Beginning and EndingSupportive Therapy for the Kidney (BEST Kidney)Study of ARF in the ICU reported that CRRT wasthe initial modality of choice for RRT support inthe ICU used in 80% of treatments, distantly fol-lowed by IHD (17%).1

Preference for CRRT over IHD in the manage-ment of more severely ill patients with AKI hasbeen attributed to a presumed survival advantageconveyed by the continuous modality.39 Severalrational theories justifying this presumption havebeen proffered. For example, in light of CRRT’sdemonstrated ability to enhance clearance of cer-tain humoral inflammatory cytokines, its use, par-ticularly in high volume, has the potential torestore immunomodulatory balance and improvesurvival.40,41 Others have opined that CRRT’s pre-sumed survival advantage is related to its greaterhemodynamic stability over IHD. However, meta-analyses of earlier trials comparing survival of un-selected critically ill patients with AKI assigned toeither IHD or CRRT and adjusted for severity ofillness have not been able to substantiate thepresumed superiority of CRRT.42,43 Recent pro-spective evidence has also challenged theadvantages of CRRT, particularly with regard tosuperiority of hemodynamic stability.44–46 In addi-tion, observational and several prospective RCTs


explicitly comparing intermittent and continuousmodalities have failed to confirm the expectedsurvival advantage of CRRT.44,45,47,48

Reasonable criticisms countering the RCTs thatcompared modalities have been made. First, thedose of dialysis prescribed in both the IHD andthe CRRT arms of earlier studies were belowwhat has subsequently been recommended asa minimally acceptable level, thus prohibitinga fair comparison between adequate IHD and ad-equate CRRT.47,49 Second, a relatively high cross-over rate between modalities became evident,potentially affecting each study’s analytic de-sign.47 Third, randomization failure occurred inone study resulting in differences in baseline cova-riates that were independently associated withmortality and biased in favor of IHD.47 The resultsof the more recent RCTs comparing modalities inunselected critically ill patient populations havebeen similarly challenged for high crossover rates,as well as limited trial power and nonstandardiza-tion of dialysis protocols.44,48

In an attempt to address the limitations of theprevious RCTs comparing intermittent and contin-uous modalities of RRT, a large multicenter study(‘‘Hemodiasafe’’) was conducted that used stan-dardized dialysis protocols and targeted doses ofRRT to meet current recommendations. Hemody-namic stability was optimized for patients treatedwith IHD through the use of extended time, cool di-alysate, sodium modeling, bicarbonate dialysate,and isovolemic initiation of IHD.45 Again, no differ-ence in 60-day survival (33% versus 32% of CCRTversus IHD) or of renal recovery was found. A lowrate (6%) of modality crossover was maintained bythe use of policies guiding modality switches andno difference in the incidence of adverse events,including hypotension, was identified.

The premise of CRRT modality superiority wasfurther refuted recently by the PICARD investiga-tors. A subgroup analysis of patients with AKI re-quiring RRT derived from this multicenterobservational study compared survival by initialRRT modality assignment.50 Despite adjustmentfor potential confounding variables and modalityselection via a propensity score approach, CRRTwas in fact associated with increased 60-day mor-tality as compared with IHD.

How can these negative results be reconciledwith the widely accepted premise of the superiorityof CRRT? First, earlier studies comparing modali-ties compared results obtained with more contem-porary CRRT to those obtained from historicalcontrols treated with IHD.47,51 Such comparisonsmay have erroneously attributed the survival ad-vantage conferred by improvements in nonmodal-ity aspects of care, such as the development of

biocompatible membranes and better criticalcare services, to modality selection instead. Sec-ond, developments in IHD technology to improveits effectiveness and tolerability have perhapsbeen underappreciated. An unexplained improve-ment in the survival rate of the IHD cohort wasnoted during the course of the Hemodiasafe Studythat was not attributable to changes in patientcharacteristics or center effects, suggesting thatimprovements in the standard of care for IHD sub-jects had occurred.45 Closer scrutiny revealed thatan unrecognized drift toward greater delivery of di-alysis had occurred that may have modified therelative benefit of one modality versus another. Infact, an increasing quantity of estimated dialysisdose was also found in the CRRT arm of the He-modiasafe study, but in the absence of direct mea-surement of IHD dose, further comparison ofrelative dose change is not possible. Third, thetheoretic advantage of CRRT conferred by its abil-ity to remove cytokines has been questioned.While CRRT has been shown to be capable of re-moving cytokines, the high cytokine generationrate, particularly during sepsis, may require an im-practical increase in effluent volume or in mem-brane flux to realize a physiologically relevantimpact on the inflammatory cascade and to effec-tively restore immunohomeostasis.52,53

The lack of a clear advantage of conventionalCRRT over IHD, in conjunction with its increasedcosts (nearly twofold that of IHD), have led to anexpanding interest in hybrid modalities that exploitthe best attributes of both intermittent and contin-uous therapies.47 Among the recently evolvingcontinuous therapies, continuous high-flux dialy-sis (CHFD) and/or high volume hemofiltration(HVHF) have the potential to better facilitate re-moval of middle molecular weight solutes thatmay be particularly important in the managementof sepsis-associated AKI.53–55 The High Volumein Intensive Care (IVOIRE) study (NCT 00241228)is an ongoing HVHF trial examining 28-day mortal-ity in patients treated with standard (35 mL/kg/h)versus high-volume (70 mL/kg/h) continuous he-mofiltration and is scheduled for completion inthe near future.

Preliminary studies of hybridized intermittenttherapies have also been recently reported. A pilotstudy comparing the effect of super high-flux/highmolecular weight cutoff (HCO) IHD to standardhigh-flux (HF) IHD in 10 septic patients found thatHCO-IHD achieved better diffusive clearance ofcytokines and decreased plasma cytokine levelsas compared with HF-IHD.56 Accelerated venove-nous hemofiltration (AVVH) is another evolving in-termittent therapy, which performs hemofiltrationdiscontinuously, ie, daily but within a compressed

Crowley & Peixoto36

time frame. A retrospective review of the experi-ence of one center’s use of AVVH demonstratedthat large volumes of hemofiltration (36 L) couldbe delivered in a compressed/intermittent timeframe (9 hours), and effectively mitigate the draw-back of continuous modalities, which is the needfor anticoagulation.57

Sustained low-efficiency dialysis (SLED) orextended daily dialysis (EDD) is probably themost familiar hybrid therapy known to intensivistsand nephrologists, having been described asa modified form of IHD nearly 20 years ago.58

Most often delivered by dual-capacity IHD ma-chines, it is typically prescribed from 6 to 12 hoursper treatment with modified blood and dialysateflow rates (70–250 and 70–300 mL/min, respec-tively).59 Excellent small molecular weight soluteclearance and arguably equivalent hemodynamicstability as compared with CRRT have been re-ported.60,61 Conceivably, the future of extracorpo-real renal support for AKI may be a hybrid of hybridtechniques! Daily SLED using HCO membranes ina diffusive fashion or delivery of HVHF over a com-pressed period could potentially achieve excellentsmall and middle molecular weight solute clear-ance, without anticoagulation and without sacrific-ing hemodynamic tolerability. Because there islimited information on the combined clinical useof these hybrid therapies, the precise role of thesevarious modalities in the management of AKI re-mains to be defined.

In summary, no consensus concerning thechoice of RRT modality for the treatment of AKI ex-ists. Although CRRT is gaining in popularity in theUnited States and worldwide, practice patternsvary substantially by region. No evidence currentlysupports the superiority of one continuous modal-ity over another, or of continuous over intermittenttreatment. Choice of RRT modality, therefore,should most properly remain determined by thepreference and experience of the prescribing phy-sician, and the technologic, fiscal, and nursing re-sources available to deliver RRT. An ongoingmulticenter international observational ICU study,the Dose Response Multicenter International (Do-Re-Mi) trial, is designed to further elaborate oncurrent RRT practice patterns and how RRT ischosen and administered.62 Hybrid therapieshold promise for combining the best of all extra-corporeal modalities.

Dialysis Dose

Historically, trials examining the impact of renalsupport in critical illness have focused on modula-tion of ‘‘dose’’ of treatment. In the early years ofRRT, comparisons were made between historical

cohorts of patients with AKI who were initiatedon RRT and maintained at advanced degrees ofazotemia (BUN >200 mg/dL) and those treatedand maintained at more modest levels of azotemia(BUN 100–150 mg/dL).63–65 Because mortalitywas drastically reduced by earlier initiation andgreater reduction of azotemia, the timing of initia-tion and dose of RRT required was adjusted tomaintain the patient at progressively lower levelsof BUN.66–68

More recently, instead of prescribing RRT toachieve an absolute BUN level, quantification ofdialysis dose has used the concept of urea clear-ance, exported from the field of chronic hemodial-ysis. For IHD it is described by a unit-lessparameter: KT/Vurea. Arguably, the physiology ofthe patient with AKI nullifies several of the mathe-matical assumptions underlying the calculation ofKT/Vurea, however in the absence of a more repre-sentative measure of dialysis dose, KT/Vurea hasbeen used by default. When considering dose forintermittent modalities, frequency of treatmentmust also be considered. In CRRT, the measure-ment of dose of renal support is different, but sim-pler than in IHD. In CRRT, because of the lowblood and dialysate flow rates used, the ultrafil-trate generated is isotonic to blood with respectto urea nitrogen. Therefore, effluent volume isused as a surrogate for urea clearance. Obviously,direct comparison of urea removal measured byeffluent volume in CRRT to KT/Vurea measured inIHD is limited.

In the past decade, five single-center clinical tri-als involving critically ill patients with AKI requiringrenal support have examined the effect of dose,quantified by KT/Vurea or effluent volume, on mor-tality.69–73

The first and only IHD trial compared daily ver-sus alternate day IHD and reported a mortalitybenefit conferred by daily treatment.69 This trialhas been criticized on several accounts, includingnonrandom treatment assignment, the floatingend point of ‘‘14 days after the last session of dial-ysis,’’ the limited representativeness of the en-rolled population, and suboptimal per treatmentdialysis dose, leading to the more limited conclu-sion that suboptimal delivery of dialysis results insuboptimal outcomes. Importantly, this study didhighlight the exceptional difficulty in achieving sol-ute clearance in the critically ill noting an averageKT/Vurea of less than 1 in both treatment arms de-spite a minimum prescribed dose equal to 1.2.

Four other studies have examined the effect ofdiffering weight-based doses of CRRT onoutcomes. The first study suggested that CRRTeffluent volume of 35 mL/kg/h was superior to20 mL/kg/h with respect to survival 15 days after


discontinuation of therapy and that no incrementalbenefit was achieved from prescribing a highereffluent volume (ie, > 35 mL/kg/h).70 Similarly,another investigation reported improved 28-daysurvival by augmenting CVVH of 25 mL/kg/h byan additional 18 mL/kg/h of dialysate suggestingthat additional small solute clearance was impor-tant in modifying outcomes.72

Other trials, however, have come to theopposite conclusion. A two-center randomizedcontrolled trial comparing outcome of critically illoliguric patients treated with ‘‘high-’’ (45 mL/kg/h)versus ‘‘low-’’ (20 mL/kg/h) volume CVVH found nodifference in 28-day mortality between groups.71

Although criticized for being potentially underpow-ered, this study’s results were confirmed recentlyby other investigators who found no difference in30-day survival between patients treated withCVVHDF at 35 versus 20 mL/kg/h.73

The only multicenter trial designed to examinethe effect of dose or ‘‘intensity’’ of renal supporton mortality in the critically ill with AKI was recentlypublished.74 The Acute Renal Failure Trial Network(ATN) Study was a very large trial that uniquelyused integrated strategies of renal support mirror-ing customary practice rather than restricting RRTto a single modality, which would have limited thegeneralizability of its results.4,75 Also distinct instudy design was the per-treatment IHD delivereddose, set at a target KT/Vurea of 1.2, which washigher than in other comparable studies. Hemody-namically stable patients (Cardiovascular SOFA %2) were treated with IHD, whereas hemodynami-cally unstable patients (Cardiovascular SOFA > 2)were assigned to CVVHDF or SLED based on localpractice. Subjects were randomized to either ofthe two interventional arms: the intensive manage-ment strategy (IMS) consisted of six times weeklyIHD or SLED or CVVHDF at 35 mL/kg/h; the lessintensive management strategy (LIMS) consistedof thrice-weekly IHD or SLED or CVVHDF at 20mL/kg/h. No significant difference in hospital- or60-day survival was found, nor of any othersecondary end point including recovery of renalfunction or rate of nonrenal organ failure.

The results of the ATN Study should not be mis-construed to mean that ‘‘dose’’ does not matter.Instead they suggest that dose of RRT beyonda delivered KT/Vurea of 1.2 thrice weekly for pa-tients managed with IHD or SLED, and beyonda total effluent flow rate of 20 mL/kg/h for patientsmanaged with CVVHDF is not helpful in reducingoverall mortality nor is it associated with improvedrenal recovery in survivors.74

Ongoing clinical trials that may contribute addi-tional understanding of the impact of RRT dose onpatient outcomes include the Randomized

Evaluation of Normal versus Augmented Level ofrenal replacement therapy in ICU (RENAL) Study(NCT 00221013), a large multicenter RCT beingconducted in Australia of CVVHDF in severe AKIin ICU comparing 90-day mortality of patientstreated with an ‘‘augmented’’ effluent flow rate of40 mL/kg/h versus patients treated with a ‘‘normal’’flow rate of 25 mL/kg/h. Approximately one third ofthe prespecified total trial population (N 5 1500)has been enrolled and trial completion is antici-pated in late 2008.

Another multicenter trial, the IVOIRE study (NCT00241228), is under way in France. It is a prospec-tive RCT comparing outcomes of patients treatedwith high-volume (70 mL/kg/h) to ‘‘standard’’ vol-ume (35 mL/kg/h) CVVH in patients with septicshock and AKI. Initiated in 2005, enrollment of a to-tal of 460 patients is expected by the end of 2008.

Surprisingly, despite the body of trials examiningdose of RRT support, all survey and observationalstudies of RRT support published to date conveyquite clearly that the concept of ‘‘dose’’ in RRT inthe ICU has been largely ignored or at least not fullyadopted by practicing clinicians.36–38,62 Reluc-tance to adopt any targeted ‘‘dosing’’ of RRT maystem from the belief that toxicity of AKI is only par-tially related to small solute clearance, and thusprescription focused only on its modificationshould not be expected to yield substantial im-provement in outcomes. Urea clearance, regard-less of how great it is or by which modality ofrenal support it is achieved, tells us nothing aboutmiddle molecular weight or cytokine clearance orother aspects of the uremic state that may also beimportant in influencing outcomes. An excellent re-view of uremic retention solutes and their toxicityhas been recently published.76 Non-urea small mo-lecular weight solute, as well as middle and largemolecular weight solutes may not be differentiallyimpacted or may even be negatively impacted byprescriptions of RRT that further enhance ureaclearance. Also, AKI is a state of extreme inflamma-tion, and RRT of any modality may be associatedwith exacerbation rather than amelioration of thestate of inflammation.77 It is reasonably possible,therefore, that any benefit derived from enhancedurea clearance by RRT, beyond a minimum thresh-old, is offset by the counteracting increase in the in-flammatory state induced by increasing exposureto RRT.

It is becoming evident that achievement offurther improvement in outcomes of critically ill pa-tients with AKI will require alternate strategies ofcare beyond RRT modality selection or increasingthe dose of RRT beyond a minimum threshold.Recent trials examining unifactorial treatment ofcomplex disease illustrate the limitations of this

Crowley & Peixoto38

approach.78,79 AKI is a pathophysiologically com-plex disease that likely requires a multifaceted ap-proach if treatment is to be successful. Morecomprehensive strategies of RRT and facilitatedrecovery from AKI are fortunately on the horizon.80

SELECT PHARMACOLOGIC INTERVENTIONSAND FUTURE DIRECTIONS

Prior pharmacotherapeutic trials of AKI have beendisappointing. Reasons for the failure of the multi-tude of agents that have been trialed are detailedin a concise review.81 More recent appreciationof AKI as a multisystem disorder triggered bywidespread stereotypical molecular responses toinjury has caused a reconsideration of the strate-gic approach toward AKI and its treatment. Inter-ventional strategies that target pathways involvedin the early systemic response to AKI, or broadtreatment strategies to address its protean sys-temic manifestations are being actively explored.

Of the emerging pharmacologic agents for thetreatment of AKI, two that are currently in use asco-interventions in critical illness are intensiveinsulin therapy (IIT) and erythropoietin (EPO).

Intensive Insulin Therapy and GlycemicManagement in Acute Kidney Injury

Hyperglycemia has emerged as an important pre-dictor of outcomes of critical illness, including theoutcome of renal failure.82–87 The effect of thetreatment of hyperglycemia on outcomes, includ-ing the development of AKI, has thus been of greatinterest to investigators and has been a focus ofrecent studies.

One such study was a large observational trial ofa mixed ICU patient population in which a historicalICU cohort was compared with a contemporaryone of consecutively admitted patients who weretreated with insulin to achieve moderate glycemiccontrol (serum glucose < 140 mg/dL). This studydemonstrated that glycemic control achievedwith insulin treatment resulted in a startling reduc-tion in hospital mortality (from 29.3% to 10.8%)and in the incidence of acute renal failure (by75%) defined as an increase of serum creatinineby 1 mg/dL or a doubling of serum creatinine.84

The landmark study that put glycemic control onthe map was a multicenter RCT using an insulin-based protocol in the treatment of patients admit-ted to a surgical ICU in Leuven, Belgium.85 IIT toreduce serum glucose to less than 110 mg/dL,was associated with a significant reduction inICU mortality (from 8% to 4.6%) and in the inci-dence of acute renal failure requiring RRT (8.2%to 4.8%, P 5 .007).

A subsequent prospective RCT broadening theapplication of IIT to the MICU population was un-able to demonstrate a similar reduction in mortalityor need for RRT, but did demonstrate a significantreduction in newly acquired kidney injury, from8.9% to 5.9% (P 5 .04) defined as either a doublingof ICU admission serum creatinine or a peak se-rum creatinine greater than 2.5 mg/dL.86

A more recent secondary analysis of pooleddata from the two RCTs comparing conventionaltherapy to IIT scrutinized the effect of IIT on the de-velopment of renal outcomes.88 Overall, nearly20% of the combined patient cohorts developedAKI, defined as a combined end point of patientsclassified with renal dysfunction according tomodified (m) RIFLE criteria of m-RIFLE-F, I or F,or oliguria or need for RRT. Interestingly, while IITsignificantly reduced the incidence of RIFLE-I/Ffrom 7.6% to 4.5% (P 5 .0006), there was no effectof IIT on the incidence of the most severe renaloutcomes (ie, need for RRT or oliguria), except inthe surgical subset of patients where it was asso-ciated with a reduced need for RRT (4.0% versus7.4% , P 5 .003) and a lower incidence of oliguria(2.6% versus 5.6%, P 5 .003) . The lower acuity ofillness and premorbid disease in the surgical co-hort as compared with the medical counterparthas led to the conclusion that IIT’s beneficial effectis likely one of primary prevention of AKI ratherthan secondary reversal of established renal injury.

How IIT and glycemic control may affect renaloutcomes in critical illness is complex and incom-pletely understood. Indirectly, dysglycemia mayadversely impact renal function via accentuationof complications such as sepsis-associatedAKI.88 Directly, hyperglycemia may contribute torenal injury via accentuation of the acute inflamma-tory response and oxidative stress of critical ill-ness.87,89,90 IIT on the other hand, may preventrenal injury through favorable effects on serumlipids, which can act as scavengers of endo-toxins.91 IIT may also reduce renal injury via mod-ulation of aberrant endothelial activation (ICAMand E-selectin) and deranged endothelial nitric ox-ide synthesis.88

IIT currently remains a controversial co-inter-vention in the management of patients with criticalillness. Earlier recommendations for glucose man-agement in the ICU92 were heavily influenced bythe Leuven trial85,92; however, aspects of the land-mark study’s design as well as its generalizabilityto all critically ill have recently been challenged.93

In addition, two European RCTs (VISEP NCT00135473 and GLUCONTROL, NCT 00220987)examining intensive glycemic control in the criti-cally ill with and without septic shock were prema-turely terminated because of safety concerns.94


Currently a large multicenter randomized trial(NICE-SUGAR; NCT 00220987) is under way to re-affirm or refute the Leuven study’s findings in a het-erogeneous critically ill population, and renaloutcome as defined by the need for RRT is a pre-determined secondary end point. Until the resultsof theses trials are available, prudence in the useof IIT and liberalization of glycemic control hasbeen advocated.

Erythropoietin

Erythropoietin (EPO) is a hematopoietic cytokinethat has been used in the management of the ane-mia of chronic kidney disease for 20 years. Recentstudies have shown however that EPO’s effectsextend well beyond erythropoiesis. Erythropoietinreceptors have been identified in a wide varietyof tissues including neurons, endothelial cells, car-diac myocytes, vascular smooth muscle cells, me-sangial cells, and renal proximal tubular cells.95 Inaddition, EPO has been shown to result in a host ofhematopoietic-independent cytoprotective effectsin experimental models of neuronal, cardiac,hepatic, and acute renal ischemia: modulation ofmitogenesis, vascular repair, oxidative stress,inflammation, and apoptosis have all been de-scribed.96,97 Numerous in vitro and in vivo AKImodels reported to date have examined EPO useas a preconditioning treatment to prevent AKI,but there is also a rationale for EPO’s use as an ac-celerant in the recovery from ischemic injurythrough its vascular and anti-inflammatory ef-fects.97 As an initial exploration into the potentialrole of EPO as a renoprotective strategy in the crit-ically ill, an open-label, randomized controlled trialhas been initiated (NCT 00676234). This single-center study of medical and surgical ICU patients‘‘at risk for AKI’’ will examine the effect of EPO, ad-ministered as either a single dose of 20,000 unitsor 40,000 units, on the short-term incidence ofAKI as measured by a battery of standard andnovel biomarkers of renal injury.

Tempering enthusiasm for the use of EPO arerecent reports of an increased risk of adverse out-comes including increased cardiovascular mor-bidity and mortality in clinical trials investigatingoptimal EPO treatment of CKD patients.98,99 Anewly engineered desialated form of EPO miti-gates the adverse effects associated with erythro-poiesis. Known as asialo-EPO, this derivative hasnormal EPO receptor affinity but an extremelyshort half-life, which reduces its erythropoieticproperties while retaining its cytoprotectiveeffects.95 Parenteral injection of asialo-EPO hasbeen shown to reduce contrast-associated ische-mic renal injury in an in vivo rat model without

adverse effect on hematocrit.95 Mechanistic inves-tigations have recently shown that the asialo-EPOattenuates contrast-associated renal tubular cellapoptosis via modulation of the signal-transduc-tive JAK2/STAT5 pathway with beneficial down-stream effects on stress-responsive protein (eg,HSP70) expression and capsase-3 activation.95 Acomprehensive characterization of asialo-EPO’srenoprotective effects as well as its clinical safetyand optimal dosage in the prevention and facili-tated recovery of AKI are exciting forefronts ofinvestigation.

Bioartificial Kidney

Currently available RRT is essentially designed tofacilitate solute and water clearance. This ap-proach hardly mimics the expansive metabolicand endocrinological functions of the human kid-ney and may explain the limited impact that RRThas had on survival in patients with AKI. A visionarystrategy toward the treatment of AKI that ad-dresses these previously overlooked functions ofthe kidney is the bioartificial kidney. Also knownas the renal tubular assist device (RAD), the bioar-tificial kidney consists of a hemofiltration cartridgelined with nonautologous human renal tubule cellsalong the inner surface of hollow fibers within thedevice. Preclinical studies of the device showedthat the cells within were able to perform renaltransport and endocrine and metabolic func-tions.100,101 Subsequent large animal studies us-ing the RAD in series with a conventionalhemofilter in the treatment of sepsis-associatedAKI demonstrated a substantial reduction in multi-organ dysfunction.102 Initial results of phase I/II tri-als using the RAD in conjunction with CRRT incritically ill patients with AKI have demonstrateda near halving of mortality.103,104 Notwithstandingthe valid criticisms of the preliminary studies per-taining to design and analysis, the initial clinical re-sults are nothing short of remarkable.105 Althoughsubstantial hurdles remain before the RAD mightbe a routine therapeutic intervention in the treat-ment of AKI, the revolutionary design of the deviceand its early trial results offer prophetic evidencethat genuine biological mimicry is at hand andthat we might soon be able to positively impactthe health of critically ill patients with AKI ina meaningful way.

SummaryIn conclusion, AKI is a common illness in the hos-pitalized population and is associated with signifi-cant mortality. New diagnostic criteria using smallchanges in creatinine and urine output have beendeveloped to facilitate diagnosis and aide in

Crowley & Peixoto40

prognosis as well. A host of biomarkers are in de-velopment that will allow for effective interdictionof AKI and earlier intervention with pharmacother-apies such as IIT and modified erythropoietin toprevent and facilitate renal recovery. Renal re-placement therapy for AKI is evolving toward a hy-bridized modality approach to exploit the benefitsof different modalities. Recently completed stud-ies suggest that excessive solute removal doesnot yield incremental benefit, thus we must reeval-uate how to optimize the replacement of renalfunction beyond simply escalating dose. A morecomprehensive approximation of the normal hu-man kidney may be available shortly in the formof the bioartificial kidney.

REFERENCES

1. Uchino S, Kellum JA, Bellomo R, et al. Acute renal

failure in critically ill patients: a multinational, multi-

center study. JAMA 2005;294(7):813–8.

2. Liano F, Junco E, Pascual J, et al. The spectrum of

acute renal failure in the intensive care unit com-

pared with that seen in other settings. The Madrid

Acute Renal Failure Study Group. Kidney Int Suppl

1998;66:S16–24.

3. Guerin C, Girard R, Selli JM, et al. Initial versus

delayed acute renal failure in the intensive care

unit. A multicenter prospective epidemiological

study. Rhone-Alpes Area Study Group on Acute

Renal Failure. Am J Respir Crit Care Med 2000;

161(3 Pt 1):872–9.

4. Mehta RL, Pascual MT, Soroko S, et al. Spectrum of

acute renal failure in the intensive care unit: the

PICARD experience. Kidney Int 2004;66(4):

1613–21.

5. Mehta RL, Kellum JA, Shah SV, et al. Acute Kidney

Injury Network: report of an initiative to improve out-

comes in acute kidney injury. Crit Care 2007;11(2):

R31.

6. Ricci Z, Cruz D, Ronco C. The RIFLE criteria and

mortality in acute kidney injury: a systematic review.

Kidney Int 2008;73(5):538–46.

7. Cruz DN, Bolgan I, Perazella MA, et al. North East

Italian Prospective Hospital Renal Outcome Survey

on Acute Kidney Injury (NEiPHROS-AKI): targeting

the problem with the RIFLE Criteria. Clin J Am Soc

Nephrol 2007;2(3):418–25.

8. Barrantes F, Tian J, Vazquez R, et al. Acute kidney

injury criteria predict outcomes of critically ill

patients. Crit Care Med 2008;36(5):1397–403.

9. Coca SG, Peixoto AJ, Garg AX, et al. The prognos-

tic importance of a small acute decrement in kidney

function in hospitalized patients: a systematic re-

view and meta-analysis. Am J Kidney Dis 2007;

50(5):712–20.

10. Bagshaw SM, Laupland KB, Doig CJ, et al. Progno-

sis for long-term survival and renal recovery in crit-

ically ill patients with severe acute renal failure:

a population-based study. Crit Care 2005;9(6):

R700–9.

11. Oeyen S, Vandijck D, Benoit D, et al. Long-term

outcome after acute kidney injury in critically-ill

patients. Acta Clin Belg Suppl 2007;(2):337–40.

12. Morgera S, Kraft AK, Siebert G, et al. Long-term

outcomes in acute renal failure patients treated

with continuous renal replacement therapies. Am

J Kidney Dis 2002;40(2):275–9.

13. Ahlstrom A, Tallgren M, Peltonen S, et al. Survival

and quality of life of patients requiring acute renal

replacement therapy. Intensive Care Med 2005;

31(9):1222–8.

14. Gopal I, Bhonagiri S, Ronco C, et al. Out of hospital

outcome and quality of life in survivors of combined

acute multiple organ and renal failure treated with

continuous venovenous hemofiltration/hemodiafil-

tration. Intensive Care Med 1997;23(7):766–72.

15. Coca SG, Yalavarthy R, Concato J, et al.

Biomarkers for the diagnosis and risk stratification

of acute kidney injury: a systematic review. Kidney

Int 2008;73(9):1008–16.

16. Parikh CR, Devarajan P. New biomarkers of acute

kidney injury. Crit Care Med 2008;36(4 Suppl):

S159–65.

17. Mishra J, Dent C, Tarabishi R, et al. Neutrophil

gelatinase-associated lipocalin (NGAL) as a bio-

marker for acute renal injury after cardiac surgery.

Lancet 2005;365(9466):1231–8.

18. Parikh CR, Abraham E, Ancukiewicz M, et al. Urine

IL-18 is an early diagnostic marker for acute kidney

injury and predicts mortality in the intensive care

unit. J Am Soc Nephrol 2005;16(10):3046–52.

19. Herget-Rosenthal S, Marggraf G, Husing J, et al.

Early detection of acute renal failure by serum cys-

tatin C. Kidney Int 2004;66(3):1115–22.

20. Ahlstrom A, Tallgren M, Peltonen S, et al. Evolution

and predictive power of serum cystatin C in acute

renal failure. Clin Nephrol 2004;62(5):344–50.

21. Han WK, Waikar SS, Johnson A, et al. Urinary

biomarkers in the early diagnosis of acute kidney

injury. Kidney Int 2008;73(7):863–9.

22. Liangos O, Perianayagam MC, Vaidya VS, et al.

Urinary N-acetyl-beta-(D)-glucosaminidase activity

and kidney injury molecule-1 level are associated

with adverse outcomes in acute renal failure.

J Am Soc Nephrol 2007;18(3):904–12.

23. Rajashekar A, Perazella MA, Crowley S. Systemic

diseases with renal manifestations. Prim Care

2008;35(2):297–328, vi–vii.

24. Peixoto AJ. Birth, death, and resurrection of the

physical examination: clinical and academic per-

spectives on bedside diagnosis. Yale J Biol Med

2001;74(4):221–8.


25. Ruffing KA, Hoppes P, Blend D, et al. Eosinophils in

urine revisited. Clin Nephrol 1994;41(3):163–6.

26. Nolan CR 3rd, Anger MS, Kelleher SP. Eosinophilu-

ria—a new method of detection and definition of

the clinical spectrum. N Engl J Med 1986;

315(24):1516–9.

27. Carvounis CP, Nisar S, Guro-Razuman S. Signifi-

cance of the fractional excretion of urea in the dif-

ferential diagnosis of acute renal failure. Kidney

Int 2002;62(6):2223–9.

28. Kaplan AA, Kohn OF. Fractional excretion of urea

as a guide to renal dysfunction. Am J Nephrol

1992;12(1–2):49–54.

29. Pepin MN, Bouchard J, Legault L, et al. Diagnostic

performance of fractional excretion of urea and

fractional excretion of sodium in the evaluations of

patients with acute kidney injury with or without

diuretic treatment. Am J Kidney Dis 2007;50(4):

566–73.

30. Han WK, Bonventre JV. Biologic markers for the

early detection of acute kidney injury. Curr Opin

Crit Care 2004;10(6):476–82.

31. Nickolas TL, O’Rourke MJ, Yang J, et al. Sensitivity

and specificity of a single emergency department

measurement of urinary neutrophil gelatinase-asso-

ciated lipocalin for diagnosing acute kidney injury.

Ann Intern Med 2008;148(11):810–9.

32. Knopp EA, Cowper SE. Nephrogenic systemic

fibrosis: early recognition and treatment. Semin

Dial 2008;21(2):123–8.

33. Abu-Alfa A. The impact of NSF on the care of pa-

tients with kidney disease. J Am Coll Radiol 2008;

5(1):45–52.

34. Reilly RF. Risk for nephrogenic systemic fibrosis

with gadoteridol (ProHance) in patients who are

on long-term hemodialysis. Clin J Am Soc Nephrol

2008;3(3):747–51.

35. Pannu N, Klarenbach S, Wiebe N, et al. Alberta

Kidney Disease N: Renal replacement therapy in

patients with acute renal failure: a systematic

review. JAMA 2008;299(7):793–805.

36. Mehta RL, Letteri JM. Current status of renal re-

placement therapy for acute renal failure. A survey

of US nephrologists. The National Kidney Founda-

tion Council on Dialysis. Am J Nephrol 1999;

19(3):377–82.

37. Overberger P, Pesacreta M, Palevsky PM, et al. Man-

agement of renal replacement therapy in acute kid-

ney injury: a survey of practitioner prescribing

practices. Clin J Am Soc Nephrol 2007;2(4):623–30.

38. Ricci Z, Ronco C, D’Amico G, et al. Practice pat-

terns in the management of acute renal failure in

the critically ill patient: an international survey.

Nephrol Dial Transplant 2006;21(3):690–6.

39. Chang JW, Yang WS, Seo JW, et al. Continuous ve-

novenous hemodiafiltration versus hemodialysis as

renal replacement therapy in patients with acute

renal failure in the intensive care unit. Scand J

Urol Nephrol 2004;38(5):417–21.

40. Kellum JA, Johnson JP, Kramer D, et al. Diffusive

vs. convective therapy: effects on mediators of

inflammation in patients with severe systemic in-

flammatory response syndrome. Crit Care Med

1998;26(12):1995–2000.

41. Piccinni P, Dan M, Barbacini S, et al. Early isovolae-

mic haemofiltration in oliguric patients with septic

shock. Intensive Care Med 2006;32(1):80–6.

42. Tonelli M, Manns B, Feller-Kopman D. Acute renal

failure in the intensive care unit: a systematic

review of the impact of dialytic modality on mortality

and renal recovery. Am J Kidney Dis 2002;40(5):

875–85.

43. Kellum JA, Angus DC, Johnson JP, et al. Continu-

ous versus intermittent renal replacement therapy:

a meta-analysis. Intensive Care Med 2002;28(1):

29–37.

44. Uehlinger DE, Jakob SM, Ferrari P, et al. Compari-

son of continuous and intermittent renal replace-

ment therapy for acute renal failure. Nephrol Dial

Transplant 2005;20(8):1630–7.

45. Vinsonneau C, Camus C, Combes A, et al. Contin-

uous venovenous haemodiafiltration versus inter-

mittent haemodialysis for acute renal failure in

patients with multiple-organ dysfunction syndrome:

a multicentre randomised trial. Lancet 2006;

368(9533):379–85.

46. Misset B, Timsit JF, Chevret S, et al. A randomized

cross-over comparison of the hemodynamic re-

sponse to intermittent hemodialysis and continuous

hemofiltration in ICU patients with acute renal

failure. Intensive Care Med 1996;22(8):742–6.

47. Mehta RL, McDonald B, Gabbai FB, et al. A ran-

domized clinical trial of continuous versus intermit-

tent dialysis for acute renal failure. Kidney Int 2001;

60(3):1154–63.

48. Augustine JJ, Sandy D, Seifert TH, et al. A random-

ized controlled trial comparing intermittent with

continuous dialysis in patients with ARF. Am J Kid-

ney Dis 2004;44(6):1000–7.

49. Kellum JA, Mehta RL, Angus DC, et al. The first

international consensus conference on continuous

renal replacement therapy. Kidney Int 2002;62(5):

1855–63.

50. Cho KC, Himmelfarb J, Paganini E, et al. Survival

by dialysis modality in critically ill patients with

acute kidney injury. J Am Soc Nephrol 2006;

17(11):3132–8.

51. Bellomo R, Farmer M, Wright C, et al. Treatment of

sepsis-associated severe acute renal failure with

continuous hemodiafiltration: clinical experience

and comparison with conventional dialysis. Blood

Purif 1995;13(5):246–54.

52. De Vriese AS, Colardyn FA, Philippe JJ, et al. Cyto-

kine removal during continuous hemofiltration in

Crowley & Peixoto42

septic patients. J Am Soc Nephrol 1999;10(4):

846–53.

53. Ronco C, Kellum JA, Bellomo R, et al. Potential in-

terventions in sepsis-related acute kidney injury.

Clin J Am Soc Nephrol 2008;3(2):531–44.

54. Lonnemann G, Bechstein M, Linnenweber S, et al.

Tumor necrosis factor-alpha during continuous

high-flux hemodialysis in sepsis with acute renal

failure. Kidney Int Suppl 1999;56(72):S84–7.

55. Brunet S, Leblanc M, Geadah D, et al. Diffusive and

convective solute clearances during continuous re-

nal replacement therapy at various dialysate and

ultrafiltration flow rates. Am J Kidney Dis 1999;

34(3):486–92.

56. Haase M, Bellomo R, Baldwin I, et al. Hemodialysis

membrane with a high-molecular-weight cutoff and

cytokine levels in sepsis complicated by acute re-

nal failure: a phase 1 randomized trial. Am J Kidney

Dis 2007;50(2):296–304.

57. Gashti CN, Salcedo S, Robinson V, et al. Acceler-

ated venovenous hemofiltration: early technical

and clinical experience. Am J Kidney Dis 2008;

51(5):804–10.

58. Tam PY, Huraib S, Mahan B, et al. Slow continuous

hemodialysis for the management of complicated

acute renal failure in an intensive care unit. Clin

Nephrol 1988;30(2):79–85.

59. Fliser D, Kielstein JT. Technology insight: treatment

of renal failure in the intensive care unit with ex-

tended dialysis. Nat Clin Pract Nephrol 2006;2(1):

32–9.

60. Marshall MR, Golper TA, Shaver MJ, et al. Sus-

tained low-efficiency dialysis for critically ill patients

requiring renal replacement therapy. [erratum ap-

pears in Kidney Int 2001;60(4):1629]. Kidney Int

2001;60(2):777–85.

61. Kumar VA, Craig M, Depner TA, et al. Extended

daily dialysis: a new approach to renal replacement

for acute renal failure in the intensive care unit. Am

J Kidney Dis 2000;36(2):294–300.

62. Kindgen-Milles D, Journois D, Fumagalli R, et al.

Study protocol: the DOse REsponse Multicentre In-

ternational collaborative initiative (DO-RE-MI). Crit

Care 2005;9(4):R396–406.

63. Teschan PE, Baxter CR, O’Brien TF, et al. Pro-

phylactic hemodialysis in the treatment of acute

renal failure. Annals of Internal Medicine,

53:992–1016, 1960. J Am Soc Nephrol 1998;

9(12):2384–97.

64. Teschan PE, Baxter CR, O’Brien TF, et al. Prophy-

lactic hemodialysis in the treatment of acute renal

failure. Ann Intern Med 1960;53:992–1016.

65. Fischer RP, Griffen WO Jr, Reiser M, et al. Early di-

alysis in the treatment of acute renal failure. Surg

Gynecol Obstet 1966;123(5):1019–23.

66. Kleinknecht D, Jungers P, Chanard J, et al. Uremic

and non-uremic complications in acute renal

failure: evaluation of early and frequent dialysis

on prognosis. Kidney Int 1972;1(3):190–6.

67. Conger JD. A controlled evaluation of prophylactic

dialysis in post-traumatic acute renal failure.

J Trauma 1975;15(12):1056–63.

68. Gillum DM, Dixon BS, Yanover MJ, et al. The role of

intensive dialysis in acute renal failure. Clin Nephrol

1986;25(5):249–55.

69. Schiffl H, Lang SM, Fischer R. Daily hemodialysis

and the outcome of acute renal failure. N Engl

J Med 2002;346(5):305–10.

70. Ronco C, Bellomo R, Homel P, et al. Effects of dif-

ferent doses in continuous veno-venous haemofil-

tration on outcomes of acute renal failure:

a prospective randomised trial. Lancet 2000;

356(9223):26–30.

71. Bouman CSC, Oudemans-Van Straaten HM,

Tijssen JGP, et al. Effects of early high-volume con-

tinuous venovenous hemofiltration on survival and

recovery of renal function in intensive care patients

with acute renal failure: a prospective, randomized

trial. Crit Care Med 2002;30(10):2205–11.

72. Saudan P, Niederberger M, De Seigneux S, et al.

Adding a dialysis dose to continuous hemofiltra-

tion increases survival in patients with acute renal

failure [see comment]. Kidney Int 2006;70(7):

1312–7.

73. Tolwani AJ, Campbell RC, Stofan BS, et al. Stan-

dard versus high-dose CVVHDF for ICU-related

acute renal failure. J Am Soc Nephrol 2008;19(6):

1233–8.

74. Network VNARFT, Palevsky PM, Zhang JH, et al. In-

tensity of renal support in critically ill patients with

acute kidney injury. N Engl J Med 2008;359(1):

7–20.

75. Palevsky PM, O’Connor T, Zhang JH, et al. Design

of the VA/NIH Acute Renal Failure Trial Network

(ATN) Study: intensive versus conventional renal

support in acute renal failure. Clin Trials 2005;

2(5):423–35.

76. Vanholder R, Baurmeister U, Brunet P, et al. A

bench to bedside view of uremic toxins. J Am

Soc Nephrol 2008;19(5):863–70.

77. Himmelfarb J, McMonagle E, Freedman S, et al.

Oxidative stress is increased in critically ill patients

with acute renal failure. J Am Soc Nephrol 2004;

15(9):2449–56.

78. Chiche J-D, Angus DC. Testing protocols in the

intensive care unit: complex trials of complex inter-

ventions for complex patients [comment]. JAMA

2008;299(6):693–5.

79. Krumholz HM, Lee TH. Redefining quality—impli-

cations of recent clinical trials. N Engl J Med

2008;358(24):2537–9.

80. Humes HD. Cell therapy: leveraging nature’s thera-

peutic potential [comment]. J Am Soc Nephrol

2003;14(8):2211–3.


81. Jo SK, Rosner MH, Okusa MD. Pharmacologic

treatment of acute kidney injury: why drugs haven’t

worked and what is on the horizon. Clin J Am Soc

Nephrol 2007;2(2):356–65.

82. Gandhi GY, Nuttall GA, Abel MD, et al. Intraopera-

tive hyperglycemia and perioperative outcomes in

cardiac surgery patients. Mayo Clin Proc 2005;

80(7):862–6.

83. Krinsley JS. Association between hyperglycemia

and increased hospital mortality in a heterogeneous

population of critically ill patients. Mayo Clin Proc

2003;78(12):1471–8.

84. Krinsley JS. Effect of an intensive glucose manage-

ment protocol on the mortality of critically ill adult

patients. [see comment] [erratum appears in

Mayo Clin Proc. 2005 Aug;80(8):1101]. Mayo Clin

Proc 2004;79(8):992–1000.

85. van den Berghe G, Wouters P, Weekers F, et al. In-

tensive insulin therapy in the critically ill patients.

N Engl J Med 2001;345(19):1359–67.

86. Van den Berghe G, Wilmer A, Hermans G, et al. In-

tensive insulin therapy in the medical ICU. N Engl

J Med 2006;354(5):449–61.

87. Basi S, Pupim LB, Simmons EM, et al. Insulin resis-

tance in critically ill patients with acute renal failure.

Am J Physiol Renal Physiol 2005;289(2):F259–64.

88. Schetz M, Vanhorebeek I, Wouters PJ, et al. Tight

blood glucose control is renoprotective in critically

ill patients. J Am Soc Nephrol 2008;19(3):571–8.

89. Collier B, Dossett LA, May AK, et al. Glucose con-

trol and the inflammatory response. Nutr Clin Pract

2008;23(1):3–15.

90. Sarafidis PA, Ruilope LM. Insulin resistance, hyper-

insulinemia, and renal injury: mechanisms and im-

plications. Am J Nephrol 2006;26(3):232–44.

91. Mesotten D, Swinnen JV, Vanderhoydonc F, et al.

Contribution of circulating lipids to the improved

outcome of critical illness by glycemic control

with intensive insulin therapy. J Clin Endocrinol Me-

tabol 2004;89(1):219–26.

92. Garber AJ, Moghissi ES, Bransome ED Jr, et al.

American College of Endocrinology position state-

ment on inpatient diabetes and metabolic control.

Endocr Pract 2004;10(1):77–82.

93. Bellomo R, Egi M. Glycemic control in the intensive

care unit: why we should wait for NICE-SUGAR.

[comment]. Mayo Clin Proc 2005;80(12):1546–8.

94. Marik PE, Varon J. Intensive insulin therapy in the

ICU: is it now time to jump off the bandwagon?

Resuscitation 2007;74(1):191–3.

95. Yokomaku Y, Sugimoto T, Kume S, et al. Asialoery-

thropoietin prevents contrast-induced nephropa-

thy. J Am Soc Nephrol 2008;19(2):321–8.

96. Hochhauser E, Pappo O, Ribakovsky E, et al.

Recombinant human erythropoietin attenuates he-

patic injury induced by ischemia/reperfusion in an

isolated mouse liver model. Apoptosis 2008;13(1):

77–86.

97. Sturiale A, Campo S, Crasci E, et al. Experimental

models of acute renal failure and erythropoietin:

what evidence of a direct effect? Ren Fail 2007;

29(3):379–86.

98. Besarab A, Bolton WK, Browne JK, et al. The ef-

fects of normal as compared with low hematocrit

values in patients with cardiac disease who are re-

ceiving hemodialysis and epoetin. N Engl J Med

1998;339(9):584–90.

99. Drueke TB, Locatelli F, Clyne N, et al. Normalization

of hemoglobin level in patients with chronic kidney

disease and anemia. N Engl J Med 2006;355(20):

2071–84.

100. Humes HD, Buffington DA, MacKay SM, et al.

Replacement of renal function in uremic animals

with a tissue-engineered kidney. Nat Biotechnol

1999;17(5):451–5.

101. Humes HD, MacKay SM, Funke AJ, et al. Tissue

engineering of a bioartificial renal tubule assist

device: in vitro transport and metabolic character-

istics. Kidney Int 1999;55(6):2502–14.

102. Humes HD, Buffington DA, Lou L, et al. Cell ther-

apy with a tissue-engineered kidney reduces the

multiple-organ consequences of septic shock.

Crit Care Med 2003;31(10):2421–8.

103. Humes HD, Weitzel WF, Bartlett RH, et al. Initial

clinical results of the bioartificial kidney containing

human cells in ICU patients with acute renal failure.

Kidney Int 2004;66(4):1578–88.

104. Tumlin J, Wali R, Williams W, et al. Efficacy and

safety of renal tubule cell therapy for acute re-

nal failure. J Am Soc Nephrol 2008;19(5):

1034–40.

105. Chertow GM, Waikar SS. Toward the promise of re-

nal replacement therapy. J Am Soc Nephrol 2008;

19(5):839–40.

Documents

MEDICAL PROGRESS Uremia - anaesthetics.ukzn.ac.za · uremia but were removed from this category as their causes were discovered. Today the term “uremia” is used loosely to describe