II. ASSESSMENT OF DYSLIPIDEMIAS I - SOCHIPEnefrologia.sochipe.cl/subidos/catalogo3/DISLIPIDEMIAS II.pdf · II. ASSESSMENT OF DYSLIPIDEMIAS IN EACH OF the following guideline state-

II. ASSESSMENT OF DYSLIPIDEMIAS

IN EACH OF the following guideline state-ments, kidney transplant recipients are in-

cluded along with other patients with CKD,whether or not they have other evidence of CKD.Each guideline statement is followed by a lettergrade (in parentheses) indicating the strength ofthe recommendation (Table 7).

GUIDELINE 1

1.1. All adults and adolescents with CKD shouldbe evaluated for dyslipidemias. (B)

1.2. For adults and adolescents with CKD, theassessment of dyslipidemias should includea complete fasting lipid profile with totalcholesterol, LDL, HDL, and triglycerides. (B)

1.3. For adults and adolescents with Stage 5 CKD,dyslipidemias should be evaluated upon pre-sentation (when the patient is stable), at 2-3months after a change in treatment or otherconditions known to cause dyslipidemias;and at least annually thereafter. (B)

Associations Between Dyslipidemias and ACVDin CKD

The incidence of ACVD is very high in pa-tients with CKD (Fig 5). Therefore, the NKFTask Force on CVD and the K/DOQI WorkGroup on CKD both concluded that, in the man-agement of risk factors such as dyslipidemia,patients with CKD should be considered to be inthe highest risk category, ie, equivalent to that ofpatients with known CHD.2,4There is very strongevidence from the general population that dyslipi-demias cause ACVD, and this evidence has ledto the ATP III guidelines for evaluation andtreatment.3 It is conceivable that the pathogene-sis of ACVD is different in patients with CKD,and that dyslipidemias do not contribute toACVDin CKD. However, the relationship between dys-lipidemias and ACVD in the general populationis robust, ie, it is valid in men and women3,46,47;old and middle-aged3,46,47; smokers and non-smokers3,47; hypertensive and non-hypertensivepatients47; diabetics and nondiabetics3,48; and in-dividuals with higher or lower LDL,3,47higher or

lower total cholesterol,3,47higher or lower triglyc-erides,3,47and higher or lower HDL (Fig 4).3,47,49,50

There are no compelling reasons to assume thatdyslipidemias do not contribute to ACVD inpatients with CKD as well.

There are no randomized, controlled, interven-tion trials testing the hypothesis that dyslipi-demias cause ACVD in patients with CKD. How-ever, in an observational study of 3,716 patientsinitiating treatment for Stage 5 CKD in 1996, theuse of statins in 362 (9.7%) was independentlyassociated with lower all-cause mortality and areduction in CVD deaths during follow-up.84

Unfortunately, it is likely that the patients usingstatins had other favorable characteristics thatwere not accounted for in the adjusted analysis,but may have explained their reduced risk forCVD independent of their use of statins. There-fore, these study results are consistent with, butdo not prove, the hypothesis that dyslipidemiascontribute to ACVD in patients with CKD.

Associations Between Dyslipidemias and ACVDin Hemodialysis Patients

There are no large, prospective, observationalstudies examining the relationship betweenACVD and dyslipidemias in hemodialysis pa-tients. A number of retrospective, cross-sectionalstudies found no relationship, or—in somecases—even paradoxical correlations betweendyslipidemias and ACVD in hemodialysis pa-tients (Table 10). However, there are a number ofreasons that studies have failed to find a positiveassociation between dyslipidemias and ACVD.None of these studies was a long-term, prospec-tive, cohort study, and it is likely that illness,inflammation, and poor nutrition confounded therelationships between dyslipidemias and ACVD.Support for this notion comes from studies thathave found that at least some of the seeminglyparadoxical associations between dyslipidemiasand ACVD are, in part, explained by statisticaladjustment for markers of malnutrition and sys-temic inflammation.85-87

More than a decade ago, it was reported thatthe association between cholesterol and mortal-ity (much of which was presumably due to CVD)in hemodialysis patients took the form of aU-shaped curve.85,86 Moreover, the associationbetween low cholesterol and increased mortality

© 2003 by the National Kidney Foundation, Inc.0272-6386/03/4104-0304$30.00/0

American Journal of Kidney Diseases, Vol 41, No 4, Suppl 3 (April), 2003: pp S22-S38S22

was reduced after adjusting for levels of serumalbumin.85,86 Similarly, in a recent prospectivestudy of 1,167 hemodialysis patients, low serumcholesterol levels were associated with all-causemortality in patients with low serum albumin.87

However, in patients with normal serum albu-

min, the opposite was true; high serum choles-terol predicted mortality.87 C-reactive protein, amarker of inflammation, has been associatedwith lower serum cholesterol levels.87 Othermarkers of inflammation, eg, interleukin-6 andtumor necrosis factor-�, are also associated with

Fig 5. Causes of death among period prevalent patients 1997–1999, treated with hemodialysis, peritonealdialysis, or kidney transplantation. Data are from the USRDS 2001 Annual Data Report (www.usrds.org). Abbrevia-tions: MI, myocardial infarction; HD, heart disease.

ASSESSMENT OF DYSLIPIDEMIAS S23

low serum cholesterol levels in hemodialysispatients.88

Altogether, these data demonstrate a seem-ingly paradoxical association between low se-rum cholesterol and increased mortality in hemo-dialysis patients. They should not be interpretedto mean that dyslipidemias do not contribute tothe pathogenesis of ACVD. Rather, it is morelikely that the opposite is true, ie, that highcholesterol contributes to ACVD in hemodialy-sis patients as it does in the general population,and that other conditions accompanying low cho-lesterol (such as inflammation) account for theincreased mortality of patients with low choles-terol.

Non-Traditional Lipid Abnormalities and ACVDin Hemodialysis Patients

Several observational studies have reported apositive association between lipoprotein(a)[Lp(a)] and ACVD in hemodialysis pa-tients.18,90,91,93,96,98 In 4 of these studies, the asso-ciation was statistically significant.18,91,93,98 Inter-estingly, apolipoprotein(a) low molecular weightphenotypes, which correlate with higher levelsof Lp(a), were recently shown to be associatedwith ACVD in 440 hemodialysis patients.90 Incontrast, a small study (n � 75) of peritonealdialysis patients99 and a small study (n � 79) oftransplant patients both failed to find an associa-tion between Lp(a) and ACVD. Although otherstudies have noted elevated Lp(a) in CKD pa-tients as well,100,101 there are no studies in pa-tients with CKD, or in the general population,examining the effect of reducing Lp(a) on CVD.Thus, it is difficult to recommend routine mea-surement of Lp(a) in clinical practice.

It is also possible that other, non-traditional,atherogenic lipoprotein abnormalities may causeor contribute to ACVD in hemodialysis patients.For example, some,102-105 but not all,106 casecontrol studies examining oxidized lipoproteinsin hemodialysis patients have reported higherlevels compared to matched controls. In addi-tion, a randomized, controlled trial found thatsupplementation with vitamin E reduced recur-rent ACVD events in 196 hemodialysis patients(P � 0.014).107 However, this trial was small,and much larger trials in the general populationhave failed to show benefit from vitamin Esupplementation.108-115

Evidence from observational studies in thegeneral population has suggested that lipoproteinremnants may contribute to ACVD, especially inpatients with high triglycerides.3 There are alsocross-sectional studies reporting that hemodialy-sis patients have higher levels of remnant lipopro-teins, eg, triglyceride-enriched lipoproteins and/orsmall dense LDL, compared to controls.104,116-123

However, none of these studies reported correla-tions between levels of remnant lipoproteins withACVD, and the significance of these abnormali-ties is unknown.

Associations Between Dyslipidemias and ACVDin Peritoneal Dialysis Patients

Only 2 studies that examined the relationshipbetween dyslipidemias and ACVD in peritonealdialysis patients were identified (Table 11). Bothof these studies had major design limitations, andboth were too small to rigorously examine therelationship between dyslipidemias and ACVDin the peritoneal dialysis patient population.

Associations Between Dyslipidemias and ACVDin Kidney Transplant Recipients

Several studies have reported a positive as-sociation between total cholesterol and ACVDin kidney transplant recipients (Table 12). Un-fortunately, few of these studies examined therelationship between LDL and ACVD. Lowerlevels of HDL were associated with ACVD in3 of 4 studies. In 3 of 6 studies, higher levels oftriglycerides were associated with ACVD. Al-together these studies suggest that the relation-ship between ACVD and dyslipidemias in kid-ney transplant recipients is similar to thatobserved in the general population. However,each of these studies had design limitations; inparticular, none was truly prospective. Kidneytransplant recipients may also have non-traditional lipoprotein abnormalities that couldtheoretically contribute to ACVD.125-127 How-ever, the role of these lipoprotein abnormali-ties in the pathogenesis of ACVD in CKD, asin the general population, is unclear.

The Evaluation of Dyslipidemias in CKD

Measurements of total cholesterol, HDL, andtriglycerides are readily available in most major

ASSESSMENT OF DYSLIPIDEMIASS24

clinical laboratories. The LDL that forms thefoundation for treatment decisions in the ATP IIIGuidelines3 is generally calculated from totalcholesterol, HDL, and triglycerides using theFriedewald formula. The ATP III Guidelines alsorecommend treatment of some dyslipidemias thatmay occur with normal or low LDL. Thesedyslipidemias—often seen in association withthe metabolic, or insulin resistance syndrome

(the syndrome of obesity, hypertension, insulinresistance, and hyperlipidemia) and character-ized by increases in circulating lipoprotein rem-nants—can be most readily measured as non-HDL cholesterol, ie, total cholesterol minus HDL(Fig 6).3 All of the major treatment decisions fordyslipidemia in these guidelines, as in the ATPIII Guidelines, are based on levels of triglycer-ides, LDL, and non-HDL cholesterol.


Associations Between Dyslipidemias andKidney Disease Progression

The principal reason to evaluate dyslipidemiasin patients with CKD is to detect abnormalitiesthat may be treated to reduce the incidence ofACVD. However, there may be other reasons toevaluate and treat dyslipidemias in CKD. A num-ber of observational studies have reported thatvarious dyslipidemias are associated with de-creased kidney function in the general popula-tion and in patients with CKD (Table 13). It isimpossible to determine from these studieswhether dyslipidemias cause reduced kidneyfunction, result from reduced kidney function, orwhether other conditions such as proteinuria causeboth reduced kidney function and dyslipidemias.Each of these explanations is plausible, and onlyrandomized, controlled trials can adequately testthe hypothesis that dyslipidemias cause a declinein kidney function.

Unfortunately, there are no large, adequatelypowered, randomized, controlled trials testingthe hypothesis that treatment of dyslipidemia

preserves kidney function. However, there havebeen several small studies,137-148 and a meta-analysis of these studies.149 This meta-analysisincluded prospective, controlled trials publishedbefore July 1, 1999. Three trials published onlyin abstract form were included in this meta-analysis137,138,148; one of these studies has subse-quently been published in a peer-reviewed jour-nal.148 All patients were followed for at least 3months, but in only 5 studies were patientsfollowed for at least 1 year. Statins were used in10 studies, gemfibrozil in 1 study, and probucolin 1 study. Altogether, 362 patients with CKDwere included in the meta-analysis. The resultssuggested that the rate of decline in GFR wassignificantly less in patients treated with a choles-terol-lowering agent compared to placebo.149 Nosignificant heterogeneity in treatment effect wasdetected between the studies. However, the qual-ity of the studies was generally low, and theirsmall sample sizes and relatively short durationof follow-up make it difficult to conclude thatlipid-lowering therapies reduce the rate of de-cline in GFR in CKD. Therefore, the primary orsecondary prevention of ACVD remains the prin-cipal reason to evaluate and treat dyslipidemiasin patients with CKD.

The Prevalence of Dyslipidemias inHemodialysis Patients

The prevalence of dyslipidemias in patientswith CKD is high (Tables 14, 15, 16, and 17).Dyslipidemias in hemodialysis patients are mostoften characterized by normal LDL, low HDL,and high triglycerides. From the published litera-ture, it is difficult to discern the prevalence ofdyslipidemia in hemodialysis patients, since moststudies are relatively small and use varying defi-nitions for dyslipidemia. Therefore, the WorkGroup examined the prevalence of dyslipidemiain a large cross-section of 1,047 hemodialysispatients in the Dialysis Morbidity and MortalityStudy (Table 16). The definitions of the ATP IIIGuidelines, as well as those adopted in theseguidelines, were used. According to ATP IIIdefinitions, only 20.2% of hemodialysis patientshad normal lipid levels, ie, LDL �130 mg/dL(�3.36 mmol/L), HDL �40 mg/dL (�1.03mmol/L), and triglycerides �150 mg/dL (�1.69mmol/L). Using the definitions of the presentguidelines, 61.1% would require treatment of a

Fig 6. Example demonstrating the relative contribu-tions of VLDL and IDL remnants to non-HDL choles-terol in two hypothetical patients with normal and hightriglycerides, respectively. Although both patients Aand B have the same total and HDL cholesterol levels,for patient A with normal triglycerides, most of thenon-HDL cholesterol is LDL. However, for Patient Bwith high triglycerides, much of the non-HDL choles-terol is VLDL and IDL remnants. Units are in mg/dL. Toconvert mg/dL to mmol/L, multiply triglycerides by0.01129 and total, LDL, HDL and non-HDL cholesterolby 0.02586. Abbreviations: VLDL, very low-density li-poproteins; IDL, intermediate density lipoproteins;LDL, low-density lipoproteins; HDL, high-density li-poproteins.




dyslipidemia; 55.7% would require treatmentbased on LDL �100 mg/dL (�2.59 mmol/L),while another 5.4% with normal LDL wouldrequire treatment based on triglycerides �200mg/dL (�2.26 mmol/L) and non-HDL choles-terol �130 mg/dL (�3.36 mmol/L) (Table 16).

The Prevalence of Dyslipidemias in PeritonealDialysis Patients

The prevalence of dyslipidemia in patientstreated with peritoneal dialysis is high, and dif-fers somewhat from that in hemodialysis patients(Tables 14, 15, 16, and 17). In a cross-section of317 peritoneal dialysis patients from the DialysisMorbidity and Mortality Study (Table 17), only15.1% had normal lipid levels according to theATP III Guidelines, ie, LDL �130 mg/dL (�3.36mmol/L), HDL �40 mg/dL (�1.03 mmol/L),and triglycerides �150 mg/dL (�1.69 mmol/L).Using the definitions of the present guidelines,78.6% would require treatment of a dyslipide-mia; 73.2% would require treatment based onLDL �100 mg/dL (�2.59 mmol/L), while an-other 5.4% with normal LDL would requiretreatment based on triglycerides �200 mg/dL(�2.26 mmol/L) and non-HDL cholesterol �130mg/dL (�3.36 mmol/L) (Table 17).

The Prevalence of Dyslipidemias in KidneyTransplant Recipients

The prevalence of dyslipidemias in kidneytransplant recipients is very high (Table 14 andTable 15). Particularly common are increases intotal cholesterol and LDL. Triglycerides are of-ten increased, but HDL is usually normal.

The Frequency of Dyslipidemia Evaluation inCKD

Many factors influence the prevalence of dys-lipidemias in CKD. Changes in proteinuria, GFR,and treatment of CKD may alter lipoproteinlevels. Therefore, it is prudent to evaluate dyslipi-demias more often than is recommended in thegeneral population. Lipoprotein levels maychange during the first 3 months of hemodialysis,peritoneal dialysis, and kidney transplantation.On the other hand, waiting 3 months to measurethe first lipid profile may needlessly delay effec-tive treatment for patients who present withdyslipidemia. For patients whose lipid profile isnormal at presentation, it is reasonable to repeat




the lipid profile 3 months later, to confirm thatthe initial values were not low due to malnutri-tion or systemic disease. During the course ofkidney disease treatment, lipid levels may change.Therefore, the Work Group recommends measur-ing subsequent levels at least annually. Reasons

to repeat lipid measurements after 2-3 monthsinclude changes in kidney replacement therapymodality, treatment with diet or lipid-loweringagents, immunosuppressive agents that affect lipids(eg, prednisone, cyclosporine, or sirolimus) or otherchanges that may affect plasma lipids.


Dyslipidemias in Adolescents

Young adults (20-40 years old) with Stage 5CKD have at least a 10-fold higher risk for CVDmortality compared to the general population.184

There are limited data on ACVD in children withCKD. However, CVD accounts for approxi-mately 23% of deaths in children and adults �30years old who started treatment for stage 5 CKDas children.185 Recent data from the Pathobiologi-cal Determinants of Atherosclerosis in Youth(PDAY) study provide compelling evidence inthe general pediatric population, that initial fattystreaks seen in adolescents develop into athero-matous plaques in young adults.186 Moreover,this atherosclerotic process is believed to beaccelerated in uremia, thus putting children withStage 5 CKD at high risk for developing ACVD.Indeed, studies of arteries from children withStage 5 CKD have demonstrated early ACVDchanges.187,188

It is important to note that lipid levels in thegeneral population change with age and puberty,and differ by gender (Tables 18, 19, 20, and21).189 Very low levels at birth increase rapidly inthe first year of life to a mean total cholesterol of150 mg/dL (3.88 mmol/L), LDL 100 mg/dL(2.59 mmol/L), and HDL 55 mg/dL (1.42 mmol/

L). From ages 1-12, lipid levels remain fairlyconstant, and are slightly lower in girls thanboys. During puberty, there is a decrease in totalcholesterol, LDL, and a slight decrease in HDLin boys. After puberty, ie, by age 17, cholesteroland LDL increase to adult levels in boys andgirls. Boys continue to have a slightly lowerHDL than girls. These changes dictate that thedefinitions of dyslipidemias be different in chil-dren and adults. These guidelines define dyslipi-demias for children using lipid levels greaterthan the 95th percentile for age and gender (Tables18, 19, 20, and 21). Treatment thresholds forchildren do not differ by age and gender, butthese thresholds are different from those of adults.

There are few data documenting the preva-lence of dyslipidemias in children and adoles-cents with CKD. A search was conducted forstudies published after 1980 that included at least15 patients and reported data on the prevalenceof dyslipidemia in unselected patients with CKD.There were no studies of hemodialysis patients.Children and adolescents on peritoneal dialysisappeared to have a very high prevalence ofdyslipidemias (Table 15). Indeed, 29% to 87% ofpediatric peritoneal dialysis patients had LDL�100 mg/dL (�2.59 mmol/L). Similarly, 72% to


84% of pediatric kidney transplant recipients hadLDL �100 mg/dL (�2.59 mmol/L) (Table 15).In a longitudinal study of pediatric transplantpatients, the prevalence of hypercholesterolemiadeclined from 70.4% to 35% at 10 years, with adecrease in hypertriglyceridemia from 46.3% to15%.182 This decline in prevalence may reflectreductions in immunosuppressive medicationsand improved kidney function. Unfortunately, no

longitudinal studies have defined the long-termrisk of dyslipidemias in children with CKD, particu-larly as they survive into young adulthood.

Use of the Friedewald Formula to CalculateLDL

The Friedewald formula appears to be themost practical, reliable method for determiningLDL cholesterol in clinical practice:


LDL � Cholesterol � HDL

� (triglycerides � 5), in mg/dL,

or

LDL � Cholesterol � HDL

� (triglycerides � 2.19), in mmol/L.191

Two recent studies found the Friedewald for-mula to be reliable in dialysis patients,192,193

although other investigators reported that thepercentage error for the formula is higher inpatients with CKD compared to the general popu-lation.194 No studies have examined the accuracyof the Friedewald formula in transplant recipi-ents, or studies in other CKD patients, eg, thosewith nephrotic syndrome.

Recent data from a study in the general popu-lation suggest that the Friedewald formula mayunderestimate LDL in patients with low LDLlevels.31 Data from the general population alsosuggest that the Friedewald formula is not accu-rate when triglycerides are �400 mg/dL (�4.52mmol/L). Direct measurement of LDL with ultra-centrifugation or immunoprecipitation tech-niques is reasonably accurate when triglyceridesare 400-800 mg/dL (4.52-9.03 mmol/L), but thereare no reliable techniques for determining LDL

when triglycerides are �800 mg/dL (�9.03mmol/L). Fasting triglycerides �800 mg/dL(�9.03 mmol/L) generally indicate the presenceof hyperchylomicronemia, and the role of hyper-chylomicronemia in ACVD is unknown.

There are few studies in children, and noneincluded children with CKD. However, in 1study of children from the general population,calculating LDL using the Friedewald formulawas more reliable in correctly classifying pa-tients with high LDL than was the direct measure-ment of LDL.195

Dyslipidemias in Acute Medical Conditions

Some acute medical conditions may tran-siently alter plasma lipid levels (Table 22). Forexample, severe infections, surgery and acutemyocardial infarction are often associated withlower-than-normal lipid levels. Other conditions,for example acute pancreatitis, may be associ-ated with higher levels. In general, it is best towait until acute conditions that may alter lipidlevels have resolved before assessing dyslipi-demias for possible ACVD risk. It should benoted, however, that the lipid profile is not signifi-cantly altered within the first 24 hours after amyocardial infarction, and a lipid profile can bemeasured during this time.196-198


The Influence of Immunosuppressive Agents

Immunosuppressive medications, eg, pred-nisone, cyclosporine, and sirolimus are amongthe several potential remediable causes of dyslipi-demias in patients with CKD and after kidneytransplantation (Table 23). It is not clear howsoon these agents exert their effects on lipopro-tein metabolism, and when lipid levels reach anew steady state. However, the effects of diet andlipid-lowering agents may not be fully manifestfor 2-3 months, and—by analogy—it may bebest to measure a lipid profile 2-3 months afterstarting or stopping an immunosuppressive agentthat is known to have a major effect on lipopro-tein levels, eg, prednisone, cyclosporine, or siroli-mus.

The present guidelines are consistent withthose of the American Society of Transplantation(AST), which recommend that a lipid profileshould be measured during the first 6 monthspost-transplant, at 1 year after transplantation,and annually thereafter.209 The AST guidelinesalso suggest that changes in immunosuppressivetherapy, graft function, or CVD risk warrantadditional testing.209

GUIDELINE 2

2.1. For adults and adolescents with Stage 5CKD, a complete lipid profile should bemeasured after an overnight fast when-ever possible. (B)

2.2. Hemodialysis patients should have lipidprofiles measured either before dialysis,or on days not receiving dialysis. (B)

Fasting

Eating raises plasma triglycerides, carriedmostly in chylomicrons and very low-densitylipoprotein (VLDL), and, as a result, totalcholesterol levels also increase. The post-prandial increases in triglycerides and choles-terol are quite variable, depending on the typeof food ingested. In addition, substantial vari-ability in post-prandial lipid levels is attribut-able to inherited and acquired differences be-tween individuals. Although these differencesaffect the risk for ACVD, the relationshipbetween post-prandial lipid levels and ACVDis not as well established as the relationshipbetween fasting lipid levels and ACVD.3 Prac-tical considerations may make non-fasting mea-surements the only alternative for some pa-tients. While fasting lipid profiles are best, it isbetter to obtain non-fasting lipid profiles thanto forgo evaluation altogether. If the lipidprofile obtained in a non-fasting patient isnormal, then no further assessment is neededat that time. However, an abnormal lipid pro-file in a non-fasting patient is an indication toobtain a fasting lipid profile.

Effects of Hemodialysis and Peritoneal Dialysison Plasma Lipids

There is some evidence that the hemodialysisprocedure acutely alters plasma lipid levels.217

This may be due to hemoconcentration and/oreffects of the dialysis membrane or heparin onlipoprotein metabolism. Therefore, assessmentshould be done prior to the hemodialysis proce-dure. There are few data describing how quickly



lipoprotein levels change during the course ofperitoneal dialysis exchanges. However, it isprobably most practical to draw blood in themorning, after an overnight fast (whenever pos-sible), and with whatever peritoneal dialysis fluidis dwelling in the peritoneal cavity when theblood is drawn.

GUIDELINE 3

Stage 5 CKD patients with dyslipidemiasshould be evaluated for remediable, second-ary causes. (B)

Rationale

Causes of secondary dyslipidemias includenephrotic syndrome,218-223 hypothyroidism,224-226

diabetes,227-229 excessive alcohol ingestion,230-234

and chronic liver disease (Table 24).235-237 Medi-cations that can cause dyslipidemias include 13-cis-retinoic acid,238-240 anticonvulsants,241-243

highly active anti-retroviral therapy,244-246 beta-blockers,247 diuretics,247 androgens/anabolicsteroids,248-251oral contraceptives,252-254 cortico-steroids,212,255,256 cyclosporine,210,215,257 and siroli-mus258,259 (Table 24). The assessment of thesesecondary causes with history, physical examina-tion, and appropriate laboratory testing is recom-mended for any patient with dyslipidemia, since

effective correction of these disorders may im-prove the lipid profile.

Urine protein excretion, especially if �3 gper 24 hours, can also cause or contribute todyslipidemias.218-223 Therefore, CKD patientswho still produce urine should have proteinexcretion measured, if this has not been donerecently. In some cases, the underlying cause(s)of the proteinuria can be treated and effec-tively reversed. In other cases, angiotensin IIconverting enzyme inhibitors or angiotensin IIreceptor blockers may help reduce protein ex-cretion, and may thereby improve the lipidprofile in some patients. Clinical hypothyroid-ism can cause dyslipidemia,224-226 and evensubclinical hypothyroidism may cause mildchanges.225,260 Some of the signs and symp-toms of hypothyroidism may resemble those ofuremia, which may make the clinical diagnosisof hypothyroidism more difficult in patients withCKD. Glucose intolerance can also cause dyslipi-demias.227-229 Therefore, patients with dyslipide-mia and CKD (but without known diabetes)should be assessed with fasting blood glucoseand possibly glycosylated hemoglobin. Glyce-mic control can improve lipid profiles.

Secondary causes of dyslipidemia in childrenand adolescents, in addition to those listed inTable 24, include lipodystrophy261,262; idiopathichypercalcemia263,264; glycogen storage dis-eases265-268; cystine storage disease; Gaucher dis-ease; Juvenile Tay-Sachs disease; Niemann-PickDisease; sphingolipidoses; obstructive liver dis-ease such as biliary atresia269,270; biliary cirrho-sis; intrahepatic cholestasis; nephrotic syndrome;anorexia nervosa271,272; progeria273,274; systemiclupus erythematosus275,276; Werner syndrome; andKlinefelter syndrome. These conditions are fortu-nately rare, and require referral to appropriatetertiary care specialists.


Documents

II. ASSESSMENT OF DYSLIPIDEMIAS I - SOCHIPEnefrologia.sochipe.cl/subidos/catalogo3/DISLIPIDEMIAS II.pdf · II. ASSESSMENT OF DYSLIPIDEMIAS IN EACH OF the following guideline state-