Stenosis in hemodialysis arteriovenous fistula: Evaluation and treatment

Trainee Forum

Stenosis in hemodialysis arteriovenousfistula: Evaluation and treatment

Rodrigo P. CAMPOS, Marcelo M. DO NASCIMENTO, Domingos C. CHULA,

Daniel E. DO NASCIMENTO, Miguel C. RIELLA

Evangelic Faculty of Parana, Evangelic University Hospital of Curitiba, Curitiba, Brazil

AbstractVascular access complications are one of the main causes associated with an increase in morbidity

and mortality in stage 5 chronic kidney disease patients. The arteriovenous fistula is regarded as the

vascular access of choice for hemodialysis (HD) because of its superior patency and lower compli-

cation rates. Stenosis is considered the major cause of dysfunction of arteriovenous fistula. Despite

the relatively low thrombosis rates of arteriovenous fistula, surveillance programs are necessary for

detection of stenosis. We report a case of a HD patient who had never achieved an adequate Kt/V

since the start of maintenance HD. During the investigation, abnormal findings were found on

physical examination of the fistula, in addition to an alteration in intra-access pressure (IAP) meas-

urements. A venous stenosis was diagnosed by Doppler ultrasound and then promptly treated with

percutaneous transluminal angioplasty. The purpose of the discussion is to highlight the peculiar-

ities of arteriovenous fistulae, methods of surveillance, including physical examination, IAP, recir-

culation, and measurements of blood flow, and the importance of the correction procedures for the

stenosis.

Key words: Arteriovenous fistula stenosis, surveillance methods, stenosis correction

NEPHROLOGY TRAINING INEVANGELIC UNIVERSITY HOSPITAL OFCURITIBA, BRAZIL

Two years of residency in Internal Medicine is a require-ment for acceptance in a Nephrology Training Program,

which usually lasts another 2 years. Certification in Ne-

phrology is not mandatory when the Nephrology Training

Program is accredited by the Department of Health and

Education. However, the Brazilian Society of Nephrology

certifies nephrologists through a national board examina-

tion. This suffices for entering private practice, although

progression in academic medicine generally requires a

Masters or a PhD degree. The first year of nephrologytraining usually involves rotations through clinical ne-

phrology and teaching activities: journal clubs, and clin-

ical and pathology conferences. In the second year, in

addition to the foregoing, there is training in transplan-

tation, peritoneal dialysis, and hemodialysis (HD). Train-

ees learn to place central venous lines and peritoneal

catheters for acute HD and peritoneal dialysis, respec-

tively. Kidney biopsies are also learned under supervision.The division oversees around 500 patients from 3

dialysis clinics, and 130 of these patients are treated with

chronic peritoneal dialysis. Our transplant program per-

forms between 75 and 90 kidney transplants a year and

has cumulatively performed close to 1000 procedures.

Acute complications from this pool of dialysis and

posttransplant patients are managed at our University

Hospital.

Correspondence to: M. C. Riella, MD, PhD, FACP, CatholicUniversity of Parana Bruno Filgueira, 369, 171 andar, 80240-220 Curitiba, Parana, Brazil.E-mail: [email protected]

Hemodialysis International 2006; 10:152–161

r 2006 The Authors. Journal compilation r 2006 International Society for Hemodialysis152

With one of the largest emergency rooms in town, ourUniversity Hospital receives several sick patients, providing

a good opportunity for management of acute renal failure.

Training in HD

In the first year of fellowship, the trainee is involved with

HD treatment for inpatients, including those in intensive

care units. In our unit, the most commonly used methodto treat acute renal failure is extended daily HD followed

by continuous veno-venous HD. Last year, we reached an

average of 60 HD procedures per month in the intensive

care units.

The training in outpatient HD occurs in the second

year of the Nephrology fellowship and lasts for 1 year.

The university division involves 120 in-hospital outpa-

tients and 250 patients in an outpatient unit. In this ro-tation, the trainee is under the supervision of the chief

HD attending, who is also responsible for sessions and

orders in unit. Twice a week, the trainee is obligated to

organize conferences about aspects of maintenance HD,

bone disease, cardiovascular disease, anemia, nutrition,

and vascular access. In this same rotation the trainee is

responsible for inpatients who are undergoing mainte-

nance HD.

Interventional nephrology trainingprogram

An interventional nephrologist is a nephrologist who per-

forms interventional procedures essential to the manage-

ment of patients with renal disease. Initially, nephrologists

only performed kidney biopsies and inserted temporary

dialysis catheters. Later, they started to insert peritonealdialysis catheters, and recently, started performing inter-

ventional radiologic procedures including permanent

tunneled catheter placement, angioplasty, and thromboly-

tic procedures in AV fistula and AV graft. The subspecialty

originated, has developed, and continues to grow with

the support of the American Society of Diagnostic and

Interventional Nephrology and the Interventional Ne-

phrology Committee of the International Society of Ne-phrology.

In 2005, a formal Interventional Nephrology Training

Program was offered at our institution and is perhaps the

first of its kind in Latin America. It was offered to and

accepted by 3 renal fellows who had completed their 2-

year Nephrology Fellowship. The program, intended to

last 1 year, involves 4 modules: ultrasonography, vascular

access, peritoneal dialysis access, and urology.

The ultrasound rotation consists of both lectures andhands-on training. It includes ultrasound physics and in-

strumentation and basic ultrasound interpretation, with

the remainder of the time devoted exclusively to perfor-

mance and interpretation of kidney, bladder, and renal

transplant ultrasounds, and kidney biopsy procedures. It

has an optional rotation on color Doppler ultrasound

training for vascular access, renal arteries, and renal trans-

plants.The vascular access module involves all aspects of vas-

cular access in this population: placement of temporary

and permanent venous catheters, methods of surveil-

lance, fistulography, and treatment of failing and failed

arteriovenous fistulae by percutaneous transluminal

angioplasty (PTA). Native arteriovenous fistula (AVF)

constitutes around 80% of vascular access seen in our

country.During the PD access rotation, the trainee will learn the

3 techniques of placing permanent PD catheters: dissec-

tion (surgical technique); peritoneoscopic techniques us-

ing a small peritoneoscope to inspect the abdomen and a

surrounding spiral guide to advance the catheter into the

abdomen and the cuff into the musculature; and blind

techniques using a needle, guide wire, dilator, and split

sheath (Seldinger techniques), or using a trocar and arigid guide.

In the urology rotation, which should be incorporated

into all clinical nephrology training because of its impor-

tance in solving rapidly nephrourologic problems, the

trainee learns about nephrostomy techniques (radios-

copic or ultrasound guided), endourological procedures,

and extracorporeal shock wave lithotripsy.

CASE REPORT

A 47-year-old man on maintenance HD was referred to

our Interventional Nephrology Team in August 2005 be-cause of a high negative prepump pressure during HD

sessions. A radiocephalic fistula in the right forearm had

been placed in April of the same year and HD had been

initiated in May 2005. An adequate Kt/V had never been

achieved since the start of HD. The AVF was being punc-

tured by the ‘‘buttonhole technique’’ with a 15-gauge

needle but blood flows above 350 mL/min led to a col-

lapse of bloodlines.

Clinical evaluation

On physical examination of his AVF, a hard ‘‘waterham-

mer’’ pulse was found in the initial segment of the vein.

When his arm was elevated above the level of the heart,

Stenosis in hemodialysis arteriovenous fistula

Hemodialysis International 2006; 10:152–161 153

there was only a partial collapse of the vein segment (Fig-

ure 1). On examination of the entire length of the fistula,

it was clear that the pulse as well as the caliber of the veindiminished abruptly at a point downstream from the an-

astomosis. At this point, an increase in the intensity of the

thrill was found. The AVF was occluded manually several

centimeters from the site of the anastomosis and no aug-

mentation of the pulse was observed.

In our unit, we also routinely measure intra-access

pressure (IAP) and venous dynamic pressure (VDP) every

2 weeks. The IAP is measured through a digital pressuremeter and is normalized for mean arterial pressure (MAP)

in the arterial (aIAP/MAP) and venous (vIAP/MAP) punc-

ture sites. Venous dynamic pressure is measured accord-

ing to the K/DOQI Protocol.1 The aIAP/MAP and vIAP/

MAP from the last measurement were 0.12 and 0.11,

respectively, and the VDP was 90 mmHg. Upon repeat

evaluation, they were 0.12, 0.10, and 95 mmHg, respec-

tively. Both tests (physical exam and decreased IAPs) sug-gested some venous stricture, possibly in the initial

segment of the vein (juxta-anastomotic region).

Radiological evaluation and procedure

On the same day as the evaluation of the access pressures,

we evaluated the AV fistula by a color Doppler ultrasound

(Esaote Megas CVX, Genoa, Italy) with a linear 10-5 MHz

probe. A stenosis was found 3 cm downstream from theanastomosis, causing more than a 50% reduction in in-

ternal diameter (significant stenosis) (Figure 2). The ex-

amination revealed a blood flow of 310 mL/min.

Figure 1 Radio-cephalic AV fistula on right forearm. The area inside the circles indicates the juxta-anastomotic area. The arrowshows the vein downstream from the juxta-anastomotic area, without arm elevation (A) and with arm elevation (B). In (B) isthere a collapse of the vein. Neither in (A) nor in (B) is there a collapse of juxta-anastomotic area.

Figure 2 Doppler ultrasound of AV fistula. The exam reveals a stenosis (arrow) with a 1.2 cm long stenosis with greater than50% reduction in lumen diameter. In the area of stenosis is seen a turbulent blood flow.

Campos et al.


On the next day, an angiogram was performed. The AVfistula was cannulated using a micropuncture needle. A

micropuncture wire was inserted into the vein and was

used to introduce a 5 F dilator. Contrast was injected

through this dilator to visualize the fistula and draining

veins. The entire venous drainage up through the central

circulation was evaluated. By occluding the fistula

downstream (proximal) from the tip of the dilator, the

vein, anastomosis, and distal artery were visualized (ret-rograde injection). There was no demonstrable arterial or

central vein stenosis. The angiogram revealed a stenosis

(greater than 50% decrease in lumen diameter) 3 cm

downstream from the anastomosis (Figure 3A). During

the course of the same examination, a percutaneous an-

gioplasty (PTA) was performed. The guidewire was

passed retrogradely in the direction of the anastomosis.

A 4 mm balloon was inflated to provide 30% overdilata-tion of the vein. This resulted in successful dilatation of

the venous stenosis (Figure 3B). Immediately after the

PTA, the blood flow observed by color Doppler ultra-

sound was 860 mL/min. During the next HD session, the

aIAP/MAP and vIAP/MAP increased to 0.25 and 0.23, re-

spectively. The next month, the patient achieved a Kt/V of

1.22, and the AV fistula has been functioning well since

then.

DISCUSSION

The native AVF is regarded as the vascular access of

choice for HD patients because of its superior patency

and lower complication rates, once it has fully matured.1

Despite the relatively low AVF thrombosis rate, existing

guidelines advocate surveillance programs for early de-tection of stenosis and its correction before thrombosis

occurs, to reduce morbidity and prolong access patency.1

Schwab et al.2 reported a decrease in AVF thrombosis rate

from 0.16 to 0.07 per patient per year using a surveillance

program and correction of AVF stenosis by PTA.

The main purpose of vascular access monitoring and

surveillance programs is the early identification of steno-

sis, because it is the main cause of AVF dysfunction andthrombosis. A critical reduction of blood flow is often the

final event that leads to the thrombosis of the fistula. The

pathophysiology underlying the occurrence of stenosis is

complex. It includes cellular proliferation, microvessel

formation, and cytokine expression (perhaps the key fac-

tor) by smooth muscle cells, endothelial cells, and ma-

crophages. The cytokines result in further activation and

proliferation of these cell types, ending in venous ne-ointimal hyperplasia.3 Potential mediators that have been

suggested to play a role in this process include basic

Figure 3 (A) Radiocephalic AV fistula angiogram with juxta-anastomotic stenosis, 3 cm moving up from the anastomosis. (B)Successful angioplasty with less than 30% residual lesion. Arrow indicates location of stenosis. Cephalic vein (CV) and radialartery (RA).



fibroblast growth factor (bFGF), platelet-derived growthfactor (PDGF), vascular endothelial growth factor (VEGF),

and extracellular matrix (ECM) proteins.3–5 Even though

these mechanisms have been studied largely in models of

AV grafts, it is possible that the same mechanisms may be

involved in the occurrence of stenosis of the AVF.

An early AVF failure is a term used to characterize a

fistula that has never been cannulated or has failed within

3 months of use. The major cause of early fistula failure isjuxta-anastomotic stenosis (JAS).6 The segment of the

vein mobilized for anastomosis is susceptible to stenosis,

either because of technical issues related to surgical pro-

cedure (kinking, axial rotations, adventitial bands), vein

devascularization, increased vein turbulence, or shear

stress.7,8 Other causes of early AVF failure may include

acessory veins, atherosclerotic disease, arterial stenosis,

anastomotic stenosis, and fibrotic veins. Beathard et al.6

have reported that venous stenosis was the cause in 78%

of all cases with early AVF failure and of these, 43% were

in the JAS location. Some studies, using other definitions

of early failure, reported that the incidence of early failure

was more common in forearm fistulas than in upper arm

fistulas9,10 and in brachiocephalic fistulas than in bra-

chiobasilic fistulas.11

Late failure is defined as the failure of an AVF that oc-curs after 3 months. Stenosis is still the most common

cause. Fortunately, venous stenosis in the downstream

segment does not occur in AVF with the same degree of

frequency as in grafts. Unlike the case with grafts, venous

downstream stenosis in AVF develops more centrally at

areas of vein bifurcation, pressure points, and in associ-

ation with venous valves.12

The NKF-K/DOQI guidelines for vascular access rec-ommend the use of vascular adequacy parameters and

physical examination for monitoring AVF and AV grafts.1

The methods for surveillance of AVF are summarized in

Table 1. Unexplained decreases in delivered dialysis dose,

as measured by Kt/V or urea reduction ratio, may be as-

sociated with vascular access dysfunction. An AVF, that

does not support a pump flow above 350 mL/min be-

cause of decreased flow will become ineffective, and re-circulation problems will occur. Prolonged bleeding after

removal of the needles from puncture sites may occasion-

ally be an indicator of elevated IAP (in the absence of ex-

cessive anticoagulation).

Physical examination is a forgotten tool for vascular

access monitoring in HD units.13 The observation that an

experienced nurse of the HD team can predict the ulti-

mate adequacy of a fistula with 80% accuracy simplythrough physical examination,14 and that the major caus-

es of AV fistula failure can be identified by physical ex-

amination,15,16 suggests that this is an easy and cost-

effective method of monitoring AV fistula. Everyone on

the HD team should be trained to recognize fistula prob-

lems on physical examination. In our HD unit ‘‘fistula

rounds’’ are performed every month. As part of this prac-

tice, it is necessary to obtain and record the informationabout the date of creation of the AV fistula, its location,

vessels involved, and past history of treatment (PTA or

surgical) in the patients chart.

The first step of such an examination is to evaluate the

anastomosis. The 2 most important components of an-

astomosis examination is the thrill, which is an indicator

of flow, and the pulse, which is an indicator of down-

stream resistance.16 Another component that can be eval-uated is the bruit. Normally, the thrill and the bruit are

prominent near the anastomosis and they are both pre-

sent in systole as well as early diastole (continuous). The

presence of stenosis reduces the blood flow and this, in

turn, reduces the strength of the thrill. If stenosis is

present at a point away from the anastomosis, the intra-

access resistance rises and the flow may occur only during

systole. In this situation, the thrill and bruit will be foundonly in systole (not continuous). With regard to the pulse,

it must be soft. If it is forceful, like a ‘‘water-hammer’’

pulse, this often suggests the presence of stenosis. The

intensity of the hyperpulsatility is proportional to the se-

verity of the stenosis.

The second step is to examine the entire length of the

AV fistula. Moving up the vein away from the an-

astomosis, the thrill and bruit gradually diminish. Anincrease in the strength of the thrill or a new palpable

thrill downstream from the anastomotic site suggests the

Table 1 Methods of surveillance for arteriovenous fistula(AVF)

Clinical evaluationPhysical examinationKt/VBleeding time

RecirculationStatic access pressure

Arterial intra-access pressureVenous intra-access pressure

Dynamic venous pressureBlood flow measurement

Doppler ultrasounda

Ultrasound dilutiona

Hematocrit dilutionThermal dilutionConductance dilution

aThe most used methods for measurement of blood flow.

Campos et al.


presence of a stenosis. The pulse should be soft and com-pressible as in the anastomosis. The pulse becomes hard

in the presence of a body stenosis. An important mane-

uver to evaluate the pulse is to elevate the arm above the

level of the heart. Normally, the pulse may collapse with

arm elevation. If it does not, the IAP must be high and a

significant stenosis may be the cause. In this case only the

segment upstream from the stenosis does not collapse but

the downstream segment does collapse. This is a goodclue to find the location of the stenosis.

An important cause of early AV fistula dysfunction is

the presence of accessory veins. They must be distin-

guished from collateral veins, which are pathological and

develop in the presence of stenosis. The development of

AV fistula depends on the inflow pressure and the resist-

ance of draining veins. This resistance is decreased in the

presence of side branch vessels (accessory veins), thuslimiting the maturation of the AV fistula. Accessory veins

may be single or multiple, but not all apparently decrease

blood flow. An accessory vein that is less than one-fourth

the diameter of the main vessel is unlikely to prove to be

significant.12 When the fistula is occluded proximally, the

thrill at the fistula should disappear. If it does not, an

accessory vein may be providing an outflow channel. In

this setting, palpation of the main vein below the site ofthe manual occlusion will reveal a thrill at the site of the

accessory vein.8

The third and last step on physical examination is to

check for pulse augmentation. With one hand, the body

of the AV fistula is manually occluded proximally and

with the other hand the pulse is palpated next to the an-

astomosis. With this maneuver, the pulse must become

hyperpulsatile. If the pulse augments poorly, it is an in-dicator of low inflow. In this case, the arteries, an-

astomosis, and juxta-anastomotic region should be

evaluated. In cases that the pulse is already hyperpulsa-

tile, the maneuver of pulse augmentation can be used to

assess the severity of the stenosis. If there is no increase in

pulse intensity, it means that the stenosis is severe, com-

parable with total occlusion. In contrast, when the inten-

sity increases, the stenosis must be of lesser degree.Physical signs as swelling in the access arm and mul-

tiple subcutaneous collateral veins at the neck, shoulder,

and upper chest are highly suggestive of central vein

stenosis. It is important for the examiner to realize that

not all central vein stenoses cause these signs. It generally

appears soon after access placement and is associated

with a previous history of central venous catheter.

As for surveillance in AV grafts, the NFK-K/DOQI rec-ommends the use of intra-access (static) and dynamic

venous pressures, and measurements of recirculation or

flow for detecting AV fistula in risk of thrombosis.1 Theproblem is that the major studies using these methods of

surveillance were conducted in patients with AV graft

until the publication of NKF-K/DOQI guidelines, and

there are some characteristics that differentiate AV fistula

from AV graft.

In AV fistula, the IAP falls to about 20% of the systemic

arterial pressure in the initial segment used for arterial

needle puncture. However, in a well-functioning AV graft,the systemic arterial pressure is dissipated across the 2

anastomoses: 45% at the arterial and 25% at the venous

end. In dysfunctional AV fistula, the IAPs tend to be lower

than in grafts, making it more difficult to detect an ap-

propriate increase. Notwithstanding the differences in

pressure profiles between AV fistula and AV graft, access

flow rates tend to be quite similar when compared with

similar anatomical locations.17 When a stenosis developsin an AV fistula, the IAP may be stable because collateral

channels (accessory or collateral veins) permit exit of the

inflowing blood. Consequently, the flow in upstream seg-

ment may not change. Thus, it is important to identify

these side branch veins in physical examination. By con-

trast, when stenosis develops in any segment of AV graft

the flow invariably decreases, because there are no col-

lateral channels to dissipate the inflowing blood.As discussed above, theoretical considerations suggest

that the measurement of IAP in AVF may not be as useful

as for AV grafts. However, this has not been validated in

clinical studies. For a reliable measure of IAP, it must be

normalized to MAP(IAP/MAP).18 It should be measured

both at the arterial puncture site (aIAP/MAP) and venous

puncture site (vIAP/MAP), because vIAP/MAP alone may

fail to detect lesions located between the needles and inthe juxta-anastomotic area. The indirect technique for

measuring IAP uses HD machine equipment.18 This tech-

nique uses the pressures from arterial prepump and ve-

nous postpump drip chambers when the blood pump is

turned off. The height distance from the AVF to the upper

meniscus of the blood drip chamber is measured, con-

verted into mmHg, and then added to the IAP value.

Nevertheless, this methodology is prolonged, requiring11 separate steps to assure accuracy.1 To simplify the

process, the IAP can be measured directly. A device,

called Access Alerts from Medisystems (Medisystems

Corporation, Seattle, WA, USA), achieves this as it joins

the sterile AVF needle to an aneroid pressure meter via a

tubing (Figure 4A). Alternatively, as mentioned earlier, we

measure IAP using a digital pressure meter with a filter to

avoid contamination of the puncture needle (Figure 4B).As distinct from the AV grafts (where the vast majority of

lesions are located at the venous anastamosis), the most



common lesions in AVF are in the juxta-anastomotic re-

gion or in the body. It then follows that while the vIAP/

MAP may be as important as aIAP/MAP for the detection

of stenosis in AV grafts, in the case of AVF vIAP/MAP will

be probably normal in the absence of outflow stenosis.

Thus, the measurement of aIAP/MAP is more important

in the prediction of stenosis in AVF. Besarab et al.19,20

proposed normal values of IAP/MAP in AVF: 0.13 to 0.43

at the arterial site and 0.08 to 0.34 at the venous site

(Table 2). However, there are no studies in AVF using IAP

as a method of surveillance for detection and preemptive

correction of stenosis. Thus, IAP monitoring cannot be

recommended as the only method for surveillance of AVF

at this time. Perhaps, a hybrid program with other meth-

ods may be the choice.21

Dynamic venous pressure can be measured during the

first 2 to 5 min of a HD session after reducing the pump

flow to 200 mL/min. Although commonly used, it should

be abandoned as its utility is limited, as it can change with

changes in hematocrit as well as the gauge of the needle.

Moreover, its value lies in the detection of lesions that

cause elevation of resistance at the outflow segment and,

thus, it is more useful as a method of surveillance for AV

grafts than AVF. Measurement of recirculation, on the

other hand, becomes a more useful screening tool for AV

fistula than in AV graft because blood flow in AV fistula,

unlike AV graft, can decrease to a level less than the pre-scribed HD machine pump flow (o300 mL/min), while

maintaining access patency.22

Quantitative measurements of blood flow have proven

to be the most sensitive predictor method of surveillance

for impeding thrombosis in AVF.2,23–27 The NFK-K/DOQI

recommends that vascular access with blood flow less

than 600 mL/min or less than 1000 mL/min that has de-

creased by more than 25% over 4 months should be re-ferred for a fistulogram.1 Unfortunately, this report did

not make a distinction between AV fistula and AV graft. It

is well known that AV fistula can remain patent even with

flows below 300 mL/min. The Canadian Guidelines for

vascular access recommend investigation of an AV fistula

Figure 4 Direct measurements of intra-access pressure using an aneroid pressure meter (A) and a digital pressure meter (B).Picture (A) was reprinted with permission from Medisystems Corporation.

Table 2 Criteria for referring patient for angiography/intervention using intra-access pressure in AVF (Adapted with per-mission from Besarab A et al. Detection of access strictures and outlet stenoses in vascular accesses. Which test is best? ASAIO J1997;43:M543-47)

Stenosis degree aIAP/MAP vIAP/MAP

o50%450% in:

0.13 to 0.43 0.08 to 0.34

Venous outflow segment � 0.44 or 0.35Mid-segment � 0.44 and o0.35Arterial inflow segment o0.131clinical signs clinical signs

Pressures are normalized for mean arterial pressure at the arterial puncture site (aIAP) and at the venous puncture site (vIAP). An increase inthe ratio of 0.2 units over time is also indication for evaluation.

Campos et al.


with flows less than 500 mL/min or a fall in access flowgreater than 20% of baseline.28 A comprehensive analysis

of the relationship between flow, stenosis, and thrombosis

is lacking in AV fistula and is essential to the determina-

tion of an optimal access protocol.

There are several noninvasive techniques for measuring

blood flow: Doppler ultrasound, ultrasound dilution,

hematocrit dilution, thermal dilution, and conductance

dilution. Doppler ultrasound can be used to measure thevelocity of blood flow and if the cross-sectional diameter

is known, the volume of blood flow can be calculated.

Nevertheless, this method may have inaccuracies because

of operator error, turbulent blood flow, and variability in

the cross-sectional diameter of AVF. Its advantage is that

this examination is useful in the evaluation of vascular

access anatomy and in determining the location and grade

of stenosis. Some studies have reported that blood flowmeasured by Doppler ultrasound is a reliable indicator of

subsequent short-term thrombosis risk in AV grafts and

AV fistula.29–32 In the study by Bay et al.33 the procedure

proved beneficial only for AV grafts.

The most widely used technique to measure blood flow

in HD vascular access is the ultrasound dilution method

using the Transonic HD01 Systems (Transonic Inc., Ith-

aca, NY, U.S.A.). The technique was originally describedby Krivitsky et al.34 and then validated in many studies,

as a safe method for the detection of stenosis and conse-

quently is recommended as a method of surveillance to

reduce the thrombosis rate.2,23–27,35 In this technique,

the access blood flow is measured during an HD session.

To perform the access flow measurement, the blood

pump is stopped and the arterial and venous bloodlines

are reversed from their normal position. Photometric flowsensors, which are connected to an electronic flow meter,

are clipped onto the reversed arterial and venous blood-

lines. The HD pump is restarted at a fixed blood flow,

usually 200 mL/min, and a saline bolus (5 mL) is injected

into the reversed venous bloodline. The saline mixes with

the blood flowing through the access and is detected by

the sensor on the arterial line. The data are collected on a

laptop computer, and blood flow values are calculatedand recorded. It basically measures the blood flow by the

difference of saline concentration between the 2 sensors,

correcting for a time factor.

The preemptive treatment of stenosis in AV grafts and

AV fistulas (before their thrombosis) contributes to the

reduction in the morbidity and mortality of patients who

need chronic renal replacement therapy. After thrombo-

sis, the salvage of vascular access becomes more difficult.With the passage of time, the thrombus can become more

resistant to thrombectomy. A temporary central catheter

raises the risk of infections and hospitalizations. Theseprocedures also increase costs. All these disadvantages

have an enormous effect on the quality of life of end-stage

renal disease patients.

The NFK-K/DOQI guidelines recommend that each di-

alysis center should determine which procedure for the

correction of stenosis (PTA or surgery) is best for the pa-

tients, based on the expertise at that center.1 There is no

randomized prospective study comparing the correctionof AV fistula stenosis by PTA or surgery. One retrospective

study comparing both methods reported a prolonged

patency rate with PTA over surgical correction in the

treatment of stenosis and occluded AV fistula.36 Uncon-

trolled, published data do not indicate a preferred tech-

nique for the treatment of AV fistula stenosis.6,27,37–45

Access stenosis should be treated if stenosis is � 50% and

is associated with abnormalities such as previous throm-botic episodes, elevated IAP, abnormal recirculation, ab-

normal physical signs, unexplained decrease in adequacy

dialysis parameters, and decreased blood flow.1 The pro-

cedure is classified as successful if there is less than a 30%

residual lesion according to NFK-K/DOQI.1

Percutaneous transluminal angioplasty has become a

standard treatment for the correction of stenosis in AVF

and AV graft.46–48 The technique has been demonstratedto be safe, easily performed, and effective. It has some

advantages over surgical procedure: it is an outpatient

procedure; it does not prohibit the immediate use of the

access for dialysis; it has minimal to no blood loss; hos-

pitalization is avoided; and rarely is there discomfort to

the patient. The only disadvantage is that there are some

lesions that do not respond to treatment and the results

obtained are not permanent.49

Increasingly, PTAs are being performed by interven-

tional nephrologists who have shown that, with appro-

priate training, they can perform a major portion of the

procedures that are necessary for access management. A

report by Beathard and Litchfield50 with 14,067 proce-

dures demonstrated that with academic training pro-

grams, interventional nephrologists can perform these

procedures effectively and safely.In conclusion, AVF dysfunction is an important cause

of inadequate HD. Therefore, early detection of AVF dys-

function must be part of the HD patient’s care routine,

especially in those who present an inadequate Kt/V. In

this presentation, we reported a case whose investigation

of AVF dysfunction, initially suspected by the presence of

abnormalities in physical examination and IAP, was con-

firmed by the detection of stenosis on Doppler ultrasoundand fistulogram. The early detection of AVF dysfunction

by physical examination and improvements of Kt/V and



blood flow observed after correction of AVF stenosisthrough PTA show the relevance of establishing proto-

cols of AVF surveillance in HD units.

Manuscript received November 2005; revised December

2005.

REFERENCES

1 III. NKF-K/DOQI clinical practice guidelines for vascularaccess: Update 2000. Am J Kidney Dis. 2001; 37:S137–S181.

2 Schwab SJ, Oliver MJ, Suhocki P, McCann R. Hemodial-ysis arteriovenous access: Detection of stenosis and re-sponse to treatment by vascular access blood flow. KidneyInt. 2001; 59:358–362.

3 Roy-Chaudhury P, Kelly BS, Miller MA, et al. Venous ne-ointimal hyperplasia in polytetrafluoroethylene dialysisgrafts. Kidney Int. 2001; 59:2325–2334.

4 De Marchi S, Falleti E, Giacomello R, et al. Risk factorsfor vascular disease and arteriovenous fistula dysfunctionin hemodialysis patients. J Am Soc Nephrol. 1996; 7:1169–1177.

5 Roy-Chaudhury P, Kelly BS, Zhang J, et al. Hemodialysisvascular access dysfunction: From pathophysiology tonovel therapies. Blood Purif. 2003; 21:99–110.

6 Beathard GA, Arnold P, Jackson J, Litchfield T. Aggressivetreatment of early fistula failure. Kidney Int. 2003;64:1487–1494.

7 Vassalotti JA. Arteriovenous fistula stenosis: New termi-nology. Semin Dial. 2004; 17.

8 Kian K, Vassalotti JA. The new arteriovenous fistula: Theneed for earlier evaluation and intervention. Semin Dial.2005; 18:3–7.

9 Miller PE, Tolwani A, Luscy CP, et al. Predictors of ad-equacy of arteriovenous fistulas in hemodialysis patients.Kidney Int. 1999; 56:275–280.

10 Feldman HI, Joffe M, Rosas SE, Burns JE, Knauss J,Brayman K. Predictors of successful arteriovenousfistula maturation. Am J Kidney Dis. 2003; 42:1000–1012.

11 Oliver MJ, McCann RL, Indridason OS, Butterly DW,Schwab SJ. Comparison of transposed brachiobasilic fis-tulas to upper arm grafts and brachiocephalic fistulas.Kidney Int. 2001; 60:1532–1539.

12 Beathard GA. A practitioner’s resource to hemodialysisarteriovenous fistulas. Fistula First. www.esrdnetwork.org, 2003

13 Beathard GA. Physical examination: The forgotten tool.In: Gray R, Sands J eds. Multidisciplinary Approach forHemodialysis Access. New York: Lippincott, Williams &Wilkins; 2002; 111–118.

14 Robbin ML, Chamberlain NE, Lockhart ME, et al. Hemo-dialysis arteriovenous fistula maturity: US evaluation.Radiology. 2002; 225:59–64.

15 Beathard GA. Physical examination of the dialysis vas-cular access. Semin Dial. 1998; 11:231–236.

16 Beathard GA. An algorithm for the physical examina-tion of early fistula failure. Semin Dial. 2005; 18:331–335.

17 Besarab A. Advances in end-stage renal diseases 2000.Access monitoring methods. Blood Purif. 2000; 18:255–259.

18 Besarab A, Frinak S, Sherman RA, et al. Simplified meas-urement of intra-access pressure. J Am Soc Nephrol. 1998;9:284–289.

19 Besarab A, Lubkowski T, Frinak S, Ramanathan S, Esco-bar F. Detecting vascular access dysfunction. ASAIO J.1997; 43:M539–M543.

20 Besarab A, Lubkowski T, Frinak S, Ramanathan S, Esco-bar F. Detection of access strictures and outlet stenoses invascular accesses. Which test is best? ASAIO J. 1997;43:M543–M547.

21 Besarab A. Preventing vascular access dysfunction:Which policy to follow. Blood Purif. 2002; 20:26–35.

22 Besarab A, Sherman R. The relationship of recircula-tion to access blood flow. Am J Kidney Dis. 1997; 29:223–229.

23 McCarley P, Wingard RL, Shyr Y, Pettus W, Hakim RM,Ikizler TA. Vascular access blood flow monitoring reduc-es access morbidity and costs. Kidney Int. 2001; 60:1164–1172.

24 Tonelli M, Jindal K, Hirsch D, Taylor S, Kane C, HenbreyS. Screening for subclinical stenosis in native vesselarteriovenous fistulae. J Am Soc Nephrol. 2001; 12:1729–1733.

25 Tessitore N, Bedogna V, Gammaro L, et al. Diagnosticaccuracy of ultrasound dilution access blood flow meas-urement in detecting stenosis and predicting thrombosisin native forearm arteriovenous fistulae for hemodialysis.Am J Kidney Dis. 2003; 42:331–341.

26 Branger BGC, Granolleras C, Dauzat M. Frequence dethromboses de fistules arterio-veineuses pour hemodia-lyse: Aport de deux methodes de surveillance: Le Dop-pler et la diluition des ultrasons. Nephrologie. 2004;25:204–209.

27 Tessitore N, Lipari G, Poli A, et al. Can blood flow sur-veillance and pre-emptive repair of subclinical stenosisprolong the useful life of arteriovenous fistulae? A ran-domized controlled study. Nephrol Dial Transplant. 2004;19:2325–2333.

28 Ethier JH, Lindsay RM, Barre PE, Kappel JE, Carlisle EJ,Common A. Clinical practice guidelines for vascular ac-cess. Canadian Society of Nephrology. J Am Soc Nephrol.1999; 10(Suppl. 13):S297–S305.

29 Strauch BS, O’Connell RS, Geoly KL, Grundlehner M,Yakub YN, Tietjen DP. Forecasting thrombosis of vascular

Campos et al.


access with Doppler color flow imaging. Am J Kidney Dis.1992; 19:554–557.

30 Sands J, Young S, Miranda C. The effect of Doppler flowscreening studies and elective revisions on dialysis accessfailure. ASAIO J. 1992; 38:M524–M527.

31 May RE, Himmelfarb J, Yenicesu M, et al. Predictivemeasures of vascular access thrombosis: A prospectivestudy. Kidney Int. 1997; 52:1656–1662.

32 Schwarz C, Mitterbauer C, Boczula M, et al. Flow mon-itoring: Performance characteristics of ultrasound dilu-tion versus color Doppler ultrasound compared withfistulography. Am J Kidney Dis. 2003; 42:539–545.

33 Bay WH, Henry ML, Lazarus JM, Lew NL, Ling J, LowrieEG. Predicting hemodialysis access failure with colorflow Doppler ultrasound. Am J Nephrol. 1998; 18:296–304.

34 Krivitski NM. Theory and validation of access flow meas-urement by dilution technique during hemodialysis. Kid-ney Int. 1995; 48:244–250.

35 Krivitski NM, MacGibbon D, Gleed RD, Dobson A. Ac-curacy of dilution techniques for access flow measure-ment during hemodialysis. Am J Kidney Dis. 1998; 31:502–508.

36 Dapunt O, Feurstein M, Rendl KH, Prenner K. Translu-minal angioplasty versus conventional operation in thetreatment of haemodialysis fistula stenosis: Results froma 5-year study. Br J Surg. 1987; 74:1004–1005.

37 Palder SB, Kirkman RL, Whittemore AD, Hakim RM,Lazarus JM, Tilney NL. Vascular access for hemodialysis.Patency rates and results of revision. Ann Surg. 1985;202:235–239.

38 Glanz S, Gordon DH, Butt KM, Hong J, Lipkowitz GS.The role of percutaneous angioplasty in the managementof chronic hemodialysis fistulas. Ann Surg. 1987;206:777–781.

39 Zibari GB, Rohr MS, Landreneau MD, et al. Complica-tions from permanent hemodialysis vascular access. Sur-gery. 1988; 104:681–686.

40 Schwab SJ, Raymond JR, Saeed M, Newman GE, DennisPA, Bollinger RR. Prevention of hemodialysis fistulathrombosis. Early detection of venous stenoses. KidneyInt. 1989; 36:707–711.

41 Gmelin E, Winterhoff R, Rinast E. Insufficient hemodi-alysis access fistulas: Late results of treatment withpercutaneous balloon angioplasty. Radiology. 1989;171:657–660.

42 Oakes DD, Sherck JP, Cobb LF. Surgical salvage of failedradiocephalic arteriovenous fistulae: Techniques and re-sults in 29 patients. Kidney Int. 1998; 53:480–487.

43 Beathard GA. Percutaneous angioplasty for the treatmentof venous stenosis: A nephrologists view. Semin Dial.1995; 8:166–170.

44 Arnold WP. Improvement in hemodialysis vascular ac-cess outcomes in a dedicated access center. Semin Dial.2000; 13:359–363.

45 Turmel-Rodrigues L, Pengloan J, Baudin S, et al. Treat-ment of stenosis and thrombosis in haemodialysis fistu-las and grafts by interventional radiology. Nephrol DialTransplant. 2000; 15:2029–2036.

46 Schwab SJ, Saeed M, Sussman SK, McCann RL, StickelDL. Transluminal angioplasty of venous stenosis in po-lytetrafluoroethylene vascular access. Kidney Int. 1987;32:395–398.

47 Glanz S, Gordon D, Butt KM, Hong J, Adamson R, Scla-fani SJ. Dialysis access fistulas: Treatment of stenoses bytransluminal angioplasty. Radiology. 1984; 152:637–642.

48 Beathard GA. Percutaneous transvenous angioplasty inthe treatment of vascular access stenosis. Kidney Int.1992; 42:1390–1397.

49 Asif A, Merrill D, Briones P, Roth D, Beathard GA. Hemo-dialysis vascular access: Percutaneous interventions bynephrologists. Semin Dial. 2004; 17:528–534.

50 Beathard GA, Litchfield T. Effectiveness and safety ofdialysis vascular access procedures performed byinterventional nephrologists. Kidney Int. 2004; 66:1622–1632.



Documents

Stenosis in hemodialysis arteriovenous fistula: Evaluation and treatment