Year : 2004 | Volume
: 15 | Issue : 3 | Page : 305--320
Screening for Prophylactic Intervention in Failing Hemodialysis Grafts and Fistulas
Brian M Murray
Department of Medicine, University at Buffalo, The State University of New York, USA
Brian M Murray
Renal Division, 462 Grider Street, Buffalo, NY 14215
|How to cite this article:|
Murray BM. Screening for Prophylactic Intervention in Failing Hemodialysis Grafts and Fistulas.Saudi J Kidney Dis Transpl 2004;15:305-320
|How to cite this URL:|
Murray BM. Screening for Prophylactic Intervention in Failing Hemodialysis Grafts and Fistulas. Saudi J Kidney Dis Transpl [serial online] 2004 [cited 2014 Apr 19 ];15:305-320
Available from: http://www.sjkdt.org/text.asp?2004/15/3/305/32980
The maintenance of suitable vascular access continues to represent a major challenge in the care of patients on chronic hemodialysis.  The preferred form of access is the natural arteriovenous fistula, since the synthetic arteriovenous graft is associated with higher rates of thrombosis and infection.  The dialysis patients have a limited number of sites where vascular access can be placed. Exhaustion of such sites can result in the long-term use of central catheters for access. Maximizing the working life of a particular access site is therefore an important consideration, which should improve the ultimate outcome of the individual dialysis patient. Thrombosis of fistulas and grafts is generally preceded by the development of stenosis or narrowing in the vessel. , In the case of fistulas, this can occur anywhere along the fistula vein or draining central veins.  In the case of the synthetic graft, the most frequent site is the anastomosis between the graft and the draining vein; stenoses can also occur within the graft, especially at the sites of repeated venipuncture or in the draining and central veins. 
After thrombosis, the graft or fistula can often be salvaged by either radiological or surgical thrombectomy, though recurrent thrombosis is frequent and ultimately leads to the loss of the graft.  It is generally felt that routine surveillance of grafts to detect developing stenoses, when coupled with prophylactic angioplasty, could potentially reduce the frequency of thrombotic events and consequently prolong the life of the access. Indeed, monthly surveillance of grafts and fistulas is a part of the recommendations of the most recent dialysis outcome quality initiative (DOQI) guidelines of the National Kidney Foundation. , The concept of access surveillance is based on the following assumptions:
1) Most episodes of access failure are preceded by the development of stenosis in the access
2) The development of stenosis in the graft prior to thrombosis can be detected by monitoring techniques and is predictive of thrombosis and finally,
3) Prophylactic intervention (angioplasty or surgery) delays thrombosis of that access.
Although the advisability of such surveillance and interventional programs is widely accepted, there is actually little prospective randomized data to substantiate them. Considerable controversy remains as to the optimal mode of surveillance, the frequency with which such tests should be performed and the threshold values that should precipitate referral for intervention.
Most accesses that thrombose have evidence of stenosis. Our experience in 17 PTFE grafts that required radiological thrombolysis revealed significant stenoses, mostly at the venous anastomosis.  However, stenoses are common even in well-functioning grafts. Gadallah et al.  examined 38 PTFE grafts with digital subtraction angiography and Doppler ultrasound without regard to suspicion of stenosis, and found evidence of a significant stenosis (>50% of diameter) in 50%. In the case of 25 arteriovenous fistulas studied prospectively by duplex and color-flow ultrasonography at 3 months, stenoses were detected in all 25 5. Whether or not a particular stenosis is a forerunner of thrombosis probably depends on the type of access, the location of the stenosis and it's rapidity of progression. Unfortunately, we have little prospective data addressing these issues. One also has to recognize the role of external factors such as sleeping on the access arm, severe hypotensive episodes and needle trauma, in determining the actual timing of thrombosis. Clearly, the mere presence of a stenosis does not necessarily imply imminent thrombosis. While a technique that detects all access stenoses greater than 50% would likely have a high sensitivity in predicting thrombosis, it might also carry a high false-positive rate. If this test was used for angioplasty referral, the high false positive rate would result in the performance of "unnecessary" angioplasties. Therefore, when selecting an access monitoring tool (see below), it is important to consider its ability to predict access thrombosis as well as its ability to detect access stenosis.
Finally, there is the assumption that prophylactic angioplasty of a stenosed access will delay thrombosis and thus prolong the work ing life of the access site leading to better longterm secondary patency. Potential factors that might prevent such an outcome include a high incidence of recurrent stenosis, often within a few months of the initial angioplasty requiring repeat intervention.  Furthermore, unnecessary angioplasty could actually accelerate the rate of progression of a stable stenosis and occasional access losses may occur due to procedure complications (e.g. vein rupture). Clearly such issues can only be definitively addressed by prospective randomized blinded studies of which unfortunately there are relatively few.
Many studies have used thrombotic rates to assess the efficacy of access surveillance. Thrombotic rates are important because an episode of thrombosis may result in missed dialyses and the need for temporary catheter placement. However, a reduction in thrombotic rate does not necessarily result in the prolongation of survival of a particular access site. This is illustrated in [Figure 1]. In scenario A, graft surveillance with prophylactic angioplasty reduces the incidence of thrombosis and prolongs survival of the access site - the desired result. This end result would justify the additional required procedures (three angioplasties versus one thrombectomy). In scenario B, however, graft survival is not prolonged despite a decrease in the thrombotic rate (from 1 to 3 episodes). Instead, surveillance merely serves to increase the number of interventions (5 angioplasties versus two thrombectomies), prior to the final episode of thrombosis as suggested by Paulson et al.  This emphasizes the importance of determining the efficacy of graft surveillance techniques in terms of graft survival as well as thrombotic rates.
In the remainder of this review I intend to review the following issues:
1) What methods of access surveillance are available and which best predicts access thrombosis?
2) What is the preferable method of prophylactic intervention?
3) What is the evidence that access surveillance and preemptive intervention actually prolongs the useful life of arteriovenous grafts and fistulas ?
a) Physical Examination
Physical examination remains the simplest and most obvious technique of access assessment available but is not as rigorously used as it should be. In modern day dialysis units in the United States, cannulation is often done by dialysis technicians with relatively little dialysis experience. The rounding physician may have difficulty doing a detailed access examination on a patient who is already on dialysis. The needles are often heavily taped to avoid them being dislodged and this obscures examination of the access. Examination should include inspection for obvious narrowings or aneurysmal swelling in addition to auscultation at the arterial, mid-portion and venous ends of the access. Prolonged bleeding after needle withdrawal which may be due to elevated pressures in a stenosed graft/fistula is also a clinical clue to access dysfunction. Few objective determinations of the effectiveness of clinical examination have been performed. In terms of detecting stenosis, Thomsen and Stenport  reported that physical examination carried 50% sensitivity and 100% specificity for detecting natural fistula stenoses. On the other hand, Trerotola et al. reported that the presence of a thrill at the arterial, midpoint and venous sections of a graft suggested flow rate greater than 450ml/min. 
b) Doppler Studies of the Access
Doppler interrogation of the access should be an ideal method of monitoring vascular access in that it provides both anatomical and functional information and also can give a close approximation of access blood flow rate. ,, Studies have shown that Doppler examination of AV fistulas just two weeks after fashioning are predictive of subsequent fistula failure.  Doppler techniques appear both sensitive and specific in detecting stenoses in both grafts ,, and fistulas.  The lower the blood flow rate estimated by Doppler the more likely was the graft to thrombose. ,, Sonography can also provide additional information including the degree of aneurysmal formation, graft-vein fistula formation and the presence of fluid collections around the access, but it requires specially trained technicians and is much more expensive than alternative monitoring techniques, which means it is performed every 3 months. This interval may not be frequent enough to detect stenoses before thrombosis occurs, particularly if the progression of stenosis is rapid.
c) Dynamic Venous pressure (DVP)
This technique was first advocated by Schwab et al.  It is based on the concept that most arteriovenous grafts are lost due to venous outlet stenosis, which results in progressively increasing hydraulic pressures in the venous limb of the access.  DVP consists of recording at each dialysis treatment the venous pressure in the dialysis circuit under standardized conditions (usually 200ml/min pump speed and zero ultrafiltration). The value obtained varies, not only with the blood pump speed, but also dialysis needle gauge and placement as well as the type of dialysis machine and tubing.  The initial studies by Choudhury et al  correlated dynamic venous pressures with the results of venography in 46 patients with PTFE grafts studied prospectively; 83% were found to have a stenosis. They found that a critical venous pressure of 145 mm Hg (for a pump speed of 250 ml/min) correlated with the presence of a stenosis. At a pump speed of 300 ml/min, the critical venous pressure was 170 mm Hg. In our own unit (Fresenius 2000K machine with 15 gauge needles), the threshold for angioplasty referral is set at a DVP greater than 125 mmHg on three consecutive treatments. However, it is not possible to establish uniform thresholds for detecting stenosis in all dialysis patients. Instead individual units must develop their own criteria. Alternatively, one can follow the progressive rise of DVP in the same patient under similar conditions. An additional drawback to this technique is that it will not detect stenoses that occur upstream of the venous needle, either within the access or close to the arterial anastomosis, which is a frequent site of problems in the natural arteriovenous fistula. 
d) Static Venous Pressures (SVP)
In an effort to circumvent the problems of reproducibility referred to above, Besarab at al. ,, developed the concept of measuring static venous pressures. With the blood pump stopped, the pressure measured by the venous drip chamber transducer is recorded. The difference between the pressure measured here and the pressure in the venous limb of the access is a function of the difference in height of the access below the drip chamber, and if this is known, the appropriate correction can be made. Usually the derived pressure (SVP) is then expressed as a fraction of the mean arterial pressure (MAP) as the normalized intra-access pressure (IAP) where IAP = SVP/ MAP. A value of >0.5, again on three consecutive occasions, is considered an indication for referral. Alternatively, intragraft pressure at the venous (and arterial) needle sites can be measured directly before initiation of dialysis, by attachment of small disposable transducers (Access Alert Disposable Filter, Henry Schein, Inc. Melville, NY) to the ends of the lines attached to the fistula/shunt needles after insertion.  The initial studies suggested that the measurements of intra-access pressure had both high sensitivity (91%) and specificity (91%) for detection of significant stenoses in both arteriovenous PTFE grafts and fistulas.  However, the widespread adoption of static techniques has been hindered by the unwillingness of dialysis units to invest either the staff time or financial outlay involved. Again, as with DVP, if only the venous needle pressure is measured then, again, arterial and intra-access stenoses may be missed. Specific criteria for angioplasty referral for this technique are detailed elsewhere. , Most recently, this group has developed an alternative, more user friendly approach, to measure venous pressure during dialysis "the Dynamic Venous Access Pressure Ratio Test" that has been incorporated into a product called VascAlert (VascAlert LLC, Chicago, IL). This consists of a software package that utilizes data already being collected by the dialysis machine to calculate a static pressure ratio.  They found this technique to be 70% sensitive and 88% specific for "access failure" over the next three months. "Access failure" was defined as individuals who required surgical or radiological intervention for an access event e.g. an obviously low access flow ( e) Access Blood Flow (ABF) Measurements
Developing blockages within an access with increasing intra-access pressure will eventually result in the reduction of access blood flow once they have reached a certain degree of stenosis.  The measurements of access blood flow not only have the potential to identify a failing access but can also provide data that sheds light on the functional performance of a graft or fistula and the efficacy of a therapeutic intervention such as angioplasty or surgical revision. An estimate of access flow can be obtained from the Doppler studies,  but the preferred method for measuring access flow remains the ultrasound dilution technique, first reported by Krivitski et al.  Several studies have suggested that for arteriovenous grafts, an absolute flow of ,,, or a 20% decrease in access flow  identifies grafts at high risk of thrombosis. For arteriovenous fistulas, which have a lower tendency to clot, the data is not as clear. K/DOQI guidelines do not suggest a different threshold for referral of fistulas than for grafts.  However, a lower threshold may be indicated because the presence of stenosis in a fistula is less predictive of short-term thrombosis than in a graft.  Tonelli et al  reported that 82% of fistulas with ABF  Miranda and Sands,  measured arterial blood flow (ABF) during Doppler ultrasound and suggested that an ABF  recently reported that a flow of  There is also the initial financial outlay to purchase the equipment and the ongoing expense of the technician time involved (10-15mins). ABF should also be subject to variations in mean arterial pressure; studies have reported that both MAP and ABF fall as dialysis session progresses. ,, Therefore, ABF measure ments are best made during the first hour of dialysis. Even then, several investigators have shown variability in the measurement of ABF from one treatment to the next that is felt to reflect the fact that the degree of hemodynamic stability can vary from one treatment to the next, even during the early part of it.  It is best to avoid measuring ABF on a day in which a patient presents either 15mmHg above or below their usual level. The needle placement for grafts, which represent a single channel, should be at least 2 inches apart; the same is not true of fistulas.  The arterial and venous needles can be placed in two different non-communicating branches of the fistula due to the presence of early venous collaterals close to the arterial anastomosis; this can result in failure to measure flow correctly.  For an accurate ABF, the arterial needle must always be in the main trunk of the fistula and it is advisable to record needle placements, as distance from the anastomosis, with each measurement to assure maximal reproducibility.
In addition to measuring access flow, the same equipment can also be used, without line reversal, to measure true access recirculation without the cardiopulmonary component.  The measurement of recirculation has not been found particularly useful in predicting access thrombosis.  However, it can be very useful in arteriovenous fistulas to identify an access that needs intervention in order to improve access flow at higher pump speeds; thus improving dialysis delivery and clearances.  Such diagnostic flow studies could be done initially in all grafts and fistulas and subsequently in those that are not achieving target clearances. The measurement of access flow could also be useful in the detection and clinical decision making in situations where excessive access flow causes complications such as high-output cardiac failure or ischemic steal syndrome.
While ultrasound was the initial method to be approved for measuring access flow, several other potential techniques have been proposed in order to make surveillance less complicated and/or costly. They were based on differential conductivity,  ionic dialysance,  transcutaneous hematocrit sensor,  hematocrit dilution,  and glucose infusion.  Most descriptions of these techniques have involved comparisons with ABF measured by the ultrasound dilution technique and have generally showed good correlations (r values 0.96-0.95) though with a tendency to overestimate ABF by 83-107ml/min with greater coefficients of variation (9.4-10.4% versus 6.4-8.9% for ultrasound dilution). Currently, ultrasound dilution remains the gold standard for measuring access flow. It also provides the opportunity with minor modifications to measure cardiac output and systemic hemodynamics. 
Which Test to Use?
The question of which of these available techniques is optimal in predicting access thrombosis has been the matter of much debate. Bay et al.  found that an ABF of  in a prospective study, compared DVP, Doppler ultrasound and ABF (by ultrasound dilution) as predictors of thrombosis in 220 vascular accesses (both fistulae and grafts). Only decreasing access flow was significantly predictive of access thrombosis in a multivariate analysis and they suggested that the risk increased precipitately if flow was below 600ml/min. In comparisons of ABF with SVP, Besarab et al.  also concluded that access flow was superior, principally because of it's ability to detect intragraft strictures not picked up by the pressure technique.
On the other hand, the utility of access flow has however been challenged, principally by Paulson, ,, who argued that the measurement of access flow lacks sufficient sensitivity and specificity as a surveillance technique in predicting graft thrombosis. He generally preferred ultrasound for stenosis detection and angioplasty referral.
[Table 1] summarize the findings of several prospective studies that have attempted to assess the predictive value of ABF measurements in predicting short-term thrombosis in both grafts [Table 1]A and fistulas [Table 1]B. As can be seen, ABF  Two studies also assessed the sensitivity and specificity of a 20% change in ABF (∆Qa) to predict access thrombosis. Paulson et al  found little evidence that adding the ∆Qa to the ABF  In contrast, our own study (Murray and Singh, unpublished data) suggested that ∆Qa was actually better than ABF  have argued that the clinical occurrence of graft thrombosis is frequent enough for the ABF test to sufficiently alter the post test probability of an event and influence clinical decision making.
As mentioned earlier, the most recent K/DOQI vascular access guidelines  suggest either access blood flow 20% or IAP>0.50 as the "preferred" parameters in the surveillance: the dynamic venous pressures and access recirculation were considered as "acceptable" methods of surveillance.
Graft stenoses, once detected by surveillance monitoring, can be corrected either with balloon angioplasty or by surgical revision. ,,
Prophylactic surgical approaches can vary from revision of the anastomosis, replacement of a damaged segment or may require that the use of the graft be suspended temporarily. Since the presence of stenosis is usually confirmed first by fistulography, an attempt will usually be made to angioplasty any detected stenosis; the finding of more than one stenosis is the rule rather than the exception. If angioplasty is only partially successful in dilating a stenosis, then consideration can be given to placement of a stent. However, any stent placement should not overlap tributary veins or eliminate potential future access sites.  Therefore, they are used only occasionally and usually for central vein occlusions.
The outcomes of both radiological and surgical interventions can also be affected by the technical skills of the operator, which may account for the differences in the reported outcomes. Most published series for both surgical and radiological interventions report procedures performed to salvage thrombosed grafts or pooled studies done for both thrombectomy and prophylactic angioplasty. Turmel-Rodrigues et al  reported in a retrospective analysis a high rate of technical success in prophylactic dilatation of both fistulas and grafts with better long-term patency in the case of fistulas. Similar results have been reported by Beatherd et al.  with secondary patency rates of 66.4% at 6 months and 44.5% at one year. The results were felt to be comparable to those achieved by surgery. Radiological techniques are usually easier to schedule particularly in the setting of emergency thromboses and thus may result in fewer hospitalization days.  Therefore, in the case of prophylactic interventions, the majority of such studies are usually done in the radiological suite with surgery being reserved for those who fail angioplasty or with lesions such as long segment outflow stenoses that do not respond well to angioplasty. With the advent of access flow measurement, it is now possible to assess the functional as well as the anatomical results of angioplasty.
We recently reported our experience with 47 prophylactic angioplasties of PTFE graft stenoses with pre and post angioplasty measurements of access flow.  All procedures were considered technically successful with less than 20% residual stenosis of the most severe lesion. The mean flow increased from 596 ± 41ml/min prior to the procedure to 922 ± 48 ml/min post angiography. Dynamic venous pressures fell from 168 ± 4 to 144 ± 4 mmHg. Thirty seven (74%) grafts experienced a significant increase (>10%) in post angioplasty flow ranging from 12% to 87%. In the remaining 26%, access flow either fell or did not improve significantly despite an apparently "successful" result at the end of the radiological procedure. In the majority, the highest post angioplasty flow was obtained on the first post angioplasty measurement and there was thereafter a tendency for flow and venous pressure to revert back towards baseline over the next three months. Before angioplasty, 34 grafts had an access flow lower than 600ml/ min, whereas after angioplasty only eight continued to exhibit flow rates in that range; the six-month secondary patency was 80%. Nine grafts required re-intervention within six months of the initial angioplasty because of either insufficient increase in flow post angioplasty or a fall in access flow to previous interventional thresholds. To date, thirteen patients have required at least one repeat angioplasty with a mean time to reangioplasty of 160 days.
Similar results have been reported by Schwab et al  who also found that 21% of grafts did not achieve an increase of >20% in ABF despite "successful" angioplasty. In their study, which included mainly grafts but also seven fistulas, the time to re-angioplasty was 5.8 months for grafts and 11.4 months for fistulas.
Finally, van der Linden et al  reported the results of 98 angioplasties for 65 AVG and 33 AVF in 60 patients. Access flow doubled in both grafts and fistulas and the degree of stenosis decreased from 65 ± 3 to 17 ± 2% for grafts and from 72 ± 5 to 23 ± 7% for fistulas. However, the reduction of stenosis did not correlate with flow as noted by Ahya et al.  The post angioplasty flow decreased faster in grafts than fistulas so that six-month unassisted patency rates were higher for fistulas (50%) than grafts (25%). Factors that appear to determine the long-term patency of the arteriovenous grafts after elective angioplasty include the magnitude of residual stenosis,  post angioplasty intragraft to systemic pressure ratio  and post angioplasty access blood flow. 
Recently, it has become possible to measure access blood flow with a modified Swan-Ganz catheter during the angioplasty procedure.  This may prove an extremely useful device to provide the radiologist with immediate functional feedback on the procedure and potentially improve long-term outcomes. None of these studies included a control group of patients whose lesions were not angioplastied or underwent surgical revision instead. Thus, despite the reported excellent secondary patency one cannot definitively conclude that prophylactic angioplasty was beneficial and/ or superior to surgical revision.
Does Access Surveillance and Prophylactic Intervention Prolong Access Survival?
Schwab et al,  who measured the dynamic venous pressure reported a decreased thrombotic rate in patients whose grafts underwent angiography/plasty. Besarab et al.  reported similar findings using the static venous pressures. Despite the evidence of benefit provided by these studies, both were retrospective, non-randomized and used historical controls.
We recently reported that an access surveillance program based on access blood flow monitoring and prompt referral for angioplasty significantly reduced thrombosis and prolonged secondary patency of arteriovenous grafts.  This was a retrospective study comparing graft outcomes at two different dialysis units. Unit A (intervention unit) monitored access flows monthly in all patients with AV grafts and prophylactic angiography was recommended to the patient's nephrologist if 1) ABF fell below 600ml/min or 2) ABF fell to  were the first to publish a randomized study of prophylactic angipoplasty. They compared patients with PTFE grafts who were randomized either to an interventional group who had quarterly ultrasound with angioplasty in case where a stenosis of greater than 50% was found or to a control group who received no intervention. They found no prolongation of graft survival in the monitored patients. There are several possible explanations for this finding. Firstly, the numbers involved were small and included (as do most studies) both newly placed grafts and those with a history of previous interventions. Secondly, it may take longer than 12 months to see a difference in graft survival as many grafts that thrombose can be salvaged. This group subsequently reexamined their data, confining the analysis only to patients with new (rather than pre-existing grafts), and were able to find evidence of benefit in this subset. 
Sands et al.  compared thrombotic rates (but not graft survival) in patients with PTFE grafts (n=35) randomized to either access flow monitoring or venous pressure monitoring with again, referral for angioplasty, if certain thresholds were met; a third control group received no surveillance. They reported significantly lower thrombotic rates in the grafts that received monthly access monitoring and found that ABF monitoring tended to result in lower thrombotic rates than VP monitoring.
Smits et al  published the results of two small randomized studies in which they compared outcomes of grafts monitored by weekly static/ dynamic VP or monthly ABF (group A), or VP and the combination of VP and ABF (group B). The numbers were relatively small (group A n=53, group B n=68) and the followup was short (average of about 6 months). They failed to find any superiority of ABF over VP or of adding ABF to VP, but again the study was underpowered and too short to be considered a true test of this hypothesis.
Most recently, Ram et al.  published a prospective randomized controlled trial in which patients with PTFE grafts received monthly measurements of ABF and quarterly Doppler assessments for graft stenosis. They were then randomized to one of three groups. The control group was referred for angioplasty based on clinical criteria only. The flow group was referred either for clinical criteria or if ABF was 50% was found on Doppler. The rate of preemptive angioplasty was higher in the stenosis group (0.65/patient/ year), than in the flow (0.34) and control group (0.22). No difference in graft survival was seen. Potential criticisms of the study include again the small numbers in each arm, and the inclusion of grafts with previous interventions. Probably of more significance, however, is the failure to include a fall in ABF as a criterion for angioplasty referral in the flow group. Interestingly the angioplasty rate in the flow group (0.34 per patient/year) was only slightly higher than that (0.22 per patient/ year) in the control group and much lower than the 1.47 per patient/year that we reported in our cohort in which angioplasty was based either on an ABF of  There is evidence that angioplasty is more successful and its effects more durable when performed at a higher ABF. ,,
In the case of arteriovenous fistulas, Sands et al  reported that arteriovenous fistulas monitored on a monthly basis with either static venous pressures or access blood flow with angioplasty referral (IAP>0.50, ABF  performed screening fistulography on 141 patients with arteriovenous fistulas and found 62 with significant stenoses (>50% of luminal diameter). These patients were then randomized to either angioplasty or no intervention. The end point of the study was either fistula thrombosis or surgical revision for inadequate dialysis. They found a four-fold increase in median survival and a 2.87-fold decrease in risk of failure in the fistulas that were angioplastied. These results strongly support the advisability of a program for regular monitoring of fistulas for stenosis and prophylactic angioplasty.
Conclusions and Recommendations
As we have seen there is no definitive, but a considerable amount of suggestive evidence that access surveillance, followed by prophylactic angioplasty and/or surgical revision can prolong the working life of an arteriovenous graft. It can also potentially assist in optimizing the performance of arteriovenous fistulas and possibly also prolonging their survival. The optimal technique for monitoring appears to be the regular measurement of access blood flow,  complemented where indicated by measurements of access recirculation. The optimal interval for such measurements is unclear, but they should probably be done at least monthly for grafts and quarterly for fistulas with additional measurements done for specific clinical problems. The lesser interval for fistulas reflects the fact that they are less likely to clot, that access flows deteriorate less rapidly in fistulas and that they appear to clot at lower blood flows than grafts.
In our unit we use both DVP (done at each dialysis) and ABF monitoring. Our preferred approach is to perform 3 initial ABF measurements in quick succession on each new access to establish a baseline flow for that particular access and then repeat ABFs monthly (grafts) or quarterly (fistulas). Additional flow measurements may be made for clinical indications such as rising DVP, back bleeding, swollen arms or poor urea clearances. In the case of grafts, if an ABF 20%, the measurement is repeated and if confirmed a fistulogram is performed and any stenoses found are dilated.
The object of angioplasty in a graft should be to achieve an access flow as close to or greater than 1000ml/min, if possible, as the higher the flow achieved the better the longterm patency.  In the case of fistulas, we usually reserve angioplasty for fistulas with ABF confirmed at less than 400ml/min. The target should be to achieve a flow of at least 600ml/min. This approach may prolong the working life of the individual access and optimize the delivery of dialysis. , There is also evidence that it results in overall reduction of access morbidity and costs through reductions of hospitalizations, missed treatments and catheter costs. 
|1||Hakim R, Himmelfarb J. Hemodialysis access failure: a call to action. Kidney Int 1998;54:1029-44.|
|2||Woods JD, Turenne MN, Strawderman RL, Young EW, Hirth RA, Port FK, Held PJ. Vascular access survival among incident hemodialysis patients in the United States. Am J Kidney Dis 1997;30:50-7.|
|3||Swedberg SH, Brown BG, Sigley R, Wright TN, Gordin D, Nicholls SC. Intimal fibromuscular hyperplasia at the venous anastomosis of PTFE grafts in hemodialysis patients. Clinical Immunocytochemical, light and electron microscopic assessment. Circulation 1989;80:1726-36.|
|4||Etheredge EE, Haid SD, Maeser MN, Sicard GA, Anderson CB. Salvage operations for malfunctioning polytetrafluoroethylene hemodialysis access grafts. Surgery 1983;94:464-70.|
|5||Sivanesan S, How TV, Bakran A. Sites of stenosis in AV fistulae for haemodialysis access. Neph Dial Transpl 1999;14:118-20.|
|6||Beathard GA. Percutaneous angioplasty for the treatment of venous stenosis: a nephrologist's perspective. Semin Dial 1995;8: 166-70.|
|7||NKF-DOQI. Clinical Practice Guidelines for Vascular Access. Am J Kidney Dis 1997;30(Suppl 3):S150-91.|
|8||NKF-DOQI. Clinical Practice Guidelines for Vascular Access: Update 2000. Am J Kidney Dis 2001;37(Suppl 1):S137-81.|
|9||Arbabzadeh M, Mepani B, Murray BM. Why do grafts clot despite access blood flow surveillance. Cardiovasc Intervent Radial 2002;25:501-5.|
|10||Gadallah MF, Paulson WD, Vickers B, Work J. Accuracy of doppler ultrasound in diagnosing anatomic stenosis of hemodialysis arteriovenous access as compared with fistulography. Am J Kidney Dis 1998;32:273-7.|
|11||Murray BM, Rajczak S, Ali B, Herman A, Mepani B. Assessment of access blood flow after preemptive angioplasty. Am J Kidney Dis 2001;37:1029-38.|
|12||Paulson WD, Ram SJ, Birk CG, Work J. Does blood flow accurately predict thrombosis or failure of hemodialysis synthetic grafts? A meta-analysis. Am J Kidney Dis 1999;34:478-85.|
|13||Thomsen MB, Stenport G. Evaluation of clinical examination preceding surgical treatment of AV-fistula problems. Is angiography necessary?. Acta Chir Scand 1985; 151:133-7.|
|14||Trerotola SO, Scheel PJ Jr, Powe NR, et al. Screening for dialysis access graft malfunction: comparison of physical examination with US. J Vasc Interv Radiol 1996;7:1520.|
|15||Robbin ML, Oser RF, Allon M, et al. Hemodialysis access graft stenosis: US detection. Radiology 1998;208(3):655-61.|
|16||Sands J, Glidden D, Miranda C. Hemodialysis access flow measurement. Comparison of ultrasound dilution and duplex ultrasonography. ASAIO J 1996;42: M899-901,.|
|17||Lin SL, Chen HS, Huang CH, Yen TS. Predicting the outcome of hemodialysis arteriovenous fistulae using duplex ultrasonography. J Formos Med Assoc 1997; 96:864-8.|
|18||Older RA, Gizienski TA, Wilkowski MJ, Angle JF, Cote DA. Hemodialysis access stenosis: early detection with color Doppler US. Radiology 1998;207:161-4.|
|19||Dousset V, Grenier N, Douws C, Senuita P, Sassouste G, Ada L, Potaux L. Hemodialysis grafts: color Doppler flow imaging correlated with digital subtraction angiography and functional status. Radiology 1991;181:89-94.|
|20||Tordoir JH, de Bruin HG, Hoeneveld H, Eikelboom BC, Kitslaar PJ. Duplex ultrasound scanning in the assessment of arteriovenous fistulas created for hemodialysis access: comparison with digital subtraction angiography. J Vasc Surg 1989;10(2):122-8.|
|21||Schwab SJ, Raymond JR, Saeed M, Newman GE, Dennis PA , Bollinger RR. Prevention of hemodialysis fistula thrombosis. Early detection of venous stenoses. Kidney Int 1989;36:707-11.|
|22||Sands J, Young S, Miranda C. The effect of Doppler flow screening studies and elective revisions on dialysis access failure. ASAIO J 1992;38:M524-7.|
|23||Strauch BS, O'Connell RS, Geoly KL, Grundlehner M, Yakub YN, Tietjen DP. Forcasting thrombosis of vascular access with Doppler color flow imaging. Am J Kidney Dis 1992;19:554-7 .|
|24||Sullivan KL, Besarab A, Bonn J, Shapiro MJ, Gardiner GA Jr, Moritz MJ. Hemodynamics of failing dialysis grafts. [erratum appears in Radiology 1993;188(2):586]. Radiology 1993;186:867-72.|
|25||Besarab A, Sullivan KL, Ross RP, Moritz MJ. Utility of intraaccess pressure monitoring in detecting and correcting venous outlet stenosis prior to thrombosis. Kidney Int 1995;47:1364-73.|
|26||Choudhury D, Lee J, Elivera HS, Ball D, Roberts AB, Ahmed Z. Correlation of venography, venous pressure, and hemoaccess function. Am J Kidney Dis 1995;25:269-75.|
|27||Besarab A, Moritz M, Sullivan K, Dorrell S, Price JJ. Venous access pressures and the detection of intra-access stenosis. ASAIO J 1992;38(3):M519-23.|
|28||Besarab A, Lubkowski T, Frinak S. A simple method for measuring intra-access pressure. J Am Soc Nephrol. 1999;11:202A.|
|29||Besarab A, Lubkowski T, Frinak S, Ramanathan S, Escobar F. Detecting vascular access dysfunction. ASAIO J 1997; 43(5):M539-43.|
|30||Besarab A, Lubkowski T, Frinak S, Ramanathan S, Escobar F. Detection of access strictures and outlet stenoses in vascular accesses. Which test is best? ASAIO J 1997;43(5):M543-7.|
|31||Frinak S, Zasuwa G, Dunfee T, Besarab A, Yee J. Dynamic venous access pressure ratio test for hemodialysis access monitoring. Am J Kidney Dis 2002;40: 760-8.|
|32||Paulson WD. Blood flow surveillance of hemodialysis grafts and the dysfunction hypothesis. Semin Dial 2001;14:175-80.|
|33||Krivitski NM. Theory and validation of access flow measurement by dilution technique during hemodialysis. Kidney Int 1995;48:244-50.|
|34||Lindsay RM, Blake PG, Malek P, Posen G, Martin B, Bradfield E. Hemodialysis access blood flow rates can be measured by a differential conductivity technique and are predictive of access clotting. Am J Kidney Dis 1997;30:475-82.|
|35||May RE, Himmelfarb J, Yenicesu M, et al. Predictive measures of vascular access thrombosis: a prospective study. Kidney Int 1997;52:1656-62.|
|36||Wang E, Schneditz D, Nepomuceno C, Lavarias V, Martin K, Morris AT, Levin NW. Predictive value of access blood flow in detecting access thrombosis. ASAIO J 1998;44:M555-8.|
|37||Bosman PJ, Boereboom FTJ, Eikelboom BC, Kooman HA , Blankestijn PJ. Graft flow as a predictor of thrombosis in hemodialysis grafts. Kidney Int 1998;54:1726-30.|
|38||Neyra NR, Ikizler T A, May RE, et al. Change in access blood flow over time predicts vascular access thrombosis. Kidney Int 1998;54:1714-9.|
|39||Tonelli M, Jindal K, Hirsch D, Taylor S, Kane C, Henbrey S. Screening for subclinical stenosis in native arteriovenous fistulae. J Am Soc Nephrol 2001;12:1729-33.|
|40||Tonelli M, Jhangri GS, Hirsch DJ, Marryatt J, Mossop P, Wile C, Jindal KK. Best threshold for diagnosis of stenosis or thrombosis within six months of access flow measurement in arteriovenous fistulae. J Am Soc Nephrol 2003;14:3264-69.|
|41||Miranda CL, Sands JJ. Flow volumes as a predictor of haemodialysis access failure. J Vasc Technol 1998;22:73-6.|
|42||Nessitore N, Bedogna V, Gammaro L, et al. Diagnostic accuracy of ultrasound dilution access blood flow measurement in detecting stenosis and predicting thrombosis in native forearm arteriovenous fistulae for hemodialysis. Am J Kidney Dis 2003;42: 331-41.|
|43||Rehman SU, Pupim LB, Shyr Y, Hakim R, Ikizler TA. Intradialytic serial vascular access flow measurements. Am J Kidney Dis 1999;34:471-7.|
|44||De Soto DJ, Ram SJ, Faiyaz R, Birk CG, Paulson WD. Hemodynamic reproducibility during blood flow measurements of hemodialysis synthetic grafts. Am J Kidney Dis 2001;37:790-6.|
|45||Schneditz D, Fan Z, Kaufman A, Levin NW. Stability of access resistance during hemodialysis. Nephrol Dial Transpl 1998;13:739-44.|
|46||Lindsay RM , Bradfield E, Rothera C, Kianfar C, Malek P, Blake PG. A comparison of methods for the measurement of hemodialysis access recirculation and access blood flow rate. ASAIO J 1998;44:62-7.|
|47||Coyne DW, Delmez J, Spence G, Windus DW. Impaired delivery of hemodialysis prescriptions: an analysis of causes and an approach to evaluation. J Am Soc Nephrol 1997;8:1315-8.|
|48||Mercadal L, Hamani A, Bene B, Petitclerc T. Determination of access blood flow from ionic dialysance: theory and validation. Kidney Int 1999;56:1560-5.|
|49||Steuer RR, Miller DR, Zhang S, Bell DA, Leypoldt JK. Noninvasive transcutaneous determination of access blood flow rate. Kidney Int 2001;60:284-91.|
|50||Yarar D, Cheung AK, Sakiewicz P, et al. Ultrafiltration method for measuring vascular access flow rates during hemodialysis. Kidney Int 1999;56:1129-35.|
|51||Ram SJ, Magnasco A, Jones SA, et al. In vivo validation of glucose pump test for measurement of hemodialysis access flow. Am J Kidney Dis 2003;42:752-60.|
|52||Krivitski NM, Depner TA. Cardiac output and central blood volume during hemodialysis: methodology. Adv in Ren Replace Ther 1999;6:225-32.|
|53||Bay WH, Henry ML, Lazarus JM, Lew NL, Ling J, Lowrie EG. Predicting hemodialysis access failure with color flow doppler ultrasound. Am J Nephrol 1998;18:296-304.|
|54||Paulson WD, Ram SJ, Birk CG, Zapczynski M, Martin SR, Work J. Accuracy of decrease in blood flow in predicting hemodialysis graft thrombosis. [see comment]. Am J Kidney Dis 2000;35:1089-95.|
|55||Krivitski NM, Gantela S. Access flow measurement as a predictor of hemodialysis graft thrombosis: making clinical decisions. Semin Dial 2001;14:181-5.|
|56||Kumpe DA, Cohen MA. Angioplasty/ thrombolytic treatment of failing and failed hemodialysis access sites: comparison with surgical treatment. Prog Cardiovasc Dis 1992;34:263-78.|
|57||Brooks JL, Sigley RD, May KJ Jr, Mack RM. Transluminal angioplasty versus surgical repair for stenosis of hemodialysis grafts. A randomized study. Am J Surg 1987;153:530-1.|
|58||Turmel-Rodrigues L, Pengloan J, Bourquelot P. Interventional radiology in hemodialysis fistulae and grafts: a multidisciplinary approach. Cardiovasc Intervent Radial 2002;25;3-16.|
|59||Kumpe DA. Regarding "Prospective randomized comparison of surgical versus endovascular management of thrombosed dialysis access grafts" [Letter] J Vasc Surg 1998; 28:386-7.|
|60||Schwab SJ, Oliver MJ, Suhocki P, McCann R. Hemodialysis arteriovenous access: detection of stenosis and response to treatment by vascular access blood flow. Kidney Int 2001;59:358-62.|
|61||van der Linden J, Smits JH, Assink JH, et al. Short- and long-term functional effects of percutaneous transluminal angioplasty in hemodialysis vascular access. J Am Soc Nephrol 2002;13:715-20.|
|62||Ahya SN, Windus DW, Vesely TM, Flow in hemodialysis grafts after angioplasty: do radiologic criteria predict success? Kidney Int 2001;59:1974-8.|
|63||Lilly RZ, Carlton D, Barker J, et al. Predictors of arteriovenous graft patency after radiologic intervention in hemodialysis patients. Am J Kidney Dis 2001;37:945-53.|
|64||Murray BM. Does postangioplasty access blood flow (ABF) predict outcomes in arteriovenous grafts? J Am Soc Nephrol 2003;14:723A.|
|65||Vesely TM. Gherardini D. Gleed RD. Kislukhin V. Krivitski NM. Use of a catheterbased system to measure blood flow in hemodialysis grafts during angioplasty procedures. Journal of Vascular & Interventional Radiology. 13:371-8, 2002.|
|66||MurrayBM, Suri S,Herman A, Rajczak S, Paulson WD, Pell M. Prolongation of arteriovenous graft survival by access flow monitoring. In Henry ML (ed). Vascular Access for hemodialysis VIII, pgs 279-292, Access Medical Press, Arlington Heights, Ill, 2002.|
|67||Lumsden AB, MacDonald MJ, Kikeri D, Cotsonis GA, Harker LA, Martin LG. Prophylactic balloon angioplasty fails to prolong the patency of expanded polytetrafluoroethylene arteriovenous grafts: results of a prospective randomized study J Vasc Surg 1997;26;382-92.|
|68||Martin LG, MacDonald MJ, Kirkeri D, Costonis GA, Harker LA and Lumsden AB. Prophylactic angioplasty reduces thrombosis in virgin ePFTE arteriovenous dialysis grafts with greater than 50% stenosis: subset analysis of a prospectively randomized study. J Vasc Interv Radiol 1999;10(4):389-96.|
|69||Sands JJ, Jaybac PA, Miranda CL, Kapsick BJ. Intervention based on monthly monitoring decreases hemodialysis access thrombosis. ASAIO J 1999;45:147-50.|
|70||Smits JH, van der Linden J, Hagen EC, et al. Graft surveillance: venous pressure, access flow, or the combination? Kidney Int 2001;59:1551-8.|
|71||Ram S, Work J, Caldito GC, Eason JM, Pervez A, Paulson WD. A randomized controlled trial of blood flow and stenosis surveillance of hemodialysis grafts. Kidney Int 2003;64:272-80.|
|72||Tessitore N, Mansueto G, Bedogna V, et al. Prospective controlled trial on effect of percutaneous transluminal angioplasty on functioning arteriovenous fistulae survival. J Am Soc Nephrol 2003;14:1623-7.|
|73||Garland JS, Moist LM, Lindsay RM. Are hemodialysis access flow measurements by ultrasound dilution the standard of care for access surveillance?. Adv Ren Replace Ther 2002;9:91-8.|
|74||Windus DW, Audrain J, Vanderson R, Jendrisak MD, Picus D, Delmez JA. Optimization of high-efficiency hemodialysis by detection and correction of fistula dysfunction. Kidney Int 1990;38:337-41.|
|75||McCarley P, Wingard RL, Shyr Y, Pettus W, Hakim RM, Ikizler TA. Vascular access blood flow monitoring reduces access morbidity and costs. Kidney Int 2001;60:1164-72.|