| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ultrasonography |
1 From the Department of Radiology (M.L.R., M.E.L.), Division of Transplant Surgery (M.H.G., C.J.Y., M.H.D.), and Division of Nephrology (M.A.), University of Alabama at Birmingham, 619 19th St S, JTN350, Birmingham, AL 35249-6830; and Nephrology Associates, Chattanooga, Tenn (N.E.C.). From the 2001 RSNA scientific assembly. Received August 13, 2001; revision requested October 9; final revision received May 6, 2002; accepted May 14. Address correspondence to M.L.R. (e-mail: mrobbin@uabmc.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: US fistula evaluation results were analyzed retrospectively in 69 patients within 4 months after AVF placement; adequacy for dialysis was known in 54. Measurements included minimum venous diameter and blood flow rate. Experienced dialysis nurses examined 30 fistulas clinically. Predictors of fistula adequacy were analyzed with univariate and multivariate logistic regression. Mean fistula diameters and blood flow rates were compared by using analysis of variance or unpaired Student t tests.
RESULTS: Fistula adequacy for dialysis doubled if the minimum venous diameter was 0.4 cm or greater (89% [24 of 27]) versus less than 0.4 cm (44% [12 of 27]; P < .001). Fistula adequacy for dialysis was nearly doubled if flow volume was 500 mL/min or greater (84% [26 of 31]) versus less than 500 mL/min (43% [nine of 21]; P = .002). Combining venous diameter and flow volume increased fistula adequacy predictive value: minimum venous diameter of 0.4 cm or greater and flow volume of 500 mL/min or greater (95% [19 of 20]) versus neither criterion met (33% [five of 15]; P = .002). Women were less likely to have an adequate fistula diameter of 0.4 cm or greater: 40% (12 of 30) of women versus 69% (27 of 39; P = .015) of men. No significant differences in blood flow or minimum venous diameter were found during 2–4 postoperative months. Experienced dialysis nurses’ accuracy in predicting eventual fistula maturity was 80% (24 of 30).
CONCLUSION: US measurements of AVF at 2–4 months in patients undergoing hemodialysis are highly predictive of fistula maturation and adequacy for dialysis.
© RSNA, 2002
Index terms: Dialysis, shunts, 91.457 • Fistula, arteriovenous, 91.12984, 91.12986, 91.457, 91.7173 • Ultrasound (US), Doppler studies, 91.12984
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Many fistulas (28%–53%) never mature adequately to be usable for dialysis (2–5). In current practice, dialysis nurses and nephrologists evaluate a fistula clinically to determine whether it has matured sufficiently to use for hemodialysis. Experienced examiners can accurately determine fistula maturation. However, it would be advantageous to develop objective quantitative criteria to evaluate the suitability of fistulas for dialysis, as greater numbers of fistulas are placed in response to the DOQI guidelines. Well-defined criteria applied early after fistula placement to help identify fistulas likely to fail would be extremely useful.
Ultrasonography (US) is an excellent modality for hemodialysis access evaluation, as it is readily available, noninvasive, and inexpensive. It avoids the risks associated with iodinated contrast material and ionizing radiation. US is a sensitive modality for the evaluation of hemodialysis synthetic graft stenosis (6). It can help identify other complications associated with grafts and fistulas, including hematoma, pseudoaneurysm, and abscess (7). However, there has been little published regarding the US evaluation of hemodialysis AVF maturity.
To our knowledge, the qualitative clinical evaluation of AVF maturation has not been compared with a quantitative US evaluation. The purpose of this study was to compare various objective US criteria for AVF maturation with subsequent fistula outcomes and clinical evaluation by experienced dialysis nurses.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Patients
From October 1998 through August 2000, 122 US evaluations of hemodialysis fistulas were performed. For the purpose of our retrospective analysis, we included only those US examinations that were performed within 4 months after fistula placement (n = 95). Four months is the maximum interval in our practice in which an AVF is allowed to mature without evaluation of potential problems. We excluded US performed for any indication other than assessment of fistula maturation, such as fistula mass or hematoma evaluation (n = 22).
Of the remaining 73 postoperative US fistula results, measurements of the minimum venous diameter were available in 69 cases, which formed our study group, and blood flow rate was available in 67. In the 69 patients (39 men, 30 women; age range, 22–81 years; mean age, 54 years ± 13) with fistula maturation evaluations with measurements, 21 upper arm and 48 forearm fistulas were placed. Similar to the dialysis population in our study, there was a predominance of nonwhite patients (45 black, 24 white). Of the population in our study, 61% (42 of 69) were diabetic. All patients underwent vascular mapping prior to AVF placement with our established protocol and criteria (8). The clinical outcome (adequacy for dialysis) was known for 54 of 69 fistulas. Fifteen cases were indeterminate because the patient had not yet undergone dialysis.
Overview of US Technique
US examination of the AVF was performed with the patient in an upright seated position. The arm was positioned at 45° from the body and comfortably supported by towels on a mobile instrument stand. An initial overview of the AVF feeding artery, draining vein, and anastomotic region was obtained. Gray-scale examination of the vessel diameter was performed in both the transverse and longitudinal planes. The diameter of the draining vein was measured in the anteroposterior dimension in the transverse plane by using gentle transducer pressure. Draining venous diameter was measured at the caudal, middle, and cranial portion of the draining vein in the forearm if a forearm fistula or in the upper arm if an upper arm fistula.
These standard measurements gave an overall assessment of the AVF draining venous diameter. However, there may be focal sclerotic or narrowed portions of the vein that may be flow limiting anywhere in the draining vein. Therefore, if the smallest visual draining venous diameter was at a location other than that at which the standard measurements were obtained, its location was recorded.
Three to five blood flow measurements were obtained in the draining vein of the AVF. To decrease the possibility of encountering turbulent flow, these measurements were obtained in the middle of the forearm for a forearm AVF or the middle of the upper arm for an upper arm AVF, locations where stenoses or extremely curvy draining veins are rarely seen. Blood flow was evaluated in a straight nontapering venous segment. Because blood flow rate acquisition was made with the assumption that laminar flow is present, areas with turbulent (not laminar) flow, such as areas with stenosis, were avoided. If an area of turbulence was noted in the usual acquisition site, volume of flow was measured as far as reasonably possible away from the site, but still in the forearm for a forearm AVF and the upper arm for an upper arm AVF.
The Doppler range gate was set to encompass the entire lumen, and the diameter was measured perpendicular to the vessel walls. The angle of flow relative to the graft wall was held constant at 60° to limit one potential aspect of measurement variability. The pulse repetition frequency and gain were individually optimized to avoid aliasing. Three to four cardiac cycles were used for automatic machine determination of the time-averaged velocity across the venous lumen. The blood flow volume in milliliters per minute was calculated with the scanner by using the measured venous diameter and time-averaged velocity measurements. The blood flow measurements were averaged.
All AVF US examinations were performed with one of three units (ATL HDI 3000, Bothell, Wash; or Acuson Sequoia or XP, Mountain View, Calif), and all volume flow measurements were obtained with the first unit. After completion of US, results were recorded on a work sheet by the sonographer and/or sonologist. Experienced sonographers performed the examinations in a laboratory accredited by the American College of Radiology.
Qualitative Clinical Evaluation
The AVFs (n = 30) were clinically evaluated by one of two registered hemodialysis nurses, each with more than 5 years of experience in dialysis access techniques. The dialysis nurses were contacted immediately after US of the fistula was performed. If they were able to come to the US section, they subjectively evaluated the maturity of the fistula and the usable length of the draining vein for dialysis. This directed physical examination of the AVF was based on clinical experience. Subjective criteria included an easily palpable superficial vein of adequate diameter and a uniform thrill to palpation. The accessible draining vein needed to be relatively straight and more than 10 cm long. On the basis of these criteria, the nurses were asked to predict whether the fistula would eventually mature, and they recorded their evaluations on a work sheet. They were blinded to the US results.
The results of each study were maintained in a database for subsequent statistical analysis. The clinical definition of a successful fistula was one that supported a blood flow rate of 350 mL/min for at least six dialysis sessions in 1 month (3).
Statistical Analysis
The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of different US cutoff values of minimum venous diameter and blood flow rate with respect to eventual fistula adequacy for dialysis were calculated. Curves were generated to predict clinical adequacy of fistulas for dialysis, and they described sensitivity, specificity, and accuracy of the minimum venous diameter and blood flow rate. The cutoff values yielding the highest accuracy were used to calculate the positive and negative predictive values. A receiver operating characteristic (ROC) curve was generated for minimum venous diameter and minimum blood flow rate. The likelihood that a new fistula would have a minimum venous diameter of 0.4 cm or greater was calculated for diabetes status, sex, race, age, and fistula site (forearm vs upper arm). Predictors of fistula adequacy were analyzed by means of univariate and multivariate logistic regression. Mean fistula diameters and blood flow rates were compared by means of analysis of variance or unpaired Student t tests. Accuracy of prediction of eventual fistula maturity by the experienced dialysis nurses was determined.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A substantial proportion (28%–53%) of AVFs never mature adequately to be usable for dialysis (2–5). Whereas a graft can be used for dialysis 2 weeks after placement, the typical AVF is not used until 3 months after placement (4). An AVF is typically not evaluated for maturity until 2–3 months after placement, and nephrologists often wait up to 6 months before declaring the fistula a failure (3,11). Additional fistula evaluation with venography, or placement of a new access, is then considered.
Accordingly, the time spent waiting for the AVF to mature can be substantial. Patients typically spend this time undergoing dialysis three times a week by means of a catheter. These patients have a potentially decreased ability to undergo dialysis effectively because of suboptimal dialysis blood flow. If the AVF never matures, this is time wasted. In addition, there is increased risk of catheter infection—as many as two episodes per year—with the attendant morbidity and cost (12). Thus, the ability to predict whether an AVF is going to mature eventually is important. Early recognition that the fistula is unlikely to mature can lead to fistula evaluation and possible revision or placement of a new vascular access (13).
AVF maturation is usually assessed subjectively by means of physical examination. A mature fistula will typically have an easily palpable superficial vein of adequate diameter that facilitates easy cannulation. It will have a uniform thrill to auscultation and palpation that indicates adequate blood flow without stenosis. The accessible draining vein needs to be more than 10 cm long to allow for rotation of needle sites and adequate distance between the cannulating needles. If the draining vein is too tortuous or too deeply located in the subcutaneous tissues, it will be difficult to cannulate.
Minimum Venous Diameter
Although subjective assessment of the AVF for maturation as detailed earlier is described (14), we found little in the literature about the objective evaluation of AVF maturity. Wong et al (15) found a mean diameter of 0.58 cm ± 0.12 in the successful radiocephalic fistula at 6 weeks. However, little detail regarding diameters of unsuccessful fistulas was given. Therefore, we sought to determine whether the minimum diameter of the outflow vein was predictive of eventual fistula maturity. An adequate fistula diameter is important, as a 15-gauge hemodialysis needle is placed into the vein by means of palpation and visual inspection. Our initial hypothesis based on clinical experience was that a minimum venous size of 0.4 cm is needed for easy cannulation.
The draining vein needs to be of a certain size not only for ease in needle placement but also for carrying adequate blood flow. The optimal criterion for minimum venous diameter is a trade-off between sensitivity and specificity. It is important to correctly identify fistulas that are likely to mature, as well as those that will not. The overall accuracy of this criterion is optimal (72%–74%) when the minimum venous diameter cutoff is 0.3–0.4 cm.
Women are less likely to achieve a minimum venous diameter of 0.4 cm, a finding consistent with prior observations of inferior fistula outcomes among women (10,16). This may be in part due to smaller vessels in women (17). In contrast, patient race, age, diabetes status, and fistula location did not correlate with maturity. No differences in blood flow rate in mature fistulas were found between the various patient subsets.
Fistula Blood Flow Rate
The ability to maintain adequate blood flow during hemodialysis is another crucial determining factor in AVF maturity. In the United States, hemodialysis is typically performed at a dialysis blood flow rate of 350–450 mL/min, for 3.5–4 hours three times per week. A fistula blood flow rate less than 350 mL/min cannot sustain the desired dialysis blood flow rate and therefore results in inadequate dialysis. Blood flow rate has been measured to be at least 350–500 mL/min in normally functioning AVFs (18). Lin et al (19) found a mean blood flow rate greater than 634 mL/min in the 2nd postoperative week in 152 patients with successful radiocephalic fistulas. Mean blood flow rate in successful AVFs was reported by Wong et al (15) to be approximately 650 mL/min at 12 weeks.
Average blood flow rate in adequately functioning AVFs in our study was 780 mL/min ± 401, which is similar to those found by Lin et al (19) and Wong et al (15). The overall accuracy of the blood flow criteria is optimal (75%–76%) for values of 400–500 mL/min. The blood flow rate through the fistula must exceed the minimum acceptable dialysis blood flow rate of 350 mL/min by at least 100 mL/min to ensure successful use of the fistula for dialysis, or the vein could collapse during hemodialysis. Factors such as venous diameter (Fig 3) and depth from the skin also determine whether a fistula is successful. For this reason, not every fistula that meets the blood flow criteria is usable for dialysis.
Single versus Multiple Assessments
A potential problem with a single measurement of fistula diameter and blood flow rate is that both the diameter and the flow may increase with time. Thus, more than one measurement may be needed to accurately predict maturation potential. However, in the group of 38 patients in the study by Wong et al (15), individual fistula blood flow rate did not increase substantially in 2–12 weeks. Although we did not obtain multiple measurements in the same fistulas over time, we found no significant difference in the flow volumes of AVFs measured at 2, 3, and 4 months.
Thus, an initial US assessment at 2–3 months after surgery should help predict subsequent fistula maturity with a high degree of accuracy. Angiographic assessment of selected fistulas with borderline measurements may be useful in identifying those with anastomotic or draining venous stenosis. Multiple draining venous branches close to the anastomosis may be surgically ligated to increase the chance of maturation (13). If little chance for AVF maturation is present on the basis of the developed criteria, an alternate permanent access can be surgically placed in a timely fashion.
Experienced dialysis nurses were relatively accurate (80% [24 of 30]) in their subjective prediction of AVF maturation. Clinically, there are some fistulas that are obviously mature. The real problem in clinical evaluation is in predicting the ultimate outcome of AVFs that are not clearly mature. If the fistula appears obviously adequate for dialysis at 2 months, further evaluation is unlikely to be needed. However, if the fistula does not appear mature, or is questionable, US evaluation for adequacy should be performed. Objective US assessment is also useful when an experienced examiner is not available, which is a common experience in many hemodialysis units.
Limitations of this study include a modest referral bias. Although all patients with AVF were supposed to be referred for postoperative US evaluation, some patients were not scheduled for US or did not keep their appointment. Although roughly equivalent numbers of forearm and upper arm AVFs were placed, there were more patients with forearm fistulas in this study. This situation demonstrates a bias toward sending patients with AVFs that were difficult to evaluate—patients in whom fistula maturation was questionable—to US.
Another limitation of this study was the potential for marked variability in flow volume measurement both in the same patient and between sonographers. Although every attempt was made to control all possible sources of variability, interobserver variability was not specifically studied in this patient population. Careful attention to technique is crucial to achieving reproducible fistula blood flow measurements and has a definite learning curve. The issue of interobserver variability is currently being actively investigated at our institution.
In summary, a single US fistula evaluation of minimum venous diameter and blood flow rate at 2–4 months can be used to accurately assess the likelihood of AVF maturation. If the AVF appears unequivocally mature at physical examination at 2 months, US may be superfluous. However, if there is any doubt about fistula maturation, the fistula should be evaluated at US. This early evaluation allows the patient with US evidence of stenosis or multiple draining venous branches to undergo early surgical or radiologic treatment. If US findings indicate the fistula is unlikely to ever mature, a new permanent access can be placed expeditiously.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, M.L.R.; study concepts, M.L.R., M.A., C.J.Y., M.H.G., M.H.D.; study design, M.L.R., M.A.; literature research, M.L.R., M.E.L., M.A.; clinical studies, M.L.R., M.E.L.; data acquisition, M.L.R., N.E.C., M.A.; data analysis/interpretation, M.L.R., M.A., M.E.L., N.E.C.; statistical analysis, M.A., N.E.C.; manuscript preparation and definition of intellectual content, M.L.R., M.E.L., M.A.; manuscript editing, M.L.R.; manuscript revision/review and final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Besarab, M. Allon, and M. L. Robbin Resolved: Fistulas Are Preferred to Grafts as Initial Vascular Access for Dialysis J. Am. Soc. Nephrol., September 1, 2008; 19(9): 1629 - 1633. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ferring, J. Henderson, A. Wilmink, and S. Smith Vascular ultrasound for the pre-operative evaluation prior to arteriovenous fistula formation for haemodialysis: review of the evidence Nephrol. Dial. Transplant., June 1, 2008; 23(6): 1809 - 1815. [Full Text] [PDF]
|
|
|
|
 |
|
 W. J. Peterson, J. Barker, and M. Allon Disparities in Fistula Maturation Persist Despite Preoperative Vascular Mapping Clin. J. Am. Soc. Nephrol., March 1, 2008; 3(2): 437 - 441. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Singh, M. L. Robbin, M. E. Lockhart, and M. Allon Clinically Immature Arteriovenous Hemodialysis Fistulas: Effect of US on Salvage Radiology, December 1, 2007; 246(1): 299 - 305. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. E. Lok Fistula First Initiative: Advantages and Pitfalls Clin. J. Am. Soc. Nephrol., September 1, 2007; 2(5): 1043 - 1053. [Full Text] [PDF]
|
|
|
|
|
|
M. Allon Current Management of Vascular Access Clin. J. Am. Soc. Nephrol., July 1, 2007; 2(4): 786 - 800. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Lee, J. Barker, and M. Allon Comparison of Survival of Upper Arm Arteriovenous Fistulas and Grafts after Failed Forearm Fistula J. Am. Soc. Nephrol., June 1, 2007; 18(6): 1936 - 1941. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. E. Lok, M. Allon, L. Moist, M. J. Oliver, H. Shah, and D. Zimmerman Risk Equation Determining Unsuccessful Cannulation Events and Failure to Maturation in Arteriovenous Fistulas (REDUCE FTM I) J. Am. Soc. Nephrol., November 1, 2006; 17(11): 3204 - 3212. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Asif, P. Roy-Chaudhury, and G. A. Beathard Early Arteriovenous Fistula Failure: A Logical Proposal for When and How to Intervene Clin. J. Am. Soc. Nephrol., March 1, 2006; 1(2): 332 - 339. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Yucel, S. O. Oktar, Y. Erten, A. Bursali, and H. Ozdemir B-Flow Sonographic Evaluation of Hemodialysis Fistulas: A Comparison With Low- and High-Pulse Repetition Frequency Color and Power Doppler Sonography J. Ultrasound Med., November 1, 2005; 24(11): 1503 - 1508. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Basile, F. Casucci, and C. Lomonte Timing of first cannulation of arteriovenous fistula: time matters, but there is also something else Nephrol. Dial. Transplant., July 1, 2005; 20(7): 1519 - 1520. [Full Text] [PDF]
|
|
|
|
|
|
M. E. Lockhart, M. L. Robbin, and M. Allon Preoperative Sonographic Radial Artery Evaluation and Correlation With Subsequent Radiocephalic Fistula Outcome J. Ultrasound Med., February 1, 2004; 23(2): 161 - 168. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. A. Turmel-Rodrigues, P. Bourquelot, J. Pengloan, M. L. Robbin, M. E. Lockhart, and M. Allon Hemodialysis Arteriovenous Fistula Maturity: US Evaluation [letter] * Dr Robbin and colleagues respond: Radiology, June 1, 2003; 227(3): 906 - 907. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
