Outcomes and predictors of failure of arteriovenous fistulae for hemodialysis

Purpose Arteriovenous fistula(AVF) is preferred vascular access for hemodialysis but has primary failure in 20–60%. Studying predictors of AVF failure would help plan appropriate management.We studied AVF outcomes, clinical and vascular factors predicting their failure in patients requiring hemodialysis. Methods Retrospective study of patients with AVF creation from January 2017 to December 2018. Outcomes studied were immediate (< 72 h), primary (3 months) AVF failure, six-month/one-year patency, analyzed for predictive clinical, vascular factors as assessed using Pre-operative Doppler Ultrasound(DUS). Results Of 530 AVFs in 460 patients, DUS was done in 426/530 (80.4%), 349/460 (75.8%) were males, mean age was 53.10 ± 14.54 (18–91), 215/460(46.7%) had Diabetes mellitus(DM), 423/460(92%) hypertension. AVFs were radiocephalic in 79/530 (14.9%), brachiocephalic 418/530 (78.9%), brachiobasilic 33/530 (6.2%). AVF Immediate/Primary failure was seen in 64/530 (12.1%), 90/352 (25.6%); Patency at six months/one year in 253/352(71.8%),191/305 (62.6%), respectively. Older age had less immediate failures (AOR 0.97, CI 0.95–0.99, p 0.03). Feeding arterial diameter predicted immediate and primary failure on univariate analysis [OR 0.64 (95% CI 0.49–0.83), 0.62 (95% CI 0.47–0.89), respectively], but not multivariate. Artery diameter of > 4.0 mm had less failures [immediate (p 0.01), primary (p 0.02)], < 2.0 mm had specificity 95.9% and 95.4% for immediate, primary failure respectively. Conclusion AVF failure is 12.1%, immediately; 25.6% three months after construction, Patency at 6 months is 71.8%, one year 62.6%. Immediate failures decrease with age. Artery diameters > 4.0 mm had less, < 2.0 mm had more failures.


Introduction
Arteriovenous fistula (AVF) of native vessels as vascular access for hemodialysis is preferred over prosthetic grafts and central venous catheters because, at comparable flow rates, it is associated with lower mortality, infections and cardiovascular events [1]. Maturation of AVF depends on adequacy of vessels and time allowed before use. Since this is a lifeline for patients, pre-operative evaluation is important. Doppler ultrasonography (DUS) is standard for vascular evaluation as it evaluates both structure and function of peripheral vessels and, hence, is useful for preoperative planning of AVF [2]. However, guidelines do not recommend its use in all patients prior to AVF construction and definite vascular parameters have not been established [3]. We studied early outcomes of AVFs in our center and clinical and vascular factors predicting their failure.

Study design: retrospective cohort study
Patient population and clinical resources: The study was approved by the institutional ethics committee. Case records of consecutive patients who had AVF construction at our center between Jan 2017 and December 2018 were perused for demographic details, type of AVF and pre-operative DUS measurements. Patients younger than 18 years were excluded. Comorbidities noted were Ischemic heart disease(IHD), diabetes mellitus(DM), hypertension, cerebrovascular accidents(CVA), and peripheral arterial disease(PVD) using standard definitions. Vascular characteristics noted were anteroposterior (AP) diameter of brachial, ulnar and radial arteries; venous AP diameter, compressibility in basilic and cephalic veins. Physical examination was carried out by the operating urologist, pulses at elbows and wrists, and the superficial veins in the forearm and upper arm were assessed.
The DUS examination was conducted by a radiologist using a Phillips Affiniti 70 Ultrasound system with a 6-13 Megahertz linear probe with angle of insonation less than 60 degrees. Coronal views were used for measurements of diameter of all blood vessels and longitudinal views for measuring blood flow velocities with patients sitting upright with arm extended below heart level.

Vein assessment
Gray-scale Doppler ultrasound was used to assess for obvious scarring from previous venepuncture, for compressibility of the vein from the wrist to the axilla, and for measurement of vein diameter. The cephalic and basilic veins were routinely mapped.

Arterial measurements
Gray-scale Doppler ultrasound is used to assess for obvious vascular scarring or calcification. Moderate vascular calcification did not preclude AVF creation, although severe radial artery calcification was regarded as a contraindication to RAVF. The radial or brachial arterial diameters were measured at wrist and antecubital fossa, respectively, during systole. Peak arterial velocity was measured in both radial and brachial arteries using color Doppler ultrasound in the longitudinal plane, approximately 5 cm from the wrist crease and antecubital fossa, respectively, to allow a good straight segment of blood vessel for analysis.
AVFs were created by a single urologist trained in the procedure. Patient follow-up was achieved in outpatient or through telephonic contact of their health care givers.
Outcomes: Primary outcomes were immediate failure defined as an AVF that has either no appearance of or loss of bruit or thrill within 72 h of creation and primary failure defined as an AVF that was inadequate for hemodialysis at 3 months' post creation, including immediate failure, failure to mature or early thrombosis [4]. Secondary outcomes were cumulative survival at 6 and 12 months defined as the time of access creation or placement until access abandonment or achievement of a censored event (death, transfer to peritoneal dialysis, transplantation, and end of study period), and included all surgical and endovascular interventions [5]. Hypothesis tested was effect of clinical factors and target arterial and venous diameters as measured by DUS on immediate and primary failure of AVF. Statistical analysis: Statistical analysis was performed using SPSS version 23 (IBM Corp., Armonk, NY, USA). Descriptive data were presented as frequency (n) and percentages (%) for categorical data, and mean ± standard deviation (SD) for continuous variables. Comparison of proportions was done by chi-square test and continuous data by t-test. To identify factors predictive of immediate and primary AVF failure, univariate and multivariate logistic regression analyses were done. Crude and adjusted odds ratios were calculated, with 95% Confidence Intervals (CI). Receiver operating characteristic curves were generated to determine the predictive value of pre-operative target artery and vein diameters for immediate and primary failures. P values of < 0.05 were considered as statistically significant.

Results
Of 530 AVFs created in 460 patients, pre-operative duplex ultrasound was available for 426/530(80.4%) in 377/460 (82%) patients. Baseline characteristics are in Table 1 Fig. 1 Outcomes of Arteriovenous Fistulae). Association of DUS with patient characteristics for radiocephalic and upper arm AVFs is given in Table 2, and Predictive factors for AVF outcomes in Table 3. Overall males (p = 0.04) had higher arterial diameters. In upper arm males (p < 0.005), age more than 65 years (p = 0.005), those with IHD (p = 0.04) had higher arterial diameters and DM was associated with both higher arterial and venous diameters (both p < 0.001). Increasing age was associated with less immediate failures(AOR 0.97,CI 0.95-0.99, p 0.03).Feeding artery diameter of > 4.0 mm was significantly associated with less immediate (p 0.01) and primary failures (p 0.02). On receiver operating curve analysis (Fig. 2  A retrospective study [5] of 599 patients with 289 AVFs and 310 arteriovenous grafts(AVG) identified female gender to be associated with higher requirement of interventions in AVFs before successful cannulation and requirement of prior intervention and older age to be the factors predicting failure after a successful cannulation. Contrary to expectation, younger age was associated with more immediate failures of AVF in our study and presence of DM did not influence overall outcomes. This may be representative of our population but further prospective studies with stratification for age are warranted. Similar to abovementioned study, gender and presence of Ischemic heart disease, cerebrovascular accident did not influence outcomes probably indicating overall influence of vessel size. Since vessel size and patency may to some extent be assessed by physical examination role, Doppler ultrasound may be superfluous except in special circumstances. We found patency rates of 87.9%, 74.4%, 71.8%, 62.6% immediately, at three and six months and one year, respectively, with 2.5% and 5.4% requiring assistance to maintain patency at six months and one year, respectively. This is somewhat similar to the results presented in a systematic review [6] including 380 studies which reported primary unassisted, primary assisted and secondary patency rates of 64%,73%,79%, respectively, with 26% AVFs maturing at six months and 21% abandoned before use.
Studies comparing use of pre-operative Doppler ultrasound with clinical exam have found no differences in primary AVF patency, primary failure, requirement of postoperative intervention [7,8] but a higher secondary patency rates [9]. Hence, KDOQI clinical practice guideline for vascular access recommends use of pre-operative DUS in patients at high risk of AVF failure such as advanced age, obesity, female gender, peripheral vascular disease, and coronary artery disease; rather than routine vascular mapping [3]. This is based on a small randomized control study comparing selective and routine vessel mapping which did not show a difference in AVF failure at 90 days [10]. These results are similar to our findings which show extremes of arterial diameter are predictive of patency. Studies assessing utility of Doppler assessment are sparse and parameters not defined. Multicenter dialysis access consortium fistula study reported a primary failure of 60% in spite of routine use of Doppler ultrasound pre-operatively [11]. A meta-analysis of five randomized controlled trials with 574 patients showed lower immediate AVF failure rates in patients who underwent upper extremity Doppler ultrasonographic examination prior to surgery but this did not offer a significant advantage over clinical examination [12]. One study recommended routine use of Doppler based on comparison with a retrospective cohort showing reduced primary failure due to a multidisciplinary experienced team performing the AVF [13]. Primary AVF failure rate in our study was 25.6%; it typically ranges from 23 to 40% [14,15].Wilmink et al. noted that vessel diameter did not predict AVF functionality and should not be used to avoid construction [16]. Arterial diameter of > 4.0 mm was associated with better outcomes in our study. In a study by Farrington et al., arterial diameter was observed to be an important predictor for both unassisted and overall AVF maturation [17]. Farber et al. demonstrated that early thrombosis risk was higher in women than in men, for forearm than upper arm fistulas, for radial than brachial artery fistulas, and for fistulas constructed from smaller caliber arteries or veins, although the trend for veins surprisingly did not extend below vessels 3 mm in diameter [18].
In our study, there was a trend for feeding artery diameter in predicting immediate and primary failure, those with arterial diameters > 4.0 mm had lesser and < 2.0 mm had more failures, respectively. Majority of our diabetic patients have undergone upper arm AVFs, which is likely why they appear

Conclusion
AVF failure is seen in 12.1% immediately after construction and in 25.6% at three months. AVF patency rates are 71.8% at 6 months and 62.6% at one year. Artery diameters of > 4.0 mm had less and < 2.0 mm had more failures. Younger age is associated with higher immediate failures.
Author contribution All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by SVR, RAP and IRR. The first draft of the manuscript was written by SVR and RAP, review, editing and supervision by SPN, AC, SVS and MVB. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding Open access funding provided by Manipal Academy of Higher Education, Manipal.

Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
Code availability Not applicable.

Conflict of interest
The authors have no conflicts of interest to declare that are relevant to the content of this article.
Ethical approval This retrospective chart review study involving human participants was in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The Hospital Ethics committee of Kasturba Medical College Manipal MAHE (IRB approval number 212/2019) approved this study.

Informed consent Informed consent was obtained from all individual participants included in the study.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.