Guidelines of the National Kidney Foundation recommending aggressive pursuit of autogenous fistulae for dialysis access in lieu of prosthetic arteriovenous grafts have stimulated a renewed interest in transposed brachial-basilic fistulae as an alternative technique for upper arm access in patients who may not be candidates for a lower arm radial-cephalic or forearm brachial-cephalic fistula. We hypothesized that in our safety-net population, where radial-cephalic and brachial-cephalic often are not possible, brachial-basilic would provide patency rates superior to arteriovenous grafts and equivalent to radial-cephalic and brachial-cephalic fistulae. We analyzed retrospectively our most recent 2.5-year experience with dialysis access procedures at our metropolitan safety-net hospital. Procedures were grouped as follows: radial-cephalic, brachial-cephalic, brachial-basilic, and arteriovenous grafts. The access outcomes measured were primary failure, time to use, need for intervention, and primary as well as secondary patency. Differences in age, sex, race, renal function (Modification of Diet in Renal Disease), baseline diagnoses (diabetes mellitus, hypertension, coronary artery disease, and peripheral vascular disease), as well as the number of previous accesses, were adjusted in the analysis. Logistic regression was used to identify independent predictors of primary failure, and Kaplan-Meier plots assessed differences in primary patency rates. A log of the time variables was used to approximate normal distribution. In all, 193 patients were included in this study as follows: radial-cephalic, 75 (39%) patients; brachial-cephalic, 35 (18%) patients; brachial-basilic, 33 (17%) patients; and arteriovenous grafts, 50 (26%) patients. Primary patency means differed marginally between groups (P = .08), and when grafts were excluded from the analysis, no difference was found between primary patency in all autogenous fistula techniques (P = .88). Kaplan-Meier plots showed that when analyzing the first 35 weeks, a significantly lower primary patency among graft recipients early after the procedure was noted, and a higher performance of BB after 20 weeks was noted (log-rank P = .05, Wilcoxon P = .004). Furthermore, secondary patency did not vary significantly between groups (P = .62). Radial-cephalic were more likely to fail primarily when compared with the other access groups (P = .03), and in a univariate analysis, underlying hypertension was associated with a lower risk of primary failure (P = .01) compared with other diagnoses. A logistic regression stepwise selection showed that the underlying diagnoses of peripheral vascular disease, diabetes mellitus, or coronary artery disease were associated with a greater risk of primary failure compared with those with HTN (P = .001; oddsratio, 4.05; 95% confidence interval, 1.71-9.59), as well as the presence of a previously failed access (P= .04; odds ratio, 2.39; 95% confidence interval, 1.08-5.67). In a safety-net population, our results suggest that 2-stage brachial-basilic transposition fistulae provide patency rates equivalent to brachial-cephalic and radial-cephalic fistulae and superior to grafts. Although 2 procedures are required, brachial-basilic fistulae provide a reliable access and should be considered the next choice when radial-cephalic and/or brachial-cephalic are not possible.