Background: AV fistula (AVF) is the safest of vascular accesses with lowest infection rates; yet only 23% patients used AVF during 1997 in USA. The lower prevalence of AVF among diabetics on hemodialysis (HD) places them at a higher risk of vascular-access-related septicemia (VRS) and ensuing mortality. In this study we assessed the outcome of VRS after maximizing the frequency of native AVF in this largest growing population on HD. Methods: Study included 218 patients, 63 diabetics and 155 nondiabetics on HD, through,July 1996 to July 2000 when National Kidney Foundation-Dialysis Outcome and Quality Initiative (NKF-DOQI) set goal was accomplished with overall 72% of functioning AVF (57.2% diabetics and 78.1% nondiabetics) through joint efforts of nephrologists and vascular surgeons. Results: Overall, 10.6% patients per year developed VRS through 125 episodes, over 10,464 patient-months, recording 1.19 episodes per 100 patient-months. In the diabetic group, 13.87% patients per year had VRS during 44 episodes with 1.45 episodes per 100 patient-months while 1.08 episodes per 100 patient-months were recorded in nondiabetics with 9.35% per year having VRS during 81 episodes. Collectively, catheters recorded 1.5 folds higher VRS episodes in diabetic than in nondiabetic group. Mortality of 9.28% per year in diabetic group as compared to that of 6.45% per year in nondiabetic group [RR-1.436, 95% CI (0.778-2.651)] was observed, while overall mortality of 7.5% per year recorded is a good deal lower than 12-22% reported. Conclusion: The NKF-DOQI set aim of dialyzing over 50% patients through AVF is attainable in diabetics as well. Optimizing AVF is a viable approach to lessen VRS related mortality in diabetics on HD. Our continued dependence on vascular catheters is largely responsible for higher mortality in diabetics than nondiabetics on HD due to lack of cagily established pre-ESRD program for diabetics. Copyright (C) 2002 S. Karger AG, Basel.