Background: Needle infiltration of arteriovenous fistulae is a common problem in US hemodialysis units. This study evaluated the frequency of fistula infiltration, its risk factors, and clinical consequences of this complication. Methods: Using a prospective computerized vascular access database, we identified all patients with a major fistula infiltration sufficiently severe to prolong catheter dependence for dialysis. These patients were compared with a control group without fistula infiltration. We also quantified subsequent access outcomes in patients with infiltrations. Results: During a 5-year period, 47 patients had a major fistula infiltration, representing a 5.2% annual rate. On multiple variable logistic regression analysis, the likelihood of fistula infiltration was associated strongly with patient age (odds ratio, 1.039/1-year increment; 95% confidence interval, 1.016 to 1.062; P = 0.0007). Fistula infiltration was not associated with sex, race, diabetic status, peripheral vascular disease, body mass index, or fistula location. New fistulas (<6 months in age) were more likely in patients with infiltrations compared with patients without infiltrations (43.5% versus 20.5%; odds ratio, 2.98; 95% confidence interval, 1.61 to 5.54; P = 0.0004). Each major fistula infiltration resulted in a mean of 2.4 diagnostic tests, surgery appointments, or interventions. Fistula thrombosis occurred in 12 patients (26%). Median prolongation of catheter dependence for dialysis in patients with major infiltrations was 97 days. Conclusion: Needle infiltration of fistulae is more common in older patients and with new fistulae. These infiltrations result in numerous procedures, as well as prolongation of catheter dependence for more than 3 months. © 2006 National Kidney Foundation, Inc.