Association of Race and Ethnicity with Vascular Access Type Selection and Outcomes

Abstract

Black patients with end-stage renal disease (ESRD) represent 30.5% of the prevalent ESRD population in the United States, despite only accounting for 18% of the total population. Black patients are less likely to have pre-ESRD care compared with their white counterparts and are 3-4 times more likely to progress from chronic kidney disease to ESRD than whites, suggesting that black patients are particularly vulnerable to disparities in outcomes related to hemodialysis and ESRD. The objective of this study is to examine the association of race with outcomes of hemodialysis access and selection of arteriovenous fistula (AVF) versus arteriovenous graft (AVG). Patients with chronic kidney disease who initiated dialysis through a tunneled hemodialysis catheter (THC) were identified in the Optum's De-identified Clinformatics® Data Mart (OptumInsight, Eden Prairie, MN) claims database (2011-2017). The odds of AVF versus AVG creation and the odds of repeat vascular access creation were analyzed using logistic regression. Time from initial AVF/AVG to THC removal and time to repeat AVF/AVG were analyzed using Cox proportional hazards. About 7,584 vascular access patients met the inclusion criteria: 5,852 (77%) AVF and 1,732 (23%) AVG. Median follow-up was 583days overall (range, 1-2,543), 589days among AVF patients (range, 1-2,543), and 260days among AVG patients (range, 1-2,529). Between races, there was no clinically significant variation in characteristics or comorbidities, with the exception of a much lower rate of obesity among Asians. Black patients had 36% lower odds of AVF index versus AVG index (P<0.001). Patients 70 years or older and patients with diabetes had lower odds of AVF index, whereas men and obese patients had greater odds of receiving AVF. Overall, graft patients were 73% more likely to have a shorter time to THC removal than fistula patients, but Hispanic graft patients were 25% more likely to have a shorter time to THC removal than whites. Patients with diabetes, patients with cardiac arrhythmia, and obesity were more likely to have a longer time to THC removal. About 1,589 (21%) patients underwent a repeat vascular access creation during the follow-up period: 19% of whites (n=802), 26% of blacks (n=483), 19% of Hispanics (n=250), and 19% of Asians (n=54) (P<0.001). Multivariate analysis demonstrated that black patients had 58% greater odds of requiring a second access than white patients (P<0.001). Graft patients, patients 70 years or older, and men had lower odds of repeat access. Black patients were 45% more likely to have a shorter time until second access creation. Graft patients, patients aged 70 years or older, and men were more likely to have a longer time until second access. Patients with obesity were more likely to have a shorter time until second access. This study's findings suggest that after initial vascular access, compared with whites, blacks have no difference in time to index access success, but their access fails earlier and more frequently, independent of access type, age, and comorbidities. Given blacks constitute 30.5% of the hemodialysis population in the United States, it is imperative that future research investigate the root causes of these disparities.