Racial Differences in the Competing Risks of Mortality and ESRD After Acute Myocardial Infarction

Background: The prevalence of earlier stage chronic kidney disease is lower for African Americans than whites in the United States. This is counterintuitive given the known 4-fold greater incidence of end-stage renal disease (ESRD) in African Americans. We describe racial differences in the rate of progression to ESRD and address the competing risk of mortality. Study Design: Retrospective analysis of Cooperative Cardiovascular Project data. Setting & Participants: 127,736 Medicare beneficiaries 65 years and older admitted to 4,545 hospitals with acute myocardial infarction between February 1994 and June 1995, with follow-up data for ESRD and mortality through June 2004. Predictors: African American versus white race, estimated glomerular filtration rate (eGFR), and their interaction; other characteristics at hospital admission. Outcomes & Measurements: Time to ESRD using Cox proportional hazards models. Results: Mean age was 77.1 years, with 8,278 African Americans (6.5%) and 49.9% women. Mean baseline eGFRs were 61.4 ± 31.4 and 57.0 ± 25.6 mL/min/1.73 m² (P < 0.001) for African Americans and whites, respectively. Of 2,161 patients (1.7%) progressing to ESRD (incidence, 3.75/1,000 person-years), 14.9% were African American. The adjusted hazard ratio for ESRD (African Americans versus whites) was 1.90 (95% confidence interval, 1.78 to 2.03); African Americans were at significantly increased risk of incident ESRD at each baseline eGFR stage (P for interaction < 0.001). Racial differences in incident ESRD were not accounted for by differences in mortality. Limitations: Retrospective analysis, residual bias from unmeasured factors, baseline eGFR determined from serum creatinine levels at the time of acute hospitalization. Conclusions: Within a nationally representative sample of Medicare patients with acute myocardial infarction, African Americans had an increased 10-year risk of ESRD regardless of baseline kidney function that was not accounted for by differences in pre-ESRD mortality.