Dialysis requirement, long-term major adverse cardiovascular events (MACE) and all-cause mortality in hospital acquired acute kidney injury (AKI): a propensity-matched cohort study

Background Dialysis-requiring acute kidney injury (D - AKI) is common in hospitalized patients. Many patients survive the immediate post AKI period, thus at risk of suffering long-term sequelae of AKI. Prior studies examining long term outcomes lack non-dialyzed AKI control groups. Without non-dialyzed AKI control group, these studies cannot provide relevant information on long-term risks or benefits associated with dialysis intervention following AKI. Methods The study cohort comprises of adults admitted to the University of Virginia Medical Center between January 1, 2002 and December 31, 2012 with baseline eGFR ≥60 ml/min per 1.73 m 2 , who developed AKI during hospitalization and survived beyond 30 days of the AKI event. Follow up was done until MACE, death or through Dec 31, 2013 (n = 11,779). AKI was defined according to KDIGO definition. MACE was defined as subsequent admission for Myocardial Infarction (MI), cerebrovascular disease (CVD) and heart failure using ICD 9-CM codes. The date of MACE was defined as the date of the first qualifying event. Demographic and premorbid clinical variables were used to generate propensity score. Patients who had temporary dialysis were matched with those managed conservatively according to propensity score in a ratio of 1:3. Results After the propensity score match, the adjusted hazard ratio for MACE, all-cause mortality and composite end point “all-cause mortality or MACE” in dialyzed versus non dialyzed patients were 1.081 (95 % CI 0.848–1.378), 1.107 (95 % CI 0.869–1.410) and 1.107 (95 % CI 0.880–1.307), respectively. Conclusion Management of AKI with temporary dialysis in hospitalized patients with baseline eGFR of ≥60 ml/min per 1.73 m 2 was NOT associated with an increased risk for subsequent admission for MACE or all-cause mortality. Clinicians may not need to worry that the dialysis procedure itself may confer additional risk for long-term MACE and all-cause mortality in AKI patients with normal pre-hospitalization GFR.