Modification of Outcomes After Acute Kidney Injury by the Presence of CKD

Abstract

Acute kidney injury (AKI) in hospitalized patients is associated with poor outcomes; however, it is unclear how relationships between AKI and clinical outcomes vary with baseline kidney function. Population-based cohort. Adults in Alberta, Canada, who were hospitalized between January 1, 2003, and December 31, 2006, with at least 1 serum creatinine measurement during hospitalization and 1 outpatient creatinine measurement within 6 months preceding admission. Baseline kidney function, defined as mean estimated glomerular filtration rate (eGFR) of all outpatient creatinine measurements within 6 months before the index hospitalization, and AKI, defined using consensus criteria. Death during the index hospitalization and death or end-stage renal disease (ESRD) after hospitalization. AKI occurred in 18.3% of the 43,008 hospitalized patients in the cohort. Risk of AKI increased with decreasing eGFR (8.9% with eGFR ≥60 mL/min/1.73 m(2) vs 68.9% with eGFR <30 mL/min/1.73 m(2)). In multivariable Cox models, AKI of any severity was associated with death during the index hospitalization across all levels of eGFR, with an HR of 2.99 (95% CI, 2.59-3.44) in patients who had the least severe AKI across all eGFR strata up to an HR of 10.62 (95% CI, 8.78-12.82) in patients with baseline eGFR >60 mL/min/1.73 m(2) and the most severe AKI. The risk of death or ESRD decreased after discharge, with the highest risk of ESRD after AKI noted in patients with eGFR <30 mL/min/1.73 m(2) (17.0% in the AKI group vs 5.6% in the non-AKI group; P < 0.01). The study cohort is restricted to patients who had outpatient serum creatinine values available. AKI of any severity increases the risk of death both during hospitalization and after discharge. Although the risk of developing ESRD after AKI is greatest in patients with baseline eGFR <30 mL/min/1.73 m(2), this is exceeded by the risk of death.