Trends in acute kidney injury, associated use of dialysis, and mortality after cardiac surgery, 1999 to 2008

Abstract

Background: The development of acute kidney injury (AKI) after cardiac surgery is associated with significant mortality, morbidity, and cost. The last decade has seen major changes in the complexity of cardiac surgical candidates and in the number and type of cardiac surgical procedures being performed.

Methods: Using data from the Nationwide Inpatient Sample, we determined the annual rates of AKI, AKI requiring dialysis (AKI-D), and inpatient mortality after cardiac surgery in the United States in the years 1999 through 2008.

Results: Inpatient mortality with AKI and AKI-D decreased from 27.9% and 45.9%, respectively, in 1999 to 12.8% and 35.3%, respectively, in 2008. Compared with 1999, the odds of AKI and AKI-D in 2008, adjusted for demographic and clinical factors, were 3.30 (95% confidence interval [CI]: 2.89 to 3.77) and 2.23 (95% CI: 1.78 to 2.80), respectively. Corresponding adjusted odds of death associated with AKI and AKI-D were 0.31 (95% CI: 0.26 to 0.36) and 0.47 (95% CI: 0.34 to 0.65.) Taken together, the attributable risks for death after cardiac surgery associated with AKI and AKI-D increased from 30% and 5%, respectively, in 1999 to 47% and 14%, respectively, in 2008.

Conclusions: In sum, despite improvements in individual patient outcomes over the decade 1999 to 2008, the population contribution of AKI and AKI-D to inpatient mortality after surgery increased over the same period.

Fig. 1

Temporal trends in (A) diagnosed acute kidney injury (AKI), (B) acute kidney injury requiring dialysis (AKI-D), and (C) in-hospital mortality, based on 601,328 discharges identified with ICD-9 coding for cardiopulmonary bypass and coronary artery bypass graft surgery or open-heart valve repair or replacement. Adjusted models included covariates on age, sex, surgery type, heart failure, diabetes mellitus, hypertension, pulmonary disease, peripheral vascular disease, cerebrovascular disease, obesity, sepsis, use of intraaortic balloon pump, and mechanical ventilation. (Solid line = adjusted; dotted line = unadjusted.)

Fig. 1

Temporal trends in (A) diagnosed acute kidney injury (AKI), (B) acute kidney injury requiring dialysis (AKI-D), and (C) in-hospital mortality, based on 601,328 discharges identified with ICD-9 coding for cardiopulmonary bypass and coronary artery bypass graft surgery or open-heart valve repair or replacement. Adjusted models included covariates on age, sex, surgery type, heart failure, diabetes mellitus, hypertension, pulmonary disease, peripheral vascular disease, cerebrovascular disease, obesity, sepsis, use of intraaortic balloon pump, and mechanical ventilation. (Solid line = adjusted; dotted line = unadjusted.)

Fig. 2

Mortality trends among patients undergoing open-heart surgery complicated by acute kidney injury (AKI) and acute kidney injury requiring dialysis (AKI-D), based on (A) 46,195 discharges identified with ICD-9 coding for AKI and (B) 4,923 discharges identified with ICD-9 coding for AKI-D. Adjusted models included covariates on age, sex, surgery type, heart failure, diabetes mellitus, hypertension, pulmonary disease, peripheral vascular disease, cerebrovascular disease, obesity, sepsis, use of intraaortic balloon pump, and mechanical ventilation. (Solid line = adjusted; dotted line = unadjusted.)

Fig 3

Trends in the attributable risk percent of death from acute kidney injury (AKI), based on 95,950 deaths after cardiopulmonary bypass and coronary artery bypass graft surgery or open-heart valve repair or replacement. (Solid line = AKI; dotted line = AKI with dialysis [AKI-D].)