Readmission and Mortality After Hospitalization With Acute Kidney Injury

Abstract

Rationale & objective: Acute kidney injury (AKI) carries high rates of morbidity and mortality. This study quantified various short- and long-term outcomes after hospitalization with AKI.

Study design: Retrospective propensity score (PS)-matched cohort study.

Setting & participants: Optum Clinformatics, a national claims database, was used to identify patients hospitalized with and without an AKI discharge diagnosis between January 2007 and September 2020.

Exposure: Among patients with prior continuous enrollment for at least 2years without AKI hospitalization, 471,176 patients hospitalized with AKI were identified and PS-matched to 471,176 patients hospitalized without AKI.

Outcome(s): All-cause and selected-cause rehospitalizations and mortality 90 and 365 days after index hospitalization.

Analytical approach: After PS matching, rehospitalization and death incidences were estimated using the cumulative incidence function method and compared using Gray's test. The association of AKI hospitalization with each outcome was tested using Cox models for all-cause mortality and, with mortality as competing risk, cause-specific hazard modeling for all-cause and selected-cause rehospitalization. Overall and stratified analyses were performed to evaluate for interaction between an AKI hospitalization and preexisting chronic kidney disease (CKD).

Results: After PS matching, AKI was associated with higher rates of rehospitalization for any cause (hazard ratio [HR], 1.62; 95% CI, 1.60-1.65), end-stage renal disease (HR, 6.21; 95% CI, 1.04-36.92), heart failure (HR, 2.81; 95% CI, 2.66, 2.97), sepsis (HR, 2.62; 95% CI, 2.49-2.75), pneumonia (HR, 1.47; 95% CI, 1.37-1.57), myocardial infarction (HR, 1.48; 95% CI, 1.33-1.65), and volume depletion (HR, 1.64; 95% CI, 1.37-1.96) at 90 days after discharge compared with the group without AKI, with similar findings at 365 days. Mortality rate was higher in the group with AKI than in the group without AKI at 90 (HR, 2.66; 95% CI, 2.61-2.72) and 365 days (HR, 2.11; 95% CI, 2.08-2.14). The higher risk of outcomes persisted when participants were stratified by CKD status (P<0.01).

Limitations: Causal associations between AKI and the reported outcomes cannot be inferred.

Conclusions: AKI during hospitalization in patients with and without CKD is associated with increased risk of 90- and 365-day all-cause/selected-cause rehospitalization and death.

Keywords: Acute kidney injury; cardiovascular disease; congestive heart failure; death; epidemiology; morbidity; outcomes; rehospitalization; sepsis.

Figure 1.

Selection and propensity score-matching process to derive the study cohort of cases (patients with an AKI discharge diagnosis) and controls (patients without an AKI discharge diagnosis) from Optum claims data, 2007 to 2020. We use a 1:1 matching algorithm. There were 140,485 eligible AKI patients that were not matched because we were unable to find non-AKI patients that would match to them.* Patients having two years of continuous Optum enrollment without a prior hospitalization with AKI and being discharged alive from the index hospitalization. Abbreviations: AKI, acute kidney injury; ESRD, end-stage renal disease.

Figure 2.

Cumulative incidence curves for rehospitalization up to 365 days (panels A. and B.) and death up to 12 months (panels C. and D.) after hospitalization with an AKI discharge diagnosis compared to propensity score-matched control patients hospitalized without an AKI diagnosis, stratified by pre-existing CKD status (panels A. and C. with CKD; B. and D. without CKD).

Figure 2.

Cumulative incidence curves for rehospitalization up to 365 days (panels A. and B.) and death up to 12 months (panels C. and D.) after hospitalization with an AKI discharge diagnosis compared to propensity score-matched control patients hospitalized without an AKI diagnosis, stratified by pre-existing CKD status (panels A. and C. with CKD; B. and D. without CKD).

Figure 2.

Cumulative incidence curves for rehospitalization up to 365 days (panels A. and B.) and death up to 12 months (panels C. and D.) after hospitalization with an AKI discharge diagnosis compared to propensity score-matched control patients hospitalized without an AKI diagnosis, stratified by pre-existing CKD status (panels A. and C. with CKD; B. and D. without CKD).

Figure 2.

Cumulative incidence curves for rehospitalization up to 365 days (panels A. and B.) and death up to 12 months (panels C. and D.) after hospitalization with an AKI discharge diagnosis compared to propensity score-matched control patients hospitalized without an AKI diagnosis, stratified by pre-existing CKD status (panels A. and C. with CKD; B. and D. without CKD).

Figure 3.

Patients hospitalized with an AKI discharge diagnosis have a higher risk for death, rehospitalization, ESRD, volume depletion, heart failure, myocardial infarction (MI), pneumonia, and sepsis compared to propensity-score matched control patients hospitalized without an AKI diagnosis at 90 days (panel A) and 365 days (panel B).

Figure 3.

Patients hospitalized with an AKI discharge diagnosis have a higher risk for death, rehospitalization, ESRD, volume depletion, heart failure, myocardial infarction (MI), pneumonia, and sepsis compared to propensity-score matched control patients hospitalized without an AKI diagnosis at 90 days (panel A) and 365 days (panel B).