Original Investigation
Intermediate and Long-term Outcomes of Survivors of Acute Kidney Injury Episodes: A Large Population-Based Cohort Study

https://doi.org/10.1053/j.ajkd.2016.05.018Get rights and content
Under a Creative Commons license
open access

Background

The long-term prognosis after acute kidney injury (AKI) is variable. It is unclear how the prognosis of AKI and its relationship to prognostic factors (baseline kidney function, AKI severity, prior AKI episodes, and recovery of kidney function) change as follow-up progresses.

Study Design

Observational cohort study.

Setting & Participants

The Grampian Laboratory Outcomes Morbidity and Mortality Study II (GLOMMS-II) is a large regional population cohort with complete serial biochemistry and outcome data capture through data linkage. From GLOMMS-II, we followed up 17,630 patients hospitalized in 2003 through to 2013.

Predictors

AKI identified using KDIGO (Kidney Disease: Improving Global Outcomes) serum creatinine criteria, characterized by baseline kidney function (estimated glomerular filtration rate [eGFR] ≥ 60, 45-59, 30-44, and <30 mL/min/1.73 m2), AKI severity (KDIGO stage), 90-day recovery of kidney function, and prior AKI episodes.

Outcomes

Intermediate- (30-364 days) and long-term (1-10 years) mortality and long-term renal replacement therapy.

Measurements

Poisson regression in time discrete intervals. Multivariable Cox regression for those at risk in the intermediate and long term, adjusted for age, sex, baseline comorbid conditions, and acute admission circumstances.

Results

Of 17,630 patients followed up for a median of 9.0 years, 9,251 died. Estimated incidences of hospital AKI were 8.4% and 17.6% for baseline eGFRs  60 and <60 mL/min/1.73 m2, respectively. Intermediate-term (30-364 days) adjusted mortality HRs for AKI versus no AKI were 2.48 (95% CI, 2.15-2.88), 2.50 (95% CI, 2.04-3.06), 1.90 (95% CI, 1.51-2.39), and 1.63 (95% CI, 1.20-2.22) for eGFRs  60, 45 to 59, 30 to 44, and <30 mL/min/1.73 m2, respectively. Among 1-year survivors, long-term HRs were attenuated: 1.44 (95% CI, 1.31-1.58), 1.25 (95% CI, 1.09-1.43), 1.21 (95% CI, 1.03-1.42), and 1.08 (95% CI, 0.85-1.36), respectively. The excess long-term hazards in AKI were lower for lower baseline eGFRs (P for interaction = 0.01).

Limitations

Nonprotocolized observational data. No adjustment for albuminuria.

Conclusions

The prognostic importance of a discrete AKI episode lessens over time. Baseline kidney function is of greater long-term importance.

Index Words

Acute kidney injury (AKI)
chronic kidney disease (CKD)
baseline kidney function
survival
mortality
epidemiology
outcomes
prognosis
AKI recovery
acute on chronic kidney disease

Cited by (0)