Long-term risk of adverse outcomes after acute kidney injury: a systematic review and meta-analysis of cohort studies using consensus definitions of exposure

• Published in: Kidney international Vol. 95; no. 1; pp. 160 - 172
• Main Authors: See, Emily J., Jayasinghe, Kushani, Glassford, Neil, Bailey, Michael, Johnson, David W., Polkinghorne, Kevan R., Toussaint, Nigel D., Bellomo, Rinaldo
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.01.2019
• Subjects: acute kidney injury; Acute Kidney Injury - complications; Acute Kidney Injury - diagnosis; Acute Kidney Injury - mortality; Acute Kidney Injury - therapy; chronic kidney disease; death; Disease Progression; end-stage kidney disease; Health Care Rationing - organization & administration; Humans; Kidney Failure, Chronic - epidemiology; Kidney Failure, Chronic - etiology; major adverse kidney event; Outcome Assessment (Health Care) - statistics & numerical data; Risk Assessment; Risk Factors; Severity of Illness Index; Survival Analysis; Survival Rate; Time Factors; chronic kidney disease; acute kidney injury; major adverse kidney event; end-stage kidney disease; death
• ISSN: 0085-2538; 1523-1755; 1523-1755
• DOI: 10.1016/j.kint.2018.08.036

Reliable estimates of the long-term outcomes of acute kidney injury (AKI) are needed to inform clinical practice and guide allocation of health care resources. This systematic review and meta-analysis aimed to quantify the association between AKI and chronic kidney disease (CKD), end-stage kidney disease (ESKD), and death. Systematic searches were performed through EMBASE, MEDLINE, and grey literature sources to identify cohort studies in hospitalized adults that used standardized definitions for AKI, included a non-exposed comparator, and followed patients for at least 1 year. Risk of bias was assessed by the Newcastle-Ottawa Scale. Random effects meta-analyses were performed to pool risk estimates; subgroup, sensitivity, and meta-regression analyses were used to investigate heterogeneity. Of 4973 citations, 82 studies (comprising 2,017,437 participants) were eligible for inclusion. Common sources of bias included incomplete reporting of outcome data, missing biochemical values, and inadequate adjustment for confounders. Individuals with AKI were at increased risk of new or progressive CKD (HR 2.67, 95% CI 1.99-3.58; 17.76 versus 7.59 cases per 100 person-years), ESKD (HR 4.81, 95% CI 3.04-7.62; 0.47 versus 0.08 cases per 100 person-years), and death (HR 1.80, 95% CI 1.61-2.02; 13.19 versus 7.26 deaths per 100 person-years). A gradient of risk across increasing AKI stages was demonstrated for all outcomes. For mortality, the magnitude of risk was also modified by clinical setting, baseline kidney function, diabetes, and coronary heart disease. These findings establish the poor long-term outcomes of AKI while highlighting the importance of injury severity and clinical setting in the estimation of risk.