The influence of electronic AKI alert on prognosis of adult hospitalized patients: a systematic review and meta-analysis

BACKGROUND: Acute kidney injury (AKI) is a critical yet frequently under diagnosed condition in hospitalized patients, impacting morbidity and mortality. Electronic alerts for AKI aimed to assist physicians in early diagnosis and intervention, though evidence for their effectiveness is inconsistent. MATERIALS AND METHODS: A systematic search was conducted in PubMed, the Cochrane Central Register of Controlled Trials, Cochrane Library, and Web of Science from inception to November 2024. Eligible studies included randomized controlled trials (RCTs), before-and-after analyses, and stepped-wedge designs involving hospitalized patients. The primary outcomes were mortality and renal replacement therapy (RRT) rates, Secondary outcomes included hospital length of stay (LoS), AKI progression and recovery. Care-centered outcomes encompassed nephrologist consultation, nephrotoxic medication discontinuation and medication review. Subgroup analysis examined the impact of response intensity, hospital type and geographic region on these outcomes. RESULTS: Twenty-two studies involving 170,696 participants were included: 8 RCTs (n = 21,710) and 14 non-RCTs or observational studies (n = 148,986). RCTs showed no effect on mortality (RR 1.02; 95% CI 0.97-1.07) or LoS (mean difference 0.04; 95% CI - 0.13 to 0.22) but a significant increase in RRT use (RR 1.13; 95% CI 1.02-1.26) with AKI alert systems. Non-RCTs, however, reported reduced mortality (RR 0.92; 95% CI 0.88-0.96), less AKI progression (RR 0.85; 95% CI 0.77-0.94), enhanced kidney recovery (RR 1.65; 95% CI 1.56-1.75), increased nephrotoxic drug discontinuation (RR 1.20; 95% CI 1.13-1.28), and higher drug review rates (RR 1.19; 95% CI 1.17-1.21), with no impact on RRT use (RR 1.08; 95% CI 0.87-1.36). Subgroup analysis revealed an increased in-hospital mortality in low response intensity (RR 1.15; 95% CI 1.00-1.32), reduced mortality in moderate response intensity (RR 0.93; 95% CI 0.89-0.97), and unclear effects in high response intensity (RR 0.88; 95% CI 0.70-1.09). AKI alert was also favored in teaching hospitals and in several regions (Europe, North America and South America). CONCLUSION: The efficacy of AKI alerts remains inconclusive. Current evidence do not support or refute their effectiveness. Variability in response intensity, hospital type and geographic region may help explaining discrepancies, underscoring the need for further research to optimize AKI alert systems with more effective action in clinical practice.