Including urinary output to define AKI enhances the performance of machine learning models to predict AKI at admission

Acute kidney injury (AKI) is a prevalent and detrimental condition in intensive care unit patients. Most AKI predictive models only predict creatinine-triggered AKI (AKICr) and might underperform when predicting urine-output-triggered AKI (AKIUO). We aimed to describe how admission AKICr prediction models perform in all AKI patients. Three types of models were trained: 1) pAKIany, predicting AKI based on creatinine or urine output, 2) pAKIUO, predicting AKI based only on urine output, and 3) pAKICr, predicting AKI based only on creatinine. We compared model performance and predictive features. The pAKIany models had the best overall performance (AUROC 0.673–0.716) and the most consistent performance across three patient cohorts grouped by type of AKI trigger (min AUROC of 0.636). The pAKICr models had fair performance in predicting AKICr (AUROCs 0.702–0.748) but poor performance predicting AKIUO (AUROCs 0.581–0.695). The predictive features for the pAKICr models and pAKIUO models were distinct, while top features for the pAKIany models were consistently a combination of those for the pAKICr and pAKIUO models. Ignoring urine output in the outcome during model training resulted in models that are unlikely to predict AKIUO adequately and may miss a substantial proportion of patients in practice. •We aimed to describe differences in AKI prediction models based on urine output vs. creatinine-triggered AKI.•Serum creatinine models could not predict AKI based on urine output and used differing predictive features.•Ignoring urine output in AKI prediction resulted in inadequate performance and missed a substantial proportion of patients.