Fig. 2
From: A universal predictive and mechanistic urinary peptide signature in acute kidney injury
AKI prediction in CBP surgery patients based on clinical pre- and per-operative features. A. ROC curves and corresponding AUROC [95% confidence interval] of published clinical scores for the prediction of AKI (KDIGO 1, 2, or 3) in the CBP surgery validation cohort. B. Parameters and associated coefficients of a local clinical model defined in the derivation cohort. The clinical score is calculated as follows: logit(p(AKI)) = Ca x Age (Years) + Ch x Hypertension (0/1) + Cg x eGFR (mL/min.1.73m2) + Ck x Kidney Graft Recipient (0/1) + Cs x Valvular Surgery (0/1) + Cc x CPB Duration (min) + I. C. Association of the local clinical score with the development of AKI in the validation cohort (all stages, left; according to KDIGO stages, right). * p < 0.05; ** p < 0.01; *** p < 0.001. D. ROC curves and corresponding AUROC [95% confidence interval] of the local clinical score compared to baseline eGFR for the prediction of all stages of AKI in the validation cohort. The AUROCs of the local clinical score and eGFR were significantly different (Delong test; p value = 0.007). ROC, receiver operating characteristic curve; AUROC, area under the receiver operating characteristic curve; CI, confidence interval; AKI, acute kidney injury; CBP, cardiac bypass; eGFR, estimated glomerular filtration rate