Fig. 4
From: A universal predictive and mechanistic urinary peptide signature in acute kidney injury
Independent validation of the predictive value of the urinary peptide signature for early AKI diagnosis after CBP surgery. A Association of the urinary peptide-based score with the development of AKI (all stages, left; according to KDIGO stages, right) in the CBP surgery validation cohort (n = 480). *** p < 0.001. B Spline plot of the association between the peptide-based score and the risk of developing AKI. A univariate logistic generalized additive model was used. The black line indicates the estimated risk of AKI with respective 95% confidence intervals. The spikes show the distribution of the peptide-based scores. C ROC curves with corresponding AUROC and 95% confidence intervals for the 204 peptides-based score, the local clinical score (and reference urinary AKI biomarkers NGAL and TIMP2*IGFBP7 for the prediction of AKI (all stages) in the validation cohort. D Odds ratios and corresponding 95% confidence intervals in the validation cohort using a multivariate logistic regression model including the local clinical score, reference urinary AKI biomarkers NGAL and TIMP2*IGFBP7, the 204 peptides-based score or a combination of the local clinical and peptide-based scores as a qualitative value according to the selected threshold (optimal Youden index in the derivation cohort). ROC, receiver operating characteristics curve; AUROC, area under the receiver operating characteristics curve; CI, confidence interval; Thr, threshold