Skip to main content
. 2020 Sep 6;73(1):209–221. doi: 10.1007/s13304-020-00878-4

Table 4.

Prediction of 90-day graft loss in the training and validation sets

Scores Training set (N = 1262) Scores Validation set (N = 520)
AUC ± SE 95% CIs p value AUC ± SE 95% CIs p value
CCI 0.94 ± 0.01 0.92–0.96  < 0.001 CCI 0.77 ± 0.08 0.62–0.93  < 0.001
D-MELD 0.60 ± 0.02 0.56–0.65  < 0.001 BAR 0.57 ± 0.06 0.45–0.68 0.36
MELD 0.60 ± 0.02 0.56–0.65  < 0.001 EAD 0.57 ± 0.07 0.43–0.71 0.35
BAR 0.60 ± 0.02 0.55–0.64  < 0.001 D-MELD 0.56 ± 0.08 0.41–0.70 0.43
EAD 0.58 ± 0.02 0.53–0.62 0.001 MELD 0.47 ± 0.07 0.33–0.61 0.70
CCI cut-off Sens Spec DOR CCI cut-off Sens Spec DOR
12.2 (25th) 98.4 27.7 23.6 12.2 82.4 32.1 2.2
29.6 (50th) 96.8 56.1 38.7 29.6 76.5 52.1 3.5
47.3 (75th) 88.8 85.9 48.3 47.3 64.7 79.3 7.0
84.9 (90th) 66.3 98.7 149.4 84.9 41.2 98.6 49.3

AUC area under the curve, SE standard error, CIs confidence intervals, CCI comprehensive complication index, D-MELD donor-model for end-stage liver disease, MELD model for end-stage liver disease, EAD early allograft dysfunction, BAR balance of risk, DOR diagnostic odds ratio