Predicting disease progression in cirrhotic patients after transjugular intrahepatic portosystemic shunt implantation: A sex-stratified analysis

Table 3 Predictive performance of different models in the training set and test set

Model	Variable	Training set AUC (95%CI)	Sensitivity	Specificity	Test set AUC (95%CI)	Sensitivity	Specificity
Model 1	Creatinine, plasma ammonia, SMI, ACLF, HE	0.842 (0.763-0.921)	0.957	0.652	0.875 (0.672-1.000)	0.75	0.962
Model 2	Creatinine, plasma ammonia, SMI, ACLF, HE	0.864 (0.785-0.942)	0.864	0.707	0.809 (0.631-0.986)	1.000	0.609
Model 3	Creatinine, SMI, HE	0.804 (0.689-0.918)	0.857	0.674	0.768 (0.581-0.955)	0.833	0.696
Model 4	Creatinine, plasma ammonia, SMI, ACLF, HE	0.842 (0.761-0.923)	0.783	0.813	0.948 (0.844-1.00)	1.000	0.792
Model 5	Creatinine, plasma ammonia, SMI, HE	0.840 (0.760-0.921)	0.957	0.636	0.823 (0.659-0.987)	1.000	0.667

We used 5-fold cross-validation to classify male patients into five parts, among which four parts were used as the training set and one part as the validation set. AUC: Area under the receiver operating characteristic curve; CI: Confidence interval; SMI: Skeletal muscle index; ACLF: Acute-on-chronic liver failure; HE: Hepatic encephalopathy.