Development and validation of a nomogram for predicting overall survival in cirrhotic patients with acute kidney injury

Table 3 Comparison of models in predicting the death of cirrhotic patients with acute kidney injury

Datasets	Time of mortality	Predictive model	AUROC (95%CI)	P value
Training dataset	30-d	Nomogram	0.757 (0.685-0.830)	Ref.
		CTP	0.694 (0.616-0.772)	0.210
		MELD	0.669 (0.585-0.752)	0.114
		MELD-Na	0.702 (0.622-0.782)	0.332
Validation dataset	30-d	Nomogram	0.763 (0.656-0.869)	Ref.
		CTP	0.689 (0.569-0.809)	0.416
		MELD	0.745 (0.631-0.859)	0.979
		MELD-Na	0.763 (0.649-0.876)	> 0.05
Training dataset	90-d	Nomogram	0.792 (0.726-0.859)	Ref.
		CTP	0.696 (0.614 – 0.778)	0.041
		MELD	0.657 (0.574-0.740)	0.003
		MELD-Na	0.670 (0.586-0.755)	0.003
Validation dataset	90-d	Nomogram	0.817 (0.716-0.917)	Ref.
		CTP	0.667 (0.523-0.809)	0.031
		MELD	0.650 (0.515-0.785)	0.008
		MELD-Na	0.642 (0.502-0.783)	0.005
Training dataset	180-d	Nomogram	0.801 (0.726-0.870)	Ref.
		CTP	0.722 (0.630-0.814)	0.045
		MELD	0.663 (0.575-0.752)	0.008
		MELD-Na	0.680 (0.588-0.771)	0.009
Validation dataset	180-d	Nomogram	0.862 (0.765-0.958)	Ref.
		CTP	0.708 (0.540-0.878)	0.027
		MELD	0.659 (0.512-0.806)	0.002
		MELD-Na	0.652 (0.491-0.812)	0.002

CTP: Child-Turcotte-Pugh; MELD: The model for end-stage liver disease; AUROC: The area under the curve of the receiver operating characteristic curve; CI: Confidence interval.