Answers are grounded in this paper and cite their sources.
Background Portal vein thrombosis (PVT) is a common complications of liver cirrhosis. Its occurrence increases the difficulty of clinical management for patients with liver cirrhosis, accelerates disease progression, and raises mortality. Therefore, early prevention, diagnosis and identification of patients with PVT are of vital importance. This study used a meta-analysis to evaluate risk prediction models for PVT in liver cirrhosis, aiming to provide a scientific basis for future model development and clinical application. Methods We searched databases including PubMed, Embase, and Web of Science for relevant literature published from database inception to July 2026. Two researchers conducted the literature screening and data extraction, and used the Prediction model Risk Of Bias Assessment Tool (PROBAST) to evaluate the risk of bias in the prediction models. Meta-analysis was performed using Stata 14.0. Results A total of 11 studies were included, involving 8670 participants and 15 prediction models were involved. Two of these prediction models underwent external validation. The PROBAST assessment indicated that all studies were at a relatively high risk of bias, although the prediction models showed good applicability. The area under the curve (AUC) values of these prediction models ranged from 0.608 to 0.919.The pooled AUC of the 15 prediction models was 0.81 (95% confidence interval: 0.78–0.84), indicating fair discrimination. Conclusion Current evidence suggests that existing prediction models for PVT in patients with liver cirrhosis seem to demonstrate fair discriminatory performance. However, owing to substantial heterogeneity and limited validation and calibration, these models are not yet suitable for widespread clinical application. Future multicenter, prospective studies are therefore warranted.
No takes yet. Share an insight, caveat, or question.
Zhang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: