Predictors and prognostic impact of spontaneous portal vein thrombosis recanalization in cirrhosis: An individual patient data meta-analysis

Background: Portal vein thrombosis (PVT) in cirrhosis has a quite unpredictable natural history. Spontaneous recanalization may occur, but predictors and prognostic impact remain uncertain. We performed an individual patient data (IPD) meta-analysis to evaluate survival according to spontaneous recanalization and to identify predictors of recanalization. Methods: We included cohort studies of patients with cirrhosis and PVT not receiving anticoagulation and obtained de-identified IPD. All-cause mortality was analyzed using a competing-risk model (competing event: liver transplantation) adjusted for prespecified baseline covariates. Predictors of recanalization were assessed with mixed-effects ordinal logistic regression including a random study intercept. Results: Eight studies contributed 586 patients. Spontaneous recanalization occurred in 264 (47.7%). Overall recanalization was independently associated with lower mortality (aSHR 0.51; 95% CI 0.28–0.90), with the benefit driven by complete recanalization (aSHR 0.45; 95% CI 0.21–0.94). Recanalization of PVT localized in the main portal vein trunk was linked to a strong survival benefit (interaction SHR 0.39; 95% CI 0.16–0.91; p =0.02), whereas no benefit was observed after recanalization of isolated intrahepatic branch thrombosis. PVT recanalization was more likely to occur in Child–Pugh A (aOR 2.63; 95% CI 1.18–5.88), partial thrombosis (aOR 1.69; 95% CI 1.09–2.56), and isolated intrahepatic location (aOR 2.04; 95% CI 1.23–3.33). Conclusions: Spontaneous recanalization occurs in approximately half of patients with cirrhosis and PVT and is associated with improved survival, particularly when complete and involving the main portal vein trunk. These findings underscore its prognostic relevance and indirectly support anticoagulation to promote recanalization.