Aim The aim of this study was to identify the role of factor VIII (FVIII) in portal vein thrombosis (PVT) occurrence in cirrhotic patients with gastroesophageal variceal bleeding.

Methods A total of 453 cirrhotic patients with gastroesophageal varices were enrolled. Computed tomography was performed at baseline and patients were divided into PVT and non-PVT groups (n = 131 vs. 322). Individuals without PVT at baseline were followed up for the development of PVT. Time-dependent receiver operating characteristic analysis of FVIII for PVT development was performed. The Kaplan–Meier methodology was used to analyze the predictive ability of FVIII for PVT incidence at 1 year.

Results FVIII activity (177.00 vs. 153.70, p = 0.001) was significantly increased in the PVT group compared with the non-PVT group in cirrhotic patients with gastroesophageal varices. FVIII activity was positively correlated with the severity of PVT (161.50 vs. 171.07 vs. 187.05%, p = 0.001). Furthermore, FVIII activity (hazard ratio [HR]: 3.48, 95% confidence interval [CI]: 1.14–10.68, p = 0.029 in model 1; HR: 3.29, 95% CI: 1.03–10.51, p = 0.045 in model 2) was an independent risk factor of 1-year PVT development in patients without PVT at baseline, which was confirmed by two separate Cox regression analysis and competing risk models. Patients with elevated FVIII activity exhibit a higher incidence of PVT in the non-PVT group at 1 year (15.17 vs. 3.16%, p < 0.001). The predictive value of FVIII remains significant in individuals who have never received splenectomy (14.76 vs. 3.04%, p = 0.002).

Conclusion Elevated FVIII activity was potentially associated with the occurrence and the severity of PVT. It might be helpful to identify cirrhotic patients at risk of PVT.