Impact of Disease-Specific Treatment and Non-Selective Beta Blockers on Risk of PVT in Cirrhotic Patients With HCV or PBC.

BACKGROUND

Portal vein thrombosis (PVT) is a common sequela of cirrhosis. Despite regression of fibrosis/cirrhosis observed among some patients after disease-specific treatment, few studies have considered the role of treatment and response on risk of PVT among patients with cirrhosis across different liver disease states. We considered aetiological treatment and response as well as whether non-selective beta-blockers (NSBBs), statins, and anticoagulants were associated with risk of PVT.

METHODS

We used data for patients with cirrhosis (compensated or decompensated) from two large, US-based multisite cohort studies. Patients with hepatitis C (HCV) were drawn from the Chronic Hepatitis Cohort Study. Patients with primary biliary cholangitis (PBC) were drawn from the Fibrotic Liver Disease Consortium. Risk for PVT was assessed using a discrete survival model that includes both fixed covariates and time-dependent variables.

RESULTS

Among 6659 HCV patients, 274 developed PVT across ~13 years of follow-up. Significant risk factors included time-varying decompensated cirrhosis (adjusted hazard ratio [aHR] 27.41, 95% confidence interval [CI] 15.89-47.30), male sex (aHR 1.48, 95% CI 1.12-1.94), and use of NSBBs (aHR 2.07, 95% CI 1.61-2.67). Among 786 PBC patients, 67 developed PVT across ~8 years of follow-up. Use of NSBBs was the only significant PVT risk factor in these patients (aHR 2.56, 95% CI 1.52-4.31). Neither sustained virological response [SVR] after antiviral treatment (HCV patients) nor response to ursodeoxycholic acid [UDCA] treatment (PBC patients) was associated with risk of PVT.

CONCLUSIONS

In two large well-characterized samples of cirrhotic patients with HCV or PBC, disease-specific treatments were not associated with PVT risk. NSBB treatment was independently associated with > 2 times the risk of PVT in both samples, regardless of compensated or decompensated status.