Risk factors for exacerbation of gastroesophageal varices and portosystemic encephalopathy during treatment with nucleos(t)ide analogs for hepatitis B virus‐related cirrhosis

AIM: The aim of this study was to determine the risk factors for worsening of gastroesophageal varices (GEVs) and development of portosystemic encephalopathy in patients with hepatitis B virus (HBV)-related cirrhosis during nucleos(t)ide analog (NA) treatment. METHODS: One hundred and thirty-seven patients with HBV-related cirrhosis were enrolled in this retrospective cohort study. Findings of portal hemodynamics with computed tomography, liver function, and endoscopic examinations during NA treatment were assessed. RESULTS: Among 137 patients, feeding vessels for GEVs (left gastric vein, posterior gastric vein, and short gastric vein) were present in 56 (41%) patients, and extrahepatic portosystemic shunt (paraesophageal vein, paraumbilicul vein, and splenorenal shunt) were present in 36 (26%) patients at the start of NA treatment. Although NA treatment was successful, significant improvements were not observed in portosystemic collateral vessels 3 years after NA treatment and GEVs were exacerbated in 48 (35%) patients. The cumulative 5- and 10-year exacerbation rate of GEVs was 27% and 50%, respectively. By multivariate analysis, the existence of feeding vessels for GEVs at the start of NA treatment was the independent predictive factor for the exacerbation of GEVs (P < 0.001). Eight patients who had extrahepatic portosystemic shunt at the start of NA treatment developed portosystemic encephalopathy during follow-up. The 3- and 5-year incidence of that was 5% and 8%, respectively. CONCLUSIONS: The presence of portosystemic collateral vessels at the start of NA treatment increases the risk of GEVs worsening and development of portosystemic encephalopathy in patients with HBV-related cirrhosis, despite improvement of liver function and success in reducing viral loads with NA treatment.