Relevance of spontaneous portosystemic shunts detected with CT in patients with cirrhosis

Background: Cirrhosis leads to portal hypertension and to the consequent formation of spontaneous portosystemic shunts (SPSSs), leading to complications related to the diversion of portal blood into the systemic circulation, which is called portosystemic shunt syndrome. Purpose: To investigate the characteristics of patients with cirrhosis and an SPSS and secondarily to assess the prognostic impact of SPSSs on portal hypertension–related complications and transplant-free survival. Materials and Methods: A retrospective database review of patients with cirrhosis (observed from March 2015 to July 2019) was performed to identify patients with CT imaging and outcomes data. For each patient, clinical and biochemical data were collected, and the presence, types, and sizes of SPSSs were investigated with CT. Patients were followed for a mean of 27.5 months 6 22.8. Multivariable logistic analysis was used to identify the clinical characteristics associated with the presence of SPSSs (any size) and presence of SPSSs 1 cm or larger. Competitive risk analysis (Fine and Gray model) was used to identify the association between SPSSs and complications and mortality. Results: Two hundred twenty-two patients with cirrhosis (157 male, 65 female; mean age, 62 years 6 12 [standard deviation]) were evaluated. An SPSS was found in 141 of 222 patients (63.5%), and 40 of 222 (18%) had a shunt diameter of at least 1 cm. At presentation, variables independently associated with the presence of SPSSs (any size) were portal vein thrombosis (odds ratio, 5.5; P = .008) and Child-Pugh class C (odds ratio, 3.0; P = .03). Previous hepatic encephalopathy (odds ratio, 4.4; P = .001) and portal vein thrombosis (odds ratio, 5.3; P = .001) were the only variables associated with SPSSs larger than 1 cm. Patients with SPSSs of any size had higher mortality (subdistribution hazard ratio, 1.9; P , .001) and higher frequency of hepatic encephalopathy (subdistribution hazard ratio, 2.3; P = .023), gastrointestinal bleeding (subdistribution hazard ratio, 2.9; P = .039), and portal vein thrombosis (subdistribution hazard ratio, 7.6; P = .005). Conclusion: The presence of spontaneous portosystemic shunts on CT images in patients with cirrhosis was associated with higher mortality and complications, including portal vein thrombosis, hepatic encephalopathy, and gastrointestinal bleeding.