Covert hepatic encephalopathy and spontaneous portosystemic shunts increase the risk of developing overt hepatic encephalopathy

Abstract

Aim: The aim of the study was to evaluate the presence of covert hepatic encephalopathy (cHE) and its characteristics according to the presence of spontaneous portosystemic shunts (SPSS) and their influence on the development of overt hepatic encephalopathy. Methods: Secondary analysis of a multicentre study, which evaluated the association between SPSS and complications of cirrhosis. The present study population includes those patients who also underwent cHE diagnostic evaluation. Presence of SPSS was evaluated by cross-sectional imaging and quantified by total SPSS-area. Logistic and Cox-regression competing risk analyses were performed. Results: About 65 patients were included of age 58 (IQR 50-66), MELD 15 (IQR 10-20), with alcoholic liver disease 63%. Thirty-two patients (49%) had cHE, had higher MELD [16 (IQR 12-24) vs 13 (IQR 9-17), P =.027], a greater proportion of SPSS [n = 18 (56%) vs n = 8 (24%); P =.008] and a higher total cross-sectional SPSS-area [28.3 (0-94.2) vs 0 (0-14.1); P =.005]. On multivariate analysis MELD [OR 1.11 (95% CI 1.01-1.21)] and presence of SPSS [OR 3.95 (95% CI 1.22-12.80)] were independently associated to cHE at baseline. During follow-up cHE was an independent predictor of oHE [cHE: HR 6.93 (95% CI 2.64-18.20). The effect of cHE on the development of oHE was greater in patients with SPSS [only cHE: HR 5.66 (95% CI 1.82-17.62), cHE and SPSS: HR 8.63 (95% CI 3.15-23.65)]. Conclusions: cHE is independently associated to the presence of SPSS (and total cross-sectional SPSS-area) and MELD. Furthermore, the presence of SPSS seems to increase the risk of cHE of developing of overt hepatic encephalopathy.