Placement of a transjugular intrahepatic portosystemic shunt (TIPS) is an effective treatment for portal hypertension. Overt hepatic encephalopathy (oHE) is a complication after TIPS associated with increased morbidity. Elevated ratio of plasma ammonia (AMM) levels compared to the local upper limit of normal (ULN) has been associated with oHE, hepatic complications and increased mortality in patients with cirrhosis without TIPS. The role of AMM in risk stratification of post-TIPS oHE is unclear. To investigate the role of AMM in the prediction of oHE in patients receiving TIPS placement. Patients with TIPS placement were recruited within a prospective observational study protocol with follow-up (FU) visits at 1, 3, 6, and 12 months after TIPS. Post hoc analyses of AMM levels for the association with the primary (oHE) and secondary endpoints (hepatic decompensation, infections, death/liver transplantation) during the first year after TIPS placement were performed. Of 188 patients with TIPS placement, 148 patients with available baseline AMM levels were included. During follow-up, 37% (55/148) of patients developed oHE. In multivariable competing risk analysis, baseline AMM/ULN (HR 2.03 CI 1.42-2.89, p = 0.001) and Freiburg index of post-TIPS survival (FIPS) score (HR 1.52 CI 1.03-2.24, p = 0.037) were independently associated with oHE. The published cut-off AMM/ULN > 1.4 showed comparable results (HR 2.40 CI 1.24-4.65, p = 0.01). AMM at FU1 was available in 100 patients, of whom 28% (28/100) developed oHE after FU1. In multivariable competing risk analysis, AMM/ULN (HR 5.48 CI 2.37-12.67, p < 0.001), psychometric hepatic encephalopathy score (HR 0.86 0.78-0.96, p = 0.005) and FIPS (HR 3.57 CI 1.79-7.14, p < 0.001) at FU1 were independently associated with oHE after FU1. No significant association between AMM/ULN and the secondary endpoints was detected. AMM levels before TIPS are independently associated with oHE after TIPS placement. AMM levels may serve as an additional marker for risk stratification of patients. Clinical trial number NCT04801290.
Sandmann et al. (Sat,) studied this question.