Usefulness of the Stroop test in predicting decompensation events in patients with cirrhosis: A prospective cohort study.

The Stroop test is used at the point of care to identify covert hepatic encephalopathy (CHE). However, it is unclear whether performance on the Stroop test can predict decompensation in patients with cirrhosis. This prospective cohort study included patients with cirrhosis admitted to the Gifu University Hospital in Japan. CHE was assessed using the Stroop test reference value determined by the Japan Society of Hepatology. Decompensation was defined as the onset of overt hepatic encephalopathy (OHE), bacterial infection, gastrointestinal bleeding, or esophageal varices. The factors associated with decompensation were assessed using the Fine–Gray competing risk regression model. Of 290 patients, CHE was present in 25 (8.6%). During a median follow-up period of 2.4 years, decompensation events occurred in 99 (34.1%) patients, including 51 (17.6%) bacterial infections, 49 (16.9%) cases of OHE, and 23 (7.9%) episodes of gastrointestinal bleeding or esophageal varices. The incidence of decompensation was significantly higher in patients with CHE than in those without (60.0% vs. 31.7%; p < 0.001). Multivariable analysis revealed that CHE (sub-distribution hazard ratio [SHR], 1.89; 95% confidence interval [CI], 1.09–3.28) was an independent predictor of decompensation events. CHE was also a significant factor for OHE (SHR, 3.47; 95% CI, 1.73–6.98) and bacterial infection (SHR, 2.57; 95% CI, 1.14–5.77). The Stroop test can predict decompensation events and stratify the risks of OHE and bacterial infections in patients with cirrhosis.