Background & aims: Hepatic encephalopathy (HE), one of the most serious prognostic factors for mortality in alcohol-related cirrhosis (ALD cirrhosis), is not recorded in Danish healthcare registries. However, treatment of HE with lactulose, the universal first-line treatment, can be identified through data on filled prescriptions. This study aimed to investigate if lactulose can be used as a surrogate marker of HE. Methods: We used Danish healthcare registries to establish a cohort of patients with ALD cirrhosis diagnosed in 2000–2018. Lactulose users were identified using data on filled prescriptions. We computed the prevalence and cumulative incidence of lactulose use, and then matched lactulose users with cohort members who were not using lactulose. This matched dataset was used to identify predictors of lactulose initiation, and to examine the association between lactulose use and all-cause mortality. Results: We included 23,089 patients, among whom we identified 4,789 first-time lactulose users. The prevalence of lactulose usage rose to 11% within the first 6 months after ALD diagnosis and reached 19% 5 years after inclusion. The 1-year cumulative incidence of lactulose use was 31%, and predictors of initiating lactulose use were history of severe liver disease (ascites, portal hypertension, spontaneous bacterial peritonitis or gastrointestinal bleeding) or hepatocellular carcinoma. Lactulose use was associated with higher mortality (adjusted hazard ratio 1.61 [95% confidence interval 1.53; 1.69]). Conclusions: Lactulose is used by 10–20% of patients with ALD cirrhosis, primarily those with severe cirrhosis, and lactulose users have a markedly higher mortality than nonusers. We conclude that lactulose use can be used to estimate the prevalence of HE and to reduce confounding from HE, but it should not be used as a surrogate for HE in studies with HE as an outcome.
Read full abstract