Importance: Dementia and hepatic encephalopathy (HE) are challenging to distinguish clinically. Undiagnosed cirrhosis in a patient with dementia can lead to missed opportunities to treat HE.
Objective: To examine the prevalence and risk factors of undiagnosed cirrhosis and therefore possible HE in veterans with dementia.
Design, setting, and participants: A retrospective cohort study was conducted between 2009 and 2019 using data from the Veterans Health Administration (VHA) and 2 separate validation cohorts from the Richmond Veterans Affairs Medical Center. Data analysis was conducted from May 20 to October 15, 2023. Participants included 177 422 US veterans with a diagnosis of dementia at 2 or more clinic visits, no prior diagnosis of cirrhosis, and with sufficient laboratory test results to calculate the Fibrosis-4 (FIB-4) score.
Exposures: Demographic and clinical characteristics.
Main outcomes and measures: An FIB-4 score (>2.67 suggestive of advanced fibrosis and >3.25 suggestive of cirrhosis), capped at age 65 years even for those above this cutoff who were included in the analysis.
Results: Among 177 422 veterans (97.1% men; 80.7% White; mean (SD) age, 78.35 [10.97] years) 5.3% (n = 9373) had an FIB-4 score greater than 3.25 and 10.3% (n = 18 390) had an FIB-4 score greater than 2.67. In multivariable logistic regression models, FIB-4 greater than 3.25 was associated with older age (odds ratio [OR], 1.07; 95% CI, 1.06-1.09), male gender (OR, 1.43; 95% CI, 1.26-1.61), congestive heart failure (OR, 1.48; 95% CI, 1.43-1.54), viral hepatitis (OR, 1.79; 95% CI, 1.66-1.91), Alcohol Use Disorders Identification Test score (OR, 1.56; 95% CI, 1.44-1.68), and chronic kidney disease (OR, 1.11; 95% CI, 1.04-1.17), and inversely associated with White race (OR, 0.79; 95% CI, 0.73-0.85), diabetes (OR, 0.78; 95% CI, 0.73-0.84), hyperlipidemia (OR, 0.84; 95% CI, 0.79-0.89), stroke (OR, 0.85; 95% CI, 0.79-0.91), tobacco use disorder (OR, 0.78; 95% CI, 0.70-0.87), and rural residence (OR, 0.92; 95% CI, 0.87-0.97). Similar findings were associated with the FIB-4 greater than 2.67 threshold. These codes were associated with cirrhosis on local validation. A local validation cohort of patients with dementia showed a similar percentage of high FIB-4 scores (4.4%-11.2%).
Conclusions and relevance: The findings of this cohort study suggest that clinicians encountering patients with dementia should be encouraged to screen for cirrhosis using the FIB-4 score to uncover reversible factors associated with cognitive impairment, such as HE, to enhance outcomes.