Extensive testing or focused testing of patients with elevated liver enzymes

Abstract

Background & aims: Many patients have elevated serum aminotransferases reflecting many underlying conditions, both common and rare. Clinicians generally apply one of two evaluative strategies: testing for all diseases at once (extensive) or just common diseases first (focused).

Methods: We simulated the evaluation of 10,000 adult outpatients with elevated with alanine aminotransferase to compare both testing strategies. Model inputs employed population-based data from the US (National Health and Nutrition Examination Survey) and Britain (Birmingham and Lambeth Liver Evaluation Testing Strategies). Patients were followed until a diagnosis was provided or a diagnostic liver biopsy was considered. The primary outcome was US dollars per diagnosis. Secondary outcomes included doctor visits per diagnosis, false-positives per diagnosis and confirmatory liver biopsies ordered.

Results: The extensive testing strategy required the lowest monetary cost, yielding diagnoses for 54% of patients at $448/patient compared to 53% for $502 under the focused strategy. The extensive strategy also required fewer doctor visits (1.35 vs. 1.61 visits/patient). However, the focused strategy generated fewer false-positives (0.1 vs. 0.19/patient) and more biopsies (0.04 vs. 0.08/patient). Focused testing becomes the most cost-effective strategy when accounting for pre-test probabilities and prior evaluations performed. This includes when the respective prevalence of alcoholic, non-alcoholic and drug-induced liver disease exceeds 51.1%, 53.0% and 13.0%. Focused testing is also the most cost-effective strategy in the referral setting where assessments for viral hepatitis, alcoholic and non-alcoholic fatty liver disease have already been performed.

Conclusions: Testing for elevated liver enzymes should be deliberate and focused to account for pre-test probabilities if possible.

Lay summary: Many patients have elevated liver enzymes reflecting one of many possible liver diseases, some of which are very common and some of which are rare. Tests are widely available for most causes but it is unclear whether clinicians should order them all at once or direct testing based on how likely a given disease may be given the patient's history and physical exam. The tradeoffs of both approaches involve the money spent on testing, number of office visits needed, and false positive results generated. This study shows that if there are no clues available at the time of evaluation, testing all at once saves time and money while causing more false positives. However, if there are strong clues regarding the likelihood of a particular disease, limited testing saves time, money and prevents false positives.

Keywords: Alcohol; Cost-effectiveness; Hepatitis C; Liver disease; Liver function tests; NAFLD; Serum aminotransferases.