From Hepatic Enzyme Alterations to Drug Metabolic Dysfunction: A Systematic Review of Prolonged Pharmacotherapy

Abstract

Background: Prolonged exposure to medicines may produce laboratory abnormalities, clinically significant hepatic injury, or changes in hepatic drug-metabolizing enzyme activity. These phenomena overlap but are not interchangeable. Objective: To critically synthesize human evidence connecting repeated or extended pharmacotherapy with hepatic enzyme alterations, liver injury and altered drug disposition. Methods: A structured search of PubMed/MEDLINE-indexed records, accessible full-text articles and publisher databases identified clinical trials, longitudinal cohorts, comparative studies, pharmacokinetic experiments and a chronic-exposure histopathological case series. The PRISMA selection accounting presents 811 records identified, 164 duplicates removed, 647 records screened, 120 full-text reports assessed and 14 primary reports included in the domain-based narrative synthesis. Findings were organized into biochemical injury, clinically significant hepatotoxicity, enzyme induction, enzyme inhibition and cumulative structural damage; reported effect estimates and denominators were preserved. Results: In a randomized trial of 4,786 adults, liver-test elevations occurred in 43.6% with low-dose methotrexate versus 24.1% with placebo (hazard ratio 2.14, 95% confidence interval 1.93-2.37). In a 213-person methotrexate cohort, 21% developed alanine aminotransferase >1.5 times the upper limit of normal over a mean 4.3 years. A study of 1,928 tuberculosis-treatment recipients recorded drug-induced liver injury in 3.0%, whereas a randomized latent-tuberculosis trial found grade 3-4 hepatitis in 16/422 isoniazid versus 3/418 rifampin recipients. Conversely, in 232,889 propensity-matched statin initiators in each baseline ALT stratum, severe acute liver injury was uncommon, with a pooled hazard ratio of 1.15 (95% confidence interval 0.75-1.75) comparing elevated versus normal baseline ALT. CYP3A functional studies found oral midazolam exposure only 5.7% of control values in patients taking carbamazepine or phenytoin, but 12.4-fold greater after 14 days of ritonavir-boosted saquinavir. Conclusion: Hepatic biochemical abnormalities neither universally indicate hepatocellular dysfunction nor predict a uniform change in drug clearance. Extended treatment can cause hepatotoxicity, produce adaptive enzyme induction or inhibition, or, less commonly, lead to structural hepatic disease. Drug-specific longitudinal monitoring and pharmacokinetic assessment are complementary.