Age- and trimester-specific liver function test reference intervals among pregnant and non-pregnant women in Nairobi City, Kenya

Background Accurate interpretation of liver function tests (LFTs) in women requires population- and pregnancy-specific reference intervals (RIs). In Kenya, diagnostic laboratories apply non-local RIs, risking clinical misclassification and potentially leading to unnecessary investigations or missed diagnoses. This study established age- and trimester-specific RIs for healthy pregnant and non-pregnant women in Nairobi City, Kenya. Methods A descriptive cross-sectional study was conducted among 1,142 pregnant women (trimester 1: n=321; trimester 2: n=499; trimester 3: n=322) and 343 non-pregnant women aged 19–40 years. Cohorts were partitioned into 19–29, and 30–40 year groups. Reference limits 2.5 th –97.5 th percentiles) were established using non-parametric methods. Between-group differences were evaluated using Kruskal-Wallis and Mann-Whitney U tests with Bonferroni correction (ρ < 0.0083). Out-of-range proportions were calculated using current practice RIs. Results Pregnancy significantly altered most LFT parameters. Relative to non-pregnant women, pregnant women had lower ALT, AST, GGT, and total bilirubin levels (ρ < 0.001), while ALP increased progressively across trimesters (ρ < 0.001). Age-related difference were observed for selected parameters, particularly albumin (ALB), AST, and direct bilirubin. Application of current practice RIs resulted in substantial misclassification, especially for ALB (combined trimesters: 56.7% out-of-range) and ALT (30.9%). Conclusion LFT reference values differs significantly by age, pregnancy status, and trimester in this population. Continued use of non-local or non-pregnancy-adjusted RIs may lead to considerable clinical misclassification. Adoption of these locally established, age- and trimester-specific RIs could enhance diagnostic accuracy and clinical decision-making for women in Nairobi and similar settings.