Elevated liver enzyme (AST and ALT) as biomarkers for severe dengue in Nepalese patients: a cross-sectional study

Dengue remains a major global health concern, with rising incidence and severity in endemic regions. Liver involvement, particularly elevated transaminases such as aspartate aminotransferase (AST) and alanine aminotransferase (ALT), is frequently observed during infection. However, previous studies report inconsistent associations between transaminase elevation and severe dengue, particularly across diverse geographical settings. This study investigates the relationship between liver enzyme elevation and dengue severity in Nepalese patients, leveraging a pre-existing dataset to assess their potential as biomarkers for severe disease.

A cross-sectional study was conducted among 325 laboratory-confirmed dengue patients at a four tertiary care hospitals in Koshi province Nepal. Liver enzyme levels (AST and ALT) were categorized based on the upper limit of normal (AST > 48 U/L, ALT > 55 U/L), and dengue severity was classified according to the WHO 2009 guidelines. Associations between elevated transaminases and dengue with warning signs (DWS) were evaluated using Chi-square or Fisher's exact test and multivariable logistic regression, adjusted for age and sex. Spearman's rank correlations were performed to assess associations between liver enzyme level, clinical features, and disease severity. Multicollinearity was assessed using the variance inflation factor (VIF).

Among the 325 dengue patients, 40.0% had dengue with warning signs (DWS). Elevated AST was significantly associated with DWS (aOR = 2.40, 95% CI: 1.49-3.86, p = 0.0003), while elevated ALT showed no significant association. AST elevation was also correlated with gastrointestinal symptoms, such as nausea (ρ = 0.243, p < 0.0001) and vomiting (ρ = 0.151, p = 0.0064), as well as with key laboratory markers including hemoglobin (ρ = 0.407, p < 0.0001) and platelet count (ρ = 0.277, p < 0.0001). In contrast, ALT showed weaker and inconsistent correlations. Adjusted analyses confirmed AST as an independent predictor of severe dengue. Female identified as a novel risk factor for DWS, independent of age and AST levels. The logistic regression model demonstrated good fit (McFadden's R² = 0.074), with no evidence of multicollinearity detected.

Elevated AST as a significant independent predictor of dengue with warning signs, and may serve as a valuable biomarker for early risk stratification. Additionally, female sex emerged as an independent risk factor for severe disease, a novel finding with potential implications for sex-specific risk stratification. The findings highlight the importance of liver function monitoring in dengue management. Despite strengths in structured assessment and comprehensive analysis, limitations including the cross-sectional design, convenience sampling, and fewer-center setting limit causal inference and generalizability. Future studies should validate these findings through prospective, multi-center research and explore sex-based risk profiles and the prognostic utility of dynamic liver enzyme monitoring.