Urinary thiodiglycolic acid is associated with increased risk of non-alcoholic fatty liver disease in children living near a petrochemical complex

Background: Occupational and experimental studies have revealed that high vinyl chloride monomer (VCM) exposure is associated with non-alcoholic fatty liver disease (NAFLD). Epidemiological study reported that children living near a petrochemical complex have elevated exposure levels of urinary thiodiglycolic acid (TDGA), a potential VCM biomarker. However, no studies on the association of urinary TDGA exposure with NAFLD in children are available. Aim: To assess the association of pediatric NAFLD with urinary TDGA exposure in school-aged children living near a petrochemical complex. Materials and methods: In total, 261 school-aged children (aged 6–13 years) living near a petrochemical complex were recruited during October 2013 to September 2014. First morning spot urine was sampled for analyzing urinary TDGA through liquid chromatography-tandem mass spectrometry. Ultrasonography and serum alanine aminotransferase (ALT) were examined in each participant. NAFLD was diagnosed as recommended by the North American and European Society of Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN and ESPGHAN). Risk for NAFLD with urinary TDGA exposure in children was evaluated using a multivariate logistic regression model. Results: The percentage of children with NAFLD^NASPGHAN and NAFLD^ESPGHAN were 9.6% and 11.5%, respectively. Median levels (μg/g creatinine) of urinary TDGA of children with NAFLD^NASPGHAN (vs non-NAFLD^NASPGHAN) and NAFLD^ESPGHAN (vs non-NAFLD^ESPGHAN) were 118.0 (vs 96.6) and 113.1 (vs 96.5), respectively. Participants in the highest urinary TDGA quartile (Q4: ≥160.0 μg/g creatinine) had a significantly increased risk (odds ratio [OR] = 4.95; 95% confidence interval [CI] = 1.15–21.38; P = 0.032) and dose-response trend (P_trend = 0.045) for NAFLD^NASPGHAN compared with those in the lowest urinary TDGA quartile (Q1: <35.4 μg/g creatinine) after adjustment for age, gender, BMI, triglycerides, HOMA-IR and distance of elementary schools from the petrochemical complex. Participants in the Q4 had borderline significantly increased risk (OR = 3.45; 95% CI = 0.89–13.42; P = 0.074) correlated with NAFLD^ESPGHAN compared with those in the Q1 after adjustment for confounders. Conclusion: Our findings support the hypothesis that children exposed to higher urinary TDGA levels significantly increased pediatric NAFLD risk. Serum ALT levels can be a useful predictor for screening children's NAFLD in field studies. Large and longitudinal studies are warranted to elucidate the association.