OR20-3 The Relationship of Obesity Severity, Triglycerides, and Elevated ALT Levels in Adolescents

Abstract

Abstract Introduction: Non-alcoholic fatty liver disease (NAFLD) is one of the leading causes of liver disease in children due to the rising prevalence of pediatric obesity, but remains understudied at a population level. This study characterizes the prevalence and demographic predictors of elevated alanine aminotransferase (ALT), a diagnostic marker of NAFLD, in a diverse cohort of obese children. We examined whether severity of obesity and presence of dyslipidemia were associated with elevated ALT among children who met BMI criteria for obesity. Methods: This cross sectional study used data from children aged 9-17y identified with obesity (BMI ≥95th percentile) at a well-child visit between 2012-2014 in an integrated health system. Lab data for ALT and fasting triglyceride (TG) measured within 1 year of the visit were examined, using ALT cutoffs that are 2x the upper normal limit specified by national pediatric specialty guidelines; 44 mg/dL for girls and 52 mg/dL for boys (JPGN 64:319, 2017). Obesity was classified as moderate (BMI 100-119% of 95th percentile, Class I) or severe (Class II and III: BMI 120-139% and ≥140% of 95th percentile). Results: Among 12,945 adolescents (mean age 13.0 ± 2.4); 54.8% were male, 19.7% white, 10.0% black, 49.0% Hispanic, 16.5% Asian, 4.9% other; 58.3% moderately obese and 41.7% severely obese (29.1% Class II and 12.6% Class III). The proportion with elevated ALT was 7.8% overall; higher in boys than girls (10.0% vs 5.0%), higher in Hispanic (9.3%) and Asian (9.5%) compared to white (5.4%) children, but lower in black (2.2%) children. Children with severe obesity were more likely to have elevated ALT compared to those with moderate obesity (11.2% vs 5.3%, p<0.001). Using multivariable logistic regression, adjusting for age, sex, and race, class II (OR 2.0, 95% CI 1.7-2.3) and class III (OR 3.1, 95% CI 2.6-3.7) obesity were associated with increased odds of elevated ALT compared to class I obesity. This graded relationship of obesity severity and elevated ALT persisted when adjusting for TG level. Elevated TG was an independent predictor of elevated ALT (OR 1.7, 95% CI 1.4-2.1 for TG 90-129; OR 2.9, 95% CI 2.4-3.4 for TG ≥130 compared to TG <90 mg/dL). In addition, male sex (OR 1.9, 95% CI 1.7-2.2), Hispanic (OR 1.7, 95% CI 1.4-2.1) and Asian (OR 1.9, 95% CI 1.5-2.3) race/ethnicity were associated with higher odds of elevated ALT; while black race (OR 0.4, 95% CI 0.2-0.6) was associated with lower odds of elevated ALT (adjusted model including age, sex, race/ethnicity, BMI). Conclusion: This study found interesting demographic differences in elevated ALT, notably an increased risk for Hispanic and Asian youth with obesity. An association between higher obesity severity and elevated ALT was also seen, even after adjusting for TG, as well as an association between elevated TG and ALT. These findings have important implications in defining at-risk phenotypes for pediatric metabolic syndrome.