Increasing prevalence of nonalcoholic fatty liver disease among United States adolescents, 1988-1994 to 2007-2010

Abstract

Objective: To assess recent trends in nonalcoholic fatty liver disease (NAFLD) prevalence among US adolescents.

Study design: Cross-sectional data from 12 714 12-19 year olds (exclusions: chronic hepatitis, hepatotoxic medications) in the National Health and Examination Survey between 1988-1994 and 2007-2010 were used to estimate trends in suspected NAFLD, defined as overweight (body mass index ≥85th percentile) plus elevated alanine aminotransferase levels (boys >25.8 U/L; girls >22.1 U/L). Linear trends in prevalence and the independent effect of demographic indicators and adiposity on NAFLD risk were tested using regression models. Complex sampling methods and P values of <.05 were used to assess statistical significance.

Results: Suspected NAFLD prevalence (SE) rose from 3.9% (0.5) in 1988-1994 to 10.7% (0.9) in 2007-2010 (P < .0001), with increases among all race/ethnic subgroups, males and females, and those obese (P trend ≤.0006 for all). Among those obese, the multivariate adjusted odds of suspected NAFLD were higher with increased age, body mass index, Mexican American race, and male sex; the adjusted odds in 2007-2010 were 2.0 times those in 1988-1994. In 2007-2010, 48.1% (3.7) of all obese males and 56.0% (3.5) of obese Mexican American males had suspected NAFLD.

Conclusion: Prevalence of suspected NAFLD has more than doubled over the past 20 years and currently affects nearly 11% of adolescents and one-half of obese males. The rapid increase among those obese, independent of body mass index, suggests that other modifiable risk factors have influenced this trend.

Figure 1

Trends in the unadjusted prevalence of suspected NAFLD among US adolescents 12–19 years old (N = 12 714) defined using overweight plus various cutpoints for elevated ALT, including sex-specific (>25.8 U/L for boys and >22.1 U/L for girls) as well as alternative cutpoints of >40 U/L and >30 U/ L. P for linear trend <.0001 for all.

Figure 2

Trends in the adjusted prevalence of suspected NAFLD (overweight plus ALT >25.8 U/L for boys and >22.1 U/ L for girls), the prevalence of high waist circumference (>90th sex and age percentile), and mean BMI z-score among A, obese female and B, obese male adolescents in the NHANES III through NHANES 2007–2010. Estimates were adjusted for age and race/ethnicity. P for linear trend was <.0001 for the prevalence of NAFLD in both males and female and for the prevalence of high waist circumference among females. Trends in BMI z-score were not significant for either males or females.