Adolescent Nonalcoholic Fatty Liver Disease and Type 2 Diabetes in Young Adulthood

• Published in: The journal of clinical endocrinology and metabolism 2021-01, Vol.106 (1), p.e34-e44
• Main Authors: Bardugo, Aya, Bendor, Cole D, Zucker, Inbar, Lutski, Miri, Cukierman-Yaffe, Tali, Derazne, Estela, Mosenzon, Ofri, Tzur, Dorit, Beer, Zivan, Pinhas-Hamiel, Orit, Ben-Ami, Michal, Fishman, Boris, Ben-Ami Shor, Dana, Raz, Itamar, Afek, Arnon, Gerstein, Hertzel C, Häring, Hans-Ulrich, Tirosh, Amir, Levi, Zohar, Twig, Gilad
• Format: Article
• Language: English
• Subjects: Adolescence; Adolescents; Biopsy; Body mass index; Comorbidity; Diabetes; Diabetes mellitus (non-insulin dependent); Fatty liver; Liver diseases; Population studies; Sensitivity analysis; Teenagers; Type 2 diabetes
• ISSN: 0021-972X; 1945-7197
• DOI: 10.1210/clinem/dgaa753

Abstract Context The long-term risk of type 2 diabetes in adolescents with nonalcoholic fatty liver disease (NAFLD) is unclear. Objective To assess type 2 diabetes risk among adolescents with NAFLD. Design and Setting A nationwide, population-based study of Israeli adolescents who were examined before military service during 1997–2011 and were followed until December 31, 2016. Participants A total of 1 025 796 normoglycemic adolescents were included. Interventions Biopsy or radiographic tests were prerequisite for NAFLD diagnosis. Data were linked to the Israeli National Diabetes Registry. Main Outcome Measures Type 2 diabetes incidence. Results During a mean follow-up of 13.3 years, 12 of 633 adolescents with NAFLD (1.9%; all with high body mass index [BMI] at baseline) were diagnosed with type 2 diabetes compared with 2917 (0.3%) adolescents without NAFLD. The hazard ratio (HR) for type 2 diabetes was 2.59 (95% confidence interval [CI], 1.47–4.58) for the NAFLD vs. the non-NAFLD group after adjustment for BMI and sociodemographic confounders. The elevated risk persisted in several sensitivity analyses. These included an analysis of persons without other metabolic comorbidities (adjusted HR, 2.75 [95% CI, 1.48-5.14]) and of persons with high BMI; and an analysis whose outcome was type 2 diabetes by age 30 years (adjusted HR, 2.14 [95% CI, 1.02-4.52]). The results remained significant when a sex-, birth year-, and BMI-matched control group was the reference (adjusted HR, 2.98 [95% CI, 1.54-5.74]). Conclusions Among normoglycemic adolescents, NAFLD was associated with an increased adjusted risk for type 2 diabetes, which may be apparent before age 30 years.