Association of bright liver with the PNPLA3 I148M gene variant in 1-year-old toddlers

Context: Nonalcoholic fatty liver disease (NAFLD) is being increasingly diagnosed at younger ages, pointing toward an early-life origin. Objective: To evaluate the frequency and risk factors for bright liver (BL) in 1-year-old toddlers. Design: Secondary analysis of the 1-year follow-up of the Feeding Study. Exposures were child PNPLA3 and TM6SF2 gene variants; child anthropometry at birth and at 1 year of follow-up; child subcutaneous, visceral, and epicardial adipose tissue at 1 year of follow-up; maternal anthropometry at the start and at the end of pregnancy; and maternal red blood cell fatty-acid composition at the third trimester of pregnancy. Setting: General population. Participants: Among 505 mother-toddler pairs, 391 children (77%) underwent liver and abdominal ultrasonography at the 1-year follow-up. Main Outcome: BL as diagnosed by ultrasonography. Results: Seventeen (4%) of 391 toddlers had BL. Compared with the toddlers with the PNPLA 3 CC genotype, the odds (95% CI) of BL were 3.01 (1.05 to 8.64, P, 0.05) times higher in those with the PNAPLA3 CG genotype and 5.37 (1.12 to 25.77, P, 0.05) higher in those with the PNPLA3 CC genotype. We found no association between BL status and TM6SF2. Body weight, body mass index, and maternal weight gain during pregnancy were higher in BL+ than in BL2 children. Visceral adipose tissue was higher but subcutaneous adipose tissue and epicardial adipose tissue were similar in BL+ and BL2 children. Conclusions: Four percent of the Feeding Study children had BL at 1 year of age. In line with expectations, PNAPLA3 was already a predictor of BL at this early age.