Letter to the editor: Association between birth weight, preterm birth, and nonalcoholic fatty liver disease in a community-based cohort.

Abstract

To the editor, While we commend Amadou et al.1 for their fine study reporting a U-shaped association between birth weight and risk of NAFLD, we have some concerns. NAFLD was defined using the fatty liver index. This index was, however, developed as a risk stratification tool (not as a diagnostic instrument) and has a rather low positive predictive value for the diagnosis of NAFLD (67%).2 Furthermore, the fatty liver index may perform poorly in young individuals with normal body mass index,3 as in the current French study population (mean age 38 y, normal body mass index).1 Given that the fatty liver index formula is largely driven by body mass index and waist circumference, the question arises whether the found associations rather explain a higher risk of obesity and would especially miss individuals with lean NAFLD. While in women, there was a positive association between high birth weight and risk of NAFLD, the mediation analysis revealed that in men high birth weight did neither have a direct nor a total effect on the later risk of NAFLD—in contrast the indirect effect seemed even protective. Against this background, how do the authors explain the contradictive results on high birth weight and NAFLD risk in men andwomen? The authors found a slight negative association between birth weight and the Forns index, which was used as proxy for liver fibrosis. Here, we see two limitations that make the interpretation of these results challenging: first, the Forns index was originally developed to noninvasively rule-out significant fibrosis and has not been validated to be used as a continuous variable. Second, it was derived in high-risk cohorts of patients with chronic hepatitis C infection,4 while the current study applied it in a general population setting. Hence, we are eager to know if the authors could confirm these results using FIB-4 instead, which is recommended by international guidelines.5