In 2025, Santos et al. conducted a study to analyze the characteristics of dietary intake in children and adolescents and investigate how diet differed between subjects with and without metabolic dysfunction-associated steatotic liver disease (MASLD).1

We recognize the value of the study’s results, which address the increasingly important topic of MASLD in the pediatric population. The authors’ focus on Latin American children and adolescents is a valuable contribution to the medical literature, given that previous studies were mostly conducted on Asian or European populations. We also highlight the fact that the authors included a multidimensional evaluation (clinical, biochemical, anthropometric, and dietary) and examined the role of micronutrients, such as zinc, whose relation to MASLD in children is not yet fully understood.

Despite these strengths, the study by Santos et al. has some weaknesses that may affect result interpretability. The study’s sample size is very small and for carrying out a multivariate analysis (21 cases of MASLD out of 89 patients) and appears insufficient for the multivariate model performed. By including four variables the study also fails to comply with the minimum rule of more than 10 events per variable proposed by Peduzzi et al., which may produce biased estimates.2

Another important consideration is that the dietary evaluation was based on only two 24-hour dietary recalls (Tuesday and Sunday), which does not enable an accurate estimation of customary intake in the pediatric population. Studies, such as that by Padilha et al., suggest that more than 7 days are needed for adequately evaluating different types of nutrients, making a two-day assessment prone to measurement bias.3 The study by Santos et al. does not specify the recruitment period or exact dates of participant inclusion. In addition, according to its results, the variables analyzed (LDL cholesterol, zinc, and protein intake) were not associated with the development of MASLD. However, several studies demonstrate biologic plausibility and support their direct link to MASLD,4,5 signifying that said lack of association could be the result of limited statistical power and dietary measurement errors.