A Simple Diet Score Predicts Fatty Liver Risk in Overweight Kids

Nearly 4 in 10 overweight children enrolled in a Tehran obesity clinic had metabolic dysfunction-associated fatty liver disease (MAFLD), a condition once considered an adult problem. A cross-sectional study published in Health Science Reports by Kamrani F and colleagues tested whether a diet score designed to support gut microbial diversity could predict that risk, and the results suggest it can, at least among children who have already entered puberty.

MAFLD prevalence in overweight children reaches alarming levels

MAFLD is defined as hepatic steatosis, fat accumulation in the liver, occurring alongside metabolic disturbances such as overweight or obesity. It has become the most common chronic liver disease in young people, and the numbers are sobering. Global data on pediatric MAFLD prevalence put the figure at roughly 41% among children and adolescents with overweight or obesity, with wide variation by geography. A separate review pegged the rate at 47% among children with obesity specifically, framing it as one of the fastest-growing concerns in pediatric medicine.

In the Tehran cohort of 505 children and adolescents (mean age 10.0 years, 52.9% boys, all with a BMI-for-age Z-score of 1 or above), 38.8% met the diagnostic criteria for MAFLD on ultrasound. Children with MAFLD had significantly higher waist circumference, liver enzymes including ALT, AST, and GGT, fasting insulin, and insulin resistance scores compared to those without it.

How a gut microbiota diet score connects to liver fat

The gut-liver axis sits at the heart of this story. When the microbial community in the gut is disrupted, a state called dysbiosis, the intestinal barrier can become leaky. Bacterial products such as lipopolysaccharides then travel to the liver through the portal vein, triggering inflammation and insulin resistance. Reviews of gut microbiota composition in children with MASLD have documented reduced populations of beneficial bacteria and elevated pro-inflammatory strains compared to healthy controls, and broader research on the gut microbiome's role in pediatric liver disease points to altered microbial metabolites as a key driver.

The Dietary Index for Gut Microbiota (DI-GM), originally developed by Kase and colleagues using urinary biomarkers of microbial diversity, was designed to capture how well a person's diet supports a healthy gut ecosystem. It has 12 components: eight that score positively (fermented dairy, chickpeas, soybeans, whole grains, dietary fiber, cranberries, broccoli, and coffee) and four that penalize (refined grains, red meat, processed meat, and a diet where fat exceeds 40% of total energy). Scores run from 0 to 12; higher is better. The Iranian team adapted this index to a pediatric population using sex-specific median intakes from a validated 147-item food frequency questionnaire completed with parental help.

Each point increase in DI-GM cut MAFLD odds by 20%

Children with MAFLD scored significantly lower on the DI-GM than those without it: 6.4 versus 7.0 on average. In fully adjusted logistic regression models, each one-unit increase in DI-GM was associated with a 20% reduction in the odds of MAFLD (OR = 0.80). Comparing the top quartile of DI-GM scores to the bottom quartile, the odds dropped by 58% (OR = 0.42). A dose-response analysis using restricted cubic splines found the association was nonlinear, with the steepest decline in MAFLD odds occurring at the lower to mid-range of the score distribution, suggesting early dietary improvements may carry the greatest return.

Higher DI-GM scores tracked with greater intakes of whole grains, legumes, fruits, and vegetables, and with lower intakes of refined grains and red meat. These diets also delivered more fiber, calcium, iron, zinc, and magnesium. On the mechanistic side, plant-rich eating patterns are known to favor butyrate-producing bacteria such as Faecalibacterium prausnitzii and Roseburia, which help maintain the gut barrier and dampen inflammation.

Mediation analysis added texture to the finding. ALT explained 20.0% of the DI-GM-to-MAFLD association, GGT explained 19.2%, and AST explained 6.7%, meaning a meaningful share of the dietary effect runs through liver enzyme levels. Insulin resistance (HOMA-IR) showed a borderline significant mediation effect of 10.7%, while triglycerides and BMI-for-age Z-score did not mediate significantly.

Puberty, sex, and the threshold that matters most

Two subgroup findings complicate the picture. First, the inverse association between DI-GM and MAFLD was statistically significant only among pubertal children, not prepubertal ones. The authors suggest that pubertal hormonal shifts, particularly transient insulin resistance, may amplify the metabolic consequences of a poor diet during this window, making dietary quality more consequential precisely when many adolescents eat the worst. Second, there was a significant interaction by sex (p < 0.01), though the directionality is not fully resolved in this sample.

The team also proposed a practical threshold: a DI-GM score of 8 or higher may be needed to observe meaningfully lower MAFLD odds in overweight and obese children, a ceiling somewhat above what was observed in American adult populations. That gap hints at higher dietary demands during growth, or at the possibility that the index itself needs pediatric recalibration.

Open questions about causality and generalizability

This study has real strengths, including a large clinical sample, ultrasound-confirmed steatosis read by a blinded radiologist, detailed dietary assessment with caregiver involvement, and formal mediation analysis. But cross-sectional designs cannot establish causality: diet and liver disease were measured at the same point in time, so the direction of the relationship remains uncertain.

Other limits deserve mention. Self-reported dietary data carry recall bias even with parental help. Antibiotic use, which can reshape the microbiome substantially, was not assessed. Two DI-GM components, avocado and green tea, had to be imputed because local consumption was too low to calculate a meaningful median. The DI-GM itself was validated using urinary enterodiol and enterolactone as proxy biomarkers for microbial diversity, not direct microbiome sequencing, so it is a surrogate measure. And the sample was drawn entirely from overweight and obese children in Tehran, limiting how far findings can travel to leaner populations or different food cultures.

For now, dietary interventions remain the first-line clinical approach for pediatric fatty liver disease, given the absence of approved pharmacological options. This study adds a specific, scoreable tool to that conversation, one that clinicians and researchers could use to stratify dietary risk in young patients. Randomized trials testing whether deliberately raising DI-GM scores through structured diet programs can reverse hepatic steatosis in children are the logical next step.

Evidence rating

EMERGING — 80/100

Caveats:

• Cross-sectional design: this study shows association, not causation — diet was measured at a single point, so the direction of the relationship cannot be confirmed.
• Dietary intake was self-reported via food frequency questionnaire with parental involvement, which is susceptible to recall bias and misreporting.
• The DI-GM score uses surrogate biomarkers (urinary enterodiol and enterolactone) as proxies for gut microbial diversity rather than direct microbiome sequencing, so the gut microbiota link is inferred, not measured.
• The sample is limited to overweight and obese Iranian children enrolled in a national obesity registry, which may limit how broadly the findings apply to other populations or weight categories.
• Hepatic steatosis was diagnosed by ultrasound, not liver biopsy, meaning mild cases and fibrosis stage could not be assessed.
• The DI-GM index was originally validated in American adults; its calibration for a pediatric Iranian population has not been independently established.
• Antibiotic use — which can substantially alter the gut microbiome — was not assessed.
• Funding was not disclosed by the authors.

Cross-sectional study of 505 overweight and obese Iranian children and adolescents (aged 7–18). Identifies an inverse association between DI-GM dietary score and MAFLD odds, with partial mediation through liver enzymes; the design cannot establish causation. The DI-GM index has been validated in adult populations but not yet independently calibrated for pediatric cohorts, and gut microbiota diversity was inferred from dietary proxies rather than direct sequencing. Not yet replicated in an independent pediatric sample.

References

1. Farzam Kamrani, Ali Nikparast, Ali Tabatabaeyan, Elahe Etesami, Pejman Rohani, Golaleh Asghari (2026). Dietary Index for Gut Microbiota and the Odds of Metabolic Dysfunction‐Associated Fatty Liver Disease in Overweight and Obese Children and Adolescents: A Cross‐Sectional Study. Health Science Reports. https://doi.org/10.1002/hsr2.72640
2. Shuangzhen Jia, Xiaolin Ye, Tianwei Wu, Zhaoxia Wang, Jie Wu (2025). Global prevalence of metabolic dysfunction-associated fatty liver disease in children and adolescents with overweight and obesity: a systematic review and meta-analysis. BMC Gastroenterology. https://doi.org/10.1186/s12876-025-04314-y
3. Swathilakshmi Venu, Manu Raj (2026). Paediatric metabolic dysfunction-associated steatotic liver disease (MASLD): a growing health concern in the age of childhood obesity. BMJ Paediatrics Open. https://doi.org/10.1136/bmjpo-2025-003846
4. The association between gut microbiota and metabolic dysfunction-associated steatotic liver disease (MASLD): a case-control study. NIH. https://pubmed.ncbi.nlm.nih.gov/41068660/
5. Bhagyalakshmi Nair, Adithya Jayaprakash Kamath, Aswathy R. Devan, Rajesh Gopalakrishna, Ashok R. Unni, Lekshmi R. Nath (2026). Gut microbiota and pediatric metabolic dysfunction–associated steatotic liver disease: clinical evidence and therapeutic implications. Frontiers in Pediatrics. https://doi.org/10.3389/fped.2026.1773294
6. ManinderSingh Setia (2016). Methodology series module 3: Cross-sectional studies. Indian Journal of Dermatology. https://doi.org/10.4103/0019-5154.182410
7. Piper Sandel, Lawrence Ma, Helen Wang, Eric A. Pasman (2023). You Are What You Eat: A Review on Dietary Interventions for Treating Pediatric Nonalcoholic Fatty Liver Disease. Nutrients. https://doi.org/10.3390/nu15153350