The association between dietary intake and urinary metabolites among children: A systematic review

Introduction: Traditional dietary assessment tools are prone to misclassification bias in children's diets due to misreporting. Using diet-related urinary biomarkers can provide more precise measurements, offering valuable insights into children's nutritional status and health. Objective: Our aim was to conduct a systematic review to identify the association between dietary intake and urinary metabolites among children. Methods: Two independent reviewers systematically searched five electronic medical databases and grey literature. Eligible studies included both intervention studies (randomized controlled trials (RCTs), quasi-experimental studies) and observational (cohort and cross-sectional) studies. Inclusion criteria comprised original research investigating associations between subjectively measured dietary intake and food-associated urinary metabolites in healthy children ≤18 years. Data were extracted and synthesized in a narrative review. Results: Fifty-four studies (12 RCTs, 1 quasi experimental, 20 cohort, 21cross-sectional) were included. Significant associations between dietary factors and their corresponding urinary metabolites were observed across most food-biomarker pairs, though effect sizes varied substantially (r=0.3–0.60) and several associations were weak to moderate in magnitude: animal-based foods and protein sources (n=15), plant-based foods and dietary patterns (n=14), sugars, sweeteners and caffeinated beverages (n=10), processed and ultra-processed foods (n=7), micronutrients. Multi-metabolite panels (hippurate + proline betaine + potassium) outperformed single biomarkers for Mediterranean dietary pattern assessment. The methodological approaches varied across study design, dietary assessment methods, urine collection protocols and metabolite analysis techniques. Conclusion: The findings suggest that urinary metabolomics is a promising yet modest objective biomarker tool for assessing dietary exposure in children, complementing traditional dietary assessment methods. Although associations between urinary biomarkers and self-reported dietary intake were weak to moderate, these estimates reflect measurement error in both methods. In this review, controlled feeding studies consistently showed stronger associations when dietary measurement error was minimized, suggesting that biomarker performance may be limited in part by the imprecision of self-reported comparators. Current evidence remains limited by heterogeneous methods and insufficient validation across diverse pediatric populations and age groups. Most urinary biomarkers applied in pediatric research have been adopted from adult validation studies. Therefore, further studies using standardized methodologies and objective dietary reference methods in age-appropriate populations are needed.