Abstract
BACKGROUND: Protein intake during complementary feeding plays a vital role in childhood growth. However, the effects of varying protein quantity and source on obesity and related metabolic outcomes remain uncertain. OBJECTIVES: This study aimed to evaluate the effect of protein intake (quantity and source) on childhood obesity [weight-for-length Z-score (WLZ) > 2], body composition [fat mass (FM), fat-free mass (FFM)], and metabolic markers (insulin, insulin-like growth factor-1) in children aged 6-23 mo, including outcomes beyond 24 mo where available. METHODS: A systematic search was conducted in MEDLINE, Cochrane CENTRAL, Embase, Web of Science, and CINAHL (January 2000-May 2024) for randomized controlled trials that compared high with low protein intake (diets with higher protein defined as an absolute difference of ≥5 g/d between groups), animal compared with plant-based protein, and meat compared with dairy. Random-effects meta-analyses were performed, and heterogeneity was assessed using I(2) statistics. Owing to the lack of binary obesity outcome data, WLZ was analyzed as a continuous measure to assess shifts in weight-for-length distribution. RESULTS: Of 5817 records identified, 20 publications (from 19 trials) met the inclusion criteria, with 12 included in the meta-analysis. Diets with higher protein content did not substantially affect WLZ. Standardized mean difference (SMD) was 0.0 (95% CI: -0.09, 0.09) for low- and middle-income countries and 0.17 (95% CI: -0.12, 0.47) for high-income countries. No effects were found for FFM [SMD: -0.05 (95% CI: -0.31, 0.21)] or FM [SMD: 0.17 (95% CI: -0.41, 0.74)]. We were not able to identify any effects of protein source of any of the outcomes. Insufficient data precluded meta-analysis for metabolic markers. CONCLUSIONS: There is limited evidence of any impact of protein quantity or quality during complementary feeding on growth or body composition. Given the short-lasting exposures and follow-up times, the long-term metabolic effects of protein supplementation require further investigation. This trial was registered at PROSPERO as CRD42024550409.