Background: The changes of plasma retinol binding protein 4 (RBP4) level after a nutrition intervention can indicate the metabolic changes associated with the delivered intervention. Objective: We investigated the changes in plasma RBP4 level among 12- to 18-month-old children after a nutrition intervention and measured its association with subcutaneous adiposity, maternal characteristics, and inflammation. Methods: Data of 520 undernourished children (250 of them had length-for-age Z score LAZ <−1 to −2 and 270 had LAZ score <−2) were collected from the Bangladesh Environmental Enteric Dysfunction study conducted in Dhaka, Bangladesh. Multivariable linear regression and generalized estimation equations (GEE) modeling techniques were used to measure the association. Results: At baseline, median RBP4 level was 19.9 mg/L (interquartile range IQR: 7.96), and at the end of the intervention, it was 20.6 mg/L (IQR: 9.06). Percentage changes in plasma RBP4 level were not significantly associated (P > .05) with the percentage changes in child’s height, weight, and subcutaneous adiposity. But maternal height (regression coefficient, β = −1.62, P = .002) and milk intake (β = −0.05, P = .01) were negatively and maternal weight was positively associated (β = 0.56, P = .03) with the changes in RBP4 levels. The GEE models revealed negative association of RBP4 levels with C-reactive protein (CRP; β = −0.14, P < .05) and α-1-acid glycoprotein (AGP; β = −0.03, P < .05). Conclusion: Children whose mothers were taller experienced less increase in plasma RBP4 level, and children whose mothers had a higher weight experienced more increase in the RBP4 level from baseline. We have also found that CRP and AGP levels and intake of whole milk were negatively associated with the plasma RBP4 level.