Ready-to-use therapeutic food (RUTF) for home-based nutritional rehabilitation of severe acute malnutrition in children from six months to five years of age.

Systematic review and meta-analysis of randomized trials in children aged 6 months to 5 years with severe acute malnutrition, conducted in low- and middle-income countries including Malawi, India, Zambia, Kenya, Cambodia, Bangladesh, Sierra Leone, and Burkina Faso. Some children were stabilized in hospital before outpatient or home-based rehabilitation, and some studies included children with HIV infection.

Home-based ready-to-use therapeutic food (RUTF), generally an oral peanut-based therapeutic paste or comparable formulated RUTF, was used for rehabilitation of severe acute malnutrition in children 6 months to 5 years of age. Across trials, standard RUTF typically provided about 732 to 733 kJ/kg/day when intended to meet total daily nutritional requirements, while some studies tested RUTF as a supplement to the usual diet or compared modified RUTF formulations.

Overall, standard RUTF used to meet total daily nutritional requirements probably improves recovery from severe acute malnutrition and may increase weight gain, but effects on relapse and mortality remain uncertain. In pooled comparisons versus alternative dietary approaches, recovery improved in 6 RCTs (n=1852; RR 1.33, 1.16 to 1.54) and at the end of intervention (RR 1.41, 1.19 to 1.68), with weight gain favoring RUTF (MD 1.12 g/kg/day, 0.27 to 1.96). When RUTF was used as a supplement to the usual diet, recovery may improve and relapse may be reduced, but mortality and weight-gain effects were unclear. Across comparisons of different RUTF formulations, recovery was generally similar, with possible relapse reduction for standard RUTF, while mortality and weight gain remained uncertain.

The evidence base was often low or very low certainty, with substantial heterogeneity across trials, inconsistent outcome definitions, and variable reporting. Trials differed in formulation, dosing strategy, setting, and participant characteristics, limiting direct comparability and broader generalizability. Mortality and relapse were infrequently or imprecisely estimated, so several clinically important effects remain uncertain.