Association between anthropometric criteria and body composition among children aged 6–59 months with uncomplicated severe acute malnutrition: A prospective cohort study from South Sudan

Background Severe acute malnutrition (SAM) is diagnosed based on bilateral edema, mid‑upper arm circumference (MUAC), and/or weight‑for‑height z‑score (WHZ). These criteria do not fully overlap and may identify children with distinct biological and clinical profiles. Understanding how children with SAM identified by different criteria differ in body composition during outpatient therapeutic program (OTP) treatment may help inform treatment and post‑discharge practices. Methods We conducted a prospective cohort study of children aged 6–59 months with uncomplicated SAM admitted by MUAC only (MUAC group), WHZ only (WHZ group), or combined MUAC and WHZ criteria (Combined group) in Aweil Centre, South Sudan. Anthropometry, triceps skinfold thickness, and bioelectrical impedance analysis were collected at admission (n = 194) and discharge (n = 127). Fat mass (FM), fat‑free mass (FFM), fat mass index (FMI), fat‑free mass index (FFMI), total body water percentage (TBW), and phase angle (PhA) were assessed. Between‑group differences and within‑child changes were analyzed using linear mixed effects models adjusted for age and sex. Results At admission, children in the MUAC and Combined groups were younger and more stunted, with the Combined group showing the most severe deficits across anthropometry, FMI, and PhA. During treatment, the WHZ group exhibited the fastest recovery, with predominantly FFM accretion (82% of weight gain) and the largest increases in FFMI. In contrast, the MUAC group accumulated mainly FM (78% of weight gain) with minimal FFM gain and a decline in PhA, while the Combined group showed mixed FM and FFM recovery patterns. Conclusions Children admitted for outpatient SAM treatment by MUAC, WHZ, or combined criteria appeared to follow different recovery trajectories despite meeting standard anthropometric discharge thresholds. Body‑composition and impedance indicators suggested heterogeneity in tissue gain patterns and vulnerability profiles across admission groups, particularly in the MUAC and Combined groups. While exploratory, these findings suggest that anthropometric recovery alone may not uniformly capture the full recovery pattern across SAM admission groups, warranting the need for longitudinal studies evaluating whether body-composition differences at discharge predict subsequent clinical outcomes and post-discharge vulnerability.