Is fat mass a better predictor of 6-month survival than muscle mass among African children aged 6–59 months with severe pneumonia?

Background: Pneumonia remains the leading cause of mortality among children under 5 years. Poor nutritional status increases pneumonia mortality. Nutritional status assessed by anthropometry alone does not provide information on which body composition element predicts survival. Body composition proxy measures including arm-fat-area (AFA), arm-muscle-area (AMA), and arm-muscle-circumference (AMC) could be useful predictors. Objective: To compare the ability of fat and muscle mass indices to predict 6-month survival among children with severe pneumonia. Methods: This prospective cohort study was nested in the COAST-Nutrition trial (ISRCTN10829073, 06/06/2018) conducted between June 2020 and October 2022 in Uganda and Kenya. We included children aged 6–59 months hospitalized for severe pneumonia with hypoxemia. Children with severe malnutrition, known chronic lung or cardiac diseases were excluded. Anthropometry and clinical status were assessed at enrolment and at follow-up to day 180. We examined Receiver Operator Characteristic (ROC) curves of fat and muscle mass indices with 6-month survival as the outcome, and compared the areas under the curve (AUCs) using chi-square tests. Cox survival analysis models assessed time-to-mortality. Results: We included 369 participants. The median age was 15-months (IQR 9, 26), and 59.4% (219/369) of participants were male. The baseline measurements were: median MUAC 15.0 cm (IQR 14.0,16.0); arm-fat-area 5.6cm² (IQR 4.7, 6.8); arm-muscle-area 11.4cm² (IQR 10.0, 12.7); and arm-muscle-circumference 12.2 cm (IQR 11.5, 12.9). Sixteen (4.3%) participants died and 4 (1.1%) were lost-to-follow-up. The AUC for Arm-Fat-Area was not significantly higher than that for Arm-Muscle-Area and Arm-Muscle-Circumference [AUC 0.77 (95%CI 0.64–0.90) vs. 0.61 (95%CI 0.48–0.74), p = 0.09 and 0.63 (95%CI 0.51–0.75), p = 0.16 respectively], but was not statistically different from MUAC (AUC 0.73 (95%CI 0.62–0.85), p = 0.47). Increase in Arm-Fat-Area and Arm-Muscle-Circumference significantly improved survival [aHR 0.40 (95%CI 0.24–0.64), p = < 0.01 and 0.59 (95%CI 0.36–1.06), p = 0.03 respectively]. Survival prediction using Arm-Fat-Area was not statistically different from that of MUAC (p = 0.54). Conclusions: Muscle mass did not predict 6-month survival better than fat mass in children with severe pneumonia. Fat mass appears to be a better predictor. Effects of fat and muscle could be considered for prognosis and targeted interventions.