643 - Identifying infants under 6 months old at risk for mortality in Sierra Leone

Identifying undernutrition and at-risk infants < 6 months old is key to reducing mortality. A weight-for-length z-score < -3 has been used to identify undernourished infants; however, this definition may not identify infants at the highest risk of mortality.

Objective:

We aimed to evaluate mid-upper arm circumference (MUAC) and other anthropometric measures for identifying infants < 6 months old at risk for mortality in rural Sierra Leone.

Design/Methods:

This is a secondary analysis of data from a randomized clinical trial (NCT03079388) comparing a package of nutritional and anti-inflammatory interventions with the standard care for pregnant women with undernutrition in rural Sierra Leone. A total of 1489 pregnant women with undernutrition defined by MUAC ≤23.0 cm enrolled in the trial. Women were randomized to an intervention package or standard care.

All singleton live births with complete data were included. Weight, length, MUAC of left arm, and mortality were assessed at 6 weeks, 3 and 6 months. Anthropometric parameters were converted to z-scores using WHO growth standards. Non-parametric receiver operator characteristic curves were used to calculate 95% confidence intervals of areas under the curve and to determine the overall validity of MUAC in identifying length-for-age z-score (LAZ) < -2, weight-for-age z-score (WAZ) < -2, weight-for-length z-score (WLZ) < −2, concurrent WAZ and LAZ < -2 (WaSt), and mortality. Youden J statistic was calculated to identify the optimal MUAC cut-off for each measure. Time-to-event values for mortality stratified by the identified anthropometric cutoffs were analyzed using Cox proportional hazards regression adjusted for maternal intervention and infant sex.

Results:

Among a total of 1319 infants, MUAC was best for identifying WaSt ≤-3 with the optimal MUAC cutoff being 10.9 cm (Table 1). A total of 63/1319 (4.8%) infants in the cohort died during the study period. MUAC demonstrated moderate diagnostic accuracy for identifying infants at-risk of mortality with ≤11.2 cm as the optimal cutoff (Table 1).

WaSt ≤-3 carried the highest hazard ratio (HR) of mortality among the measures, though this did not reach the level of significance (Table 2). WAZ < -3 was the only measure with an elevated HR of mortality to reach the level of significance (HR 2.13, 95% CI 1.21 to 3.75, p=0.009) (Table 2,  Figure 1).

Conclusion(s):

A MUAC of ≤11.2 cm is associated mortality under 6 mo. WAZ < -3 identifies infants at-risk for mortality under 6 mo better than other anthropometric measures. MUAC and WAZ may be superior to WLZ to identify infants under 6 months at risk of mortality.