Characteristics and predictors of mortality in-hospital mortality following burn injury in infants in a resource-limited setting.

Abstract

Burn outcome data in infants is lacking from sub-Saharan Africa. We, therefore, sought to assess the characteristics and predictors of in-hospital burn mortality in a resource-limited setting. We performed a retrospective study of the prospectively collected Burn Injury Surveillance database from June 2011 to December 2019. We performed bivariate analysis and Poisson regression to assess risk factors for mortality in our infant burn population. 115 (7.3%) infants met inclusion criteria. The median age of 8 months (IQR: 6-10) and primarily male (n = 67, 58.8%). Most burns were from scald (n = 62, 53.9%). Infant burn mortality was 12.2%. Poisson multivariable regression to determine burn mortality risk in infants showed that increased %TBSA burns (RR 1.04, 95% CI 1.01-1.07) and flame burns (RR 3.08, 95%CI 1.16-8.16) had a higher risk of mortality. Having surgery reduced the relative risk of death for infants with burns. We show that factors that increase infant burn mortality risk include percent total body surface area burn, flame burn mechanism, and lack of operative intervention. Increasing burn operative capability, particularly for infants and other children, is imperative.