A retrospective cohort analysis of factors associated with the development of refeeding syndrome in children 0–59 months diagnosed with severe acute malnutrition in a South African setting

Abstract

BackgroundRefeeding syndrome (RFS) is a life-threatening, underdiagnosed, and under-researched complication in treating children with severe acute malnutrition (SAM). This study aimed to determine the incidence and onset of RFS and identify biochemical abnormalities, clinical signs, and complications associated with RFS development in children, 0–59 months, treated with SAM in a South African public hospital setting. MethodsA retrospective cohort study was performed on hospital medical records of children aged 0–59 months, diagnosed with SAM at Rahima Moosa Mother and Child Hospital, Johannesburg, from 1/10/2014 to 31/12/2018. The onset of RFS among children included in the study was diagnosed based on published criteria for RFS. On admission, children who developed RFS and those who did not were compared concerning biochemistry and clinical signs and symptoms. ResultsA total of 148 medical records were retrieved from the hospital archives. The diagnosis of SAM based on the World Health Organization (WHO) definition was confirmed in 126 children who were then included in the study. The median age of the 126 children (63 % male) with confirmed SAM was 11.2 months (P25:7.0 months; P75:17.0 months). The in-hospital mortality rate was 18.2 %, of which 8.7 % were retrospectively diagnosed as having developed RFS during their recorded hospital stay, despite implementing the WHO treatment guidelines for SAM. A significantly higher percentage of the children who developed RFS presented on admission with hypophosphatemia (p = 0.015), hypokalemia (p = 0.001), hyponatremia (p = 0.001), an international normalized ratio (INR) of above 1.7 (p = 0.025), diarrhea (p = 0.042), dehydration (p = 0.029) and urinary tract infection (UTI) (p = 0.041), than those who did not. Children who developed RFS stayed in hospital significantly longer than those who did not (18 vs. 12 days with a p-value of 0.003). ConclusionIn this population of children with SAM treated in a South African public hospital setting, the presence on hospital admission of low levels of electrolytes, elevated INR, dehydration, diarrhea, and UTI was significantly associated with developing RFS. Recognizing these as possible red flags for developing RFS in children admitted with SAM might contribute to improved outcomes and needs further investigation.