Refeeding syndrome in surgical patients post initiation of artificial feeding, a prospective cohort study in a low-income country

South Africa's inequitable history has contributed to a malnourished population, further aggravated by high levels of violence and economic uncertainty culminating in a population suffering the sequelae of poverty. The perceived notion is that malnutrition places the South African population at greater risk for development of refeeding syndrome. This study aimed to identify the incidence of refeeding syndrome in the South African population. All patients admitted into the surgical intensive care unit from 1 November 2019 to 30 September 2020, were screened using the National Institute for Health and Care Excellence (NICE) refeeding risk criteria. Patients started on artificial feeds with one or more risk factors were included in the study. The syndrome was confirmed using the King's College criteria and compared with the American Society for Parenteral and Enteral Nutrition (ASPEN) criteria for refeeding syndrome. 200 Patients were included in this study. The median age of the sample population was 41 years (IQR 30–58) with a male predominance (63%). All patients included had one or more risk factors and 62 (31%) of patients fulfilled the NICE criteria. The sensitivity of the NICE criteria was 33% and specificity was 70%. The most common risk factors identified were little or no nutritional intake for >5 days (55%) followed by a history of alcohol abuse, drugs including insulin, chemotherapy, diuretics or antacids (18%). Specificity values for all risk factors were >80% apart from little or no nutritional intake for >5 days which had a specificity of 64.2%. Sensitivity values for all risk factors were low. 84.5% of patients received artificial nutritional support in the form of enteral feeds, 9% parenteral and 6.5% both enteral and parenteral feeds. A total of 146 patients required electrolyte supplementation. Three patients fulfilled the King's College criteria and one subsequently died. In comparison, 25 patients fulfilled the ASPEN criteria and 3 demised. Due to the low sensitivity and specificity of the NICE criteria, we advise that it be used merely as a guideline to identify patients at risk of refeeding syndrome and one should remain vigilant in patients with any risk factors present. Due to the small number of patients who fulfilled the King's College criteria and significant difference in incidence when compared to the ASPEN definition, a conclusion regarding the accuracy of both diagnostic criteria could not be made. We recommend a review of the