OBJECTIVE: Refeeding hypophosphatemia (RH) is a potentially fatal complication in patients with anorexia nervosa (AN), and its dietary preventive strategy is not well established. We aimed to examine the association between carbohydrate content in the diet and the occurrence of RH in inpatients with AN via retrospective medical chart review. METHOD: We performed a chart review to collect data of patients with AN hospitalized at the Department of Psychosomatic Medicine of the University of Tokyo Hospital between April 1, 2012, and February 29, 2020. Receiver operating characteristic (ROC) analysis was performed to determine the cutoff point of the percentage of carbohydrate content in the diet for the occurrence of RH. Multivariate logistic regression analysis was performed with occurrence of RH as the dependent variable and the carbohydrate content of more than the identified cutoff point as the independent variable adjusting for the risk factors for RH. RESULTS: The percentage of carbohydrate content that is higher than the cutoff point obtained from the ROC analysis (58.4%) was significantly associated with the occurrence of RH, even after adjusting for variables associated with RH in univariate logistic regression analysis (age and body mass index) as well as the average daily calorie intake (odds ratio, 5.37; 95% confidence interval, 1.60-18.1; p = .0066). DISCUSSION: We identified that diets with higher carbohydrate contents were associated with RH in inpatients with AN, even after adjusting for known risk factors. Our findings may promote the development of dietary preventive strategies against RH in inpatients with AN.
OBJECTIVE: Refeeding hypophosphatemia (RH) is a potentially fatal complication in patients with anorexia nervosa (AN), and its dietary preventive strategy is not well established. We aimed to examine the association between carbohydrate content in the diet and the occurrence of RH in inpatients with AN via retrospective medical chart review. METHOD: We performed a chart review to collect data of patients with AN hospitalized at the Department of Psychosomatic Medicine of the University of Tokyo Hospital between April 1, 2012, and February 29, 2020. Receiver operating characteristic (ROC) analysis was performed to determine the cutoff point of the percentage of carbohydrate content in the diet for the occurrence of RH. Multivariate logistic regression analysis was performed with occurrence of RH as the dependent variable and the carbohydrate content of more than the identified cutoff point as the independent variable adjusting for the risk factors for RH. RESULTS: The percentage of carbohydrate content that is higher than the cutoff point obtained from the ROC analysis (58.4%) was significantly associated with the occurrence of RH, even after adjusting for variables associated with RH in univariate logistic regression analysis (age and body mass index) as well as the average daily calorie intake (odds ratio, 5.37; 95% confidence interval, 1.60-18.1; p = .0066). DISCUSSION: We identified that diets with higher carbohydrate contents were associated with RH in inpatients with AN, even after adjusting for known risk factors. Our findings may promote the development of dietary preventive strategies against RH in inpatients with AN.
Authors: Elizabeth Kumiko Parker; Victoria Flood; Mark Halaki; Christine Wearne; Gail Anderson; Linette Gomes; Simon Clarke; Frances Wilson; Janice Russell; Elizabeth Frig; Michael Kohn Journal: J Eat Disord Date: 2021-12-11