Comparison of a low carbohydrate intake and standard carbohydrate intake on refeeding hypophosphatemia in children and adolescents with anorexia nervosa: a pilot randomised controlled trial

Abstract

Background Nutritional rehabilitation for patients with anorexia nervosa involves balancing the need for weight gain whilst mitigating the risk of refeeding syndrome. Graded caloric increases and restriction of calories from carbohydrate have been used to minimise the risk of developing refeeding hypophosphatemia. There is little evidence to support the recommended nutrient composition, specifically the recommended carbohydrate intake that is safe in this population. The aim of this pilot study was to compare the effect of a low and a standard carbohydrate feeding protocol on serum phosphate levels in children and adolescents with anorexia nervosa. Methods A pilot study of 23 children and adolescents with anorexia nervosa admitted for medical stabilisation to the adolescent ward of a tertiary hospital was undertaken. Participants were commenced on an oral feeding protocol and were randomly allocated to isocaloric meal plans that were either low carbohydrate (< 40% total energy from carbohydrate) or standard carbohydrate (50–60% total energy from carbohydrate). Serum phosphate levels were monitored daily across the first week and twice weekly thereafter. Clinical status, including weight gain, was monitored throughout admission. Results 52% (n = 12) of participants were allocated to the low carbohydrate group and 48% (n = 11) were allocated to the standard carbohydrate group. No patients in either of the diet groups developed refeeding hypophosphatemia in the first seven days of admission. Weight gain during the first week was significantly higher in the standard carbohydrate diet (1.4 kg/wk ± 0.5) compared to the low carbohydrate diet (0.6 kg/wk ± 0.9), p value 0.03. Participants from both diet groups were largely orally fed with less than 10% of the total number of meals and/or snacks across both groups provided as nutrition supplement drinks, either orally or enterally. Conclusion This pilot study supports that a standard carbohydrate intake (providing 50–60% of total energy from carbohydrate) optimises nutritional rehabilitation without increasing the risk of refeeding hypophosphatemia in adolescent inpatients with anorexia nervosa. CTN: ACTRN12621000300875. Plain English Summary: People with eating disorders who are underweight or malnourished, such as patients with anorexia nervosa, are at risk of refeeding syndrome when they receive treatment and return to regular eating. Refeeding syndrome may cause fluid and electrolyte shifts. This can occur as a result of the reintroduction of carbohydrates, and can have potentially life-threatening consequences if not managed appropriately. Refeeding hypophosphatemia is one of the early markers of refeeding syndrome. This study compared patients who were provided a low carbohydrate diet (40% total energy from carbohydrate) to those who were provided a standard carbohydrate diet (50–60% total energy from carbohydrate) to see if patients from either group were more at risk of developing refeeding syndrome. No patients in either of the diet groups developed refeeding hypophosphatemia. This pilot study may help to ensure that when patients get treated for their eating disorder in hospital, they can return to a normal diet as soon as possible with close medical monitoring.

Refeeding Hypophosphatemia in Hospitalized Adolescents With Anorexia Nervosa

“…Weight gain did not differ between the two groups because total daily energy intake was similar in both groups. While findings from this trial support the hypothesis that lower carbohydrate nasogastric tube feeding may protect against refeeding hypophosphatemia, they contradict findings from the small RCT showing greater weight gain with higher carbohydrate meal-based refeeding discussed previously [34]. More evidence is needed to understand how macronutrient content and delivery methods (oral vs. enteral) impact the safety and efficacy of refeeding.…”

“…This study identified a cut-point of >58.4% carbohydrate as being associated with refeeding hypophosphatemia [33]. A pilot RCT of oral refeeding in children and adolescents with AN and atypical AN assigned patients to meal plans that were either low carbohydrate (<40% total energy from carbohydrate) or standard carbohydrate (50%e60% total energy from carbohydrate) and found that no patients in either group developed refeeding hypophosphatemia in the first seven days of admission but noted greater weight gain with higher carbohydrate feeding [34]. While the investigators concluded that providing 50%e60% of total energy from carbohydrates optimized nutritional rehabilitation without an increasing risk of refeeding hypophosphatemia, these findings should be interpreted with caution given the small sample size and the likelihood that energy intake differed between groups and participants.…”

Clinical Outcomes of Refeeding Syndrome: A Systematic Review of High vs. Low-Calorie Diets for the Treatment of Anorexia Nervosa and Related Eating Disorders in Children and Adolescents

“…In this review, 15 studies looked at weight gain as an outcome; seven compared weight gained in the high-caloric group to that of LCR protocols [ 26 , 35 , 40 , 41 , 43 - 45 ], and eight looked only at the HCR protocol [ 27 , 39 , 47 , 49 - 53 ]. In 11 of these 15 studies, the HCR group's average weekly weight gain was significantly higher than that of the LCR group, ranging from 0.82 kg to 2.7 kg [ 26 , 27 , 40 , 41 , 44 , 46 , 47 , 49 - 52 ]. The most significant weight gain was reported in Garber et al's prospective observational study, which used a quasi-experimental design to compare LCR versus HCR refeeding methods.…”

“…Almost 95% of the papers that satisfied our inclusion criteria were published during the last decade, and the majority (80%) were either randomized controlled trials (RCTs) or retrospective studies (cohort and chart reviews). Ten studies compared low-calorie diets to high-calorie diets [26,35,[37][38][39][40][41][42][43][44][45], whereas the remaining studies only looked at high-calorie diets. There was no standard protocol used.…”

“…The findings of this study, along with other studies published in the last decade, cast doubt on the conservative "start low, go slow" feeding model. In this review, 15 studies looked at weight gain as an outcome; seven compared weight gained in the high-caloric group to that of LCR protocols [26,35,40,41,[43][44][45], and eight looked only at the HCR protocol [27,39,47,[49][50][51][52][53]. In 11 of these 15 studies, the HCR group's average weekly weight gain was significantly higher than that of the LCR group, ranging from 0.82 kg to 2.7 kg [26,27,40,41,44,46,47,[49][50][51][52].…”