Abstracts

Rationale: Initiation of the Ketogenic Diet at Boston Children's Hospital has been associated with hypoglycemia (glucose <45 mg/dl) and acidosis requiring therapeutic intervention. The initiation protocol was changed to try to minimize these complications. This study compares the occurrence of these complications using standard versus new protocols. Methods: The standard protocol was initiated at a target ratio from day 1, chosen according to age, nutritional status and medical complexity. The new protocol started at a 1:1 ratio on day 1 of admission and incremented by one ratio daily until a 3:1 ratio or beta-hydroxybutyrate of ≥3 mmol/L was attained. Data was collected prospectively from 18 patients admitted for diet initiation over a 12 month period using the new protocol and compared to a historical control group of 24 patients initiated in the preceding 12 months. No fasting was required for either protocol. Results: There were no differences between the two groups with respect to age (p=0.99), pre-diet weight-for-height z-score (WHZ) (p=0.87), ratio at discharge (p=1), length of stay (p=0.18), or beta-hydroxybutyrate at first follow up visit (p=0.89). A smaller proportion of patients initiated using the new protocol experienced hypoglycemia (1/18) than using the standard protocol (10/24) (OR 0.082, CI 0.009-0.724). This difference was not explained by age, gender, WHZ, method of feeding or concurrent treatment with topiramate or zonisamide. There was no difference in the occurrence of acidosis between the two groups (p=0.76). Food refusal occurred in a lower proportion of those initiated using the new protocol when adjusted for WHZ (OR 0.028, CI 0.001-0.73). Conclusions: The occurrence of hypoglycemia and food refusal can be reduced using an incremental diet initiation protocol without extending length of hospital stay or compromising the level of ketosis following discharge. The development of less rigorous and safer initiation protocols may improve long-term diet acceptance.