Abstracts

The ketogenic diet (KD), is an effective treatment of intractable epilepsy in children. Significant side-effects such as weight loss, hypoglycemia, vomiting, acidosis and dehydration are associated with a fasting introduction (FAST). We have previously shown that seizure efficacy is maintained with a gradual non-fasting introduction (GRAD). We explore whether GRAD results in fewer and milder adverse events and is better tolerated. Children 1 to 14 years were randomized to begin the KD using either the FAST or GRAD protocol during a 6 day hospital admission. The FAST protocol began with a [lt]48 hour fast. The 4:1 ratio meals were advanced daily in 1/3 caloric increments until the full meal was tolerated. The GRAD protocol began with a full calorie a 1:1 ratio meal, which was advanced to a 4:1 ratio over 4 days. Blood glucose was assessed q 4-6 hours, and weight, electrolytes, hydration status, vomiting, bicitra and IV fluid use were assessed daily. Fisher[apos]s exact tests were used to examine the association between protocol and occurrence of adverse events, and longitudinal mixed effects models were used to look for trends in tolerability data over time. Forty-eight children were randomized, 24 to each protocol. One child dropped out of each protocol (FAST pancreatitis, GRAD GI illness). Compared to the children in the FAST group, children in the GRAD group lost significantly less weight (-0.94 [underline]+[/underline]0.83 vs. -0.28 [underline]+[/underline]0.69 kg from Day 2-6, p=0.006). Hypoglycemia (blood glucose [lt]45 mg/dL) occurred at least once in 33% of FAST vs. 4% of GRAD group (p=0.023):100% of FAST had at least one blood glucose [lt]60 mg/dL vs. 46% in GRAD group (p[lt]0.0001). A greater proportion of children in the FAST vs. GRAD group required bicitra to correct acidosis (63% vs. 29%, p[lt]0.041), and IV fluids for dehydration (63% vs. 29%, p[lt]0.041). Longitudinal mixed effects models showed significantly lower levels of glucose, bicarbonate and chloride (p[lt]0.05) in children in the FAST group, and this was particularly true for Day 2-4. Adverse events considered serious (status epilepticus, pancreatitis, extended hospital stay) occurred in two subjects in the FAST group. The GRAD protocol is less likely to cause weight loss, hypoglycemia, acidosis, dehydration than the FAST protocol during a 6 day of KD initiation, and requires fewer medical interventions. A gradual introduction of the KD is better tolerated and maintains the efficacy of the KD. (Supported by RR K-23 16074, General Clinical Research Center MO1RR00240, Nutrition Center, MRDDRC P30 HD26979 and a private donation by Ms. Catherine Brown.)