Abstracts

Rationale: The ketogenic diet (KD) is an effective therapy for medication-resistant epilepsy. The Classic (CKD) and the Medium Chain Triglyceride (MCTKD) have been shown to be equally effective in controlling seizures. While the exact mechanism of the KD is unknown, it has been observed that urinary ketone levels correlate with seizure control, suggesting a role for ketones in seizure suppression. The addition of MCT may have an independent beneficial effect. We compared ketone production and seizure control in children treated with the CKD and MCTKD.Methods: A prospective observational cohort study was initiated at the Hospital for Sick Children, Toronto, Canada in September 2013, including all consented children starting a KD for the treatment of medication-resistant epilepsy. Seizure types and frequency, diet specifics and ketone production were recorded by caregiver and collected by study staff at 1, 3 and 6 months post diet initiation. Six-month outcomes were calculated, including mean urine ketones and mean percentage of baseline seizures and compared among those children treated with the CKD or the MCTKD. Children who discontinued the diet before 6 months were excluded from analysis but are described below.Results: Twenty children initiated a KD during the study period, however 3 children (2 girls, 1 boy, mean age 5.33 yo SD±5.77,) were excluded because they either did not follow a diet or opted out of the study before 6 months. Two of these 3 children were gastrostomy tube (G-tube) fed and used the CKD. Six-month follow up data was available for 17 children. The mean age was 5.23 y.o.(SD±4.08), 13 (76%) were male and 9 (35%) were fed via G-tube. Sixteen of the 17 patients had generalized convulsive seizures. Mean urine ketone level was 7.51 mmol/L(SD±2.62) and mean percentage of baseline seizures at 6 months was 28%(SD±35.1). Of the 17 children, 11 followed the CKD and 6 the MCTKD. The groups were similar for mean age, gender and seizure types. The mean age was 4.79 y.o(SD±4.58) on CKD and 6.03 yo(SD±3.19) on the MCTKD (p=0.54). There were 9 (81.8%) boys on the CKD and 4 (66%) on the MCTKD (p=0.58). In the CKD group, 10 (91%) had generalized seizures and in the MCTKD group all 6 (100%) had generalized seizures (p=0.58). The groups differed in how they were fed; 9(81%) in the CKD group and none in the MCTKD group fed via G-tube (p=0.002). At 6-month follow-up, mean urine ketones were significantly higher in the children treated with the CKD; mean urine ketone level was 8.64 mmol/L(SD±2.05) on the CKD and 5.43mmol/L(SD±2.37) on the MCTKD(p=0.01). There was no significant difference in the change in seizure frequency between the groups. The mean percentage of baseline seizures at 6 months in the CKD group was 26.5%(SD±26.1) and 31.7%(SD±51.5) in the MCTKD group(p=0.78).Conclusions: Urinary ketone levels are lower in children treated with the MCTKD compared to those with the CKD. Despite lower ketone levels, there is no significant difference in seizure control from baseline at 6-month follow-up. MCT oil may have a beneficial effect on seizure control independent of ketone production.