Abstracts

Rationale: The ketogenic diet (KD) has been reported as an effective therapy in children with refractory epilepsy (RE). The classic KD is usually initiated with fasting, followed by a 4:1 KD ratio. Initiation of the KD without fasting and lower initiation ratios have been reported to be better tolerated in young infants. The purpose of this project was to identify prognostic factors that may influence the response to KD in children with RE. Methods: A retrospective chart review was performed for all patients on KD. Exclusion criteria included patients discontinued from the KD due to severe side effects within a few days of starting diet. Seizure outcome was graded as follows: Class I: >90% seizure reduction, Class II: 50-90% reduction, Class III: 50-25% reduction, Class IV: no significant change or <25% seizure reduction or worsening seizures. Parameters studied included: gender, seizure type (partial vs. generalized), age at seizure onset, development status (mild/moderate/severe delay), seizure frequency at baseline, number of AED used before starting the KD, fasting vs. non fasting induction and ratio of KD used. Regression Analysis was performed at 3, 6, 9 and 12 months follow ups. Results: 57 patients met inclusion criteria documentation. Ages ranged from 1.7 years to 26.8 years (mean 14.3 years ) with a male to female ratio of 1:1.2. Mean seizures at baseline were 12.9/day. Generalized seizures were seen in a vast majority (89%) of the patients. Class I and II seizure outcome was reached in 35% and 21% of patients at 3 month interval and dropped to 26% and 5% at 12 months respectively. Patients with normal or mild development delays showed a class I improvement in 54% of patients at 3 months and 46% at 12 months follow up, while children with moderate and severe delays revealed Class I outcome in 30% at 3 months and declining to 20% at 12 months (NS). Regression analysis of data revealed that at the 3 mo follow up early seizure onset and female gender were associated with less efficacy (p 0.036 and 0.002 respectively). Lower effectiveness was also seen with low baseline seizure frequency and non-fasting induction (p 0.056 and 0.065). At 6 mo follow up, higher KD ratio at starting and fasting induction were associated with better outcome (p 0.031 and 0.040). However this difference was not sustained at 9 and 12 months follow-up. The other variables tested did not reveal significant influence on seizure outcome. Conclusions: Our study confirms the effectiveness of KD in all children with RE. Moderate and severe developmental delay, young age of seizure onset, low baseline seizure count and female gender were associated with worse outcome while higher KD ratio and fasting induction were associated with improved seizure control up to 6 months follow up.