Abstracts

Rationale: Genetic mutations are an increasingly recognized cause of early-onset refractory epilepsy. There are limited data available about the efficacy and tolerability of the ketogenic diet (KD) in genetic refractory epilepsy.  We retrospectively analyzed our experience with KD to determine its efficacy and tolerability in such children Methods: We reviewed the records of 60 children in whom KD was started for genetic etiology from 2005 to 2016, out of whom, 59 had refractory epilepsy with normal MRI. Children with >50% seizure reduction were defined as responders. Outcome and tolerability were evaluated at 1, 3, 6, 12 and 24 months after KD initiation. KD was stopped in 3 within a month of initiation due to difficulty in preparing KD meals and compliance issues (N=2), diagnosis of ethylmalonic aciduria (N=1) and reflux (N=1). Results: Genetic etiologies included SCN1A mutations in 11 (18.3%), CDKL5 mutations in 10 (16.6%). Thirty-seven (63%) were females, and mean age at seizure onset was 1.4 ± 2.7 (range 0.1 – 14) years. KD was initiated at the mean age of 4.0 ± 4.3 (range 0.2 – 17) years.  Traditional KD was started in 52 cases, modified Atkin’s diet in 5 and low glycemic index diet in 2. Median follow up after KD initiation was 1.9 years (IQR 0.4-3.9). Of the 56 patients who were on KD at 1 month, only 28 remained on KD at 24 months. Five (4 non-responders, 1 responder), 10 (8 non-responders, 2 responders), 3 (1 non-responder, 2 responders), and 10 (4 non-responders, 6 responders) patients had dropped out in the 1-3, 3-6, 6-12 and 12-24months intervals, respectively. Of these 28 children who dropped out, 8 were lost to follow up while in 20, KD was stopped due to lack of efficacy (N=14), adverse effects (hypertriglyceridemia, weight gain, weight loss, N=1 each), refusal to eat, inability to achieve ketosis, and unwilling to continue with KD (N=1 each). Responder rates for those continuing on KD at each time point were 37/56 (66.4%) at 1 month, 36/51 (71%) at 3 months, 32/41 (78%) at 6 months, 31/38 (82%) at 12 months and 24/28 (86%) at 24 months. Higher concurrent number of AEDs at the time of initiation of KD was associated with a higher non-responder rate at 1 month (Unit Odds Ratio=1.70, Odds ratio=14.56). No association was found between number of prior AEDs used and severity of developmental delay with response to KD. Conversion from non-responders to responders was higher than conversion from responders to non-responders at 3 months evaluation (4/15 vs 4/36, respectively; p < 0.001) and 3-6 months evaluation (2/7 vs 4/34, respectively; p=0.002). No similar association was found at subsequent time periods. There were further 9 patients in whom KD was stopped after 24 months for lack of efficacy (N=5), adverse effects (N=2) and inability to achieve ketosis and refusal to eat (N=1 each). Conclusions: KD is an effective treatment modality in refractory genetic epilepsy, with higher chances of non-responders turning into responders after3- 6 months of initiation. Higher number of concurrent AEDs was associated with a higher rate of non-response to KD. Discontinuation of KD due to adverse effects is uncommon Funding: none