Abstracts

Rationale: To characterize the efficacy and tolerability of ketogenic diet (KD) therapy in a cohort of South Indian children with pharmacoresistant epilepsy on a traditionally carbohydrate rich, rice based diet. Methods: Children aged 0-18 years with pharmacoresistant epilepsy, enrolled in KD program from 2010 to 2015 were included in the study. Children with suspected metabolic disorders were excluded. KD was initiated and maintained as per the institutional protocol. Diet efficacy was evaluated according to reduction in seizure frequency and in the number of AED during the follow up visits and at the last contact. Reduction of more than 50% in seizure frequency was defined as good response. Duration of retention, reasons for discontinuation and adverse events were used for assessing KD tolerability. Results: 74 children were started on KD initiation process, out of which 4 failed due to intolerance and metabolic abnormalities like recurrent hypoglycemia (2), persistent hyperammoemia (1) and recurrent vomiting and diarrhoea (1). 70 children (M: F- 48:22) completed the initiation process. The median age at onset of epilepsy was 18 months (range-15days to 13years). Median age of KD initiation was 4year 2months (range-1month-18 years). 51 children (68.9%) had more than 1 seizure type at initiation of KD. 53(71.6%) children had developmental delay.90% of children were on 3 or more AED at initiation of KD. Baseline seizure frequency was >5/day in 54(80%) children. KD was initiated in ICU without fasting for 14 children with refractory status. KD ratio was 3:1 in 67 children and 4:1 in the rest. KD was continued for a mean duration of 10.43 months (range-1-44 months). Mean follow up was 14.61 months (range-1- 46 months). 3 children were lost to follow up. 56 children were on KD at 3 months, 43 at 6 months, 29 at 1year, and 17 at 1.5years after initiation. 15 patients continued the diet for more than 18months. At the last contact, 43 children (61.4%) had seizure reduction of more than 50% and 19 had reduction between 50-90%. More than 90% reduction was noted in 24 children (34.3%) and 6(8.5%) children became seizure free. 4 children expired during KD, 2 in the first month of initiation. Reduction in number of AED was successful in 6. 15 children became noncompliant during the follow up period, irrespective of the seizure outcome. While on KD, new AEDs were added in 7 children and the dosage was hiked up in 12. Lipoid pneumonia (2), recurrent hypoglycemia (1) and severe diarrhoea (1) were the major adverse events necessitating KD withdrawal. Main reasons for discontinuation of KD were poor compliance, lack of response to diet as well as relapse of seizures after initial control. Conclusions: This pragmatic study demonstrates that KD can be safely and effectively instituted in children with pharmacoresistant epilepsies on a traditionally carbohydrate rich south Indian diet. The major limitation of this study was the non standardization of the concomitant AED therapy. Funding: Nil