Abstracts

Rationale: Ketogenic diet (KGD) is effective treatment of refractory epilepsy in children. There is little experience of its use in adults. There have been few evaluations of possible KGD effect on EEG. We recently observed rapid abatement of seizures in 1 adult patient with refractory primary generalized epilepsy (PGE) 2 days after KGD initiation with a fast. The goal of the present pilot study was to determine whether (a) KGD initiation is associated with EEG changes and (b) these changes are correlate with changes in serum glucose and urine ketones. Methods: 5 patients with refractory epilepsy (3 women, 2 men, age range 24-53) were treated with KGD and underwent continuous EEEG monitoring during KGD initiation. KGD consisted of 3:1 [fat] : [protein+carbohydrate] ratio and caloric restriction of 1600 kcal/day. Patients were hospitalized for 4-5 days for KGD initiation, with a 24-48 hour fast followed by graduated daily caloric increase to 25%, 50% and 75% target caloric intake during days 2-5 of treatment and 100% caloric intake upon discharge. There was no fluid restriction: because 3/5 patients were on topiramate treatment, patients were hydrated with 2.5 L of normal saline/day. Serum glucose and urinary ketones were checked regularly during hospitalization, together with daily electrolyte checks. Long term continuous EEG monitoring was started on the first day of the fast, and was continued for 4-5 days. Xltek 32 channel monitoring system was used. The whole EEG was reviewed manually in 4/5 patients, and interictal epileptiform discharges (IEDs: spikes and sharp waves) were counted. In 1 patient, the whole original EEG was not available, and only files containing computer spike detections using Persist software were reviewed for spike determination. Results: 2 patients had primary generalized epilepsy (PGE) with generalized tonic clonic (GTC) seizures (n=2) and absence and myoclonic seizures (n=1). 2 had cryptogenic localization-related epilepsy (LRE) of temporal (n=1), and frontal (n=1) lobe foci. 1 had symptomatic frontal lobe and generalized epilepsy. Seizure duration ranged from 9-51 years, prior AEDs from 2-8, and current AEDs from 1-4, and VNS in 1. AEDs were held constant in 4/5 patients during the study. In one patient, TPM and valproate dual therapy was changed to TPM monotherapy during KGD initiation. There was a statistically significant reduction in the mean hourly IED rates from 49 IEDs/h on day 1 to 6 IEDs/h on day 4 of KGD treatment (p<0.01). Hourly IED rate declined from KGD 1 to KGD 4-5 in 4/5 patients. There was no correlation between hourly spike frequency and serum glucose levels or urine ketone levels.