Abstracts

Rationale: The modified Atkins diet (MAD) is a ketogenic diet effective in the treatment of drug-resistant epilepsy (DRE; seizures despite adequate treatment with 2 appropriate anti-seizure drugs (ASDs)). The diet stimulates ketone body production by increasing fat intake and restricting carbohydrates. Compliance in adults has been poor. The purpose of this study was to determine whether combining liquid KetoCal® formula (4:1 ratio of fat to protein and carbohydrates combined) with MAD during the first month in adults with DRE would improve compliance for 1 month as compared to MAD alone. Secondary outcome measures included presence of urinary ketosis within 1 month of starting treatment, percent seizure reduction, response rate, and long-term compliance. Methods: Eighty adults (age 18 years and older) with DRE and 4 or more quantifiable seizures per month were enrolled. All participants received training in starting a 20 gram per day net carbohydrate limit MAD. Participants were randomized to receive one 237 mL KetoCal® cartons daily in combination with MAD during the first month (treatment arm) or the second month (control arm). Participants recorded urine ketones daily until they reached ketosis (defined as ≥ 40 mg/dL) then biweekly, weight weekly, and seizures daily and followed up at 1 and 2 months. Results: Of 80 participants enrolled, 55 (69%) were women, mean age 35 years (SD±14), mean age at seizure onset 16 (SD±14) and mean ASDs tried 8 (SD±4). In the treatment arm, 35 (88%) completed 1 month of the study and 28 (70%) completed 1 month in the control arm, p=0.1. Among those that completed the first month of the trial, 27 (77%) reported achieving urinary ketosis by 1 month in the treatment arm and 21 (75%) in the control arm, p=1.0. Mean seizure reduction at 1 month in the treatment arm was 39% vs. 16% in the control arm, p=0.2. Percent of participants that responded (≥ 50% seizure reduction) was 54% in the treatment arm and 43% in the control arm, p=0.5. Thirty participants (75%) completed the 2 month study in the treatment arm as compared to 26 (65%) in the control arm, p=0.5. In participants for whom long-term follow up results were available, 16 of 28 participants (57%; 40% of the original 40 enrolled) were still on MAD from the treatment arm (mean diet duration 18 months, range 4-52 months), while only 6 of 23 (26%; 15% of the original 40 enrolled) were still on MAD (mean diet duration 15 months, range 5-40 months) from the control arm. These differences were statistically significant using an intent-to-treat analysis, p = 0.02. Conclusions: While supplementing a modified Atkins diet with a ketogenic formula in the first month on MAD did not increase the likelihood of achieving ketosis or reducing seizures, significantly more adults remained on MAD long-term following this intervention compared to control participants on MAD alone. This suggests a potential strategy for encouraging compliance with MAD in adults with intractable epilepsy. Funding: Nutricia North America