Effect of the modified Atkins diet in adults with drug-resistant epilepsy: A controlled study.

The objective of this study was to be a proof of concept of an analysis strategy to assess the effectiveness of the modified Atkins diet (MAD) in adults with drug-resistant epilepsy (DRE). This retrospective cohort study included participants aged >16 years with at least two failed antiseizure medications (ASMs). The variables were self-reported and gathered from medical records or phone calls. Two groups are compared: MAD adjuvant to ASM (diet group) and ASM only (control). To make both groups comparable, diet and control groups were matched (1:2) by predictors of seizure remission (PSR) in adults with DRE. A systematic review of the literature was conducted to establish PSR. A total of 127 patients were eligible. Forty-six participants (36 %) initiated the MAD, but only 28 (22 %) continued the diet after 3 months. The number of past and new ASMs, epilepsy type, and follow-up time were selected as PSR. Twelve patients were included in the diet group and 24 in the control group, matched by PSR. Compared with the control group, the diet group had 4.5 greater odds of achieving ≥50 % seizure frequency reduction at 1-3 months (OR 5.5; 90 % confidence interval 1.1-26.65). Mean seizure frequency and other secondary outcomes did not differ significantly between the groups. In summary, the MAD showed a modest effect over seizure control compared with only ASM. Matching by PSR could be an alternative analysis approach to determine the effectiveness of the ketogenic diet in controlled studies. Additional studies are required to validate the analysis proposed.