Abstracts

Rationale: Memory impairment is the most common cognitive complaint in people with epilepsy (PWE). Positive effects of physical exercise on psychological and physiological well-being of PWE have been reported. However, the effects of exercise on cognitive function in PWE have not been well studied. The objective of this pilot study was to examine the effects of a 6-week supervised, structured aerobic and strength-training program on verbal memory function and brain network connectivity in PWE. Methods: 16 PWE (19-42 years old; 8 male) with subjective memory complaints, =4 seizures/month in past 6 months, no abnormalities on previous MRI, and able to undergo 3T MRI were enrolled in the study. Subjects were assigned to either exercise training or observation, but were given the option to switch to the other group if they had transportation or availability limitations. PWE in the control group had the option of later transitioning to the training group. PWE in training group were required to pass a physical examination by an epilepsy specialist prior to training, and those not seizure-free >6 months prior had to wear a helmet for safety during training. PWE had 2 research visits 6-9 weeks apart that included MRI and California Verbal Learning Test (CVLT-II; Standard Trials 1-5 Total Recall, Long Delay Free Recall, and Recognition Discriminability, d’). Aerobic fitness (VO2max) and 1 repetition maximum weight (leg press, knee extension, bench press) tests occurred at both visits for training group and on visit 2 for control group. MRI data were processed with SPM/Matlab to determine changes in resting state functional connectivity (rsFC) for regions considered part of the seizure initiation and propagation network: thalamus, lateral cerebellum, hippocampus and posterior cingulate cortex (PCC). Paired t-tests assessed visit 1-2 changes for each group. Results: 11 PWE were in the observation control group; 8 returned for visit 2 (4 of the 8 transitioned to training). The remaining 5 PWE were in the training group. 1 control PWE who transitioned to training did not complete aerobic fitness testing. At visit 1, groups did not differ in age, seizure frequency, age of onset, number of antiseizure drugs, years of education or CVLT-II scores (all p>0.1). Trained PWE increased in all strength tests (all p<0.02) and showed improvement in VO₂max, although the difference was not statistically significant (p=0.08). Trained PWE improved on Standard Trials 1-5 Total Recall (p=0.038) and d’ (p=0.0025), but not control PWE (both p>0.1). There was no change in Long Delay Free Recall for either group (both p>0.1). Both groups showed decreased left hippocampus rsFC and increased cerebellum rsFC, while only trained PWE showed increased PCC rsFC with caudate/putamen and thalamus rsFC with left orbitofrontal cortex (all FDR corrected p=0.05). Conclusions: Our results show that exercise training positively affects aerobic fitness and memory function and induces changes in brain network connectivity in PWE. These pilot results are promising and provide support for a larger study. Funding: Study support was provided by the University of Alabama at Birmingham (UAB) Faculty Development Grant Program, UAB Department of Neurology, UAB Epilepsy Center, UAB Civitan International Research Center, and UAB Center for Exercise Medicine.