Rationale:
Exercise has known health benefits and potential anti-inflammatory effects, although not well-studied in epilepsy. Magnetic resonance spectroscopic imaging and thermometry (MRSI-t) holds promise for non-invasive brain temperature assessment as a proxy for neuroinflammation but has not been used to examine exercise effects. We hypothesized that people with temporal lobe epilepsy (TLE) would exhibit increased baseline (T1) temporal lobe temperature and post-endurance exercise training (EET) brain temperature decreases and improvements in cognition, quality of life, and mood.
Methods:
A total of four participants were enrolled: 1 right (RTLE-trained) and 1 left TLE (LTLE-trained) completed 12 weeks of EET; 1 RTLE and 1 LTLE were untrained. At T1 and after 12 weeks (T2), participants completed exercise testing, assessments, and 3T MRI when available. EET (adapted from MoTrPAC protocol) progressed from 15-20 min each of treadmill running and stationary cycling (week 1) to 25-30 min of each modality (weeks 4-12) at 60-80% heart rate reserve. MRIs were performed to obtain T1-weighted anatomical images and MRSI-t. The Metabolite Imaging and Data Analysis System software was used for image reconstruction and spectral processing i.e., spatial reconstruction, frequency alignment, B
0 inhomogeneity correction, co-registration of MRSI-t and T1-weighted images, AAL2 atlas registration, lipid suppression, spectral fitting, normalization, and integration with reference water (H
2O) molecules. Voxel-wise brain temperature was calculated based on the chemical shift difference between H
2O and creatine (CRE): T
CRE = (-102.61 x Δ< H2O-CRE) + 206.1° C. Maps with T
CRE > 38.0° C at T1 and T2 were used to assess brain temperature changes with and without EET.
Results:
RTLE-trained exhibited the greatest benefit from EET, with faster completion time on the Trail Making Test Part A; increased verbal learning/memory and SF-36 mental component scores, decreased adverse events profile and total mood disturbance scores at T1 compared to T2 (
Table 1). LTLE-trained exhibited only marginal benefits, while the untrained TLE showed overall minimal change or decline in performance. T
CRE elevations remained stable in RTLE-untrained (
Fig.1A) but diminished in RTLE-trained (
Fig.1B) showing T
CRE decreases localized to temporal lobes as well as bilateral cerebellum.
Conclusions:
This study demonstrates the feasibility of performing 12 weeks of moderate-high intensity EET in people with TLE and detecting brain temperature changes using MRSI-t. Future studies with larger sample sizes are needed to elucidate exercise training effects in TLE.
Funding:
This work was supported by the McKnight Brain Institute Pilot Grant (JBA), UAB Epilepsy Center funds (JBA and AS), National Rehabilitation Research Resource to Enhance Clinical Trials (REACT; P2CHD086851) Scholar Award (JBA), Civitan International Research Center, NIH Ruth L. Kirschstein National Research Service Award (F31NS129288 to AS), State of Alabama General Funds, NIH Common Fund MoTrPAC trial (U01AR071133 to MMB), and NIH National Center for Advancing Translational Research (UL1TR003096).