Impaired exercise heart rate in drug-resistant epilepsy: Evidence from volitional exercise and extraoperative cortical electrical stimulation

Objective: Functional neuroimaging shows reduced neural activity in the hippocampus (HC), insular cortex (IC) and amygdala during rapid heart rate (HR) responses to brief isometric handgrip exercise (HG). However, the specific electrophysiological mechanisms remain unclear. Hypothesis: This preliminary study tested the hypothesis that cortical stimulation (CS) would reduce HR responses to HG. The a priori expectation was that HR would increase with HG in both healthy control (Control) and drug-resistant epilepsy (DRE) groups. Methods: Two protocols were studied: 1) Control vs DRE: To assess an independent effect of DRE, HR was obtained from the electrocardiogram in groups of Control (n=19; 7F/12M; age: 6±10 years) and DRE (n=18; 5F/13M; age: 34±10 years) participants. Changes in vagal cardiac function were assessed using the root mean square of successive differences (RMSSD). We collected these variables at baseline and the first 10 seconds of HG at 40-50% of maximum voluntary contraction. Changes from baseline to HG were calculated. 2) DRE with CS: We evaluated RMSSD and HR in people with DRE (n=13; 3F/10M; age: 35±11 years; n=10 from above DRE group) at rest and during HG both without and with CS (1 Hz, 5 mA, 10 seconds; delivered by stereoelectroencephalography). Participants received CS in the HC (n=9), IC (n=5), or amygdala (n=2) (n=3 tested 2 CS sites each). The research goal of CS was to minimize regional neural deactivation in each site of interest. Results: 1) Control vs DRE: In response to HG, HR increased in Control from 64±8 beats per minute (bpm) at baseline to 70±10 bpm (p=0.001), while HR only minimally increased in DRE (70±11 bpm at baseline vs 72±11 bpm after HG) (p=0.086). The increase in HR in Control (6.30 ± 4.21 bpm) was greater than in DRE (1.45±3.39 bpm) (p< 0.001). In response to HG, RMSSD decreased in Control from 59±25 to 43±15 ms (p=0.002) and in DRE from 27±14 to 22±12 ms (p=0.002). The reduction in RMSSD in Control (-15.58±19.11 ms) was greater than in DRE (-4.96±5.73 ms) (p=0.030). 2) DRE with CS: Neither HR nor RMSSD changed from baseline pre-CS to CS alone (no HG). In response to HG alone, HR increased ~3 bpm (p=0.038), and RMSSD decreased ~6 ms (p=0.045). In response to HG with CS, HR did not change during stimulation of the HC (80±17 vs 82±15 bpm) (p=0.167), IC (77±9 vs 77±10 bpm) (p=0.370), or amygdala (82±4 vs 83±9 bpm) (p=0.782). In response to HG with CS, RMSSD also did not change with stimulation of the HC (23±19 vs 20±7 ms) (p=0.489), IC (19±12 vs 19±10 ms) (p=0.873), or amygdala (27±7 vs 25±10 ms) (p=0.679). Summary and Conclusions: Compared to Control, DRE participants were marked by impaired HR responses to HG, as well as lower baseline and lower HG-induced changes to RMSSD. These changes in HR during HG were no longer seen with CS, suggesting that the HC, IC, and amygdala have a role in cardiac function, namely vagal activation. However, we note potential limitations that include the short duration of HG, small sample size, and the very small HR response to HG in patients with DRE whereby any effect of CS may be difficult to observe. Funding: Natural Sciences and Engineering Research Council of Canada This abstract was presented at the American Physiology Summit 2026 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.