Vagal nerve stimulation (VNS) can treat intractable epilepsy but the epileptolytic mechanism is not known. We have reported a central sympathetic inhibition during VNS (Barnes et al. 2000), but other brainstem and cortical effects remain unknown. We studied the relationship between stimulus strength and an index of cardiac vagal tone (CVT) during VNS.

Seven patients (3 males and 4 females) on VNS treatment due to intractable seizures were studied. The patients gave their informed consent and the study was approved by the local Ethics Committee. The VNS system consists of a stimulator (NeuroCybernetic Prosthesis, Cyberonics Inc., Texas, USA) placed in the left chest and an electrode wound around the cervical left vagus nerve and tunnelled under the skin to the stimulator. The NeuroScope^TM system (MediFit Diagnostics Ltd, London) monitored brainstem function continuously. MedullaLab^TM (MediFit Diagnostics Ltd, London) synchronised breathing movements measured by resistance plethysmography with transcutaneous P_O2 and P_CO2, non-invasive arterial blood pressure, heart rate, index of CVT and cardiac sensitivity to baroreflex (Julu et al. 2000). Constant current pulses (duration 500 µs (250 µs in two patients), intensity 0.25-1.75 mA) delivered at 30 Hz in three cycles of 60 s on and 66 s off in the supine position. Four current settings (mA) relative to each patient’s normal therapeutic level (Ir) were used and these were Ir – 0.75, Ir – 0.5, Ir – 0.25 and Ir. The mean CVT during 60 s of stimulation was calculated. The average of three cycles at each current level was expressed as a percentage increase above the baseline CVT.

The average baseline CVT was 5.2 ± 1.0 (mean ± S.E.M., arbitrary units in the linear vagal scale). Percentage increases of CVT rose with current levels to 46 ± 18 % (mean ± S.E.M.) at Ir (significant at P < 0.04, Wilcoxon signed rank test). The average baseline heart rate (HR) was 72.6 ± 3.3 (mean ± S.E.M. beats min^-1). The greatest reduction in HR occurred at Ir (71.1 ± 3.6 beats min^-1) and the average decrease was 2.1 % (P = 0.28, not significant, Wilcoxon signed rank test). It was notable that the patient with the largest increase in CVT had epileptic discharges in the EEG during our recording.

Other investigators have also found an increase in parasympathetic activity during VNS as measured by power spectral analysis of HR variability (Kamath et al. 1992). Our data indicate that VNS increases CVT above the baseline levels in proportion to the stimulus current.

Likely pathways through which left vagus VNS may increase CVT are (1) antidromic excitation of the nucleus ambiguus by largely ventricular efferents in the left vagus nerve (Thompson et al. 1987), (2) other Aδ afferents (from the lungs and gut) to nucleus tractus solitarii, (3) pulmonary C-fibre direct excitation of nucleus ambiguus (Wang et al. 2000), and (4) coronary baroreceptor Aδ afferents to the nucleus tractus solitarii.

Barnes, A., Duncan, R., Julu, P.O.O., Lindsay, K. & Patterson, J. (2000). J. Physiol. 523.P, 248P.

Julu, P.O.O., Kerr, A.M., Hansen, S., Witt-Engerström, I., Al-Rawas, S., Engerström, L., Apartopoulos, F. & Jamal, G.A. (2000). J. Physiol. 523.P, 254P.

Kamath, M.V., Upton, A.R., Talalla, A. & Fallen, E.L. (1992). Pacing Clin. Electrophysiol. 15, 235-243.

Thompson, M.E., Felstein, G., Yavorsky, J. & Natelson, B.H. (1987). Am. J. Physiol. 253, R150-157.

Wang, Y., Jones, J.F., Jeggo, R.D., de Burgh Daly, M., Jordan, D. & Ramage, A.G. (2000). J. Physiol. 526, 157-165. abstract