The Effects of Transcutaneous Vagal Nerve Stimulation on Dynamic Cerebral Autoregulation

Abstract

PurposeTranscutaneous vagal nerve stimulation (tVNS) applied to the cervical branch of the vagus nerve is FDA‐approved to treat migraine and cluster headache. tVNS applied to the auricle branch of the vagus nerve (tVNSab) reduces muscle sympathetic nerve activity and improves cardiac parasympathetic activity after one session in healthy participants. Given the use of tVNS to treat cerebral disorders and its effects on autonomic function, we postulated that tVNSab might be beneficial for cerebral vascular function. We tested the hypotheses that tVNSab will increase resting middle cerebral artery blood flow velocity (MCAv) and improve dynamic cerebral autoregulation (dCA) during a repeated squat‐to‐stand procedure.MethodsSeven healthy participants (age: 20 ± 3 y; BMI:19 ± 7 kg/m2; 3 women) completed two randomized experimental visits on separate days. One visit consisted of tVNSab using a transcutaneous electrical nerve stimulation device (pulse width = 200 ms; pulse frequency = 30 Hz) with electrodes attached to the tragus of each ear. A time‐control visit (SHAM) was conducted with electrodes attached to the tragus but no current was applied. Mean arterial pressure (MAP; photoplethysmography), the partial pressure of end‐tidal carbon dioxide (PETCO2; capnography) and MCAv (transcranial Doppler ultrasound) were measured continuously. After 45 min of resting tVNS or SHAM, subjects performed a repeated squat‐to‐stand maneuver at a frequency of 0.05 Hz for 5 min while tVNSab or SHAM continued. Resting conductance values for the MCA (MCAc) were calculated as MCAv/MAP. MCA dCA was assessed during the squat‐to‐stand maneuver using spectral analyses to obtain coherence, gain, and phase values in the very low frequency (VLF) and low frequency (LF) bands. Spectral analysis gain was also assessed using normalized units (gainn), defined as beat to beat values divided by the mean value relative to changes in blood pressure. Data are presented as mean ± SD.ResultsThere were no differences between tVNSab and SHAM for MAP (tVNSab: 103 ± 8; SHAM: 102 ± 12 mmHg; P=0.41), PETCO2 (tVNSab: 35 ± 5; SHAM: 36 ± 2 mmHg; P=0.42), MCAv (tVNSab: 61.5 ± 9.2; SHAM: 58.7 ± 13.3 cm/s, P=0.36) or MCAc (tVNSab: 0.60 ± 0.08; SHAM: 0.59 ± 0.19 cm/s/mmHg, P=0.46) during baseline. There were no differences between conditions for coherence at VLF (tVNSab: 0.43 ± 0.14; SHAM: 0.45 ± 0.34; P=0.44) or LF (tVNSab: 0.65 ± 0.17; SHAM: 0.72 ± 0.13; P=0.06). Phase in the VLF (tVNSab: 0.35 ± 0.32; SHAM: 0.33 ± 0.36 rad; P=0.44) and LF (tVNSab: 0.28 ± 0.23; SHAM: 0.30 ± 0.16 rad; P=0.36) were not different. Gain (tVNSab: 0.56 ± 0.30; SHAM: 0.56 ± 0.54 cm/s/mmHg; P=0.50) and gainn (tVNSab: 0.99 ± 0.53; SHAM: 0.85 ± 0.84 %/mmHg; P=0.37) in the VLF were not different. However, gain (tVNSab: 0.76 ± 0.14; SHAM: 0.86 ± 0.17 cm/s/mmHg; P=0.02) and gainn (tVNSab: 1.22 ± 0.16; SHAM: 1.35 ± 0.23 %/mmHg; P=0.04) in the LF were lower during tVNSab.ConclusiontVNSab did not alter resting cerebral vascular responses. However, MCA LF gain was lower during tVNSab versus SHAM. These preliminary data indicate that dCA in a major intracranial cerebral artery might be improved during a single session of tVNSab. Further research regarding the influence of chronic tVNSab on cerebral vascular function is warranted.