A.R.M. Upton

Regional Cerebral Blood Flow in Man Manipulated by Direct Vagal Stimulation

More than 100 years ago, Gowers^ cited several patients with partial seizures who could abort the elaboration of their seizures by sensory stimulation. In 1952, repetitive central vagal stimulation in encephale isole cats was shown by Zanchetti et al.^ to eliminate completely or reduce appreciably spontaneous cortical spindles and strychnine induced spikes. More recently, repetitive vagal stimulation has been found to inhibit chemically induced seizures in dogs.^ In 1990, two groups^^ described the use of vagal stimulation in man for the treatment of intractable partial seizures. The mechanism by which vagal stimulation might control seizures is not known although it is suggested that the mechanism probably operates through the nucleus of the solitary tract and the brain stem reticular centers. We have studied regional cerebral blood flow in five patients in whom a vagal stimulator had been implanted on the left-hand side for treatment of intractable seizures. All of the patients were taking anticonvulsant medication at the time of the study and no changes were made in their treatment regimen. Regional cerebral blood flow, r CBF, was measured by positron emission tomography, PET, following bolus injection of oxygen-15 labeled water. Six sequential measurements of r CBF were made in each patient; three when the stimulator was on and three when it was off. The order in which the stimulations were done was random and the patients were not told about the state of the stimulator. A total study lasted approximately 3 hours, during which time the EEG was monitored continuously. Two of the five patients exhibited electrical evidence of seizure activity during the tests. None of the patients slept. Thirty-one continuous transaxial sections of the brain were examined simultaneously during each study with a spatial resolution of 5 mm F.W.H.M. in all directions. Data from the PET studies was analyzed by the method of Statistical Parametric Mapping introduced by Friston et al. Significant changes in r GBF (P < 0.001) were recorded in the region of the anterior thalamus [E(l,2), a, +12 Talairach and Tournoux] and in the cingulate gyrus anteriorly [D, a, +32]. The changes in thalamic and cortical blood flow were both on the same side as the vagal stimulation and were encompassed by areas of less significant (P < 0.07) change. These preliminary findings suggest that the central consequences of vagal stimulation are not as diffuse as might be expected if the effect was mediated through the brain stem reticular substance. They are also in accord with data from this laboratory where direct electrical stimulation of the anterior thalamus was shown to evoke metabolic responses in the limbic striate system.

Effect of Vagal Nerve Electrostimulation on the Power Spectrum of Heart Rate Variability in Man

The power spectrum of heart rate variability contains low frequency (LF = 0.08–0.12 Hz) and high frequency (HF = 0.18–0.30 Hz) components said to represent neurocardiac rhythms. To verify whether such a relationship exists we report a unique study where the heart rate autospectrum was determined in a 28‐year‐old epileptic male patient with an implanted vagal electrical stimulator. The stimulator was activated at 20 Hz, 300 psec pulse, and 1.25 V. Continuous ECG and respiratory waveform records were obtained over 45 minutes every 8 hours (7–8 AM; 3–4 PM; 11–12 PM) With the stimulator ON, then 24 hours OFF and then 24 hours ON again. The overall LF:HF peak ratio increased from 0.64 to 1.99 (P < 0.001) after the stimulator was turned OFF. There was a dramatic increase in the LF peak power (> 60%) and a corresponding decrease in the HF peak power (> 65%) when the stimulator was turned OFF. These values were reversed when the stimulator was turned ON again. In the early morning and late evening hours, there was a significant rightward shift in the LF peak power frequency (average 0.057 to 0.075 Hz) whenever the stimulator was ON. Otherwise, there were no significant circadian variations in any of the autospectral components. The results demonstrate an unequivocal relationship between selective vagal nerve electro stimulation and alterations in the heart rate autospectrum.

Evidence of impaired afferent vagal function in patients with diabetes gastroparesis.

Two patients, a 28‐year‐old male and a 70‐year‐old female, with chronic insulin dependent diabetes mellitus and evidence of autonomic neuropathy were studied using cortical evoked responses following esophageal balloon and electrical stimulation. Both patients had symptomatic gastroparesis, poor gastric emptying, and reduced gasfroduodenal motility including abnormal results of scintigraphy and manome‐try. There was slowing of afferent vagal conduction but good evoked potential responses were recorded even though one patient could not feel electrical stimulation of either the proximal or distal esophagus. It is improbable that the gastric symptoms are due to an afferent autonomic neuropathy, but symptoms may well be related to impairment of motor vagal pathways. Nevertheless, afferent vagal pathways are involved in severe diabetes mellitus. The clinical significance of this delay in conduction velocity of afferent pathways remains to be established.

The cerebral response to electrical stimuli in the oesophagus is altered by increasing stimulus frequencies.

Recording of cerebral evoked responses (EP) allows the assessment of visceral afferent pathways and gut–brain communication, but the optimal stimulation parameters remain to be established. The present study determined the optimal stimulation frequency of electrical stimulation of the oesophagus to elicit EP responses. In 13 healthy male volunteers (24.1u2008±u20085.9 years), a 5u2008mm stainless‐steel electrode was placed in the distal oesophagus for electrical stimulation (ES). EP were recorded from 21 scalp electrodes placed according to the 10/20 International system. ES (15u2008mA, 200u2008μs) were delivered in repeated series of 24 stimuli. Stimulus frequency was randomly altered in different series using a pseudologarithmic range (0.1, 0.2, 0.3, 0.5, and 1u2008Hz). Two series of stimuli were applied using each stimulation frequency. Two‐dimensional topographic brain maps were created using interpolation techniques at each stimulation frequency. With increasing stimulus frequency, a significant and progressive decrease of EP amplitudes was observed between frequencies of 0.1u2008Hz and 1.0u2008Hz (P1/N2: 7.6u2008±u20081.2 vs 1.4u2008±u20080.3*u2008μV, N2/P2: 17.2u2008±u20081.7 vs 4.6u2008±u20080.4*u2008μV, P2/N3: 6.9u2008±u20080.7 vs 4.2u2008±u20080.5*u2008μV; *u2008=u2008Pu2008<u20080.05). In addition, there was a significant shortening of the mean peak latency of the intercalated P2 peak (Pu2008<u20080.0005), with a similar trend for the P3 peak (Pu2008<u20080.06), with increasing stimulus frequency from 0.1–1.0u2008Hz. Topographic brain maps localized the maximal early peaks (N1,P1,N2) in the paracentral cortical region (C3, Cz, C4), whereas the later peaks (P2 to P3) were symmetrically spread over the centro‐parietal and temporal regions (Cz, Pz, T5, T4). There was no difference in the cortical location of maximal EP amplitudes with increasing stimulus frequency. In conclusion, there is a clear relationship between stimulus frequency and amplitude of EP, suggesting rapid attenuation of the cerebral autonomic neural responses with increased electrical stimulation frequency. The effect of increased frequency on peak latencies suggests an alteration of stimulus processing in the thalamocortical region due to an altered perception of stimuli. Early EP peaks originate from basal structures of primarily the dominant hemisphere, while later peaks are localized in centroparietal cortical regions.

Effects of Chronic Left Vagal Stimulation on Visceral Vagal Function in Man

We examined the effects of chronic left vagal electrostimulation on afferent and efferent gastrointestinal vagal function in eight patients. Afferent function was assessed using cortical evoked responses to electrical stimulation of the esophagus and to direct vagal stimulation using the implanted left vagal electrode. Efferent gastrointestinal vagal function was measured by examining the basal, maximal, and sham fed stimuJated gastric acid output prior to and with chronic left vagal electrostimulation. Esophageal electro‐stimulation produced a cortical evoked response consisting of three negative and three positive peaks within 400 msec after stimulation. Prior to vagal eJectrostimulation the mean conduction velocity of the afferent signal was measured at 8.72 ± 3.39 m/sec, compatible with A‐delta fibers involvement. Basal, maximal, and sham fed acid output were 1.11, 21.87, and 9.37 mmol/hour, respectively. The evoked response to esophageal electrical stimulation was not changed with chronic left vagal electrostimulation. Direct vagal stimulation also produced evoked potentials that were comparable to those obtained with esophageal stimulation. The mean conduction velocity was 6.26 ± 2.72 m/sec (NS) so that there was no evidence of loss of myelinated fibers with chronic stimulation. No differences were detected in basal (1.29 mmol/h), maximal (21.64 mmol/h), or sham fed stimulated (8.03 mmol/h) acid output, showing that vagal electrostimulafion has no effect on either total or vagally mediated acid output, an efferent vagal function. In conclusion, chronic left vagal electrostimulation has no significant adverse effect on gastrointestinal vagal function.

Neurophysiological effects of left vagal stimulation in man.

After discussion with Reese Terry of Cyberonics in 1987, we began a study of the effects of left vagal stimulation in patients with intractable complex partial seizures. Vagal stimulation may produce synchronization and desynchronization of the electroencephalogram in cats.^ This is one form of autonomic stimuiation. Despite the results of animal studies, the clinical application of this work has followed the work of Zabara, The current results of human studies of vagal stimulation are summarized in this volume and in Epiiepsia.^ Our centre has now joined the randomized blinded, parallel controlled study of the Cyberonics Neurocybernetic Prosthesis (NCP), which is approved by the Federal Drug Administration (May 24, 1990) and by Ottawa. We have enrolled seven patients in this study so far. We present the preliminary results of our studies on two patients who have undergone left vagal stimulation for treatment of intractable complex partial seizures.

Cognitive Motor Function After Electrical Stimulation of the Vagus Nerve

Chronic stimulation of the vagus nerve does not seem to produce significant differences between high frequency and low frequency stimulation groups. Individuals within each group show significant changes between prebperative assessment and after 6‐months stimulation. Some subjects showed significant improvement and some showed significant slowing of responses. Subjects who showed improvement are still considerably slower than normals, but all patients have a very long history of complex partial seizures and exposure to multiple medications. Larger homogeneous sample sizes are needed to delineate more clearly the correlation between cognitive performance, medication effects, and stimulation effects.

Acute Effects of High Frequency Vagal Nerve Stimulation on Balance and Cognitive Motor Performance in Epilepsy: Three Case Study Reports

Quantitative measures of area of sway, total sway, and cognitive function failed to show significant differences in acute (50 minute) “ON‐OFF‐ON‐OFF” studies of high frequency left vagal stimulation in three epileptic patients undergoing treatment for chronic complex partial seizures. Fluctuation in Wood levels of anticonvulsants may have been associated with some clinicaJ effects. There were no significant adverse effects of acute left vagal stimulation in these three subjects.

Electrostimulation effects of the vagus nerve on balance in epilepsy.

Preliminary results of selected postural measures in quiet standing indicate that stimulation of the vagus nerve appears not to be producing adverse effects. With this specific sample size, more testing is needed to determine long‐term effects and future data analyses will examine correlations between electroencephalogram results, drug levels, and seizure frequency. In the present study three subjects have had old injuries to hips and ankles. Two subjects had normal values for postural control prior to stimulation, while other subjects were severely abnormal. In future, studies should include larger homogeneous sample sizes, as the current subjects show marked variability in age and premorbid health backgrounds. Future work should also control more vigorously for variables such as visual input (i.e., blindfolding subjects instead of simply closing the eyes). Evaluation of postural control mechanisms will be continued to assess stability changes in these patients as seizure frequency continues to subside.

Chronic stimulation of the left vagus nerve : Cognitive motor effects

BACKGROUNDnEarly studies of cognitive motor control have shown deficits in complex reaction time tests of epileptic subjects. The purpose of this efficacy study was to determine whether chronic (28 months) stimulation of the left vagus nerve (VNS) to control seizures increased these deficits in 6 epileptic subjects with intractable complex partial seizures.nnnMETHODSnSubjects were assessed for simple reaction time, Test A, and subsequent Tests B and C which involved more complex cognitive strategies. Tests were done pre-operatively (SI) and at intervals, 6-8 weeks (S2-S3), and at 6 month intervals (S4-S6) over a 28 month period. Data were collected and collated on an Apple II E computer (Apple, Cupertino CA. U.S.A.) and on electronic switch pad. Data were analyzed using a repeated measures analysis of covariance technique with 2 within subject factors, day, and time of day.nnnRESULTSn2/11 cognitive measures showed a statistically significant difference. Error rate associated with Test A (simple reaction time) significantly decreased for the factor of day (repeated visits) p = .01. For Test C, error rates decreased in the afternoon (p = .03). This test involved the subjects ability to respond quickly to one signal while simultaneously ignoring a second signal. Data analysis of the covariate showed that the effects of VNS are weak in comparison to baseline differences and the frequency of nerve stimulation negatively predicts the number of wrong errors. High frequency stimulation results showed fewer errors than low frequency stimulation T = -2.31, p = .03.nnnCONCLUSIONnChronic stimulation of the left vagus nerve to control seizure activity does not impair cognitive motor control.