Epilrpsiu, 40( 10):1452-1454, 1999

Lippincott Williams Kr Wilkins, Inc., Philadelphia

0 lnternational League Against Epilepsy

Brief Communication

Bradycardia and Asystole with the Use of Vagus Nerve

Stimulation for the Treatment of Epilepsy: A Rare Complication
of Intraoperative Device Testing

Jorge J. Asconape, ?David D. Moore, “Douglas P. Zipes, Laura M. Hartman, and

$William H. Duffell, Jr.
Department of Neurology, “Krunnert Institute of Curdiology, Indiana University School of Medicine, Indianapolis; und fElkhart
Clinic, Elkhart, Indiana; and ;tCyberonics Inc., Houston, Texas, U.S.A.

Summary: Purposes: A 56-year-old man with mild mental Conclusions: Possible mechanisms for the bradycardia/
retardation, right congenital hemiparesis, and refractory partial asystole include stimulation of cervical cardiac branches of the
seizures was referred for vagus nerve stimulation (VNS). vagus nerve either by collateral current spread or directly by
Methods: Routine lead diagnostic testing during the surgical inadvertent placement of the electrodes on one of these
procedure (1.0 mA, 20 Hz, and 500 ks, for -17 s) resulted, branches; improper plugging of the electrodes into the pulse
during the initial two stimulations, in a bradycardia of -30 generator, resulting in erratic varying intensity of stimulation;
beatdmin. A third attempt led to transient asystole that required reverse polarity; and idiosyncratic-type reaction in a hypersus-
atropine and brief cardiopulmonary resuscitation. ceptible individual. The manufacturer reports the occurrence
Results: The procedure was immediately terminated, the de- rate in -3,500 implants for this intraoperative event to be ap-
vice removed, and the patient recovered completely. A postop- proximately one in 875 cases or 0.1%. Key Words: Vagus
erative cardiologic evaluation, including an ECG, 24-h Holter nerve-Electrical stimulation-Cardiac arrhythmia-
monitor, echocardiogram, and a tilt-table test, was normal. Epilepsy-Seizure.

Vagal nerve stimulation (VNS) was recently intro- States (1,2). We report a case of severe bradycardia with
duced as an alternative therapy for treatment of medi- transient asystole during VNS.
cally refractory partial seizures. Given the important role
of the vagus nerve in the regulation of the heart rhythm, CASE REPORT
the potential occurrence of cardiac arrhythmias during A 56-year-old man with diagnosis of refractory com-
VNS was a serious initial concern. Subsequent Holter plex partial seizures, mild mental retardation, and a con-
monitoring and analysis in humans, however, failed to genital right hemiparesis underwent a surgical procedure
document clinically relevant cardiac rhythm changes in a for implantation of a vagal nerve stimulator.
large population of patients with epilepsy treated with During the routine intraoperative lead diagnostic test
VNS (1-3). More recent studies noted slight changes in (stimulation parameters: 1.O mA, 20 Hz, and 500 ks, for
heart rate during VNS in humans (4). There were no -17 s) the patient developed a bradycardia of -30 beats/
reports of intraoperative asystole in the clinical trials (N min that quickly reversed once the stimulation stopped.
= 454) used to support market approval in the United A second attempt, as a rechallenge to observe the cardiac
effect, resulted in a similar response. The third attempt
led to transient asystole that required several chest com-
pressions and the administration of atropine. The proce-
Accepted April 8, 1999.
Address correspondence and reprint requests to Dr. J. J. AsconapC at dure was terminated, the device was removed, and the
Department of Neurology, Indiana University School of Medicine, patient recovered completely. Propofol, fentanyl, dro-
University Hospital, Room 1711, 5.50 N. University Blvd., Indianapo- peridol, and rocuronium bromide were used during the
lis, IN 46202-52.50, V.S.A. E-mail: jasconap@iupui.edu
Presented at the American Epilepsy Society Meeting in San Diego, anesthesia, and no abrupt changes or additions would
California, December 9, 1998. have accounted for this event.

I452
 BRADYCARDIA WITH VAGUS NERVE STIMULATION 1453

Medical history was significant for mild chronic ob- gested by animal studies (6). In dogs, where the cervical
structive pulmonary disease and hypertension. There was cardiac nerves originate more distally than in humans,
no history of cardiac disease. The patient was mildly stimulation of the vagus nerve in the cervical region,
depressed and had abused alcohol in the past. Medica- proximal to the cervical cardiac branches, consistently
tions included lamotrigine (LTG), 150 mg b i d . ; primi- produces bradycardia or asystole. This effect is rapidly
done (PRM), 50 mg t.i.d.; sertraline hydrochloride, 50 reversible without intervention after stopping the stimu-
mg 9.h.s.; labetalol hydrochloride, 100 mg b.i.d.; and lation, as was the case in our patient after the initial two
olanzapine, 15 mg q.h.s. stimulations. Atropine and cardiopulmonary resuscita-
Family history revealed that his father died of cancer tion were used after the third stimulation in our patient,
at age 62 years. His mother died of pneumonia in her 80s. but it is likely that the bradycardidasystole would have
Physical examination revealed a well-developed, well- reversed spontaneously once the stimulation ended with-
nourished man with mild mental retardation. Blood pres- out any specific intervention. Alternatively, our subject
sure was 106/72 mm Hg; pulse rate, 72 beatdmin. Right might have had an aberrant anatomy of the vagus nerve
and left carotid sinus massage produced only slight slow- with lower branching of the cervical cardiac nerves, re-
ing of the heart rate. Cardiac rate and rhythm were regu- sulting in stimulation of nerve bundles that later branch
lar without murmurs, gallops, or rubs. Lungs were clear into the inferior cervical cardiac branch of the vagus
to auscultation. The rest of the general physical exami- nerve.
nation was unremarkable. Neurologic examination Another possible explanation is indirect or collateral
showed a mild right spastic hemiparesis. Deep tendon stimulation of the cervical cardiac nerves. The cardiac
reflexes were increased on the right side, with a right
nerves are located next to the vagus and could be stimu-
Babinski sign.
lated when the current is applied to the vagus nerve.
A 12-lead ECG, 24-h Holter monitor, echocardiogram,
Intraoperatively, current spread may be favored by pool-
and a tilt-table test were normal. EEGs in the past have
ing of blood or saline solutions used to irrigate the sur-
shown infrequent sharp forms arising from the left tem-
gical field.
poral and centroparietal regions. A computed tomogra-
Technical malfunction of the device also needs to be
phy (CT) scan showed diffuse atrophy more prominent
considered; however, this case was submitted to the
over the left hemisphere.
manufacturer, and it was found to be functioning prop-
erly. Still, other possibilities remain. Improper connec-
DISCUSSION tion (via one or both of the setscrews) of the leads into
The bradyarrhythmia observed in our case would be the header of the pulse generator may cause wide varia-
the anticipated cardiac effect of VNS (5). The reason that tions in the impedance, resulting in erratic varying in-
left VNS in humans does not result in bradycardia more tensity of stimulation (6). In this situation, currents could
frequently is not completely understood. The most likely vary between 0.25 and 12 mA (the output limits of the
explanation is that placement of the stimulating elec- device’s battery), well above and below the programmed
trodes distal to the origin of the superior and inferior 1.0 mA for the lead diagnostic test. Such extreme and
cervical cardiac branches of the left vagus nerve avoids erratic swings in stimulation parameters may be arrhyth-
direct stimulation of a large contingent of cardiac fibers. mogenic.
The mechanism of the bradycardidasystole in our pa- Another factor to consider is the effect of polarity
tient remains largely speculative. The surgeons involved reversal. It is possible that the surgeon might have inad-
in the case did not describe any technical difficulties vertently reversed the polarity of the leads (correct con-
during the procedure nor did they notice any unusual nection: negative electrode cranial). The design of the
anatomy of the vagus nerve. An extensive cardiac evalu- device’s lead connection boots do not preclude this from
ation after the surgery failed to demonstrate any predis- occurring. Data provided by the manufacturer show that
posing conditions. reversing the polarity of the device in dogs significantly
Various possibilities should be considered. It is pos- enhances the cardiac effects of VNS (7). In this study,
sible that the electrodes were inadvertently placed on one with proper connection (negative electrode cranial), it
of the cervical cardiac branches of the vagus nerve. Nor- took -5 mA to produce a 20% bradycardia, whereas with
mally the electrodes are placed on the left vagus nerve the pins reversed, it took only 0.8 mA to obtain the same
distal to the origin of the superior and inferior cervical effect (7).
cardiac nerves, thus reducing the likelihood of a cardiac Finally, the bradymhythmia may be an idiosyncratic,
effect. During a surgical procedure, these three nerves exaggerated reaction in a nerve virgin to stimulation. It is
are very similar to visual inspection and run in close possible that stimulation at lower intensities, which was
proximity. Stimulation of the cervical cardiac branches not done in this case, might have been better tolerated.
would probably result in bradycardia or asystole, as sug- The lead diagnostic test is performed at the 1.0-mA set-

Epilepsia, Vol. 40, No. 10, 1999

1454 J . J . ASCONAPE ET AL.

ting because this current level is necessary to verify that The manufacturer reports the occurrence rate in
it is fully functional. -3,500 implants for this intraoperative event to be ap-
More information is needed to understand the mecha- proximately one in 875 cases or 0.1% (6). Physicians
nism of bradycardia and asystole in our patient. We pro- who have had or know of other similar events are en-
pose the following steps to be completed in the operating couraged to report those to the manufacturer so more can
room in case of future occurrences: verify that the elec- be learned about the possible causes and contributors to
trodes are placed on the vagus nerve; locate the cervical these events.
cardiac branches to assure that electrodes are placed dis-
tal to their origin; verify device polarity; check that the REFERENCES
leads are well seated into position and securely locked 1. The Vagus Nerve Stimulation Study Group. A randomized con-
into place with the setscrews; check for saline or blood trolled trial of chronic vagus nerve stimulation for treatment of
medically intractable seizures. Neurology 1995;45:224-30.
pool around the lead’s stimulation coils. Once all this has 2. Handforth A, DeGiorgio CM, Schachter SC, et al. Vagus nerve
been checked, consider repeating stimulation at lower stimulation therapy for partial-onset seizures: a randomized active-
settings (i.e., 0.25 mA, 250 ps, 20 Hz, for 14 s) with control trial. Neurology 1998;5 1:48-55.
3. Setty AR, Vaughn BV, Quint SR, et al. Heart period variability
recorded ECG strip. Proceed gradually to step up stimu- during vagal nerve stimulation. Seizure 1997;6:1-6.
lation according to tolerance. Consider using Silastic 4. Frei MG, Davidchack R, Lawrence KS, Osorio 1. Effects of vagal
dam to insulate against collateral stimulation of cervical stimulation on human ECG. Epilepsia 1998;39(suppl 6):200.
5. Zipes DP. Autonomic modulation of cardiac arrhythmias. In: Zipes
cardiac branches. Defer starting prolonged VNS for 2 DP, Jalife J, eds. Cardiac electrophysiology: from cell to bedside.
weeks after implantation to minimize collateral current Philadelphia: Saunders, 1994:441-53.
spread. Collateral current spread may be reduced by the 6. Data on file at Cyberonics, Inc.
7. Smith CD, Bourland JD, Foster KS, Geddes LA, Schoenlein WE.
encapsulation of the lead and stimulation coils in a col- The chronaxie and propagation velocity of canine vagus nerve
lagen matrix, which forms in the weeks after surgery. fibers in vivo. Data of file at Cyberonics, Inc.

Epiiepsia, Vol. 40, No. 10, 1999