J Neurosurg 59:146-149, 1983

Correlation of transient neurological deficit and
somatosensory evoked potentials after intracranial
aneurysm surgery
Case report
ROBERT W. MCPHERSON, M.D., ERNST F. NIEDERMEYER, M.D., RICHARD J. OTENASEK, M.D.,
AND DANIEL F. HANLEY, M.D.

Department of Anesthesiology and Critical Care Medicine, The Johns Hopkins Hospital,
Baltimore, Maryland
v" A patient is reported in whom intraoperative somatosensory evoked potential (SEP) changes occurred in
response to temporary clipping of the right middle cerebral artery. A period of l0 minutes elapsed before
changes in SEP’s in response to contralateral nerve stimulation were noted and, during the following 2 minutes,
the waves decreased in amplitude and then were unrecordable. Waves of SEP, with amplitude similar to those
recorded before clipping but with abnormal latency, returned within 45 seconds of removal of the clip, and
the latency abnormalities persisted until the end of the operation. The patient awakened promptly at the end
of the procedure with a dense left hemiparesis which resolved over 24 hours. At the end of 24 hours, the SEP’s
in response to median nerve stimulation were symmetrical in both latency and amplitude. This report
demonstrates the accuracy of intraoperative evoked potential monitoring in demonstrating alterations of
cerebral perfusion during aneurysm surgery. It also suggests that a prolonged period of observation may be
necessary to assess the effects of temporary vessel occlusion during surgery on aneurysms or arteriovenous
malformations.
KEY WORDS
9 somatosensory evoked potentials
intracranial anenrysm

p

ATIENTSwho have suffered subarachnoid hemorrhage (SAH) from an intracranial aneurysm have
been shown to have abnormalities of cerebral
blood flow and autoregulation in the involved hemisphere. 6’m3 The degree of flow abnormality appears to
correlate with the neurological deficitJ 3 It has been
further demonstrated that hypertension may be necessary to insure adequate perfusion to all areas of the
brain. 5 The perioperative management of these patients
requires maintenance of blood pressure high enough
for adequate perfusion of the brain without undue stress
on the aneurysm. Following the induction of anesthesia,
the inability to assess changes in neurological function
makes this particularly difficult.
The short-latency components of evoked potentials
have been shown to be useful in monitoring neurological function intraoperatively. Anatomical as well as
physiological abnormalities have been shown by such
146

9 intraoperative monitoring

9

monitors, ~6’2~and intraoperative manipulation, such as
moving a retractor, based on these data has been
thought to decrease neurological injury. 7’~sRecent studies have shown that changes in somatosensory evoked
potential (SEP) waveforms parallel changes in function
in patients with SAH from intracranial aneurysmJ 9
This observation, coupled with a recent demonstration
of lack of cerebral autoregulation in patients undergoing
clipping of intracranial aneurysms during induced hypotension, 6 suggests that intraoperative SEP monitoring
offers dynamic assessment of changes in neurological
function. Such monitoring might be expected to demonstrate alterations in neurological functions prior to
permanent neurological damage. An advantage of short
latency SEP recording is its resistance to anesthetic
effects.
Spontaneous scalp electroencephalography (EEG)
has been shown to be an unreliable predictor of neu-

J. Neurosurg. / Volume 59 /July, 1983

Somatosensory evoked potentials in aneurysm surgery
rological injury during induced hypotension and during
aneurysm surgery. 14One difficulty with this method of
neurological monitoring is its sensitivity to anesthetic
agents. The following patient is presented to demonstrate the accuracy and timely nature of the SEP recording in reflecting neurological dysfunction intraoperatively.
Case Report
This 58-year-old right-handed white woman suffered
a severe headache and a grand mal seizure 1 week prior
to admission. Evaluation of the patient at another hospital, including angiography and computerized tomography scanning, demonstrated a 10-mm aneurysm at
the trifurcation of the right middle cerebral artery
(MCA). At the time of presentation for aneursym clipping, the patient was classified as neurological Grade I
according to Hunt and Hess.l
Preoperative Protocol Prior to induction of anesthesia, an arterial line and central venous catheter were
placed. Gold cup electrodes were placed at C3’, C4’
over the left and right scalp region (according to the
international 10-20 system), the C-2 vertebra, and Erb’s
point bilaterally, utilizing dermal abrasion. Stimulating
needle electrodes were placed percutaneously over each
median nerve, and SEP’s were obtained prior to the
induction of anesthesia. The SEP’s were developed with
a Nicolet Med 80 physiological monitor.* Electrode
impedances were maintained at less than 2 kohm except
for sterile needle electrodes which were about 3 kohms.
Band-pass filters were 5 to 1500 Hz. A time period of
2 to 82 msec after stimulus was observed and 128
stimuli at a rate of 5.9/sec were accumulated and averaged. Anesthesia was induced with fentanyl, 25 ug/
kg, followed by 0.15 mg/kg pancuronium. Throughout
surgery, the mean arterial blood pressure (MABP) was
maintained within 10% of control, and the PaCO2
maintained between 25 and 30 mm Hg.

lntraoperative Monitoring. The waveforms obtained by median nerve stimulation are shown in Fig.
1. Reproducible SEP’s were obtained subsequently during positioning and initiation of surgery. The amplitude
of the SEP over the right hemisphere was slightly but
consistently less than that over the left hemisphere, with
symmetrical wave latencies. After the surgical preparation the C4’ (fight hemisphere) electrode was replaced
with a sterile needle electrode. During the initial parts
of surgery, SEP’s were obtained approximately every 5
minutes.
During the dissection of the aneurysm, which had an
ill defined neck, bleeding developed in the area of the
aneurysm and a temporary clip was applied to the
MCA. A period of approximately 10 minutes elapsed
in which no detectable changes were noted in the SEP
* Nicolet Med 80 physiological monitor manufactured by
Nicolet Instrument Corp., Madison, Wisconsin.

J. Neurosurg. / Volume 59 ~July, 1983

FIG. 1. The waves generated by median nerve stimulation
are shown. The waves over the left hemisphere (on stimulating
the right median nerve) were generated first and the waves
over the right hemisphere (on stimulating the left median
nerve) were generated approximately 45 seconds later. The
times shown at left are the starting times of stimulating the
right median nerve. MCA = middle cerebral artery.

(wave developed at 1053 hours as shown in Fig. 1).
During that time, the SEP wave over the right hemisphere was developed approximately every 45 seconds.
At the end of 10 minutes, the SEP over the right
hemisphere deteriorated over approximately 2 minutes
and became unrecordable (wave developed at 1056
hours, Fig. 1). The evoked response recorded at the
level of the C-2 vertebra at that time in response to
stimulation of the left nerve was unchanged in both
amplitude and latency, indicating a failure of the right
hemisphere along with integrity of the upper spinal
region and suggesting a block between the brain stem
and cortex. No change from the preclip recordings was
noted in SEP’s in response to stimulation of a right
median nerve. The clip was quickly removed from the
MCA and the SEP wave reappeared over the right
hemisphere within approximately 45 seconds. The
waveform demonstrated an amplitude similar to that
prior to clipping, with an increase in the latency of the
waves (waves developed at 1058 hours, Fig. 1). The
conduction time from the C-2 vertebra to the right
cortex was increased by approximately 20%, whereas
the conduction time from the C-2 vertebra to the left
cortex was unchanged. This asymmetry of the SEP’s
continued until the end of the procedure. Approximately 45 minutes after removal of the temporary clip
from the MCA, a clip was applied to the aneurysm. At
the time of clip application it was thought that a small
perforating artery was included in the clip. No changes
in SEP wave over either left or right hemisphere were
147

R. W. McPherson, et al.
noted with aneurysm clipping. It should be noted that
no changes in MABP or heart rate were observed during
the period when the MCA was clamped or subsequent
to removal.
Postoperative Course. The patient awakened
promptly at the end of the procedure with a dense left
hemiparesis. The only movement of the left side was
weak hip flexion. Over the following 24 hours, the
hemiparesis resolved almost completely, and SEP’s obtained by transcutaneous stimulation of the median
nerves showed symmetry of the SEP in both amplitude
and latency, although the amplitude was decreased
below the control level. Central conduction time was
equal (C-2 vertebra to the right cortex, and C-2 vertebra
to the left cortex). Over the following few days weakness
in the left side completely resolved, and, at the time of
this report, the patient’s neurological examination is
normal.
Discussion

This case is reported to demonstrate the association
of changes in the SEP’s and neurological function during aneurysm clipping. This technique represents a
reversible ischemic event that is best explained by clipping of the MCA. Although this group of patients is at
risk of ischemia due to arterial spasm and failure of
autoregulation both prior to and during surgery as well
as hypoperfusion during induced hypotension, these
events seem less likely in this situation. Attempts to
correlate EEG changes with neurological deficits during
aneurysm surgery have been unsuccessful. ~4 This may
be due either to diagnostic difficulties caused by anesthetic effects on EEG or to a subcortical site of injury.
An increased conduction time of the SEP through the
brain stem has been shown to correlate with brain stem
dysfunction.9"w Hence, the SEP might be useful in
assessing such subcortical sites of injury. Although a
clearly definable event precipitated the SEP changes,
the literature suggests that diffuse physiological changes
may cause similar changes in the SEP waveform.8
The prolonged time course before failure of the SEP
occurred is longer than might be expected from results
of experimental MCA clipping in animals. 3 This is
perhaps due to the moderately low temperature (35.5"C)
of the patient as well as the reduced cerebral metabolic
requirement for oxygen secondary to anesthesia. The
correlation of return of SEP waveform to normal with
improvement of neurological function is consistent
with the findings in animal studiesz’4 and in clinical
studies performed shortly after head injury? ~ It also
suggests that if SEP monitoring is used to assess the
effect of vessel occlusion during arteriovenous malformation surgery, as has been suggested, 17 a prolonged
period following temporary clipping may be necessary
to demonstrate physiological alteration due to surgical
intervention.
Recognition of exact time of SEP changes indicating
neurological dysfunction resulted from the following:
148

1) direct communication between the surgical and anesthesia team; 2) a monitoring technique that allows rapid
assessment of integrity of the stimulating system (Erb’s
point and the C-2 vertebra); and 3) waveform development over a short period of time (approximately 30
seconds).
Transient neurological dysfunction occurred in this
patient despite reperfusion of the affected area within 2
minutes of the loss of evoked electrical activity. Technical limitations cause this minimum time in which
abnormal waveforms can be developed, evaluated, and
verified intraoperatively. Movement artifact or electrocautery could easily have slowed or prolonged the averaging processing, and prolonged the time to recognition of SEP abnormality.
Although it is unclear how long areas of the brain
can survive lack of oxygen during anesthesia and mild
hypothermia, the transient nature of this patient’s neurological deficit suggests that impairment of oxygen
delivery was significant. Animal studies have shown
that the degree of ischemia necessary for disappearance
of the evoked potential is less than that necessary for
permanent neurological damage. 1 This suggests that
recognition of SEP changes may allow restoration of
flow prior to onset of permanent neurological deficit.
The rapid disappearance of all recognizable waves
(the N20-P23 complex) with preservation of the evoked
potential in the upper spinal cord suggests that the block
was probably in the area of the thalamus, which is
considered to be the origin of the N2o wave? 5 The
prolongation of conduction time from the C-2 vertebra
to the right cortex following clip removal, which persisted until the end of the operation, also supports a
subcortical site of injury.
This observation has importance in several clinical
situations. It demonstrates that ischemic events can be
identified in the absence ofhemodynamic changes. This
patient also demonstrates that in anesthetized patients,
functional changes associated with surgical intervention
may take a longer period to develop than is suggested
by animal studies.
References

1. Astrup J, Symon L, Branston NM, et al: Cortical evoked
potential and extracdlular K+ and H+ at critical levels of
brain ischemia. Stroke 8:51-57, 1977
2. Bennett MH, Albin MS, BuneginL, et al: Evoked potential changes during brain retraction in dogs. Stroke 8:
487-492, 1977
3. Branston NM, Symon L, Crockard HA, et al: Relationship betweenthe corticalevokedpotential and localblood
flow followingacute middle cerebral artery occlusion in
the baboon. Exp Neurol 45:195-208, 1974
4. BriefleyJB, BrownAW, Excell BJ, et al: Brain damage in
the rhesus monkey resulting from profound arterial hypotension. I. Its nature, distribution and general physiological correlates. Brain Res 13:68-100, 1969
5. Brown FD, Hanlon K, Mullan S: Treatment of aneurysmal hemiplegia with dopamine and mannitol. J Neurosurg 49:525-529, 1978
J. Neurosurg. / Volume 59 ~July, 1983

Somatosensory evoked potentials in aneurysm surgery
6. Farrar JK, Gamache FW Jr, Ferguson GG, et al: Effects
of profound hypotension on cerebral blood flow during
surgery for intracranial aneurysms. J Neurosurg 55:
857-864, 1981
7. Grundy BL, Heros RC, Tung AS, et ah Intraoperative
hypoxia detected by evoked potential monitoring. Anesth
Analg (Cleve) 60:437-439, 1981
8. Grundy BL, Lina A, Procopio PT, et al: Reversible evoked
potential changes with retraction of the eighth cranial
nerve. Anesth Analg (Cleve) 60:835-838, 1981
9. Hargadine JR, Branston NM, Symon L: Central conduction time in primate brain ischemia - - a study in baboons.
Stroke 11:637-642, 1980
10. Heilbrun MP, Olesen J, Lassen NA: Regional cerebral
blood flow studies in subarachnoid hemorrhage. J Neurosurg 37:36-44, 1972
11. Hume AL, Cant BR: Central somatosensory conduction
after head injury. Ann Neurol 10:411-419, 1981
12. Hunt WE, Hess RM: Surgical risk as related to time of
intervention in the repair of intracranial aneurysms. J
Neurusurg 28:14-20, 1968
13. Ishii R: Regional cerebral blood flow in patients with
ruptured intracranial aneurysms, d Neurosurg 50:
587-594, 1979
14. Jones TH, Chiappa KH, Young RR, et al: EEG monitoring for induced hypotension for surgery of intracranial
aneurysms. Stroke 10:292-294, 1979

J. Neurosurg. / Volume 59 ~July, 1983

15. Maugui&e F, Courjon J: The origins of short-latency
somatosensory evoked potentials in humans. Ann Neurol
9:607-611, 1981
16. No~l P, Desmedt JE: Somatosensory cerebral evoked
potentials after vascular lesions of the brain-stem and
diencephalon. Brain 98:113-128, 175
17. Owen MP, Brown RH, Spetzler R_F, et al: Excision of
intramedullary arteriovenous malformation using intraoperative spinal cord monitoring. Surg Neurol 12:
271-276, 1979
18. Raudzens PA, Shetter AG: Intraoperative monitoring of
brain-stem auditory evoked potentials. J Neurosurg
57:341-348, 1982
19. Symon L, Hargadine J, Zawirski M, et al: Central conduction time as an index of ischaemia in subarachnoid
haemorrhage. J Neurol Sci 44:95-103, 1979
20. Uematsu S, Tolo VT: Recording of the somatosensory
evoked potentials during surgery for scoliosis and midline
myelotomy to monitor spinal cord function. Electromyogr Clin Neurophysiol 21:253-266, 1981
Manuscript received November 12, 1982.
Accepted in final form January 27, 1983.
Address reprint requests to: Robert W. McPherson, M.D.,
Department of Anesthesiology and Critical Care Medicine,
The Johns Hopkins Hospital, Meyer 8-138A, 600 North Wolfe
Street, Baltimore, Maryland 21205.

149