Journal of Clinical Monitoring and Computing (2004) 18: 303–308
C©
Springer 2005
THE EFFECTS OF ISOFLURANE AND PROPOFOL ON
INTRAOPERATIVE NEUROPHYSIOLOGICAL
MONITORING DURING SPINAL SURGERY
Zhengyong Chen, MD, MSc, CNIM, DABNM
From Active Diagnostics Inc., 216 F Street, Suite 76, Davis, CA
95616, USA
Address correspondence to Zhengyong Chen, Active Diagnos-
tics Inc., 216 F Street, Suite 76, Davis, CA 95616, USA E-mail:
chenzy68@hotmail.com
Chen Z, The effects of isoflurane and propofol on intraoperative neuro-
physiological monitoring during spinal surgery
J Clin Monit 2004; 18: 303–308
ABSTRACT. Objectives. To compare the effects of isoflurane
and propofol on intraoperative neurophysiological monitoring
(IONM) during spinal surgery. Methods. Thirty-five patients
were randomly assigned to receive isoflurane (n = 17) or propo-
fol (n = 18) anesthesia. Somatosensory evoked potentials (SEPs)
following posterior tibial nerve stimulation were recorded be-
fore induction as baselines. Isoflurane concentrations and propo-
fol infusions were adjusted to obtain four pre-determined BIS
ranges: 65–55, 55–45, 45–35 and 35–25. For each range, a sta-
ble state was maintained for at least 10 min to perform IONM.
The SEP latency P40 and amplitude P40-N50, the onset latency
and amplitude of transcranial motor evoked potentials (tcMEPs),
and threshold intensity of triggered electromyographic activity
(EMG) following pedicle screw stimulation were statistically an-
alyzed. Results. Compared with baseline values, P40 latency in-
creased and P40-N50 amplitude decreased after anesthesia with
isoflurane or propofol. Isoflurane caused a dose-dependent de-
pression of SEPs, but propofol did not. TcMEPs were recordable
and stable in all patients receiving propofol in each BIS range,
but only recordable in 10 (58.8%) receiving isoflurane with BIS
>55, and 3 (17.8%) with BIS <55. No difference was noted
in triggered EMG. Conclusions. Isoflurane inhibited IONM
more than propofol. Propofol is recommended for critical spinal
surgery, particularly when motor pathway function is monitored.
KEY WORDS. Intraoperative monitoring, evoked potentials, elec-
tromyographic activity, isoflurane, propofol.
INTRODUCTION
Intraoperative neurophysiological monitoring (IONM) of
somatosensory evoked potentials (SEPs), motor evoked po-
tentials to transcranial electrical stimulation (tcMEPs), and
free-run and triggered electromyographic activity (EMG)
has been widely used to detect iatrogenic spinal cord or
nerve root injury or misplaced pedicle screws in spinal
surgery [1–6]. These IONM tests provide information re-
garding the functional integrity of the nervous system of a
patient who is anesthetized and therefore cannot be neu-
rologically examined.
The efficacy of intraoperative monitoring depends on
the quality of the neurophysiological signals recorded in
the operation theatre. However, these signals are affected
by many anesthetic agents, which may result in both false-
negative and false-positive results during monitoring.
Isoflurane and propofol are widely used inhalational and
intravenous anesthetic agents respectively. Previous studies
have reported that inhalational agents have a greater effect
on the intraoperative neurophysiological monitoring tests
than intravenous agents [7–10]. Nevertheless, these studies

304 Journal of Clinical Monitoring and Computing Vol 18 No 4 2004
did not standardize anesthetic depth. Hence, the difference
reported in the effects of the anesthesia might be due to
the difference in the anesthetic depth.
Bispectral index (BIS) is one method of monitoring
anesthetic depth and can be applied to anesthesia for spinal
surgery, the goal of which is to prevent over- or under-
dosage of anesthetics. It has been studied widely and is in
routine use as a monitor of anesthetic depth [11–13].
The aims of the study were to compare the effects of
isoflurane and propofol on cortical SEPs, tcMEPs and
triggered EMG from pedicle screw stimulation at pre-
determined BIS values during spinal surgery, and to pro-
vide some recommendations regarding use of isoflurane
and propofol during spinal surgery when IONM is used.
PATIENTS AND METHODS
Patients
After institutional ethics committee approval and with
written informed consent from each patient or from his
or her parents, 35 patients who underwent elective spinal
surgery were randomized into one of two groups the day
before surgery. Patients were allocated in blocks of ten, in
a ratio of 1:1, to receive isoflurane or propofol anaesthe-
sia using a block randomization table. Group 1 (n = 17,
10 females) received isoflurane, and group 2 (n = 18, 11
females) received propofol. Patients with existing neuro-
logical deficits, cerebral palsy, history of seizure or epilepsy,
and cardiovascular diseases were excluded from the study.
Anesthesia
Patients received oral midazolam pre-medication in the
ward prior to transfer to the operating theater. On arrival
in the operating room, intravenous access was secured and
10 µgofintravenous alfentanil was administered.
Muscle relaxant was used only during induction for in-
tubation. Train-of-four (TOF) was used to monitor muscle
relaxation, and it was kept at least 3 during neurophysio-
logical recording.
Anesthesia was induced in all patients using propofol 2
mg/kg and alfentanil 10 µg/kg. Tracheal intubation was
facilitated with atracurium 0.5–0.6 mg/kg.
In the propofol group, anesthesia was maintained with
propofol infusion, initially at 10 mg/kg/hr; this infusion
rate was titrated to maintain BIS values at different pre-
determined levels.
In the isoflurane group, anesthesia was maintained
with isoflurane delivered in an air-oxygen mix of car-
rier gas, and the end-tidal concentration was adjusted
to maintain the BIS values at different pre-determined
levels.
Neither opioids nor nitrous oxide were given during
the recording sessions. Mean arterial pressure (MAP) and
core temperature were monitored. If required, intravenous
boluses of ephedrine were used to treat hypotension and
maintain blood pressure within 20% of baseline values.
Core temperature was monitored using a nasopharyngeal
probe and maintained above 36
◦
C throughout the study
period. Thermal mattress, forced air warmers, i.v. fluid
warmers, and the room temperature were adjusted to main-
tain a stable body core temperature throughout the proce-
dure.
BIS monitoring
BIS (BIS Monitor A-2000; Aspect Medical Systems, Nat-
ick, MA, USA) were recorded before anesthesia and
throughout the procedure.
IONM was continuously monitored during surgery and
anaesthesia. Our primary time for IONM measures was
during the maintenance phase of anaesthesia but before
the commencement of surgery. Secondary measurements
were made during spine exposure, pedicle screw insertion
(where screws were used), and rod insertion (where rods
were used).
Anesthesia administration was adjusted to keep BIS val-
ues at four different pre-determined ranges, namely 65–55,
55–45, 45–35 and 35–25. At each BIS range, there was less
than 10% variation in BIS values over a 10–15-min period
of observation for neurophysiological recording.
SEPs monitoring
SEPs were elicited by interleaved stimulation of bilateral
posterior tibial nerves at the ankle using surface electrodes.
Regular pulses of 30 mA lasting 0.2 msec were delivered
at a rate of 4.7 Hz. SEPs were recorded through subder-
mal needle electrodes placed over the cortex: the active
electrode on the scalp at Cz’, in the midline 2 cm behind
the vertex Cz with a reference electrode placed at Fz. The
ground electrode was placed on the calf. Recording elec-
trode impedance was balanced and maintained at less than 3
k
. The amplifier bandpass was 20 Hz to 2 kHz with a gain
setting of 10 µV/Div. An analysis time of 100 msec was
used for each SEP waveform, and 100–200 sweeps were
averaged.
Pre-induction baseline cortical SEPs were recorded af-
ter 10 µgofintravenous alfentanil was administrated when
the patient was awake. At least two sets of averages were
checked for reproducibility for each BIS range. In each BIS
range, the averages of P40 latency and P40-N50 amplitude