ORIGINAL RESEARCH

Overview of Intraoperative Neurophysiological Monitoring During

Spine Surgery
Parastou Shilian, Gabriel Zada, Aaron C. Kim, and Andres A. Gonzalez
University of Southern California, Los Angeles, CA, U.S.A.

Summary: Intraoperative neurophysiologic monitoring has nature of these modalities will help tailor monitoring to
had major advances in the past few decades. During spine a particular procedure to minimize postoperative neurologic
surgery, the use of multimodality monitoring enables us to deficit during spine surgery.
assess the integrity of the spinal cord, nerve roots, and
peripheral nerves. The authors present a practical approach Key words: Intraoperative Neurophysiological Monitoring, Spine
to the current modalities in use during spine surgery, surgery, Somatosensory evoked potentials, Motor evoked po-
including somatosensory evoked potentials, motor evoked tentials, Electromyography.
potentials, spinal D-waves, and free-run and triggered
electromyography. Understanding the complementary (J Clin Neurophysiol 2016;33: 333–339)

O ver the past several decades, various modalities of electrophys-

iological tests have been increasingly used in the operating room
to monitor different parts of the nervous system specifically at risk, to
spinal cord function. A typical protocol includes delivery of
a peripheral stimulus and subsequent recording proximally in the
region of the somatosensory cortex, thereby monitoring the
prevent or minimize the probability of neurological injury. In integrity of major afferent pathways of the dorsal spinal cord.
addition, the information gleaned from intraoperative neurophysio- Somatosensory evoked potentials were first recorded by Dr.
logical monitoring provides guidance to the neurosurgical team Dawson in 1947, by stimulating the peripheral nerve and
throughout the course of the operation. Intraoperative neurophysi- recording responses over the somatosensory cortex (Dawson,
ological monitoring has become a useful adjunct to a variety of spinal 1947). Two decades later, this technology was applied for the
operations, including spinal tumor resection, spinal decompression first time in the operating room in an institution affiliated with the
and fusion, deformity surgery, and vascular surgery of the spinal cord. University of Southern California. In 1974, Nash et al introduced
Intraoperative monitoring has been in use since the 1970s. One the use of SSEP for intraoperative monitoring of the nervous
of the first methods used to detect neurological injury during surgery system during spine surgery.
was the wake-up test, pioneered by Vauzele (1973). As part of this Early SSEP monitoring used middle and long-latency
cursory, yet pioneering monitoring test, general anesthesia was cortical potentials but had limited use due to a high degree of
transiently discontinued after insertion of spinal hardware to perform variability, excess noise, and sensitivity to anesthesia. Nuwer and
a neurological examination. Benefits of the Stagnara wake-up test Dawson established parameters for SSEP monitoring to improve
were that it was relatively simple to perform, without the need for its reproducibility and efficacy in the operating room. These
special equipment. Although the wake-up test was functional, it had modifications involved optimizing the stimulation rate, filter
several inherent limitations, such as a potential delay in identification settings, anesthesia protocol, and scalp montages to achieve
of neurological deficits (because it was only performed at one time a more favorable signal-to-noise ratio (Nuwer and Dawson,
point), risk of accidental extubation, or other anesthesia complica- 1984).
tions, air embolization, and possible movement of surgical instru-
mentation (Hall et al., 1978; Nuwer and Dawson, 1984; Vauzelle
et al., 1973). In recent decades, the evolution of safer and more Anatomy
sophisticated real-time modalities for neurophysiological monitoring Somatosensory evoked potentials assess the integrity of the
have essentially replaced the need to perform such wake-up tests and dorsal column-medial lemniscus pathway, as initially described
have transformed the subspecialty of neurophysiological monitoring by D’Angelo et al. (1973).
and modern spinal surgery into the discipline it is today. The dorsal column–medial lemniscus pathway conducts the
sensory modalities of vibration, proprioception, and fine touch.
From the peripheral mechanoreceptors and stretch receptors,
signal travels through dorsal horn neurons. First-order neuron
SOMATOSENSORY EVOKED POTENTIALS axons ascend through the cuneatus and gracilis fasciculi and
Somatosensory evoked potentials (SSEPs) are an extremely synapse on second-order neurons in the cuneate and gracile
useful and sensitive modality for the intraoperative assessment of nuclei, respectively. These fibers cross midline as internal arcuate
Address correspondence and reprint requests to Parastou Shilian D.O., Department
fibers and ascend as the medial lemniscus en route to the
of Neurology, University of Southern California, 1520 San Pablo, Suite 3000, pathway’s second synapse in the ventral posterolateral nucleus of
Los Angeles, CA 90033; email: [email protected] the thalamus. Axons of the third-order neurons ascend through
Copyright Ó 2014 by the American Clinical Neurophysiology Society
ISSN: 0736-0258/14/3304-0333 the posterior limb of the internal capsule to the primary
DOI 10.1097/WNP.0000000000000132 somatosensory cortex.

clinicalneurophys.com Journal of Clinical Neurophysiology Volume 33, Number 4, August 2016 333
 P. Shilian, et al. Overview of IOM During Spine Surgery

assessment of spinal cord function during spinal surgery.

METHODOLOGY However, SSEP monitoring has its own set of limitations and
Stimulation sites are selected based on the spinal cord level is very sensitive to various anesthetics (particularly halogenated
of interest. The median nerve is typically monitored for high agents and nitrous oxide), as well as temperature and blood
cervical procedures, and the ulnar nerve is generally used for pressure (Eager et al., 2011; Jou, 2000; Porkkala et al., 1997;
lower cervical procedures. Even if the pathway of the upper Sloan and Heyer, 2002). Optimization of many intraoperative
extremity nerve used is outside the area at risk (i.e., thoraco- parameters and a close working relationship with the anesthesia
lumbar procedures), monitoring can provide general information team is therefore generally required during these operations.
about systemic (blood pressure, temperature), anesthetic, or Because of a low signal-to-noise ratio in the operating room,
changes due to positioning. For thoracolumbar and cervical averaging of many stimuli is usually required to provide
operations, the posterior tibial nerve at the ankle is the preferred a reproducible signal.
monitoring site. If the posterior tibial nerve is compromised (i.e., Many monitoring practitioners use an amplitude and latency
below the knee amputation or severe peripheral neuropathy), the criteria to raise alarm during SSEP monitoring. Most accepted
peroneal nerve at the knee can be used as an alternate. criteria include either a 50% decrease in SSEP waveform
A stimulation rate of approximately 5 Hz (that are not amplitude or a 10% increase in latency from baseline (Nuwer,
multiple of 60, i.e., 4.47 Hz) is optimal (Nuwer and Dawson, 2008; Nuwer et al., 1995). However, although not systematically
1984) for SSEP monitoring. A slower stimulation frequency will studied, a 30% drop in amplitude observed during a critical
yield higher amplitude responses but will slow the time to obtain portion of the surgery with an otherwise stable waveform may be
averaged sets of responses. The cortical N20 potential can be significant. Changes due to anesthesia need to be taken into
recorded at CPc-CPi or CPc-FPz (International 10-10 system), account when evaluating cortical potential amplitudes. Sub-
and P37 can be recorded at CPi-CPc, CPi-FPz, and Cz-FPz cortical responses are more resistant and can be helpful in
channels (Fig. 1). Subcortical far field potentials, including N13, distinguishing changes caused by anesthetics. Changes in SSEP
P14, N18, P31, and N34 can be recorded from C5s-FPz channel. monitoring can sometimes be offset by optimization of blood
It is advisable to use multiple channels to record cortical pressure, patient temperature, and/or anesthesia medications or
potentials. This allows monitoring to continue without interrup- dosing.
tion should one channel become compromised. Because sub- The clinical efficacy of SSEP monitoring was evaluated in
cortical channels are less susceptible to anesthesia, the reliability a multicenter outcome survey of 51,263 spinal cases performed
of SSEP and subsequent interpretation of SSEPs may be higher by 53 US surgeons as part of the Scoliosis Research Society. The
than other monitoring modalities (Sloan and Heyer, 2002). The study reported a sensitivity of 92% and specificity of 98.9% in
typical filter settings used to record SSEPs can be found in the the ability of SSEP monitoring to detect postoperative deficits
Appendix 1. and a 60% ability to reduce major neurologic deficits (Nuwer
et al., 1995). The same study also demonstrated that using SSEP
monitoring as a sole modality is associated with a false-negative
rate of 0.127%. Many studies have since calculated a sensitivity
EFFICACY OF SSEP MONITORING range of 80% to 100% in detecting postoperative deficits when
The use of SSEP monitoring has virtually eliminated the using SSEP monitoring alone. In one operative series of spinal
necessity to wake patients up during surgery and has greatly tumor resections, the sensitivity of SSEP was found to be 88%
improved the safety and ability to provide near continuous for detecting any neurological deficits and 100% for detecting
severe deficits (Wiedemayer et al., 2004).
The literature regarding SSEP monitoring demonstrates
a disparity regarding false-negative and false-positive results,
which is an inherent limitation of many of these studies. An
example of a false positive might be a case in which there are
significant changes in the monitoring, with notification of
surgeon, yet no detectable neurological deficits postoperatively.
Although these cases are usually considered to be false positives
according to the majority of the literature, this may underestimate
the test’s sensitivity because the detection and notification in
some cases may have altered the course of the surgery, thereby
preventing a deficit if the surgeon had not been notified. Perhaps,
more concerning is the specter of false-negative monitoring,
where the patient wakes up with motor deficits with no detectable
changes in SSEP monitoring. Given that SSEP are not represen-
tative of corticospinal tract function, the relative inability of
FIG. 1. Multiple channels are recommended for recording cortical SSEP monitoring to detect isolated motor deficits may be
and subcortical potentials in the upper and lower extremities. considered a limitation of SSEP monitoring, rather than a failure
Subcortical channels can be recorded from brainstem and thalamic of this test. Motor changes with unchanged SSEP signals should
sources and are less affected by anesthetics. therefore not be considered to be a false-negative result.

334 Journal of Clinical Neurophysiology Volume 33, Number 4, August 2016 clinicalneurophys.com
 Overview of IOM During Spine Surgery P. Shilian, et al.

(1939). Forty years later, in the early in the 1980’s, cortical

INTRODUCTION TO MULTIMODALITY stimulation through an intact skull was first developed by Merton
Because of their sensitivity for detecting motor deficits and and Morton (1980). This technique was later adapted to provide
the lack of a more specific modality, for two decades, SSEP intraoperative monitoring of the major components of the motor
monitoring was the only way to monitor global spinal cord pathway, the corticospinal tracts.
function. We now know that this SSEP monitoring reflects only During TcMEP monitoring, motor fibers are nonspecifically
the integrity of the dorsal column–medial lemniscus pathway. In stimulated transcranially, and responses are recorded from
recent decades, however, the advent of multimodality neuro- appropriate muscles as dictated by the surgical procedure.
monitoring techniques has provided the ability to monitor both Transcranial stimulation may recruit motor fibers from the motor
motor and sensory pathways, in addition to nerve root function. cortex or deeper structures, such as corona radiata and the
In addition to SSEP monitoring, multimodality testing often internal capsule. Most centers use a multipulse train, ranging
includes some permutation of transcranial motor evoked potential from 3 to 7 pulses, which can help overcome anesthetic
(TcMEP), continuous free running electromyography, evoked or inhibition of the anterior horn cell synapse (Husain, 2008).
triggered electromyography, dermatomal SSEP (DSSEP), and Stimulation is performed through an interpulse interval ranging
spinal D-wave monitoring. Selection of the appropriate combi- from 2 to 4 milliseconds (Legatt, 2002), typically using
nation of monitoring modalities depends on the structure(s) at corkscrew or needle electrodes. Electrodes are placed over C1
risk, as determined by an experienced neurophysiologist in and C2 (International 10-10 system) or more laterally at C3 and
conjunction with the neurosurgeon. The widespread availability C4 to optimize desired potentials. Recording is performed by
and experience gained with multimodality neurophysiologic placing intramuscular needles 2 to 3 cm apart in the desired target
monitoring have greatly improved the efficacy of this strategy muscles. At least two muscles distal to the surgical level(s) are
in the operating room. monitored, and one muscle proximal to the surgical level(s) is
typically used as a control. The monitoring of distal rather than
proximal muscles is typically favored, on account of their richer
corticospinal representation. Increasing the number of muscles
DERMATOMAL SOMATOSENSORY monitored improves the specificity and the predictive value of
EVOKED POTENTIALS motor evoked potentials (Sala et al., 2006).
Dermatomal somatosensory evoked potentials provide the Alarm criteria for TcMEP monitoring are not as well
ability to monitor nerve root function, by way of dermatomal established as with SSEP monitoring. When a robust baseline
stimulation in the cervical and lumbosacral regions. Needle or signal is present, a complete loss or abrupt decrease in amplitude
surface electrodes can be used for surface stimulation of of $ 80% has been described as a preferred method used to
dermatomes of interest, and recording is typically done at the report critical changes. The “all or none” method for TcMEP
scalp. In general, C39 and C49 (2 cm posterior to C3 and C4 of monitoring may lack sensitivity and may delay the time to
International 10-20 system) electrodes can be used for the upper notifying the surgeon of a potential injury. Other criteria, like
extremities and Cz9 and Fpz9 for the lower extremities. changes in the morphology of responses, such as a complex
There are limited studies regarding the effectiveness of multiphasic response evolving to a biphasic or a monophasic
DSSEP in monitoring nerve root function. If DSSEPs are response, can also be cause for alarm (Quinones-Hinojosa et al).
obtained, they may provide valuable information to the surgeon It is important to consider that a gradual loss of signal may
regarding nerve root function. However, many authors allude to indicate a potentially reversible anesthetic effect (Husain, 2008).
the variability of its usefulness depending on the level being Overall, routine use of TcMEP neurophysiological assessment
monitored (Owen et al., 1993; Owen et al., 1991). has improved spinal cord monitoring by effectively evaluating
Dermatomal somatosensory evoked potentials are very the integrity and functionality of the corticospinal tract during
small potentials with amplitudes averaging less than half of surgery and improving predictability of postoperative motor
median or tibial SSEPs. Furthermore, dermatomes are inherently deficits. A study by Sala et al. (2006) showed significant
variable regarding anatomical distribution, potentially compro- improvement in motor outcomes in patients undergoing TcMEP
mising the specificity of this technique. Dermatomal patterns monitoring during intramedullary spinal cord tumor removal, as
across individuals are inconsistent, and adjacent dermatomes compared with nonmonitored historical controls.
often overlap, making the recording an interpretation of DSSEP Some of the limitations inherent to using TcMEP monitoring
signals frequently challenging (Tsai et al., 1997). In addition, during spinal surgery include susceptibility to anesthetic medi-
high sensitivity of DSSEPs to anesthesia further limits their use cations (particularly inhalational and neuromuscular blockade
in the operating room (Herdmann et al., 1996; Sloan and Heyer, agents) and intrinsic variability. The use of total intravenous
2002). On account of this variability and poor reproducibility, anesthesia is sometimes preferred when responses are difficult to
DSSEPs have limited use in the operating room. obtain or demonstrate high variability. Chen et al. (2007)
reported a success rate of 94.8% for the upper extremity and
66.6% for the lower extremity in the ability to obtain reliable
MEPs. These rates drop to 81% and 39.1%, respectively, in the
TRANSCRANIAL MOTOR EVOKED POTENTIALS presence of preexisting weakness (Chen et al., 2007). Further-
Selective activation of the human cortex by direct electrical more, TcMEPs are more sensitive than SSEPs in detecting spinal
stimulation was first demonstrated by Penfield and Boldrey cord ischemia and can be helpful in preventing paraplegia if

clinicalneurophys.com Journal of Clinical Neurophysiology Volume 33, Number 4, August 2016 335
 P. Shilian, et al. Overview of IOM During Spine Surgery

appropriate measures are undertaken by the surgical team during of greater than 80% from baseline was 100% sensitive for
thoracic and thoracoabdominal aneurysm surgery (Dong et al., detecting nerve root injury. Conversely, no patients had a new
2002). motor deficit if the final MEP was at least 67% of baseline. In
Although TcMEPs are conventionally used to monitor addition, the severity and duration of injury was reduced in MEP
spinal cord function, there is evidence to suggest that they may monitored patients compared with patients from previous studies
also be used to assess nerve root function (Fig. 2). A study by (Lieberman et al., 2008). Review of the literature by MacDonald
Fan et al showed that addition of TcMEP monitoring of the et al suggests that the use of MEP monitoring for assessing nerve
deltoids and biceps, in addition to spontaneous EMG of the same root function may be limited due to radicular overlap (each nerve
muscles, provided sufficient warning for the surgeon during root supplies many different muscles and each muscle is
posterior cervical spine surgery. In patients with MEP changes in innervated by many different nerve roots), limited sampling
addition to EMG firing, a foraminotomy was performed at C4-C5 (only a small portion of motor axons are sampled with MEPs),
level to decompress the C5 nerve root. Patients undergoing this and variability (intrinsic quality of MEP, variability in amplitude,
monitoring technique generally recovered within 1 to 7 days, threshold, and morphology) (Macdonald et al., 2012).
compared with up to 2 years in patients operated on with the use
of conventional multimodality techniques (Fan et al., 2002). A
follow-up study done by Bose et al. (2007) evaluated similar
techniques during anterior cervical spine surgeries. They also SPINAL D-WAVE
demonstrated that the addition of TcMEP to spontaneous EMG Another method used to assess corticospinal tract function is
offered complementary information and improved prognostica- recording of spinal D-waves. Stimulation is given at the cortical
tion of postoperative deficits (Bose et al., 2007). Lieberman et al level, and responses are recorded from the spinal cord through an
studied MEP monitoring in 35 patients undergoing complex epidural electrode. Some of the benefits of this method include
lumbar surgery. They showed that a drop in the MEP amplitude less sensitivity to anesthesia and an ability to monitor after

FIG. 2. This is a 58-year-old man who underwent an anterior cervical discectomy and fusion for a herniated nucleus pulposus.
Intraoperatively, changes developed in the right intrinsic hand muscle motor evoked potential. The patient developed postoperative right
hand weakness suggesting a C8 or T1 root compromise.

336 Journal of Clinical Neurophysiology Volume 33, Number 4, August 2016 clinicalneurophys.com
 Overview of IOM During Spine Surgery P. Shilian, et al.

neuromuscular blockade. D-waves can be used in conjunction with

TcMEPs to provide improved prediction of postoperative out-
comes (Fig. 3). Kothbauer et al. (1998) studied 100 cases of
intramedullary spinal cord tumor resections. The authors found
that although unaffected MEPs prognosticate no motor deficit,
significant changes in the TcMEP, along with a change in the
D-wave, suggest a permanent deficit. A change in TcMEP recording,
however, with no changes in D-wave, was more likely to predict
a transient neurological deficit (Kothbauer et al., 1998).

SPONTANEOUS ELECTROMYOGRAPHY
To record spontaneous EMG activity, active and reference
electrodes are placed in the muscles of interest, depending on the
nerve roots at risk. Nerve roots of cervical, lumbar, and sacral
segments can all be monitored. Spontaneous free-run EMG shows
no activity at baseline in the presence of healthy nerve root function FIG. 4. Baseline spontaneous electromyogram is quiet with no
(Fig. 4). However, discharges can be seen with nerve stretch, blunt discharges seen.
trauma, compression, or ischemia (Nichols and Manafov, 2012).
High-frequency and/or high-amplitude trains are clinically signif- Accurate placement of pedicle screws is often challenging, mainly
icant and suggest irritation to the nerve roots (Fig. 5). Myokymic on account of anatomical and pathological variations, and the
discharges may suggest more severe damage to the nerve root. proximity of the spinal cord and nerve roots. Pedicle screw breaches
Spontaneous EMG allows monitoring of multiple nerve roots may cause compression of the spinal cord or nerve roots and have
at the same time, with immediate and continuous feedback as there been reported in as high as 10% of spinal fixation sites (Parker et al.,
is no need for signal averaging. Furthermore, this modality is not 2011). Radiologic imaging has a sensitivity of only 63% in detecting
as affected by parameters such as blood pressure and temperature a breach of the medial or inferior aspect of the pedicle wall (Maguire
as are other neurophysiological monitoring methods. Muscle et al., 1995). Triggered EMG may be a useful adjunct in detecting
relaxants, however, should be avoided whenever possible because compression of the neural elements during spinal surgery and
they can significantly attenuate EMG activity. One of the pitfalls of involves electrical stimulation of the screw while recording time-
this modality occurs with sharp nerve transection, as the EMG may locked EMG activity from the corresponding myotome. Because
not show any abnormal activity (Nelson and Vasconez, 1995). Calancie et al. (1994) demonstrated the sensitivity and reliability of
Gunnarsson et al. (2004) demonstrated that spontaneous electro- this technique, it has been commonly used to detect pedicle wall
myography has a high sensitivity (100%), yet a low specificity breaches. Trigger EMG increases the sensitivity of identifying
(23.7%) to detect postoperative neurologic deficits and is therefore
more reliable as a screening tool for nerve root function.

TRIGGERED ELECTROMYOGRAPHY
Pedicle screws are commonly used for mechanical stabilization
of the thoracic, lumbosacral, and more recently cervical spinal levels.

FIG. 3. Obtaining D-wave in addition to TcMEP during spinal cord

surgery can help with prognosis. With loss of MEP, stability of D-
waves can be used by surgeons to decide on a more aggressive FIG. 5. Trains of firing are observed in the left tibialis anterior and
resection without causing a permanent deficit. left foot.

clinicalneurophys.com Journal of Clinical Neurophysiology Volume 33, Number 4, August 2016 337
 P. Shilian, et al. Overview of IOM During Spine Surgery

The true efficacy of IOM has not been directly studied in

humans because of inability to perform double-blinded random-
ized controlled trials. Evidence-based guidelines established by
Nuwer et al, however, concluded that intraoperative monitoring
is effective in predicting postoperative paraparesis, paraplegia,
and quadriplegia during spinal surgery (Nuwer et al., 2012). This
was based on four class I and seven class II studies (based on the
AAN four-tiered evidence classification scheme). These com-
bined data resulted in a level A recommendation to alert the
surgical team about increased risk of severe adverse neurologic
outcomes in patients with reported IOM changes.

FIG. 6. Anterior lumbar interbody fusion: increased latency and

decreased amplitude in the left lower SSEP, followed by complete
loss of left N20 response. The cortical SSEP response recovered CONCLUSIONS
after replacement of the retractor (descending timeline). The ability to monitor the functional integrity of the nervous
system in real-time fashion is crucial to the very essence of surgery of
a misplaced pedicle screw in up to 93% of lumbosacral instrumen- the spinal cord and central nervous system as a whole. The discipline
tation cases (Maguire et al., 1995). Calancie et al reported that EMG of intraoperative neurophysiological monitoring has evolved rapidly
thresholds $ 10 mA in the lumbosacral spine suggest absence of over the past several decades, from rudimentary wake-up tests to
a pedicle wall breach. Another study by Raynor et al. (2007) showed advanced and sophisticated methods for multimodality monitoring.
that using a threshold of 8 mA, triggered electromyography has Selective use of various intraoperative modalities, including SSEP,
a specificity of 94% and a sensitivity of 86% in identifying a pedicle TcMEP, spinal D-wave, and EMG monitoring now provide
screw breach. With thoracic pedicle screws, Shi et al show that a methodology for complimentary and highly sensitive screening
stimulation thresholds .11 mA have a 97.5% negative-predictive of specific anatomical subsets of the nervous system. Multimodality
value. Practitioners should remain mindful that the use of muscle neurophysiological monitoring has been shown to augment the
relaxants, preexisting radiculopathy, or current shunting may course of many spinal operations and is known to prevent
increase the threshold of stimulation of lumbosacral nerve roots, and minimize the degree of major postoperative neurological deficits.
potentially resulting in false-negative results.

REFERENCES
ANTERIOR LUMBAR INTERBODY FUSION
Special consideration needs to be given to anterior lumbar Bose B, Sestokas AK, Schwartz DM. Neurophysiological detection of
iatrogenic C-5 nerve deficit during anterior cervical spinal surgery.
surgery due to the risk of vascular compromise. It is reported that up J Neurosurg Spine 2007;6:381–385.
to 57% of patients undergoing anterior lumbar interbody fusion at Brau SA, Spoonamore MJ, Snyder L, et al. Nerve monitoring changes related
the L4-L5 level are subject to left iliac artery compression (Brau to iliac artery compression during anterior lumbar spine surgery. Spine J
et al., 2003). Vascular compromise and oxygen desaturation 2003;3:351–355.
Calancie B, Madsen P, Lebwohl N. Stimulus-evoked EMG monitoring during
correlates with changes in the left lower extremity SSEP (Fig. 6). transpedicular lumbosacral spine instrumentation. Initial clinical results.
These changes are usually transient and typically resolve with Spine (Phila Pa 1976) 1994;19:2780–2786.
adjustment or removal of the retractor. If SSEP recovery is not seen, Chen X, Sterio D, Ming X, et al. Success rate of motor evoked potentials for
intraoperative neurophysiologic monitoring: effects of age, lesion
thrombosis to the corresponding iliac artery should be considered. location, and preoperative neurologic deficits. J Clin Neurophysiol
2007;24:281–285.
D’Angelo CM, Van Gilder JC, Taub A. Evoked cortical potentials in
experimental spinal cord trauma. J Neurosurg 1973;38:332–336.
MULTIMODALITY PREDICTIVE VALUE Dawson GD. Cerebral responses to electrical stimulation of peripheral nerve
A prospective study by Kelleher et al. (2008) of 1055 patients in man. J Neurol Neurosurg Psychiatry 1947;10:134–140.
Dong CCJ, Macdonald DB, Janusz MT. Intraoperative spinal cord monitoring
undergoing cervical spine surgery using multimodality monitoring during descending thoracic and thoracoabdominal aneurysm surgery.
showed a 52% sensitivity and 100% specificity for SSEP Ann Thorac Surg 2002;74:S1873–S1876; discussion S1892–S1898.
monitoring, 100% sensitivity and 96% specificity for MEP Eager M, Shimer A, Jahangiri FR, et al. Intraoperative neurophysiological
monitoring (IONM): lessons learned from 32 case events in 2069 spine
monitoring, and 46% sensitivity and 73% specificity for EMG in cases. Am J Electroneurodiagnostic Technol 2011;51:247–263.
predicting postoperative motor deficits. Multimodality monitoring Fan D, Schwartz DM, Vaccaro AR, et al. Intraoperative neurophysiologic
provides independent and complementary information in monitor- detection of iatrogenic C5 nerve root injury during laminectomy for cervical
compression myelopathy. Spine (Phila Pa 1976) 2002;27:2499–2502.
ing spinal cord function. Sutter et al studied multimodality testing Gunnarsson T, Krassioukov AV, Sarjeant R, Fehlings MG. Real-Time
in 1017 patients undergoing spine surgery. They showed that Continuous Intraoperative Electromyographic and Somatosensory
multimodality intraoperative monitoring had a sensitivity of 89% Evoked Potential Recordings in Spinal Surgery: Correlation of Clinical
and a specificity of 99%, with a false-negative rate of 0.8%. Using and Electrophysiologic Findings in a Prospective, Consecutive Series of
213 Cases. Spine (Phila Pa 1976) 2004;29:677.
multimodality monitoring can help prevent and/or minimize the Hall JE, Levine CR, Sudhir KG. Intraoperative awakening to monitor spinal
degree of neurologic deficits in patients undergoing spinal surgery. cord function during Harrington instrumentation and spine fusion.

338 Journal of Clinical Neurophysiology Volume 33, Number 4, August 2016 clinicalneurophys.com
 Overview of IOM During Spine Surgery P. Shilian, et al.

Description of procedure and report of three cases. J Bone Joint Surg Am Nuwer MR, Emerson RG, Galloway G, et al. Evidence-based guideline update:
1978;60:533–536. intraoperative spinal monitoring with somatosensory and transcranial
Herdmann J, Deletis V, Edmonds HL, Morota N. Spinal cord and nerve root electrical motor evoked potentials*. J Clin Neurophysiol 2012;29:101–108.
monitoring in spine surgery and related procedures. Spine (Phila Pa Owen JH, Bridwell KH, Lenke LG. Innervation pattern of dorsal roots and
1976) 1996;21:879–885. their effects on the specificity of dermatomal somatosensory evoked
Husain AM. A practical approach to neurophysiologic intraoperative potentials. Spine (Phila Pa 1976) 1993;18:748–754.
monitoring. New York, Demos, 2008. Owen JH, Bridwell KH, Grubb R, et al. The clinical application of neurogenic
Jou IM. Effects of core body temperature on changes in spinal somatosensory- motor evoked potentials to monitor spinal cord function during surgery.
evoked potential in acute spinal cord compression injury: an experimental Spine (Phila Pa 1976) 1991;16(6 suppl 1):S201–S205.
study in the rat. Spine (Phila Pa 1976) 2000;25:1878–1885. Parker SL, Amin AG, Farber SH, et al. Ability of electromyographic
Kelleher MO, Tan G, Sarjeant R, Fehlings MG. Predictive value of intraoperative monitoring to determine the presence of malpositioned pedicle screws
neurophysiological monitoring during cervical spine surgery: a prospective in the lumbosacral spine: analysis of 2450 consecutively placed screws. J
analysis of 1055 consecutive patients. J Neurosurg Spine 2008;8:215–221. Neurosurg Spine 2011;15:130–135.
Kothbauer KF, Deletis V, Epstein FJ. Intraoperative monitoring. Pediatr Penfield W, Boldrey E. Somatic Motor and Sensory Representation in the
Neurosurg 1998;29:54–55. Cerebral Cortex of Man as studied by Electrical Stimulation. Brain
Legatt AD. Current practice of motor evoked potential monitoring: results of 1937;60:389–443.
a survey. J Clin Neurophysiol 2002;19:454–460. Porkkala T, Kaukinen S, Häkkinen V, Jäntti V. Effects of hypothermia and
Lieberman JA, Lyon R, Feiner J, et al. The efficacy of motor evoked sternal retractors on median nerve somatosensory evoked potentials.
potentials in fixed sagittal imbalance deformity correction surgery. Spine Acta Anaesthesiol Scand 1997;41:843–848.
(Phila Pa 1976) 2008;33:E414–E424. Quinones-Hinojosa A, Lyon R, Zada G. Changes in transcranial motor evoked
Macdonald DB, Stigsby B, Homoud Al I, et al. Utility of motor evoked potentials during intramedullary spinal cord tumor resection correlate with
potentials for intraoperative nerve root monitoring. J Clin Neurophysiol postoperative motor function. Neurosurgery 2005;56:982–993.
2012;29:118–125. Raynor BL, Lenke LG, Bridwell KH, et al. Correlation between low triggered
Maguire J, Wallace S, Madiga R, et al. Evaluation of intrapedicular screw electromyographic thresholds and lumbar pedicle screw malposition:
position using intraoperative evoked electromyography. Spine (Phila Pa analysis of 4857 screws. Spine (Phila Pa 1976) 2007;32:2673–2678.
1976) 1995;20:1068–1074. Sala F, Palandri G, Basso E, et al. Motor evoked potential monitoring improves
Merton PA, Morton HB. Stimulation of the cerebral cortex in the intact outcome after surgery for intramedullary spinal cord tumors: a historical
human subject. Nature 1980;285:227. control study. Neurosurgery 2006;58:1129–1143; discussion 1129–1143.
Nelson KR, Vasconez HC. Nerve transection without neurotonic discharges Sloan TB, Heyer EJ. Anesthesia for intraoperative neurophysiologic moni-
during intraoperative electromyographic monitoring. Muscle Nerve toring of the spinal cord. J Clin Neurophysiol 2002;19:430–443.
1995;18:236–238. Sutter M, Eggspuehler A, Grob D, et al. The diagnostic value of multimodal
Nichols GS, Manafov E. Utility of electromyography for nerve root monitoring intraoperative monitoring (MIOM) during spine surgery: a prospective
during spinal surgery. J Clin Neurophysiol 2012;29:140–148. study of 1,017 patients. Eur Spine J 2007;16(Suppl 2):S162–S170.
Nuwer MR. Intraoperative monitoring of neural function. Amsterdam: Tsai RY, Yang RS, Nuwer MR, et al. Intraoperative dermatomal evoked
Elsevier, 2008. potential monitoring fails to predict outcome from lumbar decompres-
Nuwer MR, Dawson E. Intraoperative evoked potential monitoring of the sion surgery. Spine (Phila Pa 1976) 1997;22:1970–1975.
spinal cord: enhanced stability of cortical recordings. Electroencephalogr Vauzelle C, Stagnara P, Jouvinroux P. Functional monitoring of spinal cord
Clin Neurophysiol 1984;59:318–327. activity during spinal surgery. Clin Orthop Relat Res 1973;173–178.
Nuwer MR, Dawson EG, Carlson LG, et al. Somatosensory evoked potential Wiedemayer H, Sandalcioglu IE, Armbruster W, et al. False negative findings
spinal cord monitoring reduces neurologic deficits after scoliosis surgery: in intraoperative SEP monitoring: analysis of 658 consecutive neurosur-
results of a large multicenter survey. Electroencephalogr Clin Neuro- gical cases and review of published reports. J Neurol Neurosurg
physiol 2003;96:6–11. Psychiatr 2004;75:280–286.

APPENDIX 1

TABLE 1. Typical intraoperative Neuromonitoring Settings at Our Institution

Stimulation Parameters Recording Parameters
Stimulation Stimulation Pulse Number of Low-Frequency High-Frequency
Modality Rate Type Width Intensity Averages Filter Filter
SSEP 4–5 (avoiding multiples of 60) Constant current 100–200 ms 20–60 mA* 250–500 30 Hz 500–1500 Hz
TcMEP 250–500 Hz (5–7 pulses) Constant voltage 50–75 ms 100–400 V None 30 Hz 3000 Hz
EMG 2.79 Hz Constant current 100 ms 2–40 mA None 10 Hz 3000 Hz
*Adjusted to produce adequate muscle twitch.

TABLE 2. Typical Channel Setting Used at Our Institution

Modality Recording Channels
Upper SSEP CPc-CPi, Cz-FPz, A1/C5s-FPz
Lower SSEP CpPi-CPc, CPi-FPz, Cz-FPz, A1-FPz
TcMEP Various muscles in all extremities
spEMG Various muscles in all extremities
tEMG Various muscles in all extremities
c, contralateral; i, ipsilateral.

clinicalneurophys.com Journal of Clinical Neurophysiology Volume 33, Number 4, August 2016 339