Intraoperative Neurophysiological Monitoring (IONM)

Intraoperative Neurophysiological Monitoring (IONM): Lessons Learned from 32 Case Events in 2069 Spine Cases Matthew Eager, M.D.1; Adam Shimer, M.D.1; 2 Faisal R. Jahangiri, M.D., CNIM, DABNM ; 1 1,3 Francis Shen, M.D. ; Vincent Arlet, M.D. 1

University of Virginia Medical Center Department of Orthopedic Surgery Charlottesville, Virginia 2 Impulse Monitoring Inc. Columbia, Maryland 3 University of Virginia Medical Center Department of Neurosurgery Charlottesville, Virginia

Am J Electroneurodiagnostic Technol 51:247–263, 2011 © ASET, Missouri

Intraoperative Neurophysiological Monitoring (IONM): Lessons Learned from 32 Case Events in 2069 Spine Cases Matthew Eager, M.D.1; Adam Shimer, M.D.1; 2 Faisal R. Jahangiri, M.D., CNIM, DABNM ; 1 1,3 Francis Shen, M.D. ; Vincent Arlet, M.D. 1

University of Virginia Medical Center Department of Orthopedic Surgery Charlottesville, Virginia 2 Impulse Monitoring Inc. Columbia, Maryland 3 University of Virginia Medical Center Department of Neurosurgery Charlottesville, Virginia

ABSTRACT. Intraoperative neurophysiological monitoring (IONM) is becoming the standard of care for many spinal surgeries, especially those with deformity correction and instrumentation. We reviewed 2069 spine cases with multimodality IONM including somatosensory evoked potentials (SSEP), transcranial electrical motor evoked potentials (TCeMEP), and spontaneous and triggered electromyography (s-EMG and t-EMG) in a University setting over a period of four years to examine perioperative clinical findings when an IONM event was noted and to ascertain how IONM has affected our ability to avoid potential neurological injury during spine surgery. We performed a retrospective analysis of cases from 2006 to 2010 to study the frequency and cause of intraoperative events detected via IONM and the clinical outcome of the patient. There were 32 cases (1.5%) with possible intraoperative events. There were 17 (53%) cases where IONM changes affected the course of the surgery and prevented possible postoperative neurological deficits. Seven cases (41%) were due to deformity correction, five (29%) due to hypotension, four (24%) due Received: June 29, 2011. Accepted for publication: August 24, 2011.

247

248

IONM: LESSONS LEARNED FROM 32/2069 SPINE CASES

to patient positioning, and one (6%) due to a screw requiring repositioning. None of the 17 patients had postoperative motor or sensory deficits. There were four cases with false-positive IONM findings due to correctible technical issues. Three cases required surgical revision due to pedicle screw malposition. In each case, s-EMGs failed to exhibit intraoperative changes but the patient presented with postoperative radiculopathy. We believe that the use of t-EMGs may have prevented these complications. This review reinforces the importance of multimodality IONM for spinal surgery. The incidence of possible events in our series was 1.5%, and several likely postoperative neurologic deficits were avoided by intraoperative intervention. KEY WORDS. Deformity correction, electromyography (EMG), intraoperative neurophysiological monitoring (IONM), multimodality IONM, pedicle screw stimulation, postoperative deficit, scoliosis, somatosensory evoked potential (SSEP), spinal surgery, transcranial electrical motor evoked potential (TCeMEP).

INTRODUCTION Intraoperative neurophysiological monitoring (IONM) is becoming the standard of care for many spinal surgeries, especially those with deformity correction and instrumentation. During spinal surgery, various structures are placed at risk for potential injury, including the spinal cord, nerve roots, lumbar plexus, brachial plexus, and the corresponding vascular elements. The goal of IONM is to detect and reverse potential neurological injury, minimizing postoperative functional deficits. This is particularly important during deformity correction where the incidence of postoperative paralysis has been reported between 0.25% and 3.2% (Bridwell et al. 1998). Several IONM modalities are currently available for monitoring the different aspects of the central and peripheral nervous system, each offering a unique set of benefits, limitations, and sensitivity/specificity. The most frequently used modalities for spinal procedures are somatosensory evoked potentials (SSEPs), transcranial electrical motor evoked potentials (TCeMEPs), and spontaneous and triggered electromyography (s-EMG and t-EMG). To optimize the value of IONM, an interdisciplinary effort among the surgical, neuromonitoring, and anesthesia teams is imperative (Pajewski et al. 2007). Monitoring techniques for the spinal cord were pioneered in the 1970s when SSEPs were described for use during scoliosis surgery (Nash et al. 1977). Since that time, IONM has evolved and new modalities have been developed, but SSEPs remain


249

the principal modality for spinal cord monitoring (Keith et al. 1990, Khan et al. 2006, Nuwer et al. 1995). They continue to be a simple and reliable method to assess afferent conduction from the site of peripheral nerve stimulation through the dorsal columns, brainstem, and thalamus to the primary somatosensory cortex. At present, SSEPs are used extensively during a wide variety of surgical procedures. TCeMEPs have emerged as an extremely valuable and efficacious tool in IONM. They provide a direct assessment of motor function and are most reliable in detecting ischemic changes to the motor tracts during deformity correction (Deletis and Sala 2008) because SSEPs are sensitive and reliable for detecting ischemic changes only in the sensory tracts. The initial single-pulse stimulation technique was found to be highly sensitive to anesthetic effects (Merton and Morton 1980), but the technique has since been revised to utilize multi-pulse stimulation with various modifications of anesthetic regimens to maximize the ability to monitor TCeMEPs. By the early 1990s, multi-pulse transcranial electrical stimulation was developed (Burke et al. 1992) and is now widely used in a variety of spinal surgeries, including correction of spinal deformity, cervical and thoracolumbar degenerative cases, spinal trauma, and tumor resection. Over the past 20 years, segmental instrumentation and fusion using pedicle screw constructs have become the standard for spinal stabilization. s-EMG can be used intraoperatively to monitor the nearby nerve roots. This spontaneous EMG activity is recorded with electrodes placed in the muscles of interest and based on the structures at risk. Surgical manipulation such as pulling, stretching, or compressing produces neural discharges resulting in activity in the corresponding muscles. Studies using s-EMG monitoring of the deltoid muscle reported a reduction in the incidence of C5 nerve root palsy from 7.3 to 0.9% (Jimenez et al. 2005). A preventable risk of during pedicle screw placement for segmental instrumentation is a medial breach of the pedicle wall into the spinal canal. t-EMG can be performed to determine whether a screw has breached the medial or inferior pedicle wall, thus posing a risk to the nerve root exiting at that level. When a pedicle screw is accurately placed, the surrounding bone acts as an insulator, and a higher amount of electrical current is therefore required to activate the nerve root. When a medial breach occurs, the stimulation threshold is significantly reduced. Multimodality IONM takes advantage of the strengths of its various components to provide a more comprehensive assessment of the spinal cord. Moreover, the overall function of the central and peripheral nervous systems can be monitored, from the cortex to the spinal cord, nerve roots, and the peripheral nerves. The combined use of SSEPs, TCeMEPs, and EMG provide the necessary tools to accurately and reliably monitor the functional integrity of the spinal cord during a broad spectrum of routine and complex spinal surgeries, at the same time maximizing the sensitivity and specificity of the monitoring (MacDonald et al. 2003, Jahangiri et al. 2010b). The

250


real-time information should provide an added layer of security during surgical procedures in which an intraoperative complication is an imminent possibility. The aim of this study was to review a variety of spine cases at our institution with single or multimodality IONM including SSEPs, TCeMEPs, s-EMG, and t-EMG over a period of four years to examine perioperative clinical findings when an IONM event was noted intraoperatively and to ascertain how IONM has helped us in avoiding any potential neurological injury during spine surgery.

METHODS Patient Population This study was a retrospective analysis of spine cases with IONM from 2006 to 2010 at University of Virginia Medical Center, Charlottesville, Virginia. The objective was to study the frequency and cause of intraoperative events detected via IONM and the clinical outcome of the patient. Following Institutional Review Board (IRB) approval, data from a single IONM provider were collected and any cases with possible intraoperative events were isolated. All IONM was performed directly by an experienced neurophysiologist under the real-time continuous supervision of a neurologist. The intraoperative and immediate postoperative clinical documentation of cases with events were reviewed. Cases without an intraoperative event(s) were not studied. Anesthesia Protocol The sensitivity of cortical SSEPs to inhalational agents and the effect of neuromuscular blocking agents on s-EMG and t-EMG make the anesthesia protocol crucially important during these cases (Scheufler and Zentner 2002, Sloan and Heyer 2002). So anesthesia did not interfere with any IONM modality, total intravenous anesthesia (TIVA) was used with propofol, remifentanil, and lidocaine infusion. Neuromuscular blockade was used initially only for intubation and typically wore off in time for TCeMEP baselines and EMG recordings. Subsequent neuromuscular blockade levels were monitored with train of four (TOF) by stimulating the left posterior tibial nerve and recording from the corresponding abductor hallucis and extensor hallucis brevis muscles. During surgery, a train of 4/4 twitches was maintained (Figure 1). Body temperature and arterial blood pressure were constantly monitored and correlated with neurophysiological findings (Seyal and Mull 2002). Somatosensory Evoked Potentials (SSEPs) The median and/or ulnar nerves at the wrist were typically used for stimulation of upper extremity SSEPs, and the posterior tibial nerve at the ankle was used for lower


251

FIG. 1. Train of Four (TOF) data showing all four twitches present. Left: Four present TOF responses. Right: TOF in histogram view.

extremity SSEPs. Surface pad electrodes were used for stimulation and subdermal needle electrodes were used for SSEP recordings, and a baseline recording was usually collected before incision. Somatosensory evoked potentials were generated by a low intensity electrical stimulation of peripheral nerves in hands (median nerve at 30 mA and/or ulnar at 20 mA) and feet (posterior tibial nerve at 65 mA). The pulse width was set to 0.3 milliseconds and a repetition rate of 3.79 stimulations per second was used for SSEP stimulation. In order to determine the anatomic and functional integrity of different locations along the sensory pathways, the recording electrodes were placed on the patients at multiple locations along the sensory pathway (brachial plexus/popliteal fossa, brainstem, and somatosensory cortex). The filters settings for recording were set to 30 Hertz (Hz) for low frequency filter and 750 Hz (cortical) and 1500 Hz (peripheral) for high frequency filter. The recording timebase for SSEP was set to 5 msec/division for upper extremities and 10 msec/division for lower extremities. The display gain was set to 1 µV/division. At regular intervals afterward and throughout the surgical procedure, a new SSEP recording was collected and compared with the baseline trace. Alarm criteria

252


indicating a possible intraoperative event and potential deficit consisted of a 50% reduction in amplitude and/or a 10% increase in latency. Transcranial Electrical Motor Evoked Potentials (TCeMEPs) Stimulation electrodes were placed at C3 and C4 for activation of both upper and lower extremity muscle groups. C1 and C2 were used as alternative sites if more focal activation of the lower extremity muscle groups was needed. Trains of five to seven square-wave stimuli (50µsec duration) were utilized with intensities ranging from 100 to 500V. Subdermal needle electrodes for EMG and TCeMEP recordings were placed in deltoid, biceps brachii, flexor carpi ulnaris, brachioradialis, abductor pollicis brevis, and abductor digiti minimi in upper extremities and iliopsoas, quadriceps femoris, tibialis anterior, medial gastrocnemius, abductor hallucis, and extensor hallucis brevis muscles in the lower extremities. The timebase for recording TCeMEP was set to 10 msec/division for upper and lower extremities. The filter settings were set to 10 Hz for low frequency filters and 5000 Hz for high frequency filters. Two methods were used for alarm criteria indicating a possible intraoperative event: the all-or-nothing criterion and the amplitude criterion. The all-or-nothing criterion is the most widely cited and utilized method, given the inherent variability of the recorded signals in TCeMEP monitoring (Kothbauer et al. 1998). Based on this approach, a complete loss of the TCeMEP signal from a preliminary baseline recording is indicative of a possible intraoperative event. Because of the all-or-nothing nature of this interpretation, it has been proposed that this method is not sensitive enough to detect subtle complications that may still result in correctable postoperative motor deficits (Calancie and Molano 2008). A modification of the all-or-nothing approach involves measuring the compound muscle action potential (CMAP) amplitude at baseline, then measuring relative changes in amplitude to determine if an intraoperative event has occurred. Electromyography (EMG) Free run or spontaneous EMG (s-EMG) activity was recorded from upper extremity muscles (deltoid, biceps brachii, flexor carpi ulnaris, brachioradialis, abductor pollicis brevis, and abductor digiti minimi) during cervical spine surgeries and from lower extremity muscles (iliopsoas, quadriceps femoris, tibialis anterior, medial gastrocnemius, abductor hallucis, and extensor hallucis brevis) for lower thoracic and lumbo-sacral spine surgeries. The timebase for recording s-EMG was set to 50 msec/division for upper and lower extremities. The filter settings were set to 10 Hz for low frequency filters and 5000 Hz for high frequency filters. Abnormal s-EMG responses were reported to the surgeon to avoid any damage to the nerve roots or spinal cord. Abnormal s-EMG activity includes spikes, bursts, trains, and


253

neurotonic discharges. Trains and neurotonic discharges were considered alarm criteria indicating a possible event. This type of activity consists of continuous, repetitive EMG firing caused by continuous force applied to the nerve root. Trains of higher frequency and/or higher amplitude tend to represent additional nerve fiber recruitment caused by excessive force. Spontaneous EMG spikes and bursts, on the other hand, can often inform the surgeon of proximity to the nerve root (Jahangiri et al. 2010a). The parameters for pedicle screw stimulation were set to a pulse width of 100 µseconds and a repetition rate of 2.66 Hz. A monopolar electrode was used for t-EMG to directly stimulate the top of the pedicle screws at increasing current intensities. The stimulation intensity was started from 0 mA and was increased to 30.0 mA or until a CMAP was recorded from the corresponding muscles. In the lumbar spine, studies have demonstrated that a response threshold less than 10 mA is highly indicative of a medial wall breach (Calancie et al. 1994). This is the criterion that we used to distinguish a possible intraoperative t-EMG event (Figure 2C).

FIG. 2. (A) Post operative x-ray. (B) Left L4 pedicle screw medial breach. Triggered EMG was not performed during the index procedure. Post operative foot drop required a second surgery to reposition the screw. (C) Triggered EMG (T-EMG) during follow up surgery showing lower threshold responses (4 mA) from left tibialis anterior (TA) muscle.

254


RESULTS The patients consisted of 971 males and 1098 females with ages ranging from 1 to 94 years (mean: 52, median: 55 years). See Table 1 for a summary of the diagnoses and Table 2 for a summary of the surgical procedures. In our series, there were 32 cases (1.5%) with possible intraoperative events (Table 3). There were 17 cases where IONM changes affected the course of the surgery and prevented possible postoperative neurological deficits. Of these 17, seven cases were related to deformity correction, five related to hypotension, and one screw required repositioning due to a low t-EMG threshold (Figures 2A and 2B). Four cases had changes related to patient positioning and external pressure (e.g., brachial plexus stretch) (Figure 3). None of the 17 patients had postoperative motor or sensory deficits.

Table 1.

Types of spinal disorders (n = 2069).

Diagnosis Spinal Stenosis Degenerative Disc Disease Pseudoarthrosis/Spinal Degeneration Radiculopathy/Myelopathy Lesion Kyphosis Fracture Scoliosis Spondylolisthesis Herniated Disc(s) Tethered Cord Other

No. 378 225 140 63 210 61 251 299 114 163 15 150

Table 2. Types of operative procedures (n = 2069). Procedure type

No.

ACDF Anterior/Posterior Cervical Fusion Anterior/Posterior Lumbar Fusion Spinal Decompression (any level, no fusion) Single-Level TLIF or PLIF Resection of Lesion (with or without fusion) Multi-Level Cervical or Cervicothoracic Fusion Multi-Level Thoracolumbar or Lumbar Fusion Thoracic Fusion Release of Tethered Cord Other

323 59 59 121 377 210 203 532 97 15 73

ACDF – anterior cervical discectomy and fusion, TLIF – transforaminal lumbar interbody fusion, PLIF – posterior lumbar interbody fusion.

T11-L5 anterior discectomy/fusion

Posterior thoracolumbar decompression/ fusion TLIF Posterior thoracolumbar decompression/ fusion costotransversectomy T11, T12, L1 partial vertebrectomies Posterior thoracolumbar decompression/ fusion TLIF Posterior occipitocervical decompression/fusion Anterior thoracic osteotomies

9

10

11

17

16

15

14

13

12

8

7

Posterior thoracolumbar decompression/ fusion TLIF Anterior cervical corpectomy and fusion

Cervicomedullary spinal cord tumor resection Anterior thoracic discectomies/partial corpectomies Posterior lumbosacral decompression/ fusion TLIF C7-T1 anterior decompression/fusion

6

5

Posterior lumbosacral decompression/ fusion TLIF Posterior lumbar decompression/fusion

4 None

Screw checked, repositioned

None

Right upper extremity decreased SSEP Right lower extremity loss of SSEP after graft placement Right lower extremity loss MEP, SSEP stable No baseline SSEPs

Variable SSEP, MEP

Left upper extremity decreased SSEP Right upper extremity decreased SSEP Right lower extremity decreased MEP Right lower extremity decreased SSEP, MEP Left lower extremity decreased SSEP Right lower extremity decreased SSEP, MEP

None

None

Positioning effect, arm tucked None

Labile blood pressure

Stopped procedure, stage 1 of 2

Increased blood pressure



C-arm pressing on arm, removed Arm repositioned

Improved SSEPs

No deficit

No deficit

No deficit

No deficit

No deficit

No deficit

No deficit

No deficit

No deficit

Foot drop, medial L4 screw breach Left upper extremity sensory deficit No deficit

No deficit Bilateral lower extremity paresis Right lower extremity paralysis No deficit

Intraoperative Intervention Postoperative Findings Increased blood pressure None

Loss of left upper extremity SSEP None

None

Loss of MEP in right lower extremity Low S1 screw threshold

Thoracic spinal cord tumor debulking

Intraoperative Findings

3

Procedure

Posterior cervical decompression Loss of MEP Cervicothoracic spinal cord lesion biopsy Loss of MEP in lower extremities

1 2

Case

Table 3. Summary of possible intraoperative events.

IONM: LESSONS LEARNED FROM 32/2069 SPINE CASES 255

Procedure

Posterior cervicothoracic deformity correction with fusion Posterior lumbosacral decompression/ fusion Posterior lumbosacral decompression/ fusion TLIF Posterior lumbar decompression/fusion

25

Anterior thoracic discectomy and fusion

Anterior cervical corpectomy and fusion

30

31

32

Intraoperative Findings

Bilateral lower extremity loss of MEP Improved MEP after decompression Improved MEP after decompression Improved MEP after decompression

None, spontaneous EMG only

None, spontaneous EMG only

Thoracotomy, left upper extremity (down arm) loss of SSEPs Right lower extremity loss of MEP Bilateral lower extremity loss of MEP with distraction Left lower extremity decreased SSEP, loss of MEP during correction Bilateral lower extremity loss of MEP, decreased SSEPs Low screw threshold

Bilateral lower extremity SSEPs decreased with rod placement Loss of bilateral lower SSEPs (no MEPs present at baseline) Decreased SSEPs post flip

Intraoperative Intervention Postoperative Findings

None

None

None

Decreased correction

None

None

No breach, screw replaced

Correction decreased

Repositioned, large pt, procedure shortened Needle repositioned, signals reacquired Variable signal changes, returned to baseline Correction held, increased blood pressure

Improved function

Improved function

Improved function

Foot drop, medial left L4 screw breach No deficit

Screw in canal

No deficit

No deficit

No deficit

No deficit

Transient sensory changes No deficit

Rods placed, baseline SSEPs No deficit returned None No change from preoperative function None No deficit

MEP – motor evoked potential, TLIF – transforaminal lumbar interbody fusion, SSEP – somatosensory evoked potential, EMG – electromyography, PSO – pedicle subtraction osteotomy

Posterior thoracolumbar deformity correction with fusion Anterior cervical corpectomy and fusion

29

28

27

26

Posterior thoracolumbar deformity correction with fusion

Posterior then anterior cervicothoracic fusion Anterior thoracolumbar discectomies/ fusion Posterior thoracolumbar decompression/ fusion TLIF Posterior thoracic fusion

Posterior thoracolumbar decompression/ fusion, PSO T7 spinal cord tumor resection

24

23

22

21

20

19

18

Case

Table 3. Continued.

256 IONM: LESSONS LEARNED FROM 32/2069 SPINE CASES


257

FIG. 3. Somatosensory evoked potentials (SSEP) data showing the loss of median nerve SSEP responses in right cortical, transcortical stack as well as in right cervical and Erb’s responses due to patient positioning of the right arm. Reponses returned to baseline after pressure was released on left brachial plexus.

Four cases consisting of intradural cord biopsies or tumor resections had various positive IONM findings. We anticipated intraoperative events during these cases, and this was confirmed with IONM changes. There were four cases with false-positive IONM findings. One was due to a technical issue requiring needle repositioning, the second was due to a low t-EMG threshold without a pedicle breach, the third was due to decreased TCeMEP responses with stable SSEPs, and the fourth was due to decreased SSEPs after placing the patient in the prone position. None of the four patients had any postoperative deficits.

DISCUSSION IONM is routinely performed during spinal surgery to identify possible intraoperative events in a timely manner and reverse any potential postoperative neurological deficits. IONM is beneficial in a wide variety of surgical procedures (Traul et al. 2007, Sansur et al. 2007). Our study shows that multimodality IONM of spinal

258


cord sensory and motor function is feasible and provides useful data and feedback to the surgeon intraoperatively. False Positives of the Study There was one case with false-positive t-EMG findings due to a low response threshold without a pedicle breach. Fluoroscopy was used to confirm the screw placement and the screw was left due to no visible breach. Two cases had false-positive findings for evoked potentials. One was due to decreased TCeMEP responses with stable SSEPs, and the other was due to decreased SSEPs after positioning the patient (Figure 3). There are a variety of non-surgical phenomena that may lead to these false-positive findings. Factors that can affect the SSEP amplitude include halogenated agents, nitrous oxide, positioning, hypothermia, and hypotension. SSEP changes should be considered relevant if they occur close in time to a specific surgical event. Common factors that may contribute to false-positive findings for TCeMEPs include anesthesia, body temperature, blood pressure, surgical positioning, and technical issues. An inherent limitation of TCeMEP monitoring is that they may be more technically challenging to obtain. Current success rates for obtaining TCeMEPs are approximately 94.8% in the upper extremities and 66.6% in lower extremities (compared with 98% and 93% for SSEPs) (Chen et al. 2007). If preoperative motor deficits exist at the time of surgery, the ability to obtain TCeMEPs plummets to approximately 39% in the lower extremities. False positive EMG happened when the stimulation of the pedicle screws lead to a muscle response for a current threshold below 8 mA. Different factors could explain such false positive: osteoporosis, previous surgery, very small pedicle, or a previous pedicle breech. In this study, each time the stimulation of a pedicle screw was leading to a response for less than 8 mA, the screw was removed and the screw path was probed to ensure its good localization inside the pedicle. In the case of breaching the pedicle screw was redirected. Unfortunately our record did not keep a log of these false positive events after triggered EMG. False Negatives of the Study There were three cases with lumbar instrumentation and s-EMG monitoring only where there was a medial screw breach without positive intraoperative findings. In each case, t-EMGs were not used and s-EMGs were only used which failed to exhibit intraoperative changes but the patient presented with postoperative radiculopathy (foot drop). We believe the use of pedicle screw stimulation via t-EMG may have prevented these complications by showing low response thresholds (Figure 2). This would have allowed for screw repositioning during the initial surgery, avoiding postoperative foot drop and revision surgery.


259

During screw stimulation, false-negative responses can occur as a result of various factors. These include current spread and probe placement and need to be taken into consideration to ensure the accuracy of testing. Special attention should to be paid to fluid, blood, or soft tissue around the head of the screw at the time of stimulation that could potentially shunt current away from the screw. Additionally, it is important that the stimulation probe be placed directly on the top of the core of the screw and not on the crown of the poly-axial screw. Other factors to consider as false negatives, as we have experienced after completion of this study, are the lack of responses in some muscle groups like the psoas muscle from which it is harder to obtain a recording, especially in obese patients. A chronic neuropathy or a chronic nerve root compression may cause increase in pedicle screw thresholds for electrical stimulation resulting in a high screw threshold with a breached pedicle giving a false negative result. Since we focused our study on the cases where an intraoperative event occurred, it is naturally not possible to elaborate further about false-negative findings. From our experience, however, we did not have any patients with a complete spinal cord injury without an intraoperative event (Figure 4 and Figure 5). We have had three cases where the IONM was normal, and the patient woke up with radiculopathy or a partial cord syndrome. This exact number of cases could not be determined with our retrospective data set, especially that in most cases the postoperatvie deficit improved during the course of the hospitalization. We had only three cases with false negative findings. In each of these cases only s-EMG was used and no t-EMG was utilized to test the integrity of the pedicle wall after placing pedicle screws. In these three cases s-EMG was unable to demonstrate and breached pedicle wall. Therefore, we highly recommend t-EMG for pedicle screw stimulation. The use of SSEPs and TCeMEPs in combination has shown to provide a comprehensive assessment of spinal cord function. In a recent study examining the value of multimodality IONM in cervical spine surgery, SSEPs were reported to have a sensitivity of 52% and a specificity of 100% in detecting postoperative neurological deficits, and TCeMEPs were reported to have a sensitivity of 100% and a specificity of 96% (Kelleher et al. 2008). These results imply that SSEP and TCeMEP monitoring must be performed together to successfully screen for and identify impending spinal cord injury while minimizing false-positive interpretations. A large prospective study by Sutter et al. (2007) assessed the diagnostic value of multimodality monitoring in patients undergoing spinal surgery and reported a sensitivity of 89% and a specificity of 99% in the detection of postoperative neurological. In another study by Quraishi et al. (2009), the authors evaluated 102 patients undergoing correction for adult spinal deformities with the use of multimodality IONM, including SSEP, TCeMEP, and EMG, and demonstrated an overall sensitivity of 100% and specificity of 84.3%. Clearly, evidence supports the use of multimodality

260


FIG. 4. Left: Example of neuromonitoring changes during a deformity correction case. After insertion of the convex rod: decreased motor evoked potentials in the left foot of 80% amplitude, quadriceps and the tibialis anterior have dropped 50%, right side is normal for foot, quadriceps, and tibialis anterior responses. Right: Instrumentation after correction.

monitoring during cervical, thoracic, and lumbar spinal procedures and should be used during surgery to improve patient outcome. In 2002, Pelosi et al. performed combined monitoring of SSEPs and TCeMEPs in 104 of 126 procedures (82%) in 97 patients. They found significant intraoperative changes in one or both IONM modalities in 16 patients. SSEPs recovered in 8 of 8 patients and TCeMEPs in 10 of 15 (67%). There were new postoperative deficits in 6 of 16 with abnormal testing. In a 2006 study of spinal cord monitoring in scoliosis surgery by Accadbled et al. (2006), they reported 5 true-positive findings (2.6%) in 191 cases. In the 2009 study by Quraishi et al. (2009), they described 5 true-positive findings (4.95%) in 102 cases. These numbers are both higher than our number of 17 (0.8%) where IONM changes affected the course of the surgery and prevented possible postoperative neurological deficits. There are several reasons for this. We monitored all our cases including spinal deformities (22.9%) and degenerative cases (17.6%) as well as simple anterior cervical discectomy and fusion (ACDF)


261

FIG. 5. (Upper Arrows) Transcranial electrical motor evoked potential (TCeMEP) response data stack shows loss of responses during the procedure from tibialis anterior (TA) muscle, abductor hallucis (AH), and extensor hallucis brevis (EHB) muscles bilaterally at 15:17. (Lower Arrows) Responses returned to baseline in AH and EHB muscles at 15:33 after the correction was reduced.

and posterior foraminotomies. Only simple lumbar spine decompression without instrumentation does not get monitoring.

CONCLUSION This review reinforces the importance of multimodality IONM for spinal surgery. It is extremely valuable in the prevention of neurological injury. The wide array of modalities available provides a highly sensitive and specific means of monitoring the nervous system at each spinal level. The incidence of possible events in our series was 1.5%, and several likely postoperative neurologic deficits were avoided by intraoperative intervention. Knowledge of the benefits and limitations of each monitoring modality helps maximize the diagnostic value of IONM during spinal procedures. The incidence of false-negative findings with this database was very low (3/2069 – 0.0014 %) surgeries and the incidence of false-positive events was very low as well (4/2069 – 0.0019 %).

262


ACKNOWLEDGEMENTS The authors would like to thank Andrea Holmberg, CNIM; Mazhar Minhas, M.D., CNIM; and Michael McCormick of Impulse Monitoring Inc. and Richard O’Brien, M.D., FRCP (C), MBA of American Neuromonitoring Associates for their valuable contributions and help in preparing this article. Questions or comment about this article may be directed to the corresponding author, Faisal Jahangiri, M.D., CNIM, DABNM at [email protected].

REFERENCES Accadbled F, Henry P, de Gauzy JS, Cahuzac JP. Spinal cord monitoring in scoliosis surgery using an epidural electrode. Results of a prospective, consecutive series of 191 cases. Spine 2006; 31:2614–23. Bridwell KH, Lenke LG, Baldus C, Blanke K. Major intraoperative neurologic deficits in pediatric and adult spinal deformity patients. Incidence and etiology at one institution. Spine (Phila Pa 1976) 1998; 23:324–31. Burke D, Hicks R, Stephen J, Woodforth I, Crawford M. Assessment of corticospinal and somatosensory conduction simultaneously during scoliosis surgery. Electroencephalogr Clin Neurophysiol 1992; 85:388–96. Calancie B, Madsen P, Lebwohl N. Stimulus-evoked EMG monitoring during transpedicular lumbosacral spine instrumentation. Initial clinical results. Spine (Phil Pa 1976) 1994; 19:2780– 86. Calancie B, Molano MR. Alarm criteria for motor-evoked potentials: what’s wrong with the “presence-or-absence” approach? Spine (Phil Pa 1976) 2008; 33:406–14. Chen X, Sterio D, Ming X, Para DD, Butusova M, Tong T, Beric A. Success rate of motor evoked potentials for intraoperative neurophysiologic monitoring: effects of age, lesion location, and preoperative neurologic deficits. J Clin Neurophysiol 2007; 24:281–85. Deletis V, Sala F. Intraoperative neurophysiological monitoring of the spinal cord during spinal cord and spine surgery: a review focus on the corticospinal tracts. Clin Neurophysiol 2008; 119:248–64. Jahangiri FR, Sherman JH, Holmberg A, Louis R, Elias J, Vega-Bermudez F. Protecting the Genitofemoral Nerve during Direct/Extreme lateral Interbody Fusion (DLIF/XLIF) Procedures. Am J Electroneurodiagnostic Technol 2010a; 50:321–35. Jahangiri FR, Crowley RW, Persyn JJ, Kassell N, Vega-Bermudez F. Predicting surgical outcome using somatosensory evoked potentials and transcranial electric motor evoked potentials in a cervical-medullary junction hemangioblastoma. Am J Electroneurodiagnostic Technol 2010b; 50:101–10. Jimenez JC, Sani S, Braverman B, Deutsch H, Ratliff JK. Palsies of the fifth cervical nerve root after cervical decompression: prevention using continuous intraoperative electromyography monitoring. J Neurosurg Spine 2005; 3:92–97. Keith RW, Stambough JL, Awender SH. Somatosensory cortical evoked potentials: a review of 100 cases of intraoperative spinal surgery monitoring. J Spinal Disord 1990; 3:220–26. Kelleher MO, Tan G, Sarjeant R, Fehlings MG. Predictive value of intraoperative neurophysiological monitoring during cervical spine surgery: a prospective analysis of 1055 consecutive patients. J Neurosurg Spine 2008; 8:215–21. Khan MH, Smith PN, Balzer JR, Crammond D, Welch WC, Gerszten P, Sclabassi RJ, Kang JD, Donaldson WF. Intraoperative somatosensory evoked potential monitoring during cervical spine corpectomy surgery: experience with 508 cases. Spine (Phil Pa 1976) 2006; 31:E105– 13.


263

Kothbauer KF, Deletis V, Epstein FJ. Motor-evoked potential monitoring for intramedullary spinal cord tumor surgery: correlation of clinical and neurophysiological data in a series of 100 consecutive procedures. Neurosurg Focus 1998; 4:e1. MacDonald DB, Al Zayed Z, Khoudeir I, Stigsby B. Monitoring scoliosis surgery with combined multiple pulse transcranial electric motor and cortical somatosensory-evoked potentials from the lower and upper extremities. Spine 2003; 28:194–203. Merton PA, Morton HB. Stimulation of the cerebral cortex in the intact human subject. Nature 1980; 285:227. Nash CL Jr, Lorig RA, Schatzinger LA, Brown RH. Spinal cord monitoring during operative treatment of the spine. Clin Orthop Relat Res 1977; 100–05. Nuwer MR, Dawson EG, Carlson LG, Kanim LE, Sherman JE. Somatosensory evoked potential spinal cord monitoring reduces neurologic deficits after scoliosis surgery: results of a large multicenter survey. Electroencephalogr Clin Neurophysiol 1995; 96:6–11. Pajewski TN, Arlet V, Phillips LH. Current approach on spinal cord monitoring: the point of view of the neurologist, the anesthesiologist and the spine surgeon. Eur Spine J 2007; 16 Suppl 2: S115–29. Pelosi L, Lamb J, Grevitt M, Mehdian SM, Webb JK, Blumhardt LD. Combined monitoring of motor and somatosensory evoked potentials in orthopaedic spinal surgery. Clin Neurophysiol 2002; 113:1082–91. Quraishi NA, Lewis SJ, Kelleher MO, Sarjeant R, Rampersaud YR, Fehlings MG. Intraoperative multimodality monitoring in adult spinal deformity: analysis of a prospective series of one hundred two cases with independent evaluation. Spine (Phila Pa 1976) 2009; 34:1504–12. Sansur CA, Pouratian N, Dumont AS, Schiff D, Shaffrey CI, Shaffrey ME. Part II: spinal-cord neoplasms – primary tumors of the bony spine and adjacent soft tissues. Lancet Oncol 2007; 8:137–47. Scheufler KM, Zentner J. Total intravenous anesthesia for intraoperative monitoring of the motor pathways: an integral view combining clinical and experimental data. J Neurosurg 2002; 96: 571–79. Seyal M, Mull B. Mechanisms of signal change during intraoperative somatosensory evoked potential monitoring of the spinal cord. J Clin Neurophysiol 2002; 19:409–15. Sloan TB, Heyer EJ. Anaesthesia for intraoperative neurophysiologic monitoring of the spinal cord. J Clin Neurophysiol 2002; 19:430–43. Sutter M, Eggspuehler A, Muller A, Dvorak J. Multimodal intraoperative monitoring: an overview and proposal of methodology based on 1,017 cases. Eur Spine J 2007; 16 Suppl 2: S153–61. Traul DE, Shaffrey ME, Schiff D. Part I: spinal-cord neoplasms – intradural neoplasms. Lancet Oncol 2007; 8:35–45.