SPINE Volume 39, Number 24, pp E1425-E1432 ©2014, Lippincott Williams & Wilkins

DEFORMITY

How to Make the Best Use of Intraoperative Motor Evoked Potential Monitoring? Experience in 1162 Consecutive Spinal Deformity Surgical Procedures Qianyu Zhuang, MD, Shujie Wang, PhD, Jianguo Zhang, MD, Hong Zhao, MD, Yipeng Wang, MD, Ye Tian, MD, Yu Zhao, MD, Shugang Li, MD, Xisheng Weng, MD, Guixing Qiu, MD, and Jianxiong Shen, MD

Study Design. A retrospective study of 1162 consecutive patients who underwent spinal deformity surgical procedures at our spine center from January 2010 to December 2013. Objective. To develop and evaluate a protocol of intraoperative motor evoked potential (MEP) monitoring with the warning criteria we had established on the basis of our clinical experiences and the review of previous literature. Summary of Background Data. Though MEPs monitoring have become widely used in spinal deformity surgery, different alarm criteria and response protocol used in different studies compromised their comparability; Furthermore, high false-positive rate of MEP reported by previous studies has become an increasingly prominent problem that will limit its clinical use and development. Methods. The intraoperative monitoring data of 1162 consecutive patients who underwent spinal deformity surgical procedures at our spine center were retrospectively analyzed. Age, sex, diagnosis, preoperative neurological status, intraspinal anomalies, baseline MEP, and MEP change were collected. The protocol with the warning criteria we had established was used. The false-positive rate, falsenegative rate, and positive predictive value were calculated. Results. Significant intraoperative changes were seen in the MEP data in 52 (4.4%) of all the cases. In 25 cases among which, significant MEP changes were synchronously and logically associated with high-risk surgical maneuver (pedicle screw insertion, osteotomy, correction, etc.). The false-positive rate of MEP monitoring was 0.26% (3/1140), whereas the sensitivity and specificity of MEP for detection of clinically significant intraoperative cord injury were From the Department of Orthopedics, Peking Union Medical College Hospital, Beijing, P. R. China. Acknowledgment date: April 22, 2014. Revision date: August 10, 2014. Acceptance date: August 14, 2014. The manuscript submitted does not contain information about medical device(s)/drug(s). National Natural Science Foundation of China (81272054) funds were received in support of this work. No relevant financial activities outside the submitted work. Address correspondence and reprint requests to Jianxiong Shen, MD, Peking Union Medical College Hospital, 1 Shuai Fu Yuan, Beijing 100730, P. R. China; E-mail: [email protected] DOI: 10.1097/BRS.0000000000000589 Spine

100% and 99.7%, respectively. The positive predictive value of a MEP alert in terms of a new postoperative neurological deficit was 83.3%. Conclusion. Our study indicates that the appropriate use of MEP monitoring based on our protocol is able to obtain satisfying sensitivity and specificity and thus provide important information for intraoperative decision making. Key words: motor evoked potential monitoring, spinal deformity, warning criteria, false-positive rate, signal loss, response protocol, surgical status, high-risk surgical maneuvers, intraoperative cord injury, neurological deficit. Level of Evidence: 4 Spine 2014;39:E1425–E1432

I

ntraoperative motor evoked potentials (MEPs) monitoring has become widely used in spinal deformity surgical procedures.1 However, its shortcomings were criticized by many experts in recent years. The first one is the high trial-bytrial variability of muscle MEP amplitude,2 so that the criteria of somatosensory evoked potential monitoring based on amplitude and latency cannot be applied in MEP.3 This makes it difficult to define criteria for a minor degree of deterioration of the motor tract as distinct from a complete loss of the response. The second one is that the sensitivity of the MEP potential to insult the spinal cord is quite high.3 As a result, the incidence of false-positive results will increase if judgment is based purely on this potential.4 The last one is the instability of MEP especially when used in patients with neuromuscular scoliosis or preoperative neurological deficit.5 Some experts have realized that a smart warning criterion of MEP monitoring may be a good way to overcome these shortcomings.6 Unfortunately, most of the current MEP warning criteria are just based on its signal amplitude without considering surgical status and the logical relationships between them.7–11 The study by Wilson-Holden et al,5 for example, has reported that MEP has a false-positive rate as high as 27.1% in pediatric patients with spinal cord pathology undergoing spinal deformity surgery and 1.4% even in idiopathic scoliosis group as control. Such false-positive alerts www.spinejournal.com

E1425

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140542.indd E1425

22/10/14 8:41 PM

DEFORMITY

Intraoperative MEP Monitoring • Zhuang et al

could be misleading and may compel surgeons to take unreasonable risks or unnecessarily change the operative plan. Our study, therefore, aimed to develop and evaluate a protocol of intraoperative MEP monitoring with the warning criteria we had established based on our clinical experiences and the review of previous literature. We retrospectively analyzed 1162 consecutive patients who underwent spine deformity surgical procedures at our spine center. Our focus was that whether the appropriate intraoperative use of MEP could reduce its false-positive rate without compromising its truenegative rate.

General anesthesia was induced with a bolus dose of propofol (2 mg/kg) and remifentanil (1 μg/kg) combined with muscle relaxant. No additional muscle relaxants were given because they can reduce or eliminate myogenic MEP responses.12,13 Some patients were ventilated with a gas mixture of O2 and N2O in a ratio of 1:2.

MATERIALS AND METHODS

Warning Criteria

The intraoperative monitoring data of 1162 consecutive patients who underwent spinal deformity surgical procedures at our spine center from January 2010 to December 2013 were analyzed. Surgical procedures were performed by 1 of the 6 attending orthopedic spine surgeons in a single institution. On the basis of the basic spinal surgical training, electroneurophysiologist could record the MEP status at each important point during surgery. All monitoring was done in real time in the operating room, whereby communication between the electrophysiologist, surgeon, and anesthetist could be optimized when dealing with a monitoring issue. The monitoring time varied in different types of surgical procedures. For example, we usually stopped neuromonitoring as soon as bone graft fusion was started in correction of adolescent idiopathic scoliosis (AIS). But for major deformity correction with osteotomy, the neuromonitoring will be extended until wound closure considering the potential delayed effect on spinal cord. If the MEP changes were reduced to almost the edge of our warning criteria, the monitoring time should be further extended until the end of the surgery.

Our MEP warning criteria were as follows:

Transcranial Electrical Stimulation Motor Evoked Potentials

True-Negative Data remained consistent with baseline values throughout the procedure and correlated with unchanged neurological status after surgery.

Transcranial electrical stimulation-MEPs were elicited using subcutaneous needle electrodes (Axon Systems Inc., Hauppauge, NY) by stimulating of constant voltage (100– 400 V) and 5 to 7 train pulses, with duration of 200 to 400 μs. Both stimulation electrodes were inserted subcutaneously over motor cortex regions C3–C4 (10-20 System). Recording electrodes are placed into the abductor hallucis muscles in both lower extremities and the first dorsal interosseous muscles in the upper extremities (control).

Somatosensory Evoked Potentials To elicit SEP, a pair of stimulating electrodes was applied over the posterior tibial nerve behind the medial malleoli for lower extremities recording with constant current stimulation within the range of 15 to 30 mA. The recording needles were placed at Cz and Fpz positions for active and reference. Single-pulse stimulation with a frequency between 5.1 and 5.7 Hz and duration of 200 to 300 μs were applied. Responses

E1426

www.spinejournal.com

were recorded (averaged 200 stimulations) through a 30 to 300 Hz band-pass filter and waveforms were displayed in a 100-ms window.

Anesthesia Management

1. A more than 80% amplitude loss. 2. Synchronously and logically associated with high-risk surgical maneuver (pedicle screw insertion, osteotomy, correction, etc.). 3. Systemic and anesthetic factors being ruled out. Only when the data meet all the 3 criteria mentioned in the earlier text, it would be defined as a formal alert. The exact time and the surgical procedure of the signal decrease will be recorded and reported, too.

Statistical Analysis Data were analyzed with standardized statistical software (SPSS version 17.0; SPSS Inc., Chicago, IL). Categorical variables were summarized as frequencies and continuous variables as means and standard deviations. For analyses, monitoring results were divided into 5 categories as follows, and sensitivity, specificity, positive predictive value (PPV), negative predictive value of TCE-MEP alerts were also calculated for the total study population.

True-Positive An MEP alert followed by observation of a new neurological motor deficit during a wake-up test or at the end of the procedure. False-Negative Data remained consistent with baseline values throughout the procedure, but the results of the patient’s postoperative neurological examination demonstrated neurological deficit. False-Positive An MEP alert that persisted despite corrective measures and the absence of any observable new deficit during a wake-up test and/or the absence of a new deficit at the conclusion of the procedure.

November 2014

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140542.indd E1426

22/10/14 8:41 PM

DEFORMITY

Intraoperative MEP Monitoring • Zhuang et al

Indeterminate A MEP alert that normalized after corrective measures in a patient who emerged without new motor deficit. It was difficult to determine for certain whether such an event represented a true impending cord injury that was averted or a false-positive alert that spontaneously disappeared.14

TABLE 2. Patient Characteristics of Diagnosis

Distribution

No. (%) of Patients (N = 1162) Diagnosis

RESULTS

Idiopathic scoliosis

391 (33.6)

The major characteristics of the study population are displayed in Table 1, and the diagnosis distribution is showed in Table 2. Among the 1162 patients, a total of 52 cases (4.4%) presented significant intraoperative MEP changes. However, only 25 alerts in 25 different patients (Table 3) were identified on the basis of our warning criteria, Among them, 15 cases (true-positive) demonstrated neurological deficit after surgery. Only 2 had permanent neurological deficit, and all the others recovered eventually fully, approximately 12 hours to several months after operations. Three cases (false-positive) presented an alert that persisted despite all corrective measures and later demonstrated no motor function deficit during wake-up test or at the end of the surgery. In the other 7 cases (indeterminate), MEP signal recovered after the surgical team took corrective measures, and presented no new neurological deficit in the wake-up test or after surgery, On the basis of our predetermined criteria, the false-positive rate of MEP monitoring was 3/1140 (0.26%). The calculated sensitivity of transcranial motor evoked potentials (TcMEP) monitoring for clinically recognizable intraoperative spinal cord injury in this series was 100%. The specificity was 99.7%. The PPV of a MEP alert in terms of a new postoperative neurological deficit was 83.3%. A 2 × 2 contingency table is presented in Table 4. According to our statistical analysis, high-risk surgical maneuvers that were associated with MEP alert including (Figure 1): vertebral column resection (VCR) osteotomy (n = 6), hypotension (n = 4), Pedicle subtraction osteotomy (PSO) (n = 3), screw placement (n = 3), derotation (n = 2), compression (n = 2), cage placement (n = 2), Smith-Petersen osteotomy (SPO) (n = 1), distraction (n = 1), and direct contusion (n = 1). In addition, it should be noted that as many as 20 patients (80%) in a total of 25 cases with MEP alert in our

Congenital scoliosis

449 (38.6)

TABLE 1. Patient Basic Characteristics (n = 1162) Age, yr

Mean ± SD

Range (Min–Max)

19.5 ± 14.86

2–36

Male:female

1:1.83

Height, cm

147.7 ± 20.82

72–193

Weight, kg

45.5± 20.93

11–130

BMI

18.3 ± 6.23

10–37

234.7 ± 95.34

20–580

Operation time, min

SD indicates standard deviation; BMI, body mass index.

Spine

Neuromuscular scoliosis

62 (5.3)

Neurofibromatosis scoliosis

41 (3.5)

Adult scoliosis

95 (8.2)

Marfan syndrome scoliosis

25 (2.2)

Ankylosing spondylitis

42 (3.6)

Scheuermann disease

12 (1.0)

Congenital kyphosis

45 (3.9)

study (Table 3), were diagnosed as congenital scoliosis, which should therefore be considered as a high-risk diagnosis.

Case Studies The following cases are presented to display some successful monitoring cases from the study population. Case 1 A 13-year-old adolescent girl with congenital scoliosis (Figure 2A) underwent posterior hemivertebra resection (T3–T4 and L1), correction, and instrumentation (T1–L3). The pedicle screw insertion was extremely difficult due to her complicated spinal anomalies. Just after insertion of the T2 screw on the right side, the MEP signal of right side became completely loss (Figure 2D) compared with baseline (Figure 2C). After being reported the monitoring alert, the surgeon immediately ordered the intraoperative computed tomographic scan, which (Figure 2B) showed that the T2 screw was accidentally inserted into the vertebral canal and compressed the spinal cord. Then, the surgical team removed the screw immediately. MEP and somatosensory evoked potential data remained undetectable until the surgery ended. On awakening from anesthesia, she presented an obvious weakness of right lower extremity. Her neurological function returned to normal 6 days later. This case indicated that the MEP change being synchronously and logically associated with the high-risk surgical maneuver should be taken seriously. Case 2 A 14-year-old adolescent boy with a tuberculokyhposis underwent a posterior vertebral column resection (T9, T10, and T11), correction (T4–L4) (Figure 3A, E). During the surgery, it was found that the posterior column of the 3 apical vertebras had fused together due to repeated chronic tuberculosis infection. The surgical removal of posterior part at the apex was so difficult that the temporary rod, which was www.spinejournal.com

E1427

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140542.indd E1427

22/10/14 8:41 PM

Age (yr)

10

17

13

14

9

49

14

4

23

26

6

35

16

13

28

28

22

26

11

11

3

12

Patient

SPINE140542.indd E1428

1

www.spinejournal.com

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

Sex

F

F

F

M

F

M

M

F

F

F

F

F

M

F

F

M

M

F

M

M

M

F

Preoperative Neurological Deficit

1

1

1

2

1

1

2

2

2

1

2

2

2

2

1

1

2

1

2

1

1

1

Diagnosis

Idiopathic scoliosis

Congenital scoliosis

Congenital scoliosis

Congenital scoliosis

Congenital scoliosis

Congenital scoliosis

Congenital kyphoscoliosis

Congenital kyphoscoliosis

Congenital scoliosis

Congenital scoliosis

Adult scoliosis

Congenital scoliosis

Congenital kyphoscoliosis

Adult scoliosis

Congenital scoliosis

Idiopathic scoliosis

Congenital kyphoscoliosis

Congenital scoliosis

Congenital kyphoscoliosis

Congenital scoliosis

Congenital kyphoscoliosis

Congenital scoliosis

Spinal Cord Anomalies No

No

No

Compound

No

Syringomyelia

No

No

No

No

Diastematomyelia

Diastematomyelia

No

No

No

No

No

Diastematomyelia

No

HV resection

VCR

VCR

No

VCR

VCR

HV resection

SPO

VCR

PSO

HV resection

VCR

PSO

No

No

VCR

VCR

VCR

HV resection

Low-level myelomeningocele No

No

No

Osteotomy Type

No

No

Baseline Obtained Normal

Normal

Normal

Normal

Normal

Normal

Unilateral obtainable

Unilateral obtainable

Normal

Normal

Normal

Normal

Normal

Normal

Normal

Normal

Normal

Normal

Normal

Normal

Normal

Normal

Maneuver at Time of MEP Loss

Y

Y

Hemostatic sponge placement Screw insertion

Y

N

N

N

N

Partial

Partial

Y

Y

N

Y

Partial

Y

Y

Y

Y

Partial

Y

Partial

Y

Intraoperative MEP Recovery

Cage placement

Osteotomy

Osteotomy

Osteotomy

Osteotomy

Direct contusion

Osteotomy

Screw insertion

Correction

Osteotomy

Osteotomy

Correction

Correction

Screw insertion

Osteotomy

Osteotomy

Direct contusion

Osteotomy

Osteotomy

Screw insertion

Time Interval for Return (min) 5

13

7













15

25



30



5

28

15

15



20



10























+

(Continued )







+

+

+

+

+

+

+

+

+

+



+ +





− −

+







− −







− +



Postoperative Neurological Deficit

E1428

Permanent Neurological Injury

TABLE 3. The Variables for 25 Patients With MEP Alert

DEFORMITY Intraoperative MEP Monitoring • Zhuang et al

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

November 2014

22/10/14 8:41 PM

DEFORMITY

Diagnosis Preoperative Neurological Deficit

− +

Age (yr)

13

Patient

25

Sex

28

Preoperative neurological deficit: 1, no preoperative neurological deficit; 2, preoperative moderate or minimal neurological deficit; 3, preoperative severe neurological deficit. Postoperative neurological deficit: +, postoperative neurological deficit instantly; −, no postoperative neurological deficit instantly. MEP indicates motor evoked potential; PSO, pedicle subtraction osteotomy; SPO, Smith-Petersen osteotomy; VCR, vertebral column resection.

Y Osteotomy

Spinal Cord Anomalies

24

F

1

Congenital scoliosis

No

PSO

Normal

N No No Adult scoliosis 2

Osteotomy Type

14

M

No No Congenital scoliosis 2

Baseline Obtained

23

F

Normal

Correction

25

− +

+

Maneuver at Time of MEP Loss Spine

Normal

Intraoperative MEP Recovery

N

Time Interval for Return (min)

Osteotomy





Postoperative Neurological Deficit



Permanent Neurological Injury

TABLE 3. (Continued )

Intraoperative MEP Monitoring • Zhuang et al

routinely used in VCR osteotomy, was not used at the first beginning just to get a better view. The electrophysiologists found that the MEP signal of both sides suddenly dropped to complete loss, just when the last bony segment was removed from the fused posterior part of T9–T11 (Figure 3B). After being reported the MEP alert and the exact time, the surgeon concluded that the stretching force of spinal cord was the major reason of MEP signal loss. The temporary rod were therefore placed immediately (Figure 3C) and 1.5 g of methylprednisolone was administered, the MEP signal became detectable after 10 minutes and normalized 30 minutes later (Figure 3F). So, the correction and instrumentation continued (Figure 3D). When awakening from anesthesia, the patient demonstrated an obvious neurological deterioration. Since this case, the stretching force of the spinal cord during osteotomy has received high attention from the surgical team.

DISCUSSION Although more and more studies have proved the advantages of intraoperative recordings of MEPs,1,7,8,15 there remain the following problems. First, different alarm criteria and response protocol adopted by surgical teams in different studies compromised their comparability.7–11 Second, different definitions of true- and false-positive alerts result in large discrepancies in the reported sensitivity and specificity of spinal cord monitoring among previous studies.4,5,14 Third, high false-positive rate of MEP reported by previous studies has become an increasingly prominent problem that will limit its clinical use and development.2,4 Our objective, therefore, is to develop and evaluate a protocol of intraoperative MEP monitoring with the warning criteria we had established to make the best use of the advantages of MEP without compromising its true-negative rate. In our study, our warning criteria include 3 items: a more than 80% amplitude loss; synchronously and logically associated with high-risk surgical maneuver (pedicle screw insertion, osteotomy, correction, etc.); systemic and anesthetic factors being ruled out. First, based on our clinical experiences and the review of previous literature, our monitoring team chose 80% amplitude loss as the threshold criterion. Till now, warning criteria for TCE-MEPs have not been universally established and include monitoring MEP threshold,11,16,17 waveform morphology,18 and amplitude changes including 50%9, 60%,19,20 70%,21 80%,7 and complete loss.8,10,22,23 In the study by Langeloo et al,7 the author compared 3 different warning criteria applied retrospectively to the TCE-MEP recordings of 142 patients, and came to a conclusion that at least 1 amplitude decrease of at least 80% would be a sufficiently stringent warning criterion to ensure that no neurological events go undetected. Our results, being supportive to their findings, showed a high sensitivity and specificity of MEP monitoring using 80% amplitude loss as the threshold criteria. Second, our criteria placed emphasis on the association between the MEP signal loss and the surgical status. Nordwall and Wikkelso24 suggested that when spinal cord position changes during surgical correction, the signal transmission www.spinejournal.com

E1429

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140542.indd E1429

22/10/14 8:41 PM

DEFORMITY

Intraoperative MEP Monitoring • Zhuang et al

TABLE 4. Test Performance of TcMEP

Monitoring in the Patients

Transient New No New Neurological Deficit Neurological Deficit MEP alert

15 (true-positive, tp)

3 (false-positive, fp)

No MEP alert

0 (false-negative, fn)

1144 (true-negative, tn)

False-positive rate = fp/(fp + tn) = 0.26%; sensitivity = tp/(tp + fn) = 100%; specificity = tn/(fp + tn) = 99.7%. Positive predictive value = tp/(tp + fp) = 83.3%. MEP indicates motor evoked potential; TcMEP, transcranial motor evoked potentials.

through the spinal cord may be impeded. It is possible that the level of impairment may not produce a clinically relevant difference in sensory or motor function, but that it does cause a significant data change. Therefore, if the MEP signal loss is synchronously and logically associated with high-risk surgical maneuver and persists despite repeated optimization of stimulating parameters, it may represent a reversible subclinical spinal cord injury and must be taken seriously. On the contrary, for those unexplainable signal change with no logical association with surgical procedure, we can continue to observe closely and refer to somatosensory evoked potential monitoring unless the MEP signal disappear even after stimulating optimization. Third, according to our warning criteria, nonsurgical factors must be ruled out in the first place when MEP signals loss occurs. Because MEP, as being mentioned in many studies, presents too much sensitivity while lacking enough stability, it could be influenced by many factors during surgical procedures. In some previous studies,25,26 the monitoring team reported immediately to the surgeon as long as the amplitude decreased less than the predetermined threshold, without

Figure 1. High-risk surgical maneuvers associated with 25-MEP alerts. MEP indicates motor evoked potential.

E1430

www.spinejournal.com

Figure 2. A–D, Patient is a 13-year-old adolescent girl with congenital scoliosis, T1 wedge vertebra, T3–T4 and L1 hemivertebra (A), preoperative standing PA, lateral radiographs and CT scan of a 13-yearold adolescent girl (B), the T2 pedicle screw was accidentally inserted into the vertebral canal (C), reliable MEP baseline. D, MEP of right side was completely loss. MEP indicates motor evoked potential; CT scan, computed tomographic scan.

checking its relationship with surgical procedure and ruling out anesthetic or systematic reasons. These alarms could be misleading and may compel surgeons to take unreasonable risks or unnecessarily change the operative plan in the setting of a potentially false-positive alert. As mentioned in the earlier text, MEPs monitoring is characteristic of its real-time and dynamic signal reporting mechanism, which results in its high sensitivity in detecting impending spinal cord injury. After a review of 427 monitored cases, Hilibrand et al19 reported that the use of MEP monitoring is both 100% sensitive for identification of evolving motor tract injury during cervical spine surgery. The study by Kim et al14 on patients with cervical myelopathy also reported 100% sensitivity of MEP for detection of clinically significant intraoperative cord injury. Our results November 2014

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140542.indd E1430

22/10/14 8:41 PM

DEFORMITY

Intraoperative MEP Monitoring • Zhuang et al

Figure 3. A–E, Patient is a 14-year-old adolescent boy with a tuberculokyhposis who underwent a posterior vertebral column resection, correction and spinal fusion (A), preoperative standing PA, lateral radiographs and MR image of a 14-year-old adolescent boy (B), the last bony segment was removed from the fused posterior part of T9–T11 without temporary rod. C, Temporary rod was placed immediately after the MEP alert. D, Post VCR osteotomy and correction. E, Postoperative PA and lateral radiographs. F, Intraoperative MEP. MEP indicates motor evoked potential; MR image, magnetic resonance image; PA, posteroanterior.

support the very high sensitivity of MEP monitoring for intraoperative motor tract injury. However, several previous studies have stated that MEP’s instability and hypersensitivity led to high trial-by-trial variability of its amplitude and especially, high false-positive rate. In the study by Kim et al,14 the calculated specificity of TcMEP monitoring in patients with cervical myelopathy was found to be 90%. Of the 6 total TcMEP-based alerts in the series, there was 1 true-positive and 5 false-positives. The false-positive rate was as high as 9.8%, and the PPV was 17%. Compared with these reports, our study successfully controlled the false-positive rate of MEP as low as 0.26%, the false-negative rate 0%, and improved the PPV to as high as 83.3%. These satisfying results prove the efficacy and safety of the protocol of intraoperative MEP monitoring adopted in this study. It is worth noting that the classification of MEP monitoring findings was quite varied among different studies, especially for the differentiation of “false-positive” and “true-positive.” Most studies agree to define as false positive for those alerts with persistent MEP abnormalities that were not associated with either intraoperative or postoperative neurological deficits. However, the definition varied in terms of those MEP signal significant loss that normalized after corrective measures with no new motor deficit. Thuet et al27 and WilsonHolden et al5 regarded it as false-positive because there is no neurological deficit, whereas Tamkus et al4 considered it as true-positive considering that it got back to normal after intervention. Kim et al14 classified it as “indeterminate” and stated that it was impossible to know for certain whether Spine

such an event represented a true impending cord injury that was averted or a false-positive alert that spontaneously disappeared. In our study, we agreed that it would be difficult, if not impossible, to distinguish whether it is a reversible injury or an unstable change. Therefore, we adopted the classification system proposed by Kim et al14 in our analysis. One limitation of this study is that it was not a prospective randomized study. Randomized control trials are difficult to accomplish ethically with intraoperative neuromonitoring. In this study, we evaluated our experience using MEP monitoring in a consecutive series of patients who underwent different spine deformity surgery at a single highvolume spine surgery specialty hospital. Monitoring techniques, equipment, and software was kept almost the same during the 3 years. And, use of consecutive patients from the single surgical center somewhat mitigates potential selection bias. The strength of this study, on the contrary, includes its relatively large sample size and excellent results using our protocol and warning criteria. The criteria established on the basis our analysis of previous studies and clinical experience, hopefully could provide guidance for the MEP monitoring in spinal surgical procedures. Future well-designed prospective studies are needed to further validate and understand the findings of this study.

CONCLUSION Our study indicates that the appropriate use of MEP monitoring based on our protocol is able to obtain satisfying sensitivity and specificity and thus provide important information for intraoperative decision making. www.spinejournal.com

E1431

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140542.indd E1431

22/10/14 8:41 PM

DEFORMITY ➢ Key Points ‰ MEP’s instability and hypersensitivity led to high trial-by-trial variability of its amplitude and especially, high false-positive rate. ‰ An ideal MEP warning criterion should be established based on not only signal amplitude but also the surgical status and the logical relationships between them. ‰ Our warning criteria include 3 items: (1) a more than 80% amplitude loss; (2) synchronously and logically associated with high-risk surgical maneuver; and (3) systemic and anesthetic factors being ruled out. ‰ Close cooperation between surgeon and welltrained electrophysiologists is the key to avoid postoperative neurological deficit, especially during these high-risk and complicated surgical procedure.

Intraoperative MEP Monitoring • Zhuang et al

10.

11. 12. 13.

14. 15. 16. 17.

Acknowledgments

18.

Qianyu Zhuang and Shujie Wang contributed equally to this article.

References

1. Lo YL, Dan YF, Teo A, et al. The value of bilateral ipsilateral and contralateral motor evoked potential monitoring in scoliosis surgery. Eur Spine J 2008;17:S236–8. 2. Jones SJ, Harrison R, Koh KF, et al. Motor evoked potential monitoring during spinal surgery: responses of distal limb muscles to transcranial cortical stimulation with pulse trains. Electroencephalogr Clin Neurophysiol 1996;100:375–83. 3. Tamaki T, Kubota S. History of the development of intraoperative spinal cord monitoring. Eur Spine J 2007;16:S140–6. 4. Tamkus AA, Rice KS, Kim HL. Differential rates of false-positive findings in transcranial electric motor evoked potential monitoring when using inhalational anesthesia versus total intravenous anesthesia during spine surgeries. Spine J 2013;13:01484–8. 5. Wilson-Holden TJ, Padberg AM, Lenke LG, et al. Efficacy of intraoperative monitoring for pediatric patients with spinal cord pathology undergoing spinal deformity surgery. Spine 1999;24:1685–92. 6. Langeloo DD, Journee HL, de Kleuver ME, et al. Criteria for transcranial electrical motor evoked potential monitoring during spinal deformity surgery a review and discussion of the literature. Neurophysiol Clin 2007;37:431–9. 7. Langeloo DD, Lelivelt A, Louis Journee H, et al. Transcranial electrical motor-evoked potential monitoring during surgery for spinal deformity: a study of 145 patients. Spine 2003;28:1043–50. 8. Kempton LB, Nantau WE, Zaltz I. Successful monitoring of transcranial electrical motor evoked potentials with isoflurane and nitrous oxide in scoliosis surgeries. Spine 2010;35:E1627–9. 9. Hsu B, Cree AK, Lagopoulos J, et al. Transcranial motor-evoked potentials combined with response recording through compound

E1432

www.spinejournal.com

19.

20. 21. 22. 23.

24. 25. 26.

27.

muscle action potential as the sole modality of spinal cord monitoring in spinal deformity surgery. Spine 2008;33:1100–6. Sakaki K, Kawabata S, Ukegawa D, et al. Warning thresholds on the basis of origin of amplitude changes in transcranial electrical motor-evoked potential monitoring for cervical compression myelopathy. Spine 2012;37:E913–21. Calancie B, Molano MR. Alarm criteria for motor-evoked potentials: What’s wrong with the “presence-or-absence” approach? Spine 2008;33:406–14. Moller AR, Ansari S, Cohen-Gadol AA. Techniques of intraoperative monitoring for spinal cord function: their past, present, and future directions. Neurol Res 2011;33:363–70. Osburn LL. A guide to the performance of transcranial electrical motor evoked potentials. Part 1. Basic concepts, recording parameters, special considerations, and application. Am J Electroneurodiagn Technol 2006;46:98–158. Kim DH, Zaremski J, Kwon B, et al. Risk factors for false-positive transcranial motor evoked potential monitoring alerts during surgical treatment of cervical myelopathy. Spine 2007;32:3041–6. Schwartz DM, Auerbach JD, Dormans JP, et al. Neurophysiological detection of impending spinal cord injury during scoliosis surgery. J Bone Joint Surg Am 2007;89:2440–9. Calancie B, Klose KJ, Baier S, et al. Isoflurane-induced attenuation of motor evoked potentials caused by electrical motor cortex stimulation during surgery. J Neurosurg 1991;74:897–904. Calancie B, Harris W, Brindle GF, et al. Threshold-level repetitive transcranial electrical stimulation for intraoperative monitoring of central motor conduction. J Neurosurg 2001;95:161–8. Quinones-Hinojosa A, Lyon R, Zada G, et al. Changes in transcranial motor evoked potentials during intramedullary spinal cord tumor resection correlate with postoperative motor function. Neurosurgery 2005;56:982–93; discussion 982–93. Hilibrand AS, Schwartz DM, Sethuraman V, et al. Comparison of transcranial electric motor and somatosensory evoked potential monitoring during cervical spine surgery. J Bone Joint Surg Am 2004;86-A:1248–53. Lee JY, Hilibrand AS, Lim MR, et al. Characterization of neurophysiologic alerts during anterior cervical spine surgery. Spine 2006;31:1916–22. Schwartz DM, Sestokas AK, Hilibrand AS, et al. Neurophysiological identification of position-induced neurologic injury during anterior cervical spine surgery. J Clin Monit Comput 2006;20:437–44. Kim SM, Yang H, Park SB, et al. Pattern-specific changes and discordant prognostic values of individual leg-muscle motor evoked potentials during spinal surgery. Clin Neurophysiol 2012;123:1465–70. Sala F, Palandri G, Basso E, et al. Motor evoked potential monitoring improves outcome after surgery for intramedullary spinal cord tumors: A historical control study. Neurosurgery 2006;58:1129–43; discussion 1129–43. Nordwall A, Wikkelso C. A late neurologic complication of scoliosis surgery in connection with syringomyelia. Acta Orthop Scand 1979;50:407–10. Lewis SJ, Gray R, Holmes LM, et al. Neurophysiological changes in deformity correction of adolescent idiopathic scoliosis with intraoperative skull-femoral traction. Spine 2011;36:1627–38. Pastorelli F, Di Silvestre M, Plasmati R, et al. The prevention of neural complications in the surgical treatment of scoliosis: the role of the neurophysiological intraoperative monitoring. Eur Spine J 2011;20:S105–14. Thuet ED, Winscher JC, Padberg AM, et al. Validity and reliability of intraoperative monitoring in pediatric spinal deformity surgery: a 23-year experience of 3436 surgical cases. Spine 2010;35:1880–6.

November 2014

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140542.indd E1432

22/10/14 8:41 PM

How to make the best use of intraoperative motor evoked potential monitoring? Experience in 1162 consecutive spinal deformity surgical procedures.

A retrospective study of 1162 consecutive patients who underwent spinal deformity surgical procedures at our spine center from January 2010 to Decembe...
1MB Sizes 0 Downloads 5 Views