The use of transcranial motor-evoked potentials (TcMEPs) in intraoperative neurophysiologic monitoring has proliferated over the past decade. Although numerous studies1,2 support TcMEP monitoring for evaluating corticospinal tract function during surgery, few publications have addressed the potential adverse effects of transcranial stimulation. Bite injuries due to jaw muscle contractions during stimulation are a disturbing complication associated with TcMEP monitoring.3 There is the additional risk in cases of severe tongue hematoma of potential airway obstruction after tracheal extubation. We retrospectively reviewed the incidence of bite injuries associated with TcMEP monitoring performed in 17,273 consecutive cases and offer a discussion of the circumstances and related factors surrounding these events.
The study protocol was reviewed by the New England IRB and an exempt status for anonymous retrospective chart review was granted. TcMEPs were monitored in 17,273 surgical procedures during 2009 to 2010 occurring at 307 different hospitals. Monitoring was performed using an evoked potential system from 1 of 3 manufacturers: Xltek Protektor (Natus, San Carlos, CA), Cadwell Cascade or Elite (Cadwell Laboratories, Kennewick, WA), or Axon NIM-Eclipse (Medtronic Xomed, Jacksonville, FL). Transcranial stimulation was routinely performed via 13-mm subdermal needles (Ambu or Rhythmlink) placed at C3 and C4 of the international 10-20 system of electrode placement. Stimulus variables were machine-dependent. The default stimulus for the Xltek system was a pulse train of 7 pulses of 0.5-millisecond duration at 250 Hz using a constant current stimulator up to a maximum of 200 mA. In some cases, pulse duration was increased to 1 millisecond and/or stimulus pulses were increased to as many as 11. For the Cadwell and Axon systems, the stimulus was a pulse train of 7 pulses of 75-microsecond duration using a constant voltage stimulator up to a maximum of 1000 V. The stimulus pulses delivered by the Axon NIM-Eclipse machine were biphasic rather than monophasic. The advantage of biphasic stimulation is that both sides of the corticospinal system are activated with a single pulse train. Monophasic stimulation usually requires stimulating with both polarities (C3+/C4− and C4+/C3−) to activate muscle responses on each side.
Our standard protocol for TcMEP monitoring included recommending the placement of bilateral soft bite blocks to protect the patient's tongue. The technologist was required to communicate this recommendation to the anesthesia personnel and to document that they visually verified placement of the bite blocks. Ideally, the blocks would be secured between the molars such that the teeth could not close on the tongue or endotracheal tube. Additionally, it was important to confirm that the tongue and lips were free and not trapped by the teeth or bite block. The type and number of bite blocks ultimately placed by the anesthesia personnel were documented in each case. Any bite-related injuries to teeth or soft tissue (e.g., tongue or lips) discovered at the end of a procedure were reported by the technologist using a surgical incident report form.
Our protocol for anesthesia during TcMEP monitoring included recommending minimal use of inhaled anesthetics, specifically ≤0.5 minimum alveolar concentration (MAC) or a total IV anesthetic (TIVA) consisting of primarily propofol (100–200 μg/kg/min) and narcotic, particularly in patients with any preexisting weakness or myelopathy, and avoidance of neuromuscular block except during intubation.
The neuromonitoring documentation for all surgical procedures involving bite injuries during 2009 to 2010 were reviewed and analyzed. Demographic data, procedure, type and number of bite blocks, type and dosages of primary anesthetic drugs used, equipment, and stimulus variables were recorded along with comments regarding the nature and severity of the injury and treatment required. Confidence intervals (CIs) were calculated for the incidence of injury using the Wilson “score” method without continuity correction as described by Newcombe.4 Data were analyzed using SPSS 19.0 for Windows using Pearson χ2 and Fisher exact test. Significance (2-tailed) was defined as a P value <0.05.
There were 111 reported bite injuries in 109 patients of 17,273 TcMEP-monitored cases for a total incidence of 0.63% with a 95% CI of 0.52% to 0.76%. This included 54 males and 55 females 8 to 87 years of age (mean, 50.6 years). The 111 oral injuries included 88 (79.3%) tongue injuries, 22 (19.8%) lip injuries, and 1 (0.9%) tooth injury. One patient experienced both a tongue and lip injury during the same surgery. One patient who had a minor lip injury also suffered a broken incisor. The severity of the injuries ranged from minor but visible bruising or slight bleeding to deep lacerations of the tongue or lip requiring sutures. Ten of the tongue injuries and 3 of the lip injuries were reported as bruising without laceration. Twenty-five of the injuries (22.5%) were treated by suturing the tongue (24 patients) or lip (1 patient) to repair. Thus, of all TcMEP-monitored procedures, the incidence of these more severe injuries was 0.14% (95% CI: 0.09%–0.21%). Seven cases required otolaryngologic consultation before tracheal extubation, and 1 patient was left intubated overnight because of significant swelling of the tongue. Of the 307 hospitals in which TcMEPs were performed, the 109 cases of bite injury occurred in 79 hospitals and were monitored by 80 different technologists.
Procedure and Patient Positioning
Patients were positioned supine only (anterior cervical surgery) in 59 cases (54.1%), prone only (posterior spinal surgery) in 39 cases (35.8%), combined anterior and posterior in 5 cases (4.6%), and lateral in 6 cases (5.5%). Table 1 shows the comparison of noninjured versus injured patients for age, gender, and positioning. Statistical analysis showed no difference between the groups for age or gender. There was a higher percentage of injured patients who underwent anterior surgical approaches versus posterior; however, this did not reach significance (P = 0.078).
Type and Number of Bite Block
The type of bite block used was not documented in 10 cases. The use of bilateral soft bite blocks consistent with recommended protocol was the most common, occurring in 56 cases (51.3%). A single soft bite block, usually rolled gauze, was used in 29 cases (26.6%). The number of gauze squares used per bite block was not documented. A commercially available hard bite block (e.g., Bite Gard®; Hudson RCI, Research Triangle Park, NC) was used in 8 cases (7.3%), a hard plastic airway was used in 4 cases (3.7%), and no bite block was used in 2 cases (1.8%). Hard bite blocks and plastic airways were sometimes combined with use of a rolled gauze block. Table 2 shows the breakdown of type of bite block used by minor versus severe bite injury. Bite block data were not available from the entire control group for comparison. We did not routinely capture bite block data until February of 2010. However, 7090 consecutive cases from our database had documented use of bilateral bite block in 4890 cases (69%), single bite block in 2088 cases (29.4%), and a hard bite block in 112 cases (1.6%).
Type and Depth of Anesthetic
From the control group, using a sample of 8888 consecutive cases from February to December 2010, 8355 (94%) used some inhaled anesthetic (non-TIVA cases). Of the 109 injured patients, 101 (92.7%) were anesthetized with inhaled anesthesia. Eight injured patients (7.3%) had TIVA. The majority of the injured patients (77 cases) were anesthetized with a balanced technique involving inhaled anesthetic above our recommended 0.5 MAC level including 60 patients maintained with up to 1 MAC and 17 patients maintained above 1 MAC. Twenty-four patients had inhaled anesthesia maintained at or below 0.5 MAC, and 8 patients were maintained on TIVA. Among the TIVA patients, propofol concentration was ≤100 μg/kg/min for 3 patients, 100 to 200 μg/kg/min for 4 patients, and >200 μg/kg/min for 1 patient (Table 2). The percentage of injuries requiring sutures with TIVA (12.5%, 1 of 8 cases) was lower compared with use of inhaled drugs (23.7%, 24 of 101 cases); however, statistical significance could not be established (P = 0.679).
Of the 77 patients maintained above 0.5 MAC, 59 (76.6%) required maximal intensity versus 15 (62.5%) of 24 patients maintained at <0.5 MAC. Of the 8 patients maintained with TIVA, 3 (37.5%) required maximal intensity stimulation.
Table 2 shows the number of patients in which stimulus variables were kept within default values versus patients in which the stimulus variables exceeded the default number of pulses (7) in the stimulus train or the maximal possible intensity was applied. Pulse trains of 8 to 11 pulses were applied in 62 patients versus 47 at the default 7 pulses. Maximum intensity was applied in 77 cases (70.6%) versus 32 cases in which stimulation was maintained below maximal values. In 6 cases, pulse duration was increased from 0.5 to 1 millisecond. Overall, 86 of 109 cases (78.9%) involved use of maximal intensity, increased pulse duration or train length to establish TcMEPs.
Equipment and Stimulus Parameters
The equipment used among the bite-injured patients included 94 Xltek, 12 Axon, and 3 Cadwell machines. Table 3 shows the incidence of bite injuries for each machine as well as the average stimulus variables for each. The average stimulus intensity applied was 192 mA for the Xltek machines, 752 mA for the Axon machines, and 342 V for the Cadwell machines. The average number of stimulus trains applied was 57 (range: 9–265). There was a significant association between monitoring system and injury status (Pearson χ2 = 9.1, df = 2, P = 0.003). Post hoc Fisher exact test with Bonferroni correction for 3 comparisons (i.e., P values were multiplied by 3 to calculate the corrected P) showed that the Xltek system injury rate (0.6%) was significantly different from the Axon system (1.37%); corrected P = 0.039. However, there was no significant difference between the Cadwell and Axon systems; corrected P = 0.102. The Xltek and Cadwell systems were not different from each other.
Comments from the Surgical Incident Reports
The most common comment (14 cases) indicated that 1 or more of the bite blocks was observed to have shifted out of position (in some cases completely falling out) or had become flattened and ineffective during the case. Positioning of the patient was sometimes implicated in the shifted bite block. Shifting or swelling of the tongue was attributed to the injury in 8 cases. Malocclusion of the teeth or missing teeth may have contributed to bite injuries in 5 cases. Four reports indicated the bite block was initially too small or improperly placed. In 2 cases, failure of the technologist to communicate with the anesthesiologist regarding bite block placement resulted in no bite block being placed. Two reports indicated that blood was discovered in the patient's mouth, but it was unclear where the source was.
Oral trauma from biting caused by transcranial stimulation is an infrequent but distressing complication of TcMEP monitoring, occurring in 0.63% (95% CI: 0.52%–0.76%) of our study group. Lacerations requiring suture repair occurred in 0.14% (95% CI: 0.09%–0.21%) and introduced delays in tracheal extubation including 1 case requiring extended intubation because of significant tongue swelling and potential airway compromise. Contusion or laceration of the tongue or lip, tooth damage, and even a mandibular fracture5 have been attributed to jaw muscle clenching during TcMEP stimulation. Two studies have also attributed strong stimulus-induced biting to causing ruptured endotracheal tubes requiring emergency reintubation.6,7 Although rare, these serious events illustrate the need for applying protective measures to patients undergoing TcMEP monitoring.
The mechanisms of transcranial stimulus-induced jaw clenching includes 3 possible means: direct temporalis and possibly masseter muscle activation (especially with laterally placed scalp electrodes), activation of trigeminal nerve motor fibers due to deep penetrating current (which may occur with high-intensity stimulation and widely spaced stimulating electrodes), and corticobulbar tract activation.
MacDonald3 reported an estimated incidence of tongue or lip laceration to be 0.19%, or 29 reported injuries of approximately 15,000 cases, a rate quite lower than our result. However, MacDonald's estimate was a summary of primarily unpublished reports and personal communication. It is possible that these sources underreported or only reported the most severe injuries that took place. Only 3 of the 29 reported injuries came from published reports.8,9
Lieberman et al.10 reported 4 minor tongue lacerations despite using a secured gauze bite block from a group of 35 patients (11.4% incidence). Deiner and Osborn11 reported several cases of lingual hematoma without laceration with 4 × 4 gauze rolled bite blocks placed between the patient's teeth. Mahmoud et al.12 reported a necrotic tongue lesion that occurred after shifting of a soft bite block and subsequently recommended placing an additional dental guard. Davis et al.13 reported 2 cases of significant tongue and lip injury.
In a recent study of the safety of TcMEP monitoring, Schwartz et al.14 reported on the incidence of complications from 18,862 spine procedures. They reported 25 tongue lacerations, an incidence of 0.13%. Twenty-one cases were described as “self-limiting tongue lacerations or abrasions that healed without intervention,” whereas 4 others required otolaryngologic consult and suturing. This very low incidence of tongue injury can probably be explained by some important differences in method and reporting.
First, only complications that rose to the level of documentation by the anesthesiologist or surgeon were included; thus, minor bleeding or bruising not considered noteworthy would not have been included. Our reports were submitted by technologists who in some cases were clearly conservative in their reporting. For example, comments from our incident reports such as “some blood on the bottom of the tongue” or “source of bleeding was not determined” suggest minor events that may be related to tracheal intubation or suctioning and might easily be dismissed.
Second, Schwartz et al. consistently used a TIVA technique, which has been recommended by some to be favorable for obtaining TcMEPs.15–17 Given that inhaled drugs were used as part of the anesthetic regimen in approximately 94% of our TcMEP cases, higher intensity stimulation may have been necessary, particularly when inhaled drugs were used at values >0.5 MAC. Our data do not support one anesthetic regimen over the other and this topic needs future study. Although our limited data suggest a trend to greater use of maximal stimulus with higher anesthetic values, many factors influence the use of maximal stimulus variables including patient physiology, anesthesia values of both inhaled and TIVA drugs, stimulus electrode placement, scalp edema, or monitoring strategy. Although avoiding maximal stimulus values is not always possible, we recommend using the lowest intensity to obtain acceptable TcMEPs and optimizing technical and anesthetic conditions to obtain TcMEPs.
Finally, Schwartz et al. reported performing stimulation at the C1–C2 scalp locations, whereas our protocol used C3–C4. Because of the more lateral location of C3 and C4 and greater interelectrode distance, direct activation of temporalis is increased, and activation of corticobulbar pathways is also increased leading to stronger jaw contractions (Fig. 1). Our choice to use C3–C4 by default in our protocol was to optimize our chances of getting TcMEP responses18 and was not guided by the risk of bite injury.
The NIM-Eclipse device had more than twice the proportion of bite injuries compared with the other machines. In part, this may be related to its high current, high voltage capacity and the use of biphasic stimulus pulses, which is currently not available for transcranial stimulation on the Xltek or Cadwell machine. Whether this mode of stimulus actually increases the overall biting response and therefore risk of bite injury compared with traditional monophasic transcranial stimulation deserves future study. The NIM-Eclipse was also introduced into our organization midway through 2010. It is possible that user inexperience with the stimulus variables as compared with the Xltek device led to delivery of higher intensity stimuli (752 V average for the injured group) that contributed to increased injuries.
Use of partial neuromuscular blockade to reduce muscle contractions during stimulation has been suggested as a strategy to reduce risk of bite injury13; however, this may compromise the quality and interpretability of the TcMEP as well as other monitoring modalities (e.g., electromyography) that depend on the neuromuscular junction. Therefore, we do not recommend routine use of neuromuscular blocking drugs as a method of reducing risk of bite injury.
Despite consistent use of bite blocks, tongue and lip injuries still occur. This may be the result of improperly sized or improperly placed bite blocks or displacement of the bite block during the procedure. Risk factors appear to include missing or malocclusions of the teeth, swelling of the tongue, high-intensity stimulation, and positioning of the patient, which may displace the bite block. Although not statistically significant, anterior approaches tended to be more prevalent than posterior approaches in the bite-injured group suggesting that perhaps the tongue falls back in the mouth and may be more easily trapped by the teeth.
The specific type of bite block used seems to be less important than ensuring that the bite block is large enough to prevent the teeth from being able to close on the tongue and is secured properly to minimize chance of shifting. The mouth should be inspected to ensure that the tongue and lips are clear of the teeth and not trapped by the bite block. We discourage the use of plastic oral airways or the endotracheal tube as bite protection. Tooth injury is more likely to occur when biting down on hard plastic, and endotracheal tube rupture is a potential serious consequence of biting. Bite blocks should be periodically inspected, particularly after changes in patient position.
Name: Arvydas Tamkus, MD, PhD, DABNM.
Contribution: This author helped design the study, conduct the study, analyze the data, and approve the final manuscript.
Name: Kent Rice, MS, DABNM.
Contribution: This author helped analyze the data and prepare the manuscript.
This manuscript was handled by: Gregory J. Crosby, MD.
1. Hilibrand A, Schwartz D, Sethuraman V, Vaccaro A, Albert T. Comparison of transcranial electric motor and somatosensory evoked potential monitoring during cervical spine surgery. J Bone Joint Surg Am 2004;86:1248–53
2. 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
3. MacDonald DB. Safety of intraoperative transcranial electrical stimulation motor evoked potential monitoring. J Clin Neurophysiol 2002;19:416–29
4. Newcombe RG. Two-sided confidence intervals for the single proportion: comparison of seven methods. Stat Med 1998; 17:857–72
5. Calancie B, Harris W, Brindle F, Green B. Threshold-level transcranial electrical stimulation for intraoperative monitoring of central motor conduction. J Neurosurg 2001;95:161–86
6. MacDonald DB. Intraoperative motor evoked potential monitoring: overview and update. J Clin Monit Comput 2006;20: 347–77
7. Duma A, Novak K, Schramm W. Tube-in-tube emergency airway management after a bitten endotracheal tube caused by repetitive transcranial electrical stimulation during spinal cord surgery. Anesthesiology 2009;111:1155–7
8. Jones SJ, Harrison R, Koh KF, Mendoza N, Crockard HA. 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
9. 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
10. Lieberman JA, Lyon R, Feiner J, Hu SS, Berven SH. The efficacy of motor evoked potentials in fixed sagittal imbalance deformity correction surgery. Spine 2008;33:E414–24
11. Deiner SG, Osborn IP. Prevention of airway injury during spine surgery: rethinking bite blocks. J Neurosurg Anesthesiol 2009;21:68–9
12. Mahmoud M, Spaeth J, Sadhasivam S. Protection of tongue from injuries during transcranial motor-evoked potential monitoring. Paediatr Anaesth 2008;18:902–3
13. Davis SF, Kalarickal P, Strickland T. A report of two cases of lip and tongue bite injury associated with transcranial motor evoked potentials. Am J Electroneurodiagnostic Technol 2010;50:313–20
14. Schwartz DM, Sestokas AK, Dormans JP, Vaccaro AR, Hillibrand AS, Flynn JM, Li PM, Shah SA, Welch W, Drummond DS, Albert TJ. Transcranial motor evoked potential monitoring in spine surgery: is it safe? Spine 2011;36:1046–9
15. Sloan T, Heyer E. Anesthesia for intraoperative neurophysiologic monitoring of the spinal cord. J Clin Neurophysiol 2002;19:430–43
16. Chen Z. The effects of isoflurane and propofol on intraoperative neurophysiological monitoring during spinal surgery. J Clin Monit Comput 2004;18:303–8
17. Scheuffler K, Zentner J. Total intravenous anesthesia for intraoperative monitoring of the motor pathways: an integral view combining clinical and experimental data. J Neurosurg 2002;96:571–9
18. Szelenyi A, Kothbauer KF, Deletis V. Transcranial electric stimulation for intraoperative motor evoked potential monitoring: stimulation parameters and electrode montages. Clin Neurophysiol 2007;118:1586–95