During corrective surgery for spinal deformities, spinal cord monitoring of afferent and efferent nerve pathways is essential to detect and correct for possible nerve damage caused by compression or distraction of the spinal cord, indirect spinal cord lesions caused by decreased blood perfusion, or direct nerve damage caused by inserted metal implants (1,2). Standard monitoring procedures for scoliosis surgery consist of somatosensory evoked potentials (SSEPs) evoked at the foot, and evoked motor potentials (MEPs) using transcranial electrical stimulation (TCES). These procedures are very effective in detecting impending spinal cord injury during surgery for spinal deformity (1).
The voltages induced when MEPs are evoked by TCES are assumed to be destructive to any implanted brain or cranial nerve stimulator and its surrounding tissue. However, there is a lack (0.14%, almost exclusively tongue lacerations) of reported complications involved with TCES (3). Nonetheless, manufacturers of brain and cranial nerve stimulators, especially those of cochlear implants (CIs), consider the use of MEP eliciting TCES devices’ high risk and contraindicated (4–6). Similar recommendations (5,6) exist for monopolar electrical cauterization (MoEC) at or above the level of the second thoracic vertebra (T2). MoEC is assumed to cause destruction of implant electronics and auditory nerve damage. However, when tested on cadavers with a CI, there was no evidence for such damage following MoEC at the tongue or abdomen (7). The question of possible damage to a CI during TCES has only been investigated, based on our literature search, at one clinic on 2 CI patients (8). Yellin et al. (8) reported no damage. This lack of investigations contrasts with reports that children with congenital spinal deformities often present with miscellaneous problems (9), and some of them are likely to need a CI because of a comorbid hearing loss (10).
CIs are cranial acoustic nerve stimulators which tonal-topically stimulate the auditory nerve within the cochlea. An externally worn speech processor communicates with the implanted electronics transcutaneously via a transmitter coil held in place by attracting magnets, one centered in the transmitter coil and one centered in the middle of the receiver electronics coil. Thus, electrical currents at the head during TCES or MoEC could also induce receiver coil currents and magnet movement. The intended primary outcome for CIs, improved speech understanding, is successful. Mean speech discrimination at normal speech loudness level (65 dB SPL) is 75% among adult Swiss patients (11) allowing many to have unlimited choices in education or profession. Even in persons who cannot communicate orally because of CNS deficits, the presence of an auditory input “channel” improves total communication (12).
The case reported here was a 15-year-old male patient with a CI who needed corrective spinal surgery for a progressive congenital kyphoscoliosis at the thoracolumbar junction to preserve ambulation. TCES for MEPs was required to monitor efferent nerve integrity and MoEC was needed to keep blood loss low during subperiostal preparation. Both techniques are declared as contraindicated for patients by CI manufacturers (5,6), yet yielded no damage to our patient's CI. Thus, the purpose of this case study report is to increase awareness of the issues involved with using TCES and MoEC in CI patients, thereby encouraging studies defining limits for induced voltages and applicable power with these techniques. As far as we are aware, this is the first study documenting the lack of CI damage when both techniques are used on the same patient.
The patient had a Moebius syndrome (13), bilateral deafness, congenital right-sided facial paresis, cerebral apraxia with cerebellum hypotrophy, central respiratory regulation disturbance, and progressive congenital kyphoscoliosis. He was referred to the University of Basel Children's Hospital (UKBB) for corrective spinal surgery. The main deformity affected the sagittal profile of the spine with a rigid kyphosis of 70° at the lower thoracic spine. Owing to the deformity progression and its impact on sagittal balance control, the patient was on the verge of losing his ability to walk. The focus of the surgery was to restore the sagittal balance by performing a corrective posterior instrumented spinal fusion, thereby maintaining ambulation. An informed consent form was signed by the parents, including the permission to use the anonymized medical data of the patient for publication.
A CI (N24 RECA, Cochlear Ltd., Sydney, Australia) was implanted on the left side in 2006 at 6 years of age. Owing to no response to electrical stimulation, the 2 most basal of the 22 electrodes were not activated. The impedance of all electrodes was normal (see preoperative values in Fig. 1). Although speech perception could not be tested because of the patient's multiple handicaps, both parents and audiopaedagogic personnel confirmed reactions to speech and sound.
A working group from our departments assessed the risks of using MoEC for hemostasis (conventional monopolar blade (Medtronic Advanced Energy, Portsmouth, NH) and PEAK Plasma Blade (PEAK Surgical, Palo Alto, CA)), and TCES for MEP monitoring. As the patient had multiple handicaps, including mental retardation, the use of a “wake-up test” to evaluate the neurologic condition during surgery was predicted to fail and therefore to be avoided. During the wake-up test, sedatives are reduced until the patient is able to cooperate and move his legs and feet on command. Reduced movement would indicate possible motor nerve damage. This test requires certain cognitive demands that our patient could not comply with. Because the patient was able to walk and had normal sensorimotor function based on preoperative tests, MEP monitoring during spinal surgery was considered essential.
Our risk assessment focused on the advantages and safety of TCES for MEP monitoring (3) and MoEC (7) compared to the risk of CI damage based on the available evidence at the time (November 2015).
We identified the following potential risks:
- Cochlear tissue and/or CI electronics damage because of TCES or MoEC;
- Induced forces on the CI receiver-coil magnet because of TECS or MoEC causing magnet movement, as described for magnetic resonance imaging techniques (14).
- Risk of worsened facial nerve stimulation by the CI after reimplantation. This was occurring at current levels equivalent to maximum comfortable hearing levels preoperatively.
- Damage to the CI electrode array if explanted before spinal surgery and then reimplanted.
Risk 1 was given the highest priority as the total elimination of the risk would require explantation of the implant and is equivalent to risk 4. Furthermore, failed CI electronics could be detected intraoperatively and the CI replaced directly.
Risk 2 is relatively easy to eliminate as the permanent magnet of the CI electronics can be removed without explantation of the implant.
Given the lack of clear evidence for CI damage because of TCES (12) or MoEC (7) compared to the risks of explantation and reimplantation of the CI, we, together with the parents, decided to perform the corrective spinal surgery using both TCES for MEP monitoring and MoEC.
Anesthesia was performed using a ketamine-based intravenous anesthesia according to the case report of Erb et al. (15). The monitoring procedure is illustrated in Fig. 2. CI impedance and neural response telemetry (NRT) were measured, then, the externally worn speech processor was removed and the magnet of the CI receiver electronics was explanted. For MEP monitoring, the TCES technique of Andersson and Ohlin (16) was used. TCES consisting of a train of six pulses (maximum amplitude 430 V, duration 100 μs, interstimulus interval 1 ms) was applied with a Digitimer D185 stimulator (Digitimer Ltd., Welwyn Garden City, England) triggered by a Viking IV D monitoring system (Nicolet Biomedical Inc., Madison, WI). The stimulating anode electrode was placed at the Cz position and a ring of four cathode electrodes centered around the anode was placed with each electrode approximately 6–8 cm apart (17) but with the left cathode electrodes more medial – see Fig. 2. A spatial facilitating stimulus train of 10 pulses (amplitude 23.5 mA, pulse duration 0.5 ms, interpulse interval 2 ms) was applied to the medial border of the plantar arch of the foot (cathode being proximal to the anode) 60 ms before onset of TCES pulse train (see Fig. 3). The conditioning stimulus was adjusted up to 100 mA to give maximal facilitation without evoking a reflex response by itself. This facilitation technique permitted a reduction in the electrical pulse width and voltage required to elicit TCES evoked MEPs (13,16). SSEPs were measured as described in Fig. 2.
MEPs and SSEPs were registered in supine position before and after application of the Gardner-Wells tongue and the transfemoral apex pins for the intraoperative skull–femoral traction. MEPs and SSEPs were checked again after positioning the patient prone on the spine table (see Fig. 3). During surgery, the use of TCES as well as MoEC applied at the second thoracic (T2) to the forth lumbar (L4) vertebra were reduced to the minimum possible, 300 V and 130 W, respectively. All cautery was at or below the level of the clavicles (T2-T3). Throughout the corrective posterior instrumented spinal fusion from T2 to L4, MEPs and SSEPs remained stable (Fig. 3). A new permanent magnet was inserted into the CL receiver magnet pocket, the speech processor refitted, and the electrode impedance and NRT thresholds were remeasured (see Fig. 1). No changes in impedances or NRT thresholds were noted (Fig. 1). Impedance values between 6 and 10 kΩ as in Fig. 1 indicate normal CI electronic function and that no electrode is open circuited or shorted. NRT responses indicated that the acoustic nerve was responding to electrical stimulation, and also that the CI electronics were intact and hearing-related nerve tissue was not damaged.
The orthopedic outcome at 18 months showed a good clinical condition, a relevant reduction in pain, gait preservation and radiologically stable position of all metal implants, maintenance of the achieved spinal correction, and good coronal and sagittal balance. Otherwise, the patient was generally more cooperative during daily activities according to caregivers (possibly related to pain reduction). Concerning the CI outcome the patient could discriminate between different loudness levels postoperatively. This permitted use of other CI stimulation strategies (monopolar ACE) leading to decreased facial nerve stimulation as well as an improved mean threshold of 45 dB hearing level between 125 Hz and 8 kHz 18 months after the surgery compared to 65 dB preoperatively.
The use of spinal cord monitoring (MEPs) with TCES and MoEC at T2 are considered high-risk procedures by CI manufacturers (5,6). In our case, the patient underwent corrective spinal fusion surgery for a severe congenital deformity of the spine. The use of TCES to evoke MEPs is mandatory to monitor for any spinal cord damage. MoEC helps in reducing blood loss in these complex surgical procedures.
In contrast to the assumed risks, the unchanged CI impedances in our patient indicated a lack of tissue damage in the cochlea. The unchanged NRT thresholds indicated that auditory nerve stimulation and CI electronics remained intact. The finding of unchanged functionality of the CI (at least 18 months postsurgery) is indicative of a need for further studies exploring the dispersion of current during TCES and MoEC for improved risk assessment for CI patients.
A study by Edwards et al. (18) revealed very little current spread outside of the stimulation area when mounting the electrodes for TCES in a similar way to ours. Similar studies could help in setting limits to the maximum TCES amplitudes to be used, and assess the spread of current outside of the stimulation area, for example, by using different stimulation electrode configurations (18,19). In our case, the voltage level of TCES could be held as low as 305 V because of MEP amplitude enhancement with conditioning stimuli applied at the foot (16) in comparison to the 300–600 V used in the only other report of TCES stimulation during corrective spinal surgery in a CI patient (8). We demonstrated no changes in CI impedances and NRT thresholds immediately postoperatively compared to preoperatively, and therefore assumed, as other authors (20), that there was no change in CI function. It is possible that we may have missed minor changes, an adult patient might have noticed, because of the development status of our patient.
Jeyakumar et al. (7) demonstrated that the concern regarding the use of MoEC in patients with CIs is unfounded for the 50 W levels they used when applied for 30 min at the tongue and abdomen of cadavers implanted with CIs. Post MoEC there was no change in CI function. They indicated that there is a tendency to extrapolate negative effects to CIs based on detrimental effects to cardiac pacemakers. Even when MoEC was used during adenotonsillectomy on two children at levels of 36 W there was no effect on the patient's CI (20). Other authors have noted similar results during dental surgery (21). In this report, we have been able to demonstrate that both TCES of 300 V and MoEC of 130 W at the clavicles had no adverse effects on CI function postoperatively.
The crucial question is whether the reason we observed no effects of TCES and MoEC on the patient's CI was because the voltages induced between the active and ground CI electrodes by MoEC at T2 or TCES at Cz did not exceed the voltage levels the protective diodes and circuitry of the CI can withstand. An examination of the electrical amplitudes we used, the distance to the CI from the stimulation sites (see Fig. 2) and the induced voltages permitted in the CI provide some insights answering this question. CIs are designed to protect against electrostatic discharge and are typically tested to withstand induced voltages of ±10 V between the active and ground CI electrodes to show compliance with CI standard ISO14708–7 with respect to diathermy. Crucially, this testing does not determine the maximum voltage levels that could be withstood. A recent study by Jeyakumar et al. (22) using MED-EL devices does help to provide a partial answer to our question. They applied MoEC to the pectoralis muscle (at the same level as T2 we used) and on the temporalis muscle (closer to the implant than our Cz stimulation site), and found that voltages across the CI electrodes reached 22 V with 50 W of MoEC and that this voltage did not change when MoEC was increased to 100 W at both stimulation sites. That is the CI protection for electrostatic discharge worked also to protect against 100 W of MoEC. Based on a typical surface EEG electrode resistance of 1 kOhm and the TCES voltage of 300 V, we estimate that we applied 90 W at a distance of approximately 10 cm from the CI. This level is lower than the 100 W Jeyakumar et al. (22) applied closer to the implant. The level we applied at T2, 130 W was also lower when the greater distance to the CI (25 cm) is taken into account. Based on the considerations above, it would seem crucial for future studies that the voltages induced between CI active electrodes and the reference ground electrode by MoEC at T2 and TCES at Cz are compared with levels CI implants can withstand. Possibly CI manufacturers would then be in a position to revise their recommendations regarding the use of MoEC and TCES.
We thank the implant development teams of the CI manufacturers MED-EL and Cochlear Ltd. for their helpful advice. We also thank Ms. Barbara Wenger for editorial assistance.
1. Bhagat S, Durst A, Grover H, et al. An evaluation of multimodal spinal cord monitoring in scoliosis surgery: a single centre experience of 354 operations. Eur Spine J
2. Jahangiri FR, Sayegh SA, Azzubi M, et al. Benefit of intraoperative neurophysiological monitoring in a pediatric patient with spinal dysmorphism, split cord malformation, and scoliosis. Neurodiagn J
3. Schwartz D, Sestokas A, Dormans J, et al. Transcranial electric motor evoked potential monitoring during spine surgery: is it safe? Spine
4. Legatt A. Current practice of motor evoked potential monitoring: results of a survey. J Clin Neurophysiol
7. Jeyakumar A, Wilson M, Sorrel J, et al. Monopolar cautery and adverse effects on cochlear implants. JAMA otolaryngology - head & neck surgery
8. Yellin J, Wiggins C, Franco A, Sankar W. Safe transcranial electric stimulation
motor evoked potential monitoring during posterior spinal fusion in two patients with cochlear implants. J Clin Monit Comput
9. Woods LA, Haller RJ, Hansen PD, Fukumoto DE, Herman RM. Decreased incidence of scoliosis in hearing-impaired children. Implications for a neurologic basis for idiopathic scoliosis. Spine
10. Doyle C, Blake K. Scoliosis in CHARGE: a prospective survey and two case reports. Am J Med Genet A
11. Stieger C, Bodmer D, Brand Y. Conservative and surgical rehabilitation of hearing loss. Ther Umsch
12. Driver S, Jiang D. Paediatric cochlear implantation factors that affect outcomes. Eur J Paediatr Neurol
13. McClure PK, Kilinc E, Oishi S, Riccio AI, Karol LA. Mobius syndrome: a 35-year single institution experience. J Pediatr Orthop
14. Hassepass F, Stabenau V, Arndt S, et al. Magnet dislocation: an increasing and serious complication following MRI in patients with cochlear implants. RoFo
15. Erb TO, Ryhult SE, Dutimann E, Hasler C, Luetschg J, Frei FJ. Improvement of motor-evoked potentials by ketamine and spatial facilitation during spinal surgery in a young child. Anesth analg
16. Andersson G, Ohlin A. Spatial facilitation of motor evoked responses in monitoring during spinal surgery. Clin Neurophysiol
17. Rossini PM, Marciani MG, Caramia M, Roma V, Zarola F. Nervous propagation along ’central’ motor pathways in intact man: characteristics of motor responses to ’bifocal’ and ’unifocal’ spine and scalp non-invasive stimulation
18. Edwards D, Cortes M, Datta A, Minhas P, Wassermann EM, Bikson M. Physiological and modeling evidence for focal transcranial electrical brain stimulation in humans: a basis for high-definition tDCS. Neuroimage
19. Bortoletto M, Rodella C, Salvador R, Miranda PC, Miniussi C. Reduced current spread by concentric electrodes in transcranial electrical stimulation (tES). Brain Stimul
20. Tien DA, Woodson EA, Anne S. Safety of monopolar electrocautery in patients with cochlear implants. Ann Otol Rhinol Laryngol
21. Roberts S, West LA, Liewehr FR, Rueggeberg FA, Sharpe DE, Potter BJ. Impact of dental devices on cochlear implants. J Endod
22. Jeyakumar A, Jackson NM, Givens VB, Brickman TM, Arriaga MA. Does electrocautery damage cochlear implants? Otol Neurotol
Keywords:Copyright © 2019 by Otology & Neurotology, Inc. Image copyright © 2010 Wolters Kluwer Health/Anatomical Chart Company
Cochlear implant; Intraoperative neurophysiologic monitoring; Monopolar electric cauterization; Scoliosis corrective surgery; Transcranial electric stimulation