ORIGINAL STUDY
Impedance Testing on Cochlear Implants After Electroconvulsive Therapy Theodore R. McRackan, MD,* Alejandro Rivas, MD,* Andrea Hedley-Williams, AuD,* Vidya Raj, MB, ChB,† Mary S. Dietrich, MS, PhD,‡ Nathaniel K. Clark, MD,† and Robert F. Labadie, MD, PhD*
Objective: Cochlear implants (CI) are neural prostheses that restore hearing to individuals with profound sensorineural hearing loss. The surgically implanted component consists of an electrode array, which is threaded into the cochlea, and an electronic processor, which is buried under the skin behind the ear. The Food and Drug Administration and CI manufacturers contend that electroconvulsive therapy (ECT) is contraindicated in CI recipients owing to risk of damage to the implant and/or the patient. We hypothesized that ECT does no electrical damage to CIs. Methods: Ten functional CIs were implanted in 5 fresh cadaveric human heads. Each head then received a consecutive series of 12 unilateral ECT sessions applying maximum full pulse-width energy settings. Electroconvulsive therapy was delivered contralaterally to 5 CIs and ipsilaterally to 5 CIs. Electrical integrity testing (impedance testing) of the electrode array was performed before and after CI insertion, and after the first, third, fifth, seventh, ninth, and 12th ECT sessions. Electroconvulsive therapy was performed by a staff psychiatrist experienced with the technique. Explanted CIs were sent back to the manufacturer for further integrity testing. Results: No electrical damage was identified during impedance testing. Overall, there were statistically significant decreases in impedances (consistent with no electrical damage) when comparing pre-ECT impedance values to those after 12 sessions. There was no statistically significant difference (P > 0.05) in impedance values comparing ipsilateral to contralateral ECT. Manufacturer testing revealed no other electrical damage to the CIs. Conclusion: Electroconvulsive therapy does not seem to cause any detectable electrical injury to CIs. Key Words: cochlear implant, electroconvulsive therapy, impedance testing (J ECT 2014;30: 303–308)
A
pproved by the Food and Drug Administration (FDA) in 1984 for adults and in 1990 for children, cochlear implants (CIs) are neural prosthetic devices that convert acoustic signals to electrical energy, which is used directly to stimulate the auditory nerve, allowing deaf individuals the ability to perceive sound. These devices consist of 2 parts: a surgically implanted component and an externally worn process (Fig. 1). The surgically implanted component has an electrode array that has a series of individual electrical contacts (FDA-cleared devices typically have 16–22 contacts), which, when activated, From the *Department of Otolaryngology-Head Neck Surgery, †Department of Psychiatry and ‡Department of Biostatistics, Schools of Medicine and Nursing, Vanderbilt University, Nashville, TN. Received for publication October 22, 2013; accepted February 26, 2014. Reprints: Robert Labadie MD, PhD, Department of Otolaryngology, Vanderbilt University, 7209 Medical Center East-South Tower, 1215 21st Ave S, Nashville, TN 37232-8605 (e‐mail: [email protected]). Theodore R. McRackan and Alejandro Rivas contributed equally to this work. The authors have no conflicts of interest or financial disclosures to report. Copyright © 2014 by Lippincott Williams & Wilkins DOI: 10.1097/YCT.0000000000000124
Journal of ECT • Volume 30, Number 4, December 2014
stimulate discrete areas in the cochlea. Electrodes closer to the entrance of the cochlea stimulate neurons responsible for higher-frequency sounds and those placed deeper in the cochlea stimulate neurons responsible for lower-frequency sounds. Controlling the activation of the electrode array is a subcutaneously placed internal stimulator composed primarily of an integrated circuit, which processes the signal received from the externally worn processor. To transmit the signal across the skin, magnets within the internal stimulator and externally worn processor align a radiofrequency (RF) transmitter in the external processor with a corresponding RF receiver within the internal stimulator. In practice, sound is detected via the external worn process, processed (eg, filtered), transmitted via RF to the internal stimulator, which controls stimulation of individual contact within the electrode array, after which sound is perceived. For most individuals, rehabilitation with CIs allows the restoration of functional hearing. As of a December 2010 FDA report, 219,000 individuals have received CIs. Another report puts the total at more than 320,000.1 Double-digit annual growth is predicted for the foreseeable future.2 Electroconvulsive therapy (ECT) is a well-established nonpharmacological treatment modality that is indicated for several major psychiatric illnesses including major depressive disorder, bipolar disorder, catatonia, and certain subtypes of schizophrenia. It is estimated that 100,000 people a year receive ECT.3 Whereas public perception has oscillated, ECT is a safe procedure with a low complication rate ( 0.05).
Cochlear Implant Testing Each CI device underwent initial impedance testing (Cochlear Americas Custom Sound EP 3.2) before cadaveric insertion. The implants were tested again after insertion (before ECT) and then again after the first, third, fifth, seventh, ninth, and twelfth sessions. Impedance values were recorded for common ground (CG: each electrode referenced to the remaining 21 intracochlear electrodes), monopolar 1 (MP1: each electrode referenced to the external ground electrode), monopolar 2 (MP2: each electrode referenced to the plate electrode on the internal receiver/ stimulator), and monopolar 1 + 2 (MP1+2: each electrode referenced between MP1 and MP2). The implants were then sent back to Cochlear Corporation for internal testing, where evaluation of each device was performed, looking for impact failure, hermeticity failure, electronic failure, electrode array malfunction, or other nonspecific types of device failures.
MP1 Impedance Testing In the MP1 impedance testing, the mean (SD) pre-ECT value for all CIs was 24.13 (6.47) kΩ. This value decreased on average by 7.44 kΩ to 16.69 (4.87) kΩ after 12 ECT sessions (P = 0.001). This difference became statistically significant after the first ECT session (P = 0.035). The implants exposed to ipsilateral ECT had a mean (SD) impedance of 21.50 (5.92) kΩ, which fell by 7.20 kΩ to 14.03 (4.39) kΩ after 12
TABLE 1. Mean CG Impedance Values for Each CI
Implant
Statistical Analysis All statistical summaries and analyses were conducted using SPSS version 18 (IBM Corporation). Distributions of impedance values were summarized using mean and standard deviation. Overall and differential patterns of changes in impedance values within each of the 4 study conditions (CG, MP1, MP2, and MP1+2) were tested using multilevel generalized estimating equations linear modeling incorporating Bonferroniadjusted P values for post hoc pairwise tests between each time of assessment. This approach adjusts the standard errors for the correlated repeated assessments of impedance. Other than the Bonferroni-adjusted post hoc tests, a 2-tailed critical alpha of 0.05 was used for determining statistical significance.
RESULTS Before CI implantation, the first cadaver head was tested by administering ECT. This confirmed that the static and © 2014 Lippincott Williams & Wilkins
ECT Laterality
Initial Impedance (Ω)
Post-ECT Change in Impedance Impedance (Ω) (Ω)
1 Ipsilateral 15.26 3.54 2 Ipsilateral 11.02 7.79 3 Ipsilateral 12.50 12.49 4 Ipsilateral 20.57 13.08 5 Ipsilateral 51.21 13.73 6 Contralateral 8.66 7.67 7 Contralateral 10.14 7.02 8 Contralateral 12.72 11.52 9 Contralateral 25.54 10.95 10 Contralateral 30.34 17.96 Ipsilateral, mean (SD), Ω 22.11 (16.67) 10.12 (4.36) Contralateral, mean (SD), Ω 17.50 (9.80) 11.03 (4.35) All CIs, mean (SD), Ω 19.80 (13.12) 10.58 (4.14)
−11.72 −3.23 −0.01 −7.49 −37.48 −1.00 −3.12 −1.20 −14.59 −12.37 −12.00 −6.45 −9.22
Impedance values listed represent mean of the 22 electrodes for each implant.
www.ectjournal.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
305
Journal of ECT • Volume 30, Number 4, December 2014
McRackan et al
FIGURE 3. Mean CG impedance values for each CI after each post-ECT interval tested. Impedance values listed represent mean of the 22 electrodes for each implant. Note that the decrease in impedance becomes statistically significant after the third ECT session. Implants 1 to 5 were exposed to ipsilateral ECT, implants 6 to 10 were exposed to contralateral ECT.
ECT sessions. Similarly, for implants exposed to contralateral ECT, the impedance fell by 7.68 kΩ from 26.76 (6.48) to 19.08 (5.05) kΩ after 12 sessions of ECT (Table 2; Fig. 4). The differences between the 2 groups were not statistically significant (all P > 0.05).
MP2 Impedance Testing For MP2 impedance testing, similar results were observed. The mean (SD) initial pre-ECT impedance value for all CIs was 21.01 (6.71) kΩ. After 12 ECT sessions, this value fell by 6.64 kΩ to 14.37 (3.90) kΩ; P = 0.003). The decrease in impedance became statistically significant after the third ECT session (P = 0.024). For the CIs that received ipsilateral ECT, the mean (SD) impedance values fell by 5.39 kΩ from 18.99 kΩ (5.11 Ω) to 13.60 kΩ (4.87 Ω). The CIs exposed to contralateral ECT fell by 7.89 kΩ from 23.03 (8.07) kΩ to 15.14 (3.05) kΩ
FIGURE 4. Mean MP1 impedance values for each CI after each post-ECT interval tested. Impedance values listed represent mean of the 22 electrodes for each implant. Note that the decrease in impedance becomes statistically significant after the first ECT session. Implants 1 to 5 were exposed to ipsilateral ECT, and implants 6 to 10 were exposed to contralateral ECT.
(Table 3; Fig. 5). There was no statistically significant difference between the 2 groups (all P > 0.05).
MP1+2 Impedance Testing Finally, in the MP1+2 testing, the mean (SD) initial preECT for all CIs was 21.35 (7.26) kΩ. The mean (SD) value fell by 7.03 kΩ to 14.32 (4.30) kΩ after 12 sessions of ECT (P = 0.002). This decrease became statistically significant after the third ECT session (P = 0.02). With regard to ipsilateral ECT, the CIs’ initial mean (SD) impedance value of 18.91 (5.13) kΩ fell by 5.85 kΩ to 13.05 (4.80) kΩ. For CIs exposed to contralateral ECT, the mean (SD) initial impedance value of 23.80 (8.80) kΩ fell by 8.21 kΩ to 15.59 (3.24) kΩ (Table 4; Fig. 6). Once again, there was no statistically significant difference between the ipsilateral and contralateral groups at any time point or overall (all P > 0.05).
Manufacturer Internal Electrical Testing All implants were then explanted and sent to the manufacturer for internal electrical testing. This examination revealed
TABLE 2. Mean MP1 Impedance Values for Each CI TABLE 3. Mean MP2 Impedance Values for Each CI
Implant
ECT Laterality
Initial Post-ECT Change in Impedance Impedance Impedance (Ω) (Ω) (Ω)
1 Ipsilateral 26.23 8.44 2 Ipsilateral 14.49 11.25 3 Ipsilateral 16.05 15.54 4 Ipsilateral 27.49 19.26 5 Ipsilateral 23.27 17.04 6 Contralateral 21.16 24.96 7 Contralateral 18.93 15.41 8 Contralateral 28.39 15.18 9 Contralateral 31.50 17.17 10 Contralateral 33.82 22.65 Ipsilateral, mean (SD), Ω 21.50 (5.92) 14.30 (4.40) Contralateral, mean (SD), Ω 26.76 (6.48) 19.08 (5.05) All CIs, mean (SD), Ω 24.13 (6.47) 16.70 (4.87)
−17.79 −3.24 −0.51 −8.23 −6.22 3.81 −3.52 −13.20 −14.33 −11.17 −7.20 −7.68 −7.44
Impedance values listed represent mean of the 22 electrodes for each implant.
306
www.ectjournal.com
Implant
ECT Laterality
Initial Post-ECT Change in Impedance Impedance Impedance (Ω) (Ω) (Ω)
1 Ipsilateral 17.34 6.72 2 Ipsilateral 13.81 10.73 3 Ipsilateral 15.21 14.96 4 Ipsilateral 25.76 18.90 5 Ipsilateral 22.83 16.69 6 Contralateral 14.41 13.93 7 Contralateral 14.24 11.16 8 Contralateral 27.39 14.64 9 Contralateral 31.06 16.94 10 Contralateral 28.04 19.05 Ipsilateral, mean (SD), Ω 18.99 (5.11) 13.60 (4.87) Contralateral, mean (SD), Ω 23.03 (8.07) 15.14 (3.05) All CIs, mean (SD), Ω 21.01 (6.71) 14.37 (3.90)
−10.62 −3.08 −0.25 −6.87 −6.14 −0.48 −3.09 −12.76 −14.12 −9.00 −5.39 −7.89 −7.44
Impedance values listed.
© 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Journal of ECT • Volume 30, Number 4, December 2014
FIGURE 5. Mean MP2 impedance values for each CI after each post-ECT interval tested. Impedance values listed represent mean of the 22 electrodes for each implant. Note that the decrease in impedance becomes statistically significant after the third ECT session. Implants 1 to 5 were exposed to ipsilateral ECT, and implants 6 to 10 were exposed to contralateral ECT.
no change in the implant hermeticity, electronic status, or electrode array function after ECT.
DISCUSSION ECT With Cochlear Implants It is well established that patients with hearing loss, whether aided or not, are at an increased risk for depression. A recent large multicenter trial determined that for every decibel of signal-to-noise ratio reduction of hearing status, the odds for developing moderate to severe depression increased by 5%.12 This, along with the increasing frequency and indications for CI implantation, suggests that overlap in patient populations receiving CIs and ECT will increase in the future. Although ECT is presently contraindicated in CI recipients, this recommendation has been made in the absence of supporting evidence. Using both a PubMed MeSH and an OVID keyword search for “Electroconvulsive Therapy” and “cochlear TABLE 4. Mean MP1+2 Impedance Values for Each CI
Implant
ECT Laterality
Initial Post-ECT Change in Impedance Impedance Impedance (Ω) (Ω) (Ω)
1 Ipsilateral 17.33 6.69 2 Ipsilateral 13.67 10.63 3 Ipsilateral 15.08 12.55 4 Ipsilateral 25.69 18.82 5 Ipsilateral 22.76 16.59 6 Contralateral 14.47 13.96 7 Contralateral 14.24 11.19 8 Contralateral 27.33 14.56 9 Contralateral 30.95 16.90 10 Contralateral 32.02 21.35 Ipsilateral, mean (SD), Ω 18.91 (5.11) 13.05 (4.87) Contralateral, mean (SD), Ω 23.80 (8.07) 15.59 (3.06) Average All CIs, mean 21.35 (6.71) 14.32 (3.90) (SD), Ω
−10.64 −3.04 −2.54 −6.88 −6.17 −0.51 −3.05 −12.77 −14.05 −10.67 −5.85 −8.21 −7.03
Impedance values listed represent mean of the 22 electrodes for each implant.
© 2014 Lippincott Williams & Wilkins
Impedance Testing on Cochlear Implants After ECT
FIGURE 6. Mean MP1+2 impedance values for each CI after each post-ECT interval tested. Impedance values listed represent mean of the 22 electrodes for each implant. Note that the decrease in impedance becomes statistically significant after the third ECT session. Implants 1 to 5 were exposed to ipsilateral ECT, and implants 6 to 10 were exposed to contralateral ECT.
implant”, we found only 2 articles addressing this issue. The first was a letter to the editor in the Journal of ECT describing a case where a psychiatrist considered performing ECT on a treatmentrefractory patient with major depressive disorder.13 Both the consulting otolaryngologist and the CI manufacturer, however, recommended against ECT owing to the possible risk of injury to the implant and the patient. The other article is a case report that we published describing the successful use of ECT in a 17-year old adolescent boy who was implanted with a Nucleus 22 device at age 3 years.11 The patient presented with delirious mania, for which ECT is the gold standard treatment. Maximal psychopharmacologic therapy was unsuccessful in treating the patient, and it was ultimately decided to attempt ECT. The patient successfully underwent 2 treatment sessions of contralateral unilateral ECT and was eventually discharged from the hospital after returning to his baseline mental state. After each treatment, the external processor was placed back on the patient, and speech perception scores remained unchanged, which correlates to no change in functional hearing. Integrity testing was completed by the CI manufacturer during the hospital admission and revealed a normally functioning device. Four months after ECT, the patient reported pain and intermittent shocking sensation at the site of the external processor requiring ultimate replacement of the implant. The night before reimplantation, the patient had his N22 internal receiver removed to allow MRI scanning. Unfortunately, since the Nucleus 22 did not have a removable magnet, the electrode array had to be cut to allow removal of the internal receiver complete with magnet before the MRI. Although it would be unusual for electric current from the ECT applied 4 months before symptoms to cause damage to the internal receiver, this possibility cannot be excluded. Interestingly, further testing of the CI by the manufacturer after explantation revealed no hermeticity failure, electronic failure, electrode array malfunction, or any other nonspecific type of device failure. It should also be noted that the Nucleus 22 device has the highest failure rate of any CI: at least 3% in adults9 and 9% in children.14,15
ECT With Deep Brain Stimulation To date, there have been 2 case reports of the successful use of ECT in patients with DBS.9,10 Deep brain stimulators are electrical pulse generators surgically placed in the subthalamic www.ectjournal.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
307
Journal of ECT • Volume 30, Number 4, December 2014
McRackan et al
nucleus to treat conditions such as intractable tremor and Parkinson disease. Structurally, DBS electrodes are very similar to CIs except that they contain fewer electrodes and have simpler stimulation algorithms. The implantable pulse generator in DBS is placed subcutaneously in the upper thorax as opposed to the internal stimulator of CIs, which is placed behind the ear. This may result in larger electrical currents induced in CIs versus DBSs.16 Another difference between DBS and CI surgery is that a craniotomy is made in DBS surgery, whereas this is not performed in CI surgery. This absence of bone in patients with DBS may decrease resistivity between the ECT-stimulating electrodes and thus increase current spread compared to CIs. In the case reports of ECT in patients with DBS, unilateral ECT was performed in a patient with a contralateral DBS,9 and bilateral ECT was carried out in a patient with a DBS.10 Both cases demonstrated no adverse effects to the patient with regard to the DBS. Both patients’ DBS devices continued to function normally with posttreatment imaging demonstrating no device migration.
CI Impedance and ECT In our study, only one implant under one condition showed evidence of increased impedance values after ECT. This implant (number 6) had an increased impedance of 3.08 kΩ under the MP1 condition. The 3 other conditions tested with this implant all showed decreased impedance values after ECT sessions. In the 39 (97.5%) other implant/conditions tested, there was evidence of decreased impedance values. All electrodes under all conditions showed a mean decreased impedance of 7.58 kΩ with no detectable shorts. Moreover, more stringent internal testing done by the CI manufacturer showed no evidence of electrical injury. Whereas it is interesting that ECT significantly decreased the impedances in a number of implants, the clinical significance of this is unknown. Even in experimental models devoid of CIs, decreases in tissue impedances are seen in animals subjected to ECT.17 In patients with CIs, decreased CI impedance values are seen immediately after stimulation via neural response telemetry in the operating room and by regular implant use. Although the exact reason for this remains unclear, it has been shown that CIs collect proteinaceous debris or casks upon and after insertion.18 After electrical stimulation, this debris is shed, which has been shown to diminish impedance. Perhaps the most important finding from this study is that there seems to be no evidence of electrical injury to CIs after ECT when electrical current is applied with the external processor not attached. This finding occurred although the CIs in this study were subjected to the maximal amount of electrical current used for clinical ECT in both contralateral and unilateral stimulation.
Study Limitations Ideally, histologic analysis of intracochlear tissue would have been performed. This may have revealed microscopic tissue changes secondary to ECT. Owing to our anatomical gift program standards, we were unable to send fresh cadaver specimens to experts for evaluation for tissue injury. Therefore, we cannot rule out tissue damage occurring when ECT is applied to patients with CI. Further studies in animal models may be of significant use to determine whether ECT causes thermal injury and is safe to use in patients with CI. However, if ECT is to be used in patients with CI, the external receiver should be removed and the stimulating ECT electrodes should be placed as far away from the implant as possible.
308
www.ectjournal.com
CONCLUSION Electroconvulsive therapy causes no detectable damage to CI electronic circuitry in a fresh human cadaver model. Electrical stimulation from ECT is associated with statistically significant decreases in impedance values in CIs, which is consistent with a lack of electrical damage to the CI circuits. Further studies are required to examine the safety and risks of ECT in patients with CI in clinical practice. REFERENCES 1. Altman LK. Lasker Awards Go to Five Scientists and Gateses. Available at: http://www.nytimes.com/2013/09/10/health/ lasker-awards-winners-2013.html?_r=0. New York Times. September 9, 2013. Accessed January 22, 2014. 2. Anonymous. The Global Hearing Implants Market is Forecast to Exceed $2 Billion in 2017. Available at: https://www.asdreports.com/news.asp? pr_id=191. ASD Reports. January 31, 2012. Accessed January 22, 2014. 3. Anonymous. Electroconvulsive therapy. Available at: http://www.nmha. org/ect. Accessed August 28, 2013. 4. Nutall GA, Bowersox MR, Douglass SB, et al. Morbidity and mortality in the use of electroconvulsive therapy. J ECT. 2004;20:237–241. 5. Anonymous. Warnings and Precautions. Available at: http://www. advancedbionics.com/content/dam/ab/Global/en_ce/documents/ libraries/Professional%20Library/AB%20Product%20Literature/ System_Indications_Precautions/Warnings_and_Precautions.pdf. Accessed March 28, 2014. 6. Anonymous. Warnings and Precautions. Available at: http://www. cochlear.com/wps/wcm/connect/us/for-professionals/cochlear-implants/ warnings-and-precautions. Accessed March 28, 2014. 7. Anonymous. Benefits and risks of Cochlear implants. Available at: http://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/ ImplantsandProsthetics/CochlearImplants/ucm062843.htm. Accessed March 28, 2014. 8. Hartmann SJ, Saldivia A. ECT in an elderly patient with skull defects and shrapnel. Convuls Ther. 1990;6:5–171. 9. Moscarillo FM, Annunziata CM. ECT in a patient with a deep brain-stimulating electrode in place. J ECT. 2000;16:287–90. 10. Bailine S, Kremen N, Kohen I, et al. Bitemporal electroconvulsive therapy for depression in a Parkinson disease patient with a deep brain stimulator. J ECT. 2008;24:171–2. 11. Labadie RF, Clark NK, Cobb CM, et al. Electroconvulsive therapy in a cochlear implant patient. Otol Neurotol. 2010;31:64–6. 12. Nachtegaal J, Smit JH, Smits C, et al. The association between hearing status and psychosocial health before the age of 70 years: results from an internet-based national survey on hearing. Ear Hear. 2009;30:302–12. 13. Malek-Ahmadi P, Hanretta AT. Cochlear implant and ECT. J ECT. 2003;19:51. 14. Battmer RD, Linz B, Lenarz T. A review of device failure in more than 23 years of clinic experience of a cochlear implant program with more than 3,400 implantees. Otol Neurotol. 2009;3455–463. 15. Parisier SC, Chute PM, Popp AL. Cochlear implant mechanical failures. Am J Otol. 1996;17:730–4. 16. Deng ZD, Lisanby SH, Peterchev AV. Transcranial magnetic stimulation in the presence of deep brain stimulation implants: induced electrode currents. IEEE Eng Med Biol Soc Conf. 2010;6821–6824. 17. Russell RW, Pierce JF, Townsend JC. Characteristics of tissue impedance in the rat under conditions of electroconvulsive shock stimulation. Am J Physiol. 1949;156:317–21. 18. Tykocinski M, Cohen LT, Cowan RS. Measurement and analysis of access resistance and polarization impedance in implant recipients. Otol Neurotol. 2005;26:948–956.
© 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Impedance Testing on Cochlear Implants After Electroconvulsive Therapy Theodore R. McRackan, MD,* Alejandro Rivas, MD,* Andrea Hedley-Williams, AuD,* Vidya Raj, MB, ChB,† Mary S. Dietrich, MS, PhD,‡ Nathaniel K. Clark, MD,† and Robert F. Labadie, MD, PhD*
Objective: Cochlear implants (CI) are neural prostheses that restore hearing to individuals with profound sensorineural hearing loss. The surgically implanted component consists of an electrode array, which is threaded into the cochlea, and an electronic processor, which is buried under the skin behind the ear. The Food and Drug Administration and CI manufacturers contend that electroconvulsive therapy (ECT) is contraindicated in CI recipients owing to risk of damage to the implant and/or the patient. We hypothesized that ECT does no electrical damage to CIs. Methods: Ten functional CIs were implanted in 5 fresh cadaveric human heads. Each head then received a consecutive series of 12 unilateral ECT sessions applying maximum full pulse-width energy settings. Electroconvulsive therapy was delivered contralaterally to 5 CIs and ipsilaterally to 5 CIs. Electrical integrity testing (impedance testing) of the electrode array was performed before and after CI insertion, and after the first, third, fifth, seventh, ninth, and 12th ECT sessions. Electroconvulsive therapy was performed by a staff psychiatrist experienced with the technique. Explanted CIs were sent back to the manufacturer for further integrity testing. Results: No electrical damage was identified during impedance testing. Overall, there were statistically significant decreases in impedances (consistent with no electrical damage) when comparing pre-ECT impedance values to those after 12 sessions. There was no statistically significant difference (P > 0.05) in impedance values comparing ipsilateral to contralateral ECT. Manufacturer testing revealed no other electrical damage to the CIs. Conclusion: Electroconvulsive therapy does not seem to cause any detectable electrical injury to CIs. Key Words: cochlear implant, electroconvulsive therapy, impedance testing (J ECT 2014;30: 303–308)
A
pproved by the Food and Drug Administration (FDA) in 1984 for adults and in 1990 for children, cochlear implants (CIs) are neural prosthetic devices that convert acoustic signals to electrical energy, which is used directly to stimulate the auditory nerve, allowing deaf individuals the ability to perceive sound. These devices consist of 2 parts: a surgically implanted component and an externally worn process (Fig. 1). The surgically implanted component has an electrode array that has a series of individual electrical contacts (FDA-cleared devices typically have 16–22 contacts), which, when activated, From the *Department of Otolaryngology-Head Neck Surgery, †Department of Psychiatry and ‡Department of Biostatistics, Schools of Medicine and Nursing, Vanderbilt University, Nashville, TN. Received for publication October 22, 2013; accepted February 26, 2014. Reprints: Robert Labadie MD, PhD, Department of Otolaryngology, Vanderbilt University, 7209 Medical Center East-South Tower, 1215 21st Ave S, Nashville, TN 37232-8605 (e‐mail: [email protected]). Theodore R. McRackan and Alejandro Rivas contributed equally to this work. The authors have no conflicts of interest or financial disclosures to report. Copyright © 2014 by Lippincott Williams & Wilkins DOI: 10.1097/YCT.0000000000000124
Journal of ECT • Volume 30, Number 4, December 2014
stimulate discrete areas in the cochlea. Electrodes closer to the entrance of the cochlea stimulate neurons responsible for higher-frequency sounds and those placed deeper in the cochlea stimulate neurons responsible for lower-frequency sounds. Controlling the activation of the electrode array is a subcutaneously placed internal stimulator composed primarily of an integrated circuit, which processes the signal received from the externally worn processor. To transmit the signal across the skin, magnets within the internal stimulator and externally worn processor align a radiofrequency (RF) transmitter in the external processor with a corresponding RF receiver within the internal stimulator. In practice, sound is detected via the external worn process, processed (eg, filtered), transmitted via RF to the internal stimulator, which controls stimulation of individual contact within the electrode array, after which sound is perceived. For most individuals, rehabilitation with CIs allows the restoration of functional hearing. As of a December 2010 FDA report, 219,000 individuals have received CIs. Another report puts the total at more than 320,000.1 Double-digit annual growth is predicted for the foreseeable future.2 Electroconvulsive therapy (ECT) is a well-established nonpharmacological treatment modality that is indicated for several major psychiatric illnesses including major depressive disorder, bipolar disorder, catatonia, and certain subtypes of schizophrenia. It is estimated that 100,000 people a year receive ECT.3 Whereas public perception has oscillated, ECT is a safe procedure with a low complication rate ( 0.05).
Cochlear Implant Testing Each CI device underwent initial impedance testing (Cochlear Americas Custom Sound EP 3.2) before cadaveric insertion. The implants were tested again after insertion (before ECT) and then again after the first, third, fifth, seventh, ninth, and twelfth sessions. Impedance values were recorded for common ground (CG: each electrode referenced to the remaining 21 intracochlear electrodes), monopolar 1 (MP1: each electrode referenced to the external ground electrode), monopolar 2 (MP2: each electrode referenced to the plate electrode on the internal receiver/ stimulator), and monopolar 1 + 2 (MP1+2: each electrode referenced between MP1 and MP2). The implants were then sent back to Cochlear Corporation for internal testing, where evaluation of each device was performed, looking for impact failure, hermeticity failure, electronic failure, electrode array malfunction, or other nonspecific types of device failures.
MP1 Impedance Testing In the MP1 impedance testing, the mean (SD) pre-ECT value for all CIs was 24.13 (6.47) kΩ. This value decreased on average by 7.44 kΩ to 16.69 (4.87) kΩ after 12 ECT sessions (P = 0.001). This difference became statistically significant after the first ECT session (P = 0.035). The implants exposed to ipsilateral ECT had a mean (SD) impedance of 21.50 (5.92) kΩ, which fell by 7.20 kΩ to 14.03 (4.39) kΩ after 12
TABLE 1. Mean CG Impedance Values for Each CI
Implant
Statistical Analysis All statistical summaries and analyses were conducted using SPSS version 18 (IBM Corporation). Distributions of impedance values were summarized using mean and standard deviation. Overall and differential patterns of changes in impedance values within each of the 4 study conditions (CG, MP1, MP2, and MP1+2) were tested using multilevel generalized estimating equations linear modeling incorporating Bonferroniadjusted P values for post hoc pairwise tests between each time of assessment. This approach adjusts the standard errors for the correlated repeated assessments of impedance. Other than the Bonferroni-adjusted post hoc tests, a 2-tailed critical alpha of 0.05 was used for determining statistical significance.
RESULTS Before CI implantation, the first cadaver head was tested by administering ECT. This confirmed that the static and © 2014 Lippincott Williams & Wilkins
ECT Laterality
Initial Impedance (Ω)
Post-ECT Change in Impedance Impedance (Ω) (Ω)
1 Ipsilateral 15.26 3.54 2 Ipsilateral 11.02 7.79 3 Ipsilateral 12.50 12.49 4 Ipsilateral 20.57 13.08 5 Ipsilateral 51.21 13.73 6 Contralateral 8.66 7.67 7 Contralateral 10.14 7.02 8 Contralateral 12.72 11.52 9 Contralateral 25.54 10.95 10 Contralateral 30.34 17.96 Ipsilateral, mean (SD), Ω 22.11 (16.67) 10.12 (4.36) Contralateral, mean (SD), Ω 17.50 (9.80) 11.03 (4.35) All CIs, mean (SD), Ω 19.80 (13.12) 10.58 (4.14)
−11.72 −3.23 −0.01 −7.49 −37.48 −1.00 −3.12 −1.20 −14.59 −12.37 −12.00 −6.45 −9.22
Impedance values listed represent mean of the 22 electrodes for each implant.
www.ectjournal.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
305
Journal of ECT • Volume 30, Number 4, December 2014
McRackan et al
FIGURE 3. Mean CG impedance values for each CI after each post-ECT interval tested. Impedance values listed represent mean of the 22 electrodes for each implant. Note that the decrease in impedance becomes statistically significant after the third ECT session. Implants 1 to 5 were exposed to ipsilateral ECT, implants 6 to 10 were exposed to contralateral ECT.
ECT sessions. Similarly, for implants exposed to contralateral ECT, the impedance fell by 7.68 kΩ from 26.76 (6.48) to 19.08 (5.05) kΩ after 12 sessions of ECT (Table 2; Fig. 4). The differences between the 2 groups were not statistically significant (all P > 0.05).
MP2 Impedance Testing For MP2 impedance testing, similar results were observed. The mean (SD) initial pre-ECT impedance value for all CIs was 21.01 (6.71) kΩ. After 12 ECT sessions, this value fell by 6.64 kΩ to 14.37 (3.90) kΩ; P = 0.003). The decrease in impedance became statistically significant after the third ECT session (P = 0.024). For the CIs that received ipsilateral ECT, the mean (SD) impedance values fell by 5.39 kΩ from 18.99 kΩ (5.11 Ω) to 13.60 kΩ (4.87 Ω). The CIs exposed to contralateral ECT fell by 7.89 kΩ from 23.03 (8.07) kΩ to 15.14 (3.05) kΩ
FIGURE 4. Mean MP1 impedance values for each CI after each post-ECT interval tested. Impedance values listed represent mean of the 22 electrodes for each implant. Note that the decrease in impedance becomes statistically significant after the first ECT session. Implants 1 to 5 were exposed to ipsilateral ECT, and implants 6 to 10 were exposed to contralateral ECT.
(Table 3; Fig. 5). There was no statistically significant difference between the 2 groups (all P > 0.05).
MP1+2 Impedance Testing Finally, in the MP1+2 testing, the mean (SD) initial preECT for all CIs was 21.35 (7.26) kΩ. The mean (SD) value fell by 7.03 kΩ to 14.32 (4.30) kΩ after 12 sessions of ECT (P = 0.002). This decrease became statistically significant after the third ECT session (P = 0.02). With regard to ipsilateral ECT, the CIs’ initial mean (SD) impedance value of 18.91 (5.13) kΩ fell by 5.85 kΩ to 13.05 (4.80) kΩ. For CIs exposed to contralateral ECT, the mean (SD) initial impedance value of 23.80 (8.80) kΩ fell by 8.21 kΩ to 15.59 (3.24) kΩ (Table 4; Fig. 6). Once again, there was no statistically significant difference between the ipsilateral and contralateral groups at any time point or overall (all P > 0.05).
Manufacturer Internal Electrical Testing All implants were then explanted and sent to the manufacturer for internal electrical testing. This examination revealed
TABLE 2. Mean MP1 Impedance Values for Each CI TABLE 3. Mean MP2 Impedance Values for Each CI
Implant
ECT Laterality
Initial Post-ECT Change in Impedance Impedance Impedance (Ω) (Ω) (Ω)
1 Ipsilateral 26.23 8.44 2 Ipsilateral 14.49 11.25 3 Ipsilateral 16.05 15.54 4 Ipsilateral 27.49 19.26 5 Ipsilateral 23.27 17.04 6 Contralateral 21.16 24.96 7 Contralateral 18.93 15.41 8 Contralateral 28.39 15.18 9 Contralateral 31.50 17.17 10 Contralateral 33.82 22.65 Ipsilateral, mean (SD), Ω 21.50 (5.92) 14.30 (4.40) Contralateral, mean (SD), Ω 26.76 (6.48) 19.08 (5.05) All CIs, mean (SD), Ω 24.13 (6.47) 16.70 (4.87)
−17.79 −3.24 −0.51 −8.23 −6.22 3.81 −3.52 −13.20 −14.33 −11.17 −7.20 −7.68 −7.44
Impedance values listed represent mean of the 22 electrodes for each implant.
306
www.ectjournal.com
Implant
ECT Laterality
Initial Post-ECT Change in Impedance Impedance Impedance (Ω) (Ω) (Ω)
1 Ipsilateral 17.34 6.72 2 Ipsilateral 13.81 10.73 3 Ipsilateral 15.21 14.96 4 Ipsilateral 25.76 18.90 5 Ipsilateral 22.83 16.69 6 Contralateral 14.41 13.93 7 Contralateral 14.24 11.16 8 Contralateral 27.39 14.64 9 Contralateral 31.06 16.94 10 Contralateral 28.04 19.05 Ipsilateral, mean (SD), Ω 18.99 (5.11) 13.60 (4.87) Contralateral, mean (SD), Ω 23.03 (8.07) 15.14 (3.05) All CIs, mean (SD), Ω 21.01 (6.71) 14.37 (3.90)
−10.62 −3.08 −0.25 −6.87 −6.14 −0.48 −3.09 −12.76 −14.12 −9.00 −5.39 −7.89 −7.44
Impedance values listed.
© 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Journal of ECT • Volume 30, Number 4, December 2014
FIGURE 5. Mean MP2 impedance values for each CI after each post-ECT interval tested. Impedance values listed represent mean of the 22 electrodes for each implant. Note that the decrease in impedance becomes statistically significant after the third ECT session. Implants 1 to 5 were exposed to ipsilateral ECT, and implants 6 to 10 were exposed to contralateral ECT.
no change in the implant hermeticity, electronic status, or electrode array function after ECT.
DISCUSSION ECT With Cochlear Implants It is well established that patients with hearing loss, whether aided or not, are at an increased risk for depression. A recent large multicenter trial determined that for every decibel of signal-to-noise ratio reduction of hearing status, the odds for developing moderate to severe depression increased by 5%.12 This, along with the increasing frequency and indications for CI implantation, suggests that overlap in patient populations receiving CIs and ECT will increase in the future. Although ECT is presently contraindicated in CI recipients, this recommendation has been made in the absence of supporting evidence. Using both a PubMed MeSH and an OVID keyword search for “Electroconvulsive Therapy” and “cochlear TABLE 4. Mean MP1+2 Impedance Values for Each CI
Implant
ECT Laterality
Initial Post-ECT Change in Impedance Impedance Impedance (Ω) (Ω) (Ω)
1 Ipsilateral 17.33 6.69 2 Ipsilateral 13.67 10.63 3 Ipsilateral 15.08 12.55 4 Ipsilateral 25.69 18.82 5 Ipsilateral 22.76 16.59 6 Contralateral 14.47 13.96 7 Contralateral 14.24 11.19 8 Contralateral 27.33 14.56 9 Contralateral 30.95 16.90 10 Contralateral 32.02 21.35 Ipsilateral, mean (SD), Ω 18.91 (5.11) 13.05 (4.87) Contralateral, mean (SD), Ω 23.80 (8.07) 15.59 (3.06) Average All CIs, mean 21.35 (6.71) 14.32 (3.90) (SD), Ω
−10.64 −3.04 −2.54 −6.88 −6.17 −0.51 −3.05 −12.77 −14.05 −10.67 −5.85 −8.21 −7.03
Impedance values listed represent mean of the 22 electrodes for each implant.
© 2014 Lippincott Williams & Wilkins
Impedance Testing on Cochlear Implants After ECT
FIGURE 6. Mean MP1+2 impedance values for each CI after each post-ECT interval tested. Impedance values listed represent mean of the 22 electrodes for each implant. Note that the decrease in impedance becomes statistically significant after the third ECT session. Implants 1 to 5 were exposed to ipsilateral ECT, and implants 6 to 10 were exposed to contralateral ECT.
implant”, we found only 2 articles addressing this issue. The first was a letter to the editor in the Journal of ECT describing a case where a psychiatrist considered performing ECT on a treatmentrefractory patient with major depressive disorder.13 Both the consulting otolaryngologist and the CI manufacturer, however, recommended against ECT owing to the possible risk of injury to the implant and the patient. The other article is a case report that we published describing the successful use of ECT in a 17-year old adolescent boy who was implanted with a Nucleus 22 device at age 3 years.11 The patient presented with delirious mania, for which ECT is the gold standard treatment. Maximal psychopharmacologic therapy was unsuccessful in treating the patient, and it was ultimately decided to attempt ECT. The patient successfully underwent 2 treatment sessions of contralateral unilateral ECT and was eventually discharged from the hospital after returning to his baseline mental state. After each treatment, the external processor was placed back on the patient, and speech perception scores remained unchanged, which correlates to no change in functional hearing. Integrity testing was completed by the CI manufacturer during the hospital admission and revealed a normally functioning device. Four months after ECT, the patient reported pain and intermittent shocking sensation at the site of the external processor requiring ultimate replacement of the implant. The night before reimplantation, the patient had his N22 internal receiver removed to allow MRI scanning. Unfortunately, since the Nucleus 22 did not have a removable magnet, the electrode array had to be cut to allow removal of the internal receiver complete with magnet before the MRI. Although it would be unusual for electric current from the ECT applied 4 months before symptoms to cause damage to the internal receiver, this possibility cannot be excluded. Interestingly, further testing of the CI by the manufacturer after explantation revealed no hermeticity failure, electronic failure, electrode array malfunction, or any other nonspecific type of device failure. It should also be noted that the Nucleus 22 device has the highest failure rate of any CI: at least 3% in adults9 and 9% in children.14,15
ECT With Deep Brain Stimulation To date, there have been 2 case reports of the successful use of ECT in patients with DBS.9,10 Deep brain stimulators are electrical pulse generators surgically placed in the subthalamic www.ectjournal.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
307
Journal of ECT • Volume 30, Number 4, December 2014
McRackan et al
nucleus to treat conditions such as intractable tremor and Parkinson disease. Structurally, DBS electrodes are very similar to CIs except that they contain fewer electrodes and have simpler stimulation algorithms. The implantable pulse generator in DBS is placed subcutaneously in the upper thorax as opposed to the internal stimulator of CIs, which is placed behind the ear. This may result in larger electrical currents induced in CIs versus DBSs.16 Another difference between DBS and CI surgery is that a craniotomy is made in DBS surgery, whereas this is not performed in CI surgery. This absence of bone in patients with DBS may decrease resistivity between the ECT-stimulating electrodes and thus increase current spread compared to CIs. In the case reports of ECT in patients with DBS, unilateral ECT was performed in a patient with a contralateral DBS,9 and bilateral ECT was carried out in a patient with a DBS.10 Both cases demonstrated no adverse effects to the patient with regard to the DBS. Both patients’ DBS devices continued to function normally with posttreatment imaging demonstrating no device migration.
CI Impedance and ECT In our study, only one implant under one condition showed evidence of increased impedance values after ECT. This implant (number 6) had an increased impedance of 3.08 kΩ under the MP1 condition. The 3 other conditions tested with this implant all showed decreased impedance values after ECT sessions. In the 39 (97.5%) other implant/conditions tested, there was evidence of decreased impedance values. All electrodes under all conditions showed a mean decreased impedance of 7.58 kΩ with no detectable shorts. Moreover, more stringent internal testing done by the CI manufacturer showed no evidence of electrical injury. Whereas it is interesting that ECT significantly decreased the impedances in a number of implants, the clinical significance of this is unknown. Even in experimental models devoid of CIs, decreases in tissue impedances are seen in animals subjected to ECT.17 In patients with CIs, decreased CI impedance values are seen immediately after stimulation via neural response telemetry in the operating room and by regular implant use. Although the exact reason for this remains unclear, it has been shown that CIs collect proteinaceous debris or casks upon and after insertion.18 After electrical stimulation, this debris is shed, which has been shown to diminish impedance. Perhaps the most important finding from this study is that there seems to be no evidence of electrical injury to CIs after ECT when electrical current is applied with the external processor not attached. This finding occurred although the CIs in this study were subjected to the maximal amount of electrical current used for clinical ECT in both contralateral and unilateral stimulation.
Study Limitations Ideally, histologic analysis of intracochlear tissue would have been performed. This may have revealed microscopic tissue changes secondary to ECT. Owing to our anatomical gift program standards, we were unable to send fresh cadaver specimens to experts for evaluation for tissue injury. Therefore, we cannot rule out tissue damage occurring when ECT is applied to patients with CI. Further studies in animal models may be of significant use to determine whether ECT causes thermal injury and is safe to use in patients with CI. However, if ECT is to be used in patients with CI, the external receiver should be removed and the stimulating ECT electrodes should be placed as far away from the implant as possible.
308
www.ectjournal.com
CONCLUSION Electroconvulsive therapy causes no detectable damage to CI electronic circuitry in a fresh human cadaver model. Electrical stimulation from ECT is associated with statistically significant decreases in impedance values in CIs, which is consistent with a lack of electrical damage to the CI circuits. Further studies are required to examine the safety and risks of ECT in patients with CI in clinical practice. REFERENCES 1. Altman LK. Lasker Awards Go to Five Scientists and Gateses. Available at: http://www.nytimes.com/2013/09/10/health/ lasker-awards-winners-2013.html?_r=0. New York Times. September 9, 2013. Accessed January 22, 2014. 2. Anonymous. The Global Hearing Implants Market is Forecast to Exceed $2 Billion in 2017. Available at: https://www.asdreports.com/news.asp? pr_id=191. ASD Reports. January 31, 2012. Accessed January 22, 2014. 3. Anonymous. Electroconvulsive therapy. Available at: http://www.nmha. org/ect. Accessed August 28, 2013. 4. Nutall GA, Bowersox MR, Douglass SB, et al. Morbidity and mortality in the use of electroconvulsive therapy. J ECT. 2004;20:237–241. 5. Anonymous. Warnings and Precautions. Available at: http://www. advancedbionics.com/content/dam/ab/Global/en_ce/documents/ libraries/Professional%20Library/AB%20Product%20Literature/ System_Indications_Precautions/Warnings_and_Precautions.pdf. Accessed March 28, 2014. 6. Anonymous. Warnings and Precautions. Available at: http://www. cochlear.com/wps/wcm/connect/us/for-professionals/cochlear-implants/ warnings-and-precautions. Accessed March 28, 2014. 7. Anonymous. Benefits and risks of Cochlear implants. Available at: http://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/ ImplantsandProsthetics/CochlearImplants/ucm062843.htm. Accessed March 28, 2014. 8. Hartmann SJ, Saldivia A. ECT in an elderly patient with skull defects and shrapnel. Convuls Ther. 1990;6:5–171. 9. Moscarillo FM, Annunziata CM. ECT in a patient with a deep brain-stimulating electrode in place. J ECT. 2000;16:287–90. 10. Bailine S, Kremen N, Kohen I, et al. Bitemporal electroconvulsive therapy for depression in a Parkinson disease patient with a deep brain stimulator. J ECT. 2008;24:171–2. 11. Labadie RF, Clark NK, Cobb CM, et al. Electroconvulsive therapy in a cochlear implant patient. Otol Neurotol. 2010;31:64–6. 12. Nachtegaal J, Smit JH, Smits C, et al. The association between hearing status and psychosocial health before the age of 70 years: results from an internet-based national survey on hearing. Ear Hear. 2009;30:302–12. 13. Malek-Ahmadi P, Hanretta AT. Cochlear implant and ECT. J ECT. 2003;19:51. 14. Battmer RD, Linz B, Lenarz T. A review of device failure in more than 23 years of clinic experience of a cochlear implant program with more than 3,400 implantees. Otol Neurotol. 2009;3455–463. 15. Parisier SC, Chute PM, Popp AL. Cochlear implant mechanical failures. Am J Otol. 1996;17:730–4. 16. Deng ZD, Lisanby SH, Peterchev AV. Transcranial magnetic stimulation in the presence of deep brain stimulation implants: induced electrode currents. IEEE Eng Med Biol Soc Conf. 2010;6821–6824. 17. Russell RW, Pierce JF, Townsend JC. Characteristics of tissue impedance in the rat under conditions of electroconvulsive shock stimulation. Am J Physiol. 1949;156:317–21. 18. Tykocinski M, Cohen LT, Cowan RS. Measurement and analysis of access resistance and polarization impedance in implant recipients. Otol Neurotol. 2005;26:948–956.
© 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
