An Evidence-Based Algorithm for Intraoperative Monitoring During Cochlear Implantation

Transcription

1 Otology & Neurotology 33:169Y176 Ó 2012, Otology & Neurotology, Inc. An Evidence-Based Algorithm for Intraoperative Monitoring During Cochlear Implantation Maura K. Cosetti, Scott H. Troob, Jonathan M. Latzman, William H. Shapiro, John Thomas Roland Jr., and Susan B. Waltzman Department of Otolaryngology-Head and Neck Surgery, New York University School of Medicine, New York, New York, U.S.A. Objective: To generate an evidence-based algorithm for the use of intraoperative testing during cochlear implantation (CI). Study Design: Retrospective review. Setting: Tertiary referral center. Patients: A total of 277 children (aged 6 mo to 17 yr) and adults 18 years and older with normal cochlear anatomy who underwent primary and revision cochlear implantation at a single center between 2005 and 2010 were included. Intervention: Intraoperative electrophysiologic monitoring and intraoperative Stenver s view plain film radiography. Main Outcome Measure: Intraoperative testing included the following: 1) individual electrode impedance measurements; 2) neural response telemetry (tnrt) levels for electrodes E20, E15, E10, and E5; and 3) plain film radiograph assessment of electrode position. Results: No patient demonstrated abnormalities on all 3 modalities. Open or short electrodes on impedance testing were found in 6% of patients; half of these normalized when remeasured. Absent tnrt responses on 1 or more electrodes occurred in 14% of patients, although complete lack of response was rare (1.4%) and did not correlate with a dysfunctional device. Spread of excitation was performed in 1 patient and was consistent with a tip rollover. Intraoperative radiography identified tip-rollover and extracochlear electrode placement in all cases (n = 5, 1.8%) and prompted the use of the backup device. Conclusion: Immediate intraoperative determination of device functionality and optimal electrode placement is advantageous. Of the modalities tested, including electrode impedance, tnrt, and plain radiograph, only the radiographic results impacted intraoperative surgical decision making and led to the use of the backup device. Key Words: Cochlear implantsvelectrode impedancevintraoperative monitoringvintraoperative imagingv Neural response telemetryvspread of excitation. Otol Neurotol 33:169Y176, Optimal intracochlear electrode placement has long been recognized as a prerequisite for maximizing cochlear implant (CI) success. In an ideal setting, detection of incorrect electrode placement or a malfunctioning or failed receiver-stimulator would occur at the time of implantation, thus affording the opportunity for immediate action while the patient remains under anesthesia. Toward that end, a variety of intraoperative monitoring capabilities to assess device functionality and electrode Address correspondence and reprint requests to Maura K. Cosetti, M.D., NYU School of Medicine, Department of Otolaryngology, 550 First Ave, Suite 7Q, New York, NY 10016; maura.cosetti@ nyumc.org This research was supported in part by the Rienzi Foundation for Cochlear Implant Research and a donation from Shelley and Steven Einhorn. This research will be presented at the national meeting of the American Neurotology Society in Chicago, IL, on April 29, J. Thomas Roland is on the surgical advisory boards of Cochlear Corporation, Advanced Bionics, and MED-EL. William H. Shapiro is on the audiology advisory boards of Cochlear Corporation and Advanced Bionics. placement have been used in CI surgery. These include electrophysiologic measures of electrode impedance (EI), electrically evoked compound action potential (ECAP) and spread of excitation (SOE) and radiographic imaging modalities of plain film radiograph, fluoroscopy, 3-dimensional (3-D) rotational x-ray, and intraoperative computed tomography (CT) (1Y7). All current CI systems have reverse telemetric capabilities through which the implanted device is able to send functional information (EI and ECAP) back to the speech processor and software for clinical interpretation. EI is a function of the array itself and the resistive characteristics of the fluid or tissue surrounding each electrode. It provides information on individual electrode integrity, such as short or open circuits (1). In contrast, the ECAP is an indicator of the neural responsiveness of the auditory nerve (2). Each manufacturer has software allowing intraoperative ECAP measurement: neural response imaging from Advanced Bionics LLC (Los Angeles, CA, USA), neural response telemetry (NRT) from Cochlear (Sydney, Australia), and auditory nerve response telemetry from MED-EL (Innsbruk, Austria). In general, ECAP recording is rapid, 169

2 170 M. K. COSETTI ET AL. not degraded by motion artifact and unaffected by type or depth of anesthesia, making telemetry feasible in both the operating room or clinic (2,8). Intraoperatively, ECAP measurements can be used to asses the response of a patient s auditory system to electrical stimulation immediately after intracochlear insertion of the CI electrode (2,9Y11). Prior research found no significant correlation between intraoperative tnrt values and postoperative speech performance at 1 year (11). In that study, absence of intraoperative tnrt response was rare; however, it did not indicate a lack of stimulation or a dysfunctional device. Lastly, SOE has been reported as a useful intraoperative electrophysiologic measure of CI electrode placement (3). SOE measurements provide information regarding the selectivity of neural excitation fields around each electrode; when these overlap, they can suggest presence of a tip rollover. (An in-depth discussion of SOE measurements can be found in Cohen et al. [12]). Intraoperative radiographic imaging has been reported to provide confirmation of correct, intracochlear placement of the electrode array (3,7,11,13Y15). Plain radiographs using modified Stenver s view can be used to evaluate electrode location, position, and presence of tip rollover. In cases of abnormal cochlear anatomy, intraoperative fluoroscopy also has been advocated as a dynamic real-time assessment of CI electrode placement (5,7). Additional, more sophisticated imaging modalities also have been used during CI surgery, including CT and 3-D rotational x-ray, although intraoperative access to this technology is uncommon (6,7). Although previous research has reported use of these modalities for intraoperative assessment of device functionality and electrode placement, the question of when to use the backup device remains ambiguous. Comprehensive assessment of intraoperative electrophysiologic and radiographic monitoring in patients with normal cochlear anatomy has not yet been published. At the author s institution, intraoperative monitoring during CI surgery routinely consists of EI, tnrt, and plain radiograph. Measurements of SOE are performed on an individual case basis. The purpose of this study was to evaluate the contribution of each of these modalities on intraoperative decision making during cochlear implantation. METHODS Study Design A retrospective review was conducted of all patients who underwent primary and revision cochlear implantation with a Nucleus 24, Freedom, or Nucleus 512 device between 2005 and August 2010 at a single, tertiary referral institution. Children (aged 6 mo to 17 yr) and adults 18 years and older with available intraoperative data were included. In cases of bilateral implantation, data were included for both ears. Patients with structural cochlear abnormalities, such as CHARGE, Mondini or common cavity malformations, and intracochlear ossification, and cases in which fluoroscopy was used were excluded. Patients who received a double array, hybrid device or had partial insertion of a standard array also were excluded. Subjects Complete intraoperative data, including EI testing, tnrt measurements, and plain radiograph, were available for 139 adults and 138 children (total, n = 277) with normal cochlear anatomy. A variety of factors account for the lack of complete data on all patients implanted during the study period, including changes in monitoring software and charting methods. Table 1 describes demographic and intraoperative monitoring data for 277 subjects with complete data. The mean age at implantation was 55.7 years (range, 19.7Y92.2 yr) for adults and 6.2 years (range, 0.6Y18 yr) for children. Spread of data by year indicates relatively equal distribution over the 6-year study period with the most available data from 2010 (n = 61, 22%). Device types included Nucleus 24 (n = 71, 25.6%), Freedom (131, 47.3%), and 512 (n = 73, 26.3%). Procedures All operations were performed using a standard mastoidectomy and posterior tympanotomy approach. In all cases, the cochleostomy was placed anterior and inferior to the round window using a 1-mm diamond burr, and the electrode was inserted using Advanced Off-Stylet technique. All patients had full electrode array insertions. In revision implantation, the in situ electrode array was carefully severed laterally within the mastoid cavity before removal of the receiver/stimulator. With the in situ electrode in place, the receiver/stimulator of the new device was secured. To prevent any intracochlear soft tissue around the in situ electrode to collapse and obscure or obstruct the cochlear lumen, the new array was inserted as soon as the old electrode was removed. (For more details on operative techniques for revision CI surgery, please see Cosetti and Roland [16]). Intraoperative electromyographic monitoring of the facial nerve was standard in all cases. TABLE 1. Demographic and intraoperative monitoring data of pediatric and adult cochlear implant subjects N = 277 Total, n Children 138 Adults 139 Age at implantation, yr Children: mean (range) 6.2 (0.6Y18) Adults: mean (range) 55.7 (19Y92) Normal a EI, tnrt, and XR, n (%) Children 127 (92) Adults 134 (96.4) Data by year of implantation, n (%) (11.5) (13.7) (18.8) (19.5) (11.5) (22) Data by device type, n (%) Nucleus (25.6) Freedom 131 (47.3) CI (26.3) EI indicates electrode impedance; tnrt, neural response telemetry; XR, plain x-ray. a Patients with normal data were those with no open or short electrodes on impedance testing, measureable tnrt responses on all electrodes tested, and x-ray confirmation of correct, intracochlear placement of the electrode array.

3 INTRAOPERATIVE MONITORING DURING CI 171 After electrode insertion, intraoperative electrophysiologic monitoring was performed using the Custom Sound Software version 2.0 or 3.0 included with the Nucleus Freedom and Nucleus 512 device. Both EI and tnrt were performed by an experienced CI audiologist either present in the operating room (until 2007) or remotely (2007 to present). (Further detail on remote monitoring can be found in Shapiro et al [17]). EI measurements were obtained on all electrodes. EI was repeated at the discretion of the surgeon when 1 or more open or short electrodes were discovered. tnrt values were then obtained from 4 electrodes (E5, E10, E15, and E20) with additional electrodes tested when deemed prudent by the monitoring audiologist or surgeon. (In the Nucleus 24, Freedom, and 512 devices, electrodes are numbered from the cochleostomy to the apex.) Abnormal response was defined as an absent response or tnrt level of 0 on the electrode tested. Spread of excitation measurements were not performed routinely, in part because of lack of optimization within the Custom Sound Software. As in prior literature, identification of more than 1 local maximum in the SOE curve for 2 or more probe electrode positions were considered suggestive of tip rollover (Fig. 1) (3,12). Intraoperative modified Stenver s view plain radiograph was obtained in all cases and interpreted by the operating surgeon while the patient remained under general anesthesia (Fig. 2). When an abnormality of electrode location or position was detected on intraoperative radiograph, the device was removed, and a new device was placed during the same surgery. Intraoperative monitoring was performed for the new device as well. For revision cases, category of failure was categorized using the International Classification of Reliability for Implanted Cochlear Implant Receiver Stimulators (2010) (18). Manufacturer reports of explanted devices (i.e., cause of failure analyses) were used, when available, to categorize device failure per the 2010 classification. Devices explanted for medical reasons (i.e., infection, skin breakdown) were categorized as D regardless of abnormalities on postexplant testing. Device failure, category C, was used when an explanted device was deemed in-specification on manufacturer analysis, but the patient received clinical benefit from the new device. When the postexplant device analysis was not available, but in vivo integrity testing suggested the device was in-specification, reduced FIG. 1. SOE curve demonstrating 2 local maxima suggestive of a tip rollover at approximately electrode 14. The probe electrode used in this SOE is electrode 17. FIG. 2. Intraoperative modified Stenver s view plain radiograph of a right CI demonstrating malposition (tip rollover) of the electrode array. clinical benefit was documented, and performance improvement was reported with the new device, the failure was categorized as C. RESULTS Of 277 patients with complete intraoperative data, a total of 114 adults (80%) and 111 children (80%) had completely normal intraoperative monitoring, specifically normal EI with no open or short electrodes, measureable tnrt on all electrodes tested, and plain radiographic confirming correct intracochlear location and position of the electrode array. There were no initial or out-of-the-box hard failures during the study period. A total of 5 adults (3.4%) and 11 children (8%) had abnormal impedance measurements consisting of one or more open or short electrodes. Of these, 50% normalized when EIs were remeasured after approximately 4 to 6 minutes. Of the 11 children with impedance abnormalities, 1 child had 4 open electrodes, which improved to 1 after remeasurement, and another child had 3 open, which normalized to 1 when rerun. There were no children or adults with more than 4 open or short electrodes. Overall, absent tnrt measurements were found in 25 adults (18%) and 14 children (10%). There were 4 patients (1.4%) with absent tnrt on all electrodes tested: 2 adults and 2 children. Impedance measures in these patients were normal and plain radiograph confirmed correct electrode placement. These devices were not removed at the time of surgery. None of these 4 patients has experienced a device failure in 3 to 5 years of follow-up. Abnormalities on both electrophysiologic measures, EI and tnrt, were rare and occurred in only 1 subject, a 4-year-old male patient implanted with a Nucleus Freedom device. He demonstrated 1 open electrode on intraoperative measurement and no tnrt response on E20. Responses on E15, E10, and E5 were 124, 154, and 129, respectively.

4 172 M. K. COSETTI ET AL. TABLE 2. Age at implantation (yr) Intraoperative data for patients (total, n = 5) with abnormalities discovered on intraoperative radiograph Initial device Intraoperative data: initial device Intraoperative data: backup device surgery EI tnrt a XR EI tnrt XR 0.9 N512 First run: 4 electrodes open 4/4 Tip rollover + 4/4 + Second run: 1 electrode open 5.1 N /4 Tip rollover b + 4/ N /4 Tip rollover + 4/ Freedom + 4/4 Tip rollover + 4/ Freedom + 3/4, NR on E10 Electrode in the superior SCC + 3/4, NR on E10 + +, normal impedance, no open or short electrodes; +, correct, intracochlear position of the electrode array; NR indicates no response; RS, receiverstimulator; SCC, semicircular canal. a Indicates the number of measurable responses/electrodes tested. tnrt responses were obtained from 4 electrodes: E5, E10, E15, and E20. b Spread of excitation (SOE) performed on this patient demonstrated 2 local maxima in the SOE curve localized to E17. Details of patients with abnormal intraoperative radiographic findings can be seen in Table 2. Of the 277 patients, there were 5 malpositions (1.8%) found on intraoperative x-ray: 4 tip rollovers and 1 electrode in the superior semi-circular canal (SCC). Electrophysiologic testing of all 4 cases of tip rollover was normal for both the initial CI and backup device. SOE was performed on 1 patient and showed 2 local maxima in the SOE curve localized to E17. In the case of superior SCC electrode placement, tnrt could not be obtained on 1 of 4 electrodes tested (E10). Identical tnrt responses were seen after reimplantation with the backup device. Figures 3 and 4 summarize the relationship between the 3 intraoperative modalities: EI, tnrt and radiography. None of the 277 patients demonstrated abnormalities on all 3 intraoperative tests. Only patients with malpositions of the electrode array on plain radiograph underwent immediate revision and use of the backup device. An attempt was made to analyze the intraoperative data for devices that failed during the study period. As specified above, only revision patients with normal cochlear anatomy who received a Nucleus 24, Freedom, or 512 device during the study period were included. Complete intraoperative data are available for a total of 36 revision surgeries performed on 34 patients (18 adults and 16 children) during the study period. Of these, 13 patients had the initial cochlear implantation performed at our center (Table 3). Details regarding intraoperative testing, symptoms, integrity testing, and explanted device analysis can be seen in Table 3. Average time to device failure was 3.6 years (range, 0.6Y12 yr) Category of device failure was C in 12 of 13 cases with 1 revision surgery for ongoing flap infection, category D. Two patients experienced hard failures of their CI22 devices after 10 and 12 years, respectively. For 11 of the 13 revision patients, intraoperative monitoring was normal. This includes 2 patients, A.S. and L.G., who experienced device failure within 7 and 10 months after implantation, respectively. Upon reimplantation, both patients had tnrt responses on only 2 of 6 electrodes tested. Both patients demonstrate clinical benefit with their new device, performing above prerevision levels on tests of speech perception approximately 1.5 and 4 years after surgery. Among 13 revision cases, there were 2 patients (M.W. and H.B.) who had measureable tnrt levels on 3 of the 4 electrodes tested. (This pattern of tnrt responses was not unique: 20 adults [14.2%] and 12 children [8.7%] whose devices did not fail demonstrated this same pattern of tnrt responses.) Revision patients M.W. and H.B. experienced device failure at 10 months and 5 years, respectively, after surgery. Intraoperative monitoring during reimplantation was unremarkable. FIG. 3. Venn diagram of pediatric CI patients with abnormal intraoperative monitoring (total, n = 27). No patients had abnormalities on all 3 tests. Abnormal tnrt consisted of absent responses on Q1 electrodes tested. Two of the 14 patients had a complete lack of response on all electrodes tested. Abnormal EI included open/short on 1 or more electrodes tested. No patient had a complete open/short array. FIG. 4. Venn diagram of adult CI patients with abnormal intraoperative monitoring (total, n = 30). No patients had abnormalities on all 3 tests. Abnormal tnrt consisted of absent responses on 1 or more electrodes tested. Two of the 24 patients had complete lack of responses on all electrodes tested. Abnormal EI included open/short on 1 or more electrodes tested. No patient had a complete open/short array.

5 INTRAOPERATIVE MONITORING DURING CI 173 TABLE 3. Intraoperative and device failure data for patients with normal anatomy who underwent revision cochlear implantation between 2005 and 2010 Initial device Intraoperative data: Integrity test results Time to initial surgery Symptoms Category Intraoperative data: revision surgery EI tnrt a XR of failure b No. of open/short RS of failure c failure d (yr) EI tnrt XR Cochlear DAR for explanted device HB CI24RCS + 3/4 + A, B 8 + C 5 + 4/4 + Tear in silicone carrier, RS passed all tests when dry EB Freedom + 4/4 + A, B 6 + C /4 + Device passed all tests MG CI24RCS + 4/4 + A, C 6 + C 4 + 4/4 + Unable to test electrode; RS functional LG Freedom + 4/4 + A, B 7 + C /5 + Device passed all tests DL Freedom + 4/4 + C None + C /4 + Tear in silicone carrier, RS passed all tests when dry AM CI24RCS + eabr e + + A, C None + C 7 + 4/4 + Device passed all tests MP CI22 + eabr + + Hard failure CNT-Hard failure CNT-Hard failure C /4 + Tear in silicone carrier, RS passed all tests when dry JS CI24RCA + 4/4 + A, B 8 + C /4 + Unable to test electrode; RS functional AS Freedom + 4/4 + A, B 5 + C /6 + Not available MS CI22 + eabr + + A, C CNT-Hard failure CNT-Hard failure C /4 + Tear in silicone carrier, RS passed all tests when dry LT Freedom A, B 7 + C open e, rerun WNL 4/4 + Not available CS Freedom + 4/4 + D Not done Not done D 1 + 4/4 + Device passed all tests MW Freedom + 3/4 + C 1 + C /4 + Device passed all tests +, Normal impedance on all electrodes, no open or short; +, intracochlear position of the electrode array; eabr, e-auditory brainstem response testing; DAR, device analysis report for the explanted device; +, within normal limits. a Indicates the number of measurable responses/electrodes tested. tnrt responses were obtained from 4 electrodes: E5, E10, E15, and E20. If no response was obtained from the standard 4 electrodes, additional electrodes were tested at the discretion of the surgeon. b Symptoms of failure were categorized as follows: Type A, decrease in speech perception performance of at least 30%; Type B, progressive electrode deactivation; Type C, pain or tinnitus with device usage; Type D, infection around device. c As determined by the 2010 International Classification of reliability for implanted cochlear implant receiver stimulators (1). d Time (years) from first surgery to reimplantation. e Intraoperative eabr (e-auditory brainstem response) testing was performed before the introduction and use of software capable of tnrt recording.

6 174 M. K. COSETTI ET AL. Table 3 describes the results of intraoperative monitoring, integrity testing, and manufacturer device analysis of the explanted devices. In all 13 cases, intraoperative impedance measurements revealed no open or short electrodes. Ultimately, however, more than half of these 13 patients had open/short electrodes on integrity testing: 7 patients had between 5 and 8 (mean, 6.7) open or short electrodes, and 1 patient had 1 open or short electrode. There was no relationship between timing of integrity testing and number of open or short electrodes. Of the 5 patients who had no open or short electrodes on integrity testing, 2 were hard failures, and telemetry testing could not be performed because of loss of telemetric lock, 1 was explanted because of infection (Category D) and integrity testing was not performed, and an additional 2 patients had normal impedance measurement. Manufacturer device analysis of the implants with open or short electrodes before explanation did not identify abnormalities with the electrode array or contacts. For the 20 patients who underwent primary surgery at another institution and revision surgery at our center, 3 malpositions of the electrode array were discovered on plain x-ray during workup for device failure at our center. During revision surgery, incorrect cochleostomy position, specifically superior to the round window, was found on additional 6 patients (19). DISCUSSION Intraoperative monitoring during CI surgery has evolved to assist with assessment of device functionality and correct intracochlear placement of the electrode array. Results of the current study support the feasibility of intraoperative monitoring, specifically EI, NRT, and x-ray, and suggest that the majority of patients with normal cochlear anatomy have measurable responses and correct electrode placement. Of the 3 modalities, only the results of plain film x-ray impacted surgical decision making and led to the use of the backup device. A total of 5 electrode abnormalities were discovered on x-ray (1.8%): 4 tip rollovers and 1 electrode placement in the superior SCC. The incidence of electrode foldover is less than published rates in Grolman et al. (3). In both series, the surgical procedures were uneventful, and electrode placement anomalies were apparent on intraoperative imaging, ether plain radiograph or 3-D rotational x-ray, and SOE. Grolman et al. found SOE curve maxima to be abnormal in all cases of tip rollover; the present study performed SOE in 1 patient with electrode tip rollover with similar results. These data suggest that SOE performed by an experienced audiologist may be an effective intraoperative test for identification of electrode foldover; however, changes in monitoring software may be necessary to allow widespread implementation. Intraoperative plain radiograph also identified 1 case of electrode placement in the superior SCC. Published literature contains 4 prior reports of electrode placement in the SCC in patients with normal cochleovestibular anatomy (14,15,20,21). In 2 of 4 cases, intraoperative impedance and NRT showed measureable responses on all electrodes; radiologic confirmation was not performed. Postoperatively, these patients experienced vertigo with device activation prompting radiologic imaging with temporal bone CT and ultimate identification of incorrect electrode placement. Patients underwent a second revision surgery for device removal and reimplantation (15,20). In the third case report, impedances indicated intact electrode circuitry, and 3 of 4 tested electrodes had measureable tnrt responses, which was similar to our findings (14). In the last, most recent, case report, no tnrt responses were obtained intraoperatively (21). In our study and in 2 of 4 case reports, intraoperative plain film x-ray confirmed placement in the superior SCC. Electrode malposition has been cited as the cause of CI revision in 13% to 16% of cases (22,23). Among our revisions in patients with normal anatomy, preoperative workup for suspected device failure found electrode position abnormalities on plain x-ray in 3 of 15 patients implanted at other institutions. Although data from those operations are not available, it is possible that these electrode malpositions could have been identified with intraoperative x-ray after initial implantation, potentially negating the need for revision surgery. Published case reports detailing extracochlear placement of CI electrodes into the carotid canal in normal and malformed cochleas do not consistently report the use or results of any intraoperative testing (24). However, a recent report by Nevoux et al. (2010) describes electrode placement in the intratemporal carotid canal in a child with Connexin 26 mutation, normal cochlear anatomy, and a middle ear effusion at the time of implantation. Intraoperative monitoring consisted only of tnrt measurements, for which they report a complete lack of responses. Postoperative CT scan indentified the extracochlear electrode location. The child was successfully and safely reimplanted during a second surgery the same day. Although electrode malposition in the carotid canal was characterized by a total lack of tnrt response, the present and prior studies suggest that absent tnrt responses may occur in cases of a functional device in the correct location (11). Regarding EI, the overall incidence of open or short electrodes in the present study was comparable to the rate of electrode circuit failures in Nucleus devices discovered during intraoperative monitoring in Carlson et al. (5.7% versus 7.9%, respectively) (25). Abnormal intraoperative impedances can be transient, caused by air bubbles generated by electrode insertion and may resolve quickly, thus explaining the frequent normalization of these measures when run a second time. Carlson et al. also report on 3 patients who underwent reimplantation for increasing electrode deactivation and declining performance. As in the current study, these failed devices had no EI abnormalities intraoperatively. Unfortunately, in the review of our revision cases, we found no unique pattern of intraoperative monitoring. Ten of 13 patients had completely normal results on EI, tnrt, and x-ray. Of the 3 with abnormal results, all were similar

7 INTRAOPERATIVE MONITORING DURING CI 175 to patterns seen among nonrevision cases. Furthermore, patients that ultimately develop up to 8 open or short electrodes on integrity testing had no EI abnormalities intraoperatively, regardless of time to failure. Device failure, even within 2 years of implantation, was not preceded or predicted by any unique abnormalities on intraoperative testing. Although restricted to patients with normal anatomy, time to revision surgery was comparable to recent published rates in Chung et al. (23) and Marlow et al. (26). Review of ex vivo manufacturer device analysis did not correlate with intraoperative electrophysiologic monitoring. The present study found no immediate out of the box failures. Intraoperatively, this type of hard failure could be characterized by a loss of telemetric lock or an inability for the external processor to communicate with the receiver-stimulator, thus preventing telemetric functional assessment. At the study institution, difficulty with intraoperative telemetry is first addressed by checking and, if necessary, exchanging all monitoring equipment such as the cable, coil, external processor, and monitoring software. If still unresolved, the next step of the algorithm, an x-ray, is performed. Upon confirmation of correct, intracochlear placement, no additional action is taken. Although lack of telemetric lock can relate to skin flap thickness, current surgical incisions have evolved such that thinning of the flap over the device is not longer possible without extending or enlarging the postauricular incision. Additionally, even if technically feasible, flap thickness may be exacerbated by edema during surgery and be unresponsive to thinning. This, in combination with the rarity of the out of the box failures, suggests that device removal may not be prudent in cases of loss of telemetric lock. The results of this study on patients with normal anatomy suggests that current, widely accessible electrophysiologic monitoring, specifically EI and tnrt, does not impact intraoperative decision making and whether to use the backup device. However, these measurements may have other important applications. First, intraoperative tnrt measurements can provide a valid basis for initial programming, especially in difficult-to-program populations such as very young children or those with multiple disabilities. EI measurements may have future implications for atraumatic electrode insertion and preservation of residual hearing. Recent research in animal and human cadaveric models provide support for continuous impedance monitoring during electrode insertion as a method of assessing intracochlear physiology and preventing insertional trauma (27). In vivo human testing is necessary to validate this application. As the present study was restricted to patients with normal cochleovestibular anatomy, these results may not be generalizable to patients with malformed cochleas. Although the risk of extracochlear electrode placement is higher in patients with cochlear malformations, exact rates are unknown. Prior literature has advocated the use of intraoperative imaging, including fluoroscopy and CT, to assist in challenging cases. Further research is necessary to investigate the impact of both radiologic and electrophysiologic monitoring on surgical planning and decision making during challenging cases. CONCLUSION Immediate intraoperative determination of device functionality and optimal electrode placement is advantageous. Intraoperative monitoring, specifically EI, tnrt, and plain film x-ray is feasible in routine CI surgery and may be normal in the majority of patients. Of the 3 modalities, only the results of plain film x-ray impacted surgical decision making and led to the use of the backup device. REFERENCES 1. van Wermeskerken GK, van Olphen AF, Smoorenburg GF. Intraand postoperative electrode impedance of the straight and Contour arrays of the Nucleus 24 cochlear implant: relation to T and C levels. Int J Audiol 2006;45:537Y Mens LH. Advances in cochlear implant telemetry: evoked neural responses, electrical field imaging, and technical integrity. Trends Amplif 2007;11:143Y Grolman W, Maat A, Verdam F, et al. Spread of excitation measurements for the detection of electrode array foldovers: a prospective study comparing 3-dimensional rotational x-ray and intraoperative spread of excitation measurements. Otol Neurotol 2009;30:27Y Zeitler DM, Budenz CL, Roland JT Jr. Revision cochlear implantation. Curr Opin Otolaryngol Head Neck Surg 2009;17:334Y8. 5. Coelho DH, Waltzman SB, Roland JT Jr. Implanting common cavity malformations using intraoperative fluoroscopy. Otol Neurotol 2008;29:914Y9. 6. Carelsen B, Grolman W, Tange R, et al. Cochlear implant electrode array insertion monitoring with intra-operative 3D rotational X-ray. Clin Otolaryngol 2007;32:46Y Bloom JD, Rizzi MD, Germiller JA. Real-time intraoperative computed tomography to assist cochlear implant placement in the malformed inner ear. Otol Neurotol 2009;30:23Y6. 8. Crawford MW, White MC, Propst EJ, et al. Dose-dependent suppression of the electrically elicited stapedius reflex by general anesthetics in children undergoing cochlear implant surgery. Anesth Analg 2009;108:1480Y7. 9. van Dijk B, Botros AM, Battmer RD, et al. Clinical results of AutoNRT, a completely automatic ECAP recording system for cochlear implants. Ear Hear 2007;28:558Y Botros A, Psarros C. Neural response telemetry reconsidered: I. The relevance of ECAP threshold profiles and scaled profiles to cochlear implant fitting. Ear Hear 2010;31:367Y Cosetti MK, Shapiro WH, Green JE, et al. Intraoperative neural response telemetry as a predictor of performance. Otol Neurotol 2010; 31:1095Y Cohen LT, Saunders E, Richardson LM. Spatial spread of neural excitation: comparison of compound action potential and forwardmasking data in cochlear implant recipients. Int J Audiol 2004;43: 346Y Sorrentino T, Cote M, Eter E, et al. Cochlear reimplantations: technical and surgical failures. Acta Otolaryngol 2009;129:380Y Viccaro M, Covelli E, De Seta E, et al. The importance of intraoperative imaging during cochlear implant surgery. Cochlear Implants Int 2009;10:198Y Tange RA, Grolman W, Maat A. Intracochlear misdirected implantation of a cochlear implant. Acta Otolaryngol 2006;126:650Y Cosetti M, Roland JT. Cochlear implant electrode insertion. Oper Techn Otolaryngol Head Neck Surg 2010;21: Shapiro WH, Huang T, Shaw T, et al. Remote intraoperative monitoring during cochlear implant surgery is feasible and efficient. Otol Neurotol 2008;29:495Y8.

8 176 M. K. COSETTI ET AL. 18. Battmer RD, Backous DD, Balkany TJ, et al. International classification of reliability for implanted cochlear implant receiver stimulators. Otol Neurotol 2010;31:1190Y Briggs RJ, Tykocinski M, Stidham K, et al. Cochleostomy site: implications for electrode placement and hearing preservation. Acta Otolaryngol 2005;125:870Y Pau H, Parker A, Sanli H, et al. Displacement of electrodes of a cochlear implant into the vestibular system: intra- and postoperative electrophysiological analyses. Acta Otolaryngol 2005;125:1116Y Ramalingam R, Ramalingam KK, Padmaja HS. An unusual occurrence in cochlear implantation surgery: misplaced electrode. J Laryngol Otol 2009;123:e Lassig AA, Zwolan TA, Telian SA. Cochlear implant failures and revision. Otol Neurotol 2005;26:624Y Marlowe AL, Chinnici JE, Rivas A, et al. Revision cochlear implant surgery in children: the Johns Hopkins experience. Otol Neurotol 2010;31:74Y Nevoux J, Loundon N, Leboulanger N, et al. Cochlear implant in the carotid canal. Case report and literature review. Int J Pediatr Otorhinolaryngol 2010;74:701Y Carlson ML, Archibald DJ, Dabade TS, et al. Prevalence and timing of individual cochlear implant electrode failures. Otol Neurotol 2010;31:893Y Chung D, Kim AH, Parisier S, et al. Revision cochlear implant surgery in patients with suspected soft failures. Otol Neurotol 2010; 31:1194Y Campbell AP, Suberman TA, Buchman CA, et al. Correlation of early auditory potentials and intracochlear electrode insertion properties: an animal model featuring near real-time monitoring. Otol Neurotol 2010;31:1391Y8.