P rotoundly deaf patients benefit from multichannel cochlear implants.

Transcription

1 Christian Czerny1 Erich Steiner1 Wolfgang Gstoettner2 Wolf-Dieter Baumgartner2 Herwig lmhof1 Received February 20, 1997; accepted after revision May 12, Department of Radiology ) Division of Osteology), University Hospital of Vienna, Waehringer Guertel 18-20,1090Vienna, Austria. Address correspondence to C. Czerny. 2Department of Otolaryngology, University Hospital of Vienna, 1090 Vienna, Austria. AJR 1997:169: X/97/ American Roentgen Ray Society Postoperative Radiographic Assessment ofthe Combi 40 Cochlear Implant OBJECTIVE. The aims of this study were to establish a plain radiographic technique for the assessment of the postoperative appearance. position. and insertion depth of the Combi 40 cochlear implant and to correlate the radiologic findings with surgical reports. SUBJECTS AND METHODS. In an experimental study. an electrode of the Combi 40 device was inserted into the cochlea of a cadaveric skull. Digital radiographs were obtained in a modified Chausse III projection. in which the skull was placed supine on the radiography table with the infraorbitomeatal line strictly perpendicular to the film cassette. The skull was then rotated 30#{176} away from the side to be examined. and the central X-ray beam was angled I 5#{176} cephalad to the infraorbitomeatal line. On these radiographs. the point of cochleostomy was marked by a needle tip and was projected inferior to the vestibule and on a line drawn through the supenor semicircular canal and the vestibule. The appearance and position of the electrode was evaluated. An electrode was defined as completely inserted if all electrode contacts projected medial to the line drawn through the superior semicircular canal and the vestibule. We also studied cochlear implant insertion ofthe Combi 40 device in 37 patients. Postoperative digital radiographs of these patients were obtained and analyzed for the criteria as defined in the cadaveric study. In addition. the insertion depth of the electrode and the angle of insertion were measured on the radiographs. This depth was correlated with depth of insertion as estimated at surgery. RESULTS. The cadaveric study showed that the completely inserted electrode was seen on radiographs as a nonoverlapping spiral within the cochlea. All electrode contacts projected medial to the line drawn through the superior semicircular canal and the vestibule. In all 37 patients. the electrode could be seen without overlapping. According to our criteria. a completely inserted electrode was seen in 32 patients. In these patients. the insertion depth ranged from 21 to 34 mm and the angle of insertion ranged from 350#{176} to 810#{176}.In two patients. we saw a completely inserted electrode with a bend. In three patients. an incompletely inserted electrode was seen. Excellent correlation existed between the radiologic and surgical results with regard to insertion depth (r =.92). CONCLUSION. Digital radiographs obtained in the modified Chausse III projection allow clear depiction of the electrode and avoid overlapping. Such radiographs enable a reliable and accurate assessment of the position and insertion depth of the electrode of this new cochlear implant. Such images can serve as a baseline tir further radiographic examinations when extrusion or slippage of the electrode is clinically suspected. P rotoundly deaf patients benefit from multichannel cochlear implants. which considerably improve auditory performance I I The multichannel cochlear implant prosthesis consists of the receiver-stimulator. which is implanted in the postauricular region. and the electrode array. which is inserted into the scala tympani through a cochleostomy drilled ilito the cochlear promontory. Typically. the electrode is inserted into the basal turn of the cochlea. However, insertion in the apical regions of the cochlea is advantageous because it allows better stimulation of surviving neural elements in the deep-frequency range 131- Postoperative or intraoperative radiologic evaluation of intracochlear electrode position is a standard procedure in cochlear implant surgery. Postoperative assessment of cochlear implants usually is perfornied using con- AJR:169, December

2 Czerny et al. ventional radiographic techniques and CT [4-8j. Deep insertion of the electrode of the Nucleus device (Cochlear Corporation, Melbourne, Australia), as measured on radiographs, was approximately as deep as 25 mm, or 540#{176}. These data were correlated with audiologic tests, and it was suggested that deeper insertions of the electrode might be of potential benefit to the patient [3, 5]. In 1994, a new type of cochlear implant (Combi 40 device; MED-EL, Innsbruck, Austria) was introduced, designed to provide deeper insertions into the cochlea than was possible with other devices [3, 9]. For accurate radiologic assessment of deeply inserted electrodes without overlapping in one projection, the central X- ray beam should be perpendicular to the turns of the electrode with respect to the spiral turn of the cochlea. Therefore, a radiographic projection in which the central X-ray beam is angled perpendicular to the turns of the cochlea or parallel to the axis of the modiolus would be more appropriate for the accurate assessment of this new type of electrode [10-12]. To our knowledge, reports of the radiologic appearance, assessment of electrode position, and estimation of insertion depth of this new cochlear implant have not been published. The purposes of this study were to establish a plain radiographic technique for the assessment of this new cochlear implant, to present the postoperative radiologic appearance of the electrode, to assess the electrode position, to determine the depth of electrode insertion, and to correlate the radiographic findings with the surgical reports. Subjects and Methods Experimental Study To obtain a plain radiograph of the electrode of the Combi 40 cochlear implant in one projection and to assess the position of the electrode without overlapping and by which the point of cochleostomy was projected inferior to the vestibule, an original Combi 40 electrode was inserted into the cochlea of a cadaveric skull. The electrode insertion was performed by the same otolaryngologist with the cochleostomy drilled in the same location used in the patient study described later. The point of cochleostomy was marked by a needle tip. The electrode of this cochlear implant consists of eight pairs of contacts arranged in a twin-surface configuration for eight-channel stimulation. The distribution range covered by the eight pairs of contacts is 20.6 mm, and the last contact is located on the tip of the electrode. First, digital radiographs were obtained in the anteroposterior and lateral view, in the Stenver view, and in the view proposed by Marsh et al. [5]. Then, digital radiographs in a projection similar to the Chausse Ill view were obtained with the following settings [10-12]: The skull was positioned supine on the radiography table with the infraorbitomeatal line (the line connecting the infraorbital rim and the supenor aspect of the external auditory meatus) strictly perpendicular to the film cassette. The skull was then rotated 30#{176} away from the side to be examined. The central X-ray beam was angled 15#{176} cephalad to the infraorbitomeatal line and then centered to the middle ofthis line (Fig. 1). The conus was set such that the orbit and, particularly, the lens were not in the imaged area. The projection, which was the most suitable for the depiction of the intracochlear electrode without overlapping and in which the point of cochleostomy was projected inferior to the vestibule and on a line drawn through the superior semicircular canal (SSC) and the vestibule, was chosen by two radiologists in consensus and performed in the patient study. Patient Study In 37 patients, postoperative digital radiographs of the Combi 40 electrode in the modified Chausse III projection were performed using settings identical to those described in the cadaveric study. All patients (16 male patients, 2 1 female patients; mean age, years; range, 2-76 years) underwent implantation of the Combi 40 device by the same otolaryngologist. One of these patients underwent implantation of an older type of device outside our otolaryngology department. A complete insertion of the electrode was defined surgically if all electrode contacts had passed through the cochleostomy. An incomplete insertion of the electrode was defined surgically if one or more electrode contacts had not passed the Central X-ray beam Film 300 cassette Midsagittal line A cochleostomy. The depth of insertion of the electrode was estimated surgically by counting the number of contacts that passed through the cochleostomy. Deeper insertion was estimated surgically by measuring the distance from the first contact to a marker located 30.8 mm from the tip of the electrode. This marker did not prevent an insertion deeper than 30.8 mm. Image analysis of all digital radiographs was performed by two radiologists in consensus who evaluated the appearance of the electrode. the position of the electrode with respect to complete or incomplete insertion. the number of turns of the electrode within the spiral of the cochlea. the insertion depth in millimeters, and the angle of insertion. As defined in the cadaveric study, the point of cochleostomy was projected on a line drawn through the SSC and the vestibule on the radiographs obtained in the modified Chausse III projection (Figs. 2-4). An insertion was defined as complete if all electrode contacts could be depicted medial to this line. An insertion was defined as incomplete if one or more contacts were projected on this line or lateral to this line. If an electrode could not be visualized within the cochlea, it was defined as a misplaced inserted electrode. Other complications, such as a bend in the electrode, were also noted. In addition, the insertion depth of the electrode was assessed by counting the turns of the electrode within the cochlea. The depth of insertion (in millimeters) of the electrode was determined by measuring the distance from the line drawn through the SSC and the vestibule to the first contact of the electrode and adding this distance to the distance from the first contact to the last contact. The angle of insertion (in degrees) of the electrode within the cochlea was determined in the following way (Fig. 4): A line was drawn parallel to the course of the basal Central X-ray beam Film cassette Fig. 1.-Drawings show tube setting in relation to patients head, which is in supine position with infraorbitomeatal line strictly perpendicularto film cassette. A, Head is rotated 300 away from the side to be examined. B, Central X-ray beam is angled 15#{176} cephalad to infraorbitomeatal line AJR:169, December 1997

3 Radiographic Assessment of a Cochlear Implant Fig. 2.-Skull of cadaver, from 75-year-old man. Digital radiograph of skull with inserted electrode reveals that point of cochleostomy is marked by tip of needle (straight solid arrow) and is projected inferior to vestibule (arrowhead) on line (curvedarrow) drawn through superior semicircular canal and vestibule. All electrode contacts can be seen medial to cochleostomy. indicating completely inserted electrode (open arrow). Fig. 3.-Drawing corresponding to Figure 2 shows superior and lateral semicircular canal, vestibule, and cochlea with completely inserted electrode. Point of cochleostomy is seen inferior to vestibule and projected on line drawn through superior semicircular canal and vestibule. Fig. 4.-Digital radiograph of 2-yearold boy reveals completely inserted electrode (large straight arrow) without overlapping. Electrode contacts are delineated as small rectangular structures (open arrow). Point of cochleostomy (small straight arrow) is visualized inferior to vestibule and on line (curved arrow) drawn through superior semicircular canal and yestibule. Lines (arrowheads) for measurement of angle of insertion of electrode are drawn. Insertion depth of electrode is 28 mm and angle of insertion is 576#{176}. turn with respect to the electrode within the basal turn, which presented as a horizontal line. A second line was then drawn through the virtual apex of the cochlea perpendicular to the line, which was parallel to the basal turn. A third line was drawn perpendicular to the last contact and through the line running through the apex of the cochlea. The angle between these two lines was measured and added to 360#{176}, 540#{176},or 720#{176} if the last contact was beyond the line drawn through the apex of the cochlea. The angle was subtracted from 360#{176}. 540#{176}, or 720#{176} if the last contact was before that line (Fig. 4). The radiographic findings concerning a completely or incompletely inserted electrode were compared with the surgical reports. The radiographically measured insertion depth was compared with the surgically estimated insertion depth. For statistical evaluation. the Pearson conelation coefficient was used. Results On the radiographs obtained in the modifled Chausse III projection, the point of cochleostomy for the insertion of the electrode was projected inferior to the vestibule in the cadaveric study and in all patients. As defined previously, the point of cochleostomy could be visualized inferior to the vestibule and was projected on a line drawn through the superior semicircular canal and the vestibule (Figs. 2-4). In this projection, a completely inserted electrode was depicted in the basal turn as a horizontal line and as a spiral within the turns of the cochlea without overlapping of the electrode and with all eight contacts projected medial to the point of cochleostomy with respect to the line drawn through the SSC and the vestibule (Figs. 2-4). The contacts of the electrode could be depicted as small rectangular structures (Fig. 4). In the other projections performed in the cadaveric study we noted an overlapping of the electrode. and these projections were not performed in the patient study. A complete insertion of the electrode within the cochlea was seen on digital radiographs in 32 patients. and in five of these patients even the twin-surface configuration of the first two pairs of electrode contacts could be identified (Fig. 5). In 32 patients. the completely inserted intracochlear electrode was turned approximately one to 2.25 times. In these patients, the depth of insertion of the electrode varied between 2 1 and 34 mm, and the angle of insertion varied between 350#{176} and 810#{176} (Figs. 4-6). Point of cochleoston y Among patients with a completely inserted electrode with the same insertion depth measured on the radiographs, the angle of insertion varied (e.g., insertion depth = 21 mm, range of angle = #{176}) (Fig. 7). In one of the patients with a completely inserted Combi 40 electrode, parts of an older type of electrode from a previous operation performed outside our otolaryngology department were projected into the vestibule (Fig. 8). In two patients. the electrode was completely inserted. but a bend in the electrode outside the contact range of the electrode was depicted (Fig. 9). Three other patients showed an incompletely inserted electrode, less than 2 1 mm deep. with one electrode contact located in the cochleostomy in each patient (Fig. 10). Two of these patients had a maximum angle of insertion of 340#{176},and one of these patients had an angle of insertion of 280#{176}. All radiographic findings concerning a complete or incomplete insertion of the electrode were confirmed by the surgeon s report. We found no discrepancies between the surgical reports and the radiologic findings concerning completely or incompletely inserted electrodes. The radiologically evaluated insertion depth of the electrode showed excellent correlation with the surgically estimated insertion depth (r=.92)(fig. 11). Discussion The postoperative radiographic assessment ofthe stimulating electrode is important for all implant patients, to document the correct intracochlear electrode placement and to determine the depth of insertion [4, 5]. This assessment confirms the success of the surgical procedure and can serve as a baseline for further control in the event of complications, such as extrusion of the electrode, which would cause a worsening of function. Furthermore, it has AJR:169, December

4 ... Czerny et al. Fig. 5.-Digital radiograph of 65-year-old man reveals completely inserted electrode (arrow) without overlapping, with insertion depth of 22 mm and angle of insertion of 560#{176}. Twin configuration of first two pairs of electrode contacts (arrowheads) is shown. Depthoflnseition 35 (mm) as measured on rac5ographs i*..ti. ----,.4 : Ane otinsert,on Idegrees) as measured on radiographs Fig. 7.-Scatter diagram shows that various angles of insertion measured on radiographs correspond to one insertion depth in millimeters as measured on radiographs. 800 I,1.,... Fig. 6.-Digital radiograph of 66-year-old man reveals completely inserted electrode (arrow) without overlapping and with 2.25 turns of electrode within cochlea. Insertion depth of electrode is 34 mm and angle of insertion is 810#{176}. Fig. 8.-Digital radiograph of 44-year-old man reveals completely inserted electrode (arrow) with insertion depth of 28 mm and angle of insertion of 570#{176}. Parts of older type of electrode (arrowhead) resulting from previous operation performed outside our otolaryngology department are visualized in vestibule. Fig. 9.-Digital radiograph of 5-year-old girl reveals completely inserted electrode (arrow) with angle of insertion of 370#{176}. Bend in electrode (arrowhead) outside stimulating contacts is revealed. Fig. 10.-Digital radiograph of 37-year-old woman reveals incompletely inserted electrode (arrow) with one electrode contact (arrowhead) projected in cochleostomy. Angle of insertion of electrode is 340#{176} AJR:169, December 1997

5 Radiographic Assessment of a Cochlear Implant Depth of insertion (mm) as measured at surgery , Depth of insertion (mm) as measured on radiographs Fig. 11.-Scatter diagram shows correlation between insertion depth of electrodes measured on radiographs and at surgery (r=.92). been suggested that deeply inserted electrodes might influence postoperative hearing success because more deeply inserted electrodes result in lower electrical thresholds and better auditory performance [2, 5]. Previous studies assessed the Nucleus multichannel cochlear prosthesis by using conventional radiography either in the anteroposterior and lateral view, in the Stenver view, or in a so-called modified Stenver view [4, 5]. A plain radiographic technique for the assessment of the Combi 40. which has been shown to provide deep intracochlear insertion, has not yet been established [3]. In addition, to our knowledge, detailed postoperative evaluation of the radiographic appearance and of the position of electrodes of the Combi 40 device has not yet been reported in the literature, and a correlation between the radiologically measured insertion depth of this electrode and surgically estimated insertion depth has not been performed. We performed digital radiography in our study because it offers several advantages compared with conventional radiography. First, the dose-response characteristics are perfectly linear and thus subject contrast and density are captured with the same efficiency from low to high doses. Second, the stability of contrast and density in this dose range provide an improvement in image uniformity, and repeated examination because ofexposure errors is thus eliminated. Third, postprocessing of the images can enhance structures of diagnostic importance, such as the superior semicircular canal, the vestibule, and the electrode [13-15]. These advantages enabled an easier assessment of the Combi 40 electrode and of the electrode position. In addition, the use of a small conus combined with digital radiography and only one projection has the advantage of dose reduction as well as the exclusion of the lens from the radiated field. All of these advantages proved to be beneficial, particularly when children were examined. The Combi 40 electrode seems to offer the possibility of a deeper insertion compared with the Nucleus implant on the basis of our results, as compared with the results of a previous study [5]. As mentioned, deeper insertion of the electrode might result in better auditory performance [2, 3, 5]. In our patients with completely inserted electrodes, a maximum insertion of 34 mm and 8 10#{176} was found, although in a previous study deeply inserted electrodes with an insertion depth of 25 mm or 540#{176} were described [5]. In our opinion, this cannot be explained by our different imaging technique but may be the result of the mechanical properties of the electrode and of the surgical technique of our otolaryngologist who performed endoscopy of the cochlea before insertion of the electrode [9]. For the radiographic evaluation of the position and insertion depth of the Nucleus implant, a Stenver view or a modified Stenver view was used, as described in previous studies [4, 5]. However, these projections were not suitable for the radiographic assessment of the Combi 40 electrode because the electrode was overlapping, which made the accurate evaluation of electrode position and estimation of insertion depth difficult. The overlapping of the electrode in those projections might be explained by the fact that the spiral turns of the cochlea with respect to the electrode were not perpendicular to the central X-ray beam in those projections, and therefore other authors also proposed a so-called Chausse III projection for the best depiction of the spiral turns of the cochlea [10-12]. However, by obtaining digital radiographs in the cadaveric and patient study in our modified projection, we found the electrode of the Combi 40 device was not overlapping and was best depicted in all patients. In this projection, the twin-surface configuration of the first two pairs of electrode contacts could be identified in five patients. An overlapping of the electrode as noted in the other projections might have made a bend in the electrode invisible. Bends were noted in two patients in our study. The clear depiction of the intracochlear electrode without overlapping confirms the fact that the central X-ray beam might have been perpendicular to the intracochlear electrode in our proposed projection. In addition, the point of cochleostomy was always projected inferior to the vestibule and on the line drawn through the SSC and the vestibule in this projection, which offered a consistent landmark for determining the electrode position and insertion depth. This point was also confirmed by the fact that we found no discrepancies between the surgical reports and our radiologic findings concerning a complete or incomplete insertion of the electrode. Therefore, this projection seems to be reliable for the evaluation of the position and insertion depth of the electrode and for the depiction of a bend in the electrode, which was reported to be accurately revealed on CT only [4]. The visibility of the entire electrode without overlapping and the possibility of counting the contacts medial to the point of cochleostomy with respect to the line drawn through the SSC and the vestibule offered a simple method for evaluating complete or incomplete electrode insertion and for counting the number of turns of the electrode within the cochlea. The excellent correlation between the radiographically measured insertion depth and the surgically estimated insertion depth might result in the consistent projection of the cochleostomy inferior to the vestibule on all radiographs. In contrast to our results, Marsh et al. [5] found a discrepancy between the radiologically and surgically measured insertion depths. This discrepancy might be explained either by our projection, on which the point of cochleostomy was always projected inferior to the vestibule, or by the fact that our surgeon implanted the electrode keeping the insertion site in the same location in all patients. However, measuring the angle of insertion might be more useful for the prediction of auditory performance than determining the insertion depth in millimeters [5]. This angle could be determined by drawing lines as proposed in AJR:169, December

6 Czerny et al. this study. As shown on the radiographs, the basal turn with respect to the electrode in the basal turn was depicted as a horizontal line. In our opinion, this line could be taken as a reference line or baseline for measuring the angle of insertion because evaluating the angle from this line reflects precisely how many turns of the cochlea were covered by the electrode. The angle of insertion reflects more precisely the position of the electrode tip and the proximity of the electrode tip to the apex of the cochlea than the inserted distance measured in millimeters. In addition, we have found that, in patients with a completely inserted electrode of 21 mm, for example, the angle of insertion varied between 350#{176} and 450#{176}, which might be because of the fact that the electrode may not always lie at the outer wall of the scala tympani [3]. Therefore, it might be more useful to conelate the angle of insertion with audiologic tests than with the insertion depth in millimeters [5]. In conclusion, the findings of our study show that the appearance, position, and insertion depth of the electrode with no overlapping could be reliably and accurately assessed using our plain radiographic technique. Our proposed projection and method for the assessment of electrode position and for the evaluation of the depth and angle of insertion was precise and simple. It may serve as a baseline for further control should complications arise, such as extrusion or slippage of the electrode. CT evaluation of electrode position, location, and insertion depth might be unnecessary, which would decrease the radiation dose and the cost of examination. Therefore, the postoperative assessment of cochlear implants with this projection has become a standard procedure in our institution and is now performed intraoperatively on request of the surgeon, to offer the possibility of changing the insertion depth of the electrode. References 1. Burian K. Hochmair-Desoyer IJ. Eisenwort B. The Vienna cochlear implant program. Otolaryngol Clin NorthAm 1986;19: Clark AO. The University of Melboume/Cochlear Corporation (Nucleus) Program. Otolarvngol Clipi NorthAin 1986:19: Gstoettner W. Plenk H, Franz P. et al. Cochlear implant deep electrode insertion: extent of insertional trauma. Ada Otolarvngol 1997;l 17: Shpizner BA, Holliday RA, Roland JT, Cohen NL, Waltzman SB. Shapiro WH. Postoperative imaging of the multichannel cochlear implant. AJNR 1995:16: Marsh MA, Xu J, Blamey PJ, et al. Radiologic evaluation of multichannel intracochlear implant insertion depth. Am J Otol 1993:14: Rosenberg RA. Cohen NL, Reede DL. Radiographic imaging for the cochlear implant. Ann Otol Rhino! Laryngo! 1987;96: Ball JB Jr. Miller GW, Hepfner ST. Computed tomography of single-channel cochlear implants. AJNR 1986:7: Mukherji 5K, Mancuso AA, Kotzur IM, et al. CT of the temporal bone: findings after mastoidectomy, ossicular reconstruction, and cochlear implantation. AiR 1994:163: Gstoeuner W, Baumgartner WD, Franz P. Hamzavi J. Cochlear deep insertion surgery. Laryngoscope 1997;107(in press) 10. Brusis T, Moedder U. HNO Roentgen-A ujhahmetechnik und Norma!befunde. Berlin: Springer- Verlag. 1984: Zonnefeld FW. Computed tomography ofthe tempora! bone and orbit: technique ofdirect muftip!anar high-resolution CT and correlative cryosectional anatomy. Baltimore: Urban and Schwarzenbezg, 1987: Dimopoulos P. Muren C. Anatomic variations of the cochlea and relations to other temporal bone structures. Acta Radio! 1990;3 l(fasc. 5): Sonoda M, Tanako M, Miyahara J. Kato H. Computed radiography utilizing scanning laser stimulated luminescence. Radiology 1983;148: Merrin CRB, Tutton RH, Bell KA, et al. Clinical application of digital radiography: computed radiographic imaging. RadioGraphics 1985:5: BaIter S. On the work of the radiologist: separation of image capture from image display. Acta Radio! 1988:29: AJR:169, December 1997