Otology & Neurotology 35:e123Ye129 Ó 2014, Otology & Neurotology, Inc.
Comparison of Stapedotomy Minus Prosthesis, Circumferential Stapes Mobilization, and Small Fenestra Stapedotomy for Stapes Fixation *†Gu¨l Ozbilen Acar, *‡Ilkka Kiveka¨s, *§Bassem M. Hanna, *Lin Huang, *kQuinton Gopen, and *Dennis S. Poe *Department of Otolaryngology & Communication Enhancement, Boston Children’s Hospital and Harvard Medical School, Boston, Massachusetts, U.S.A.; ÞIstanbul Goztepe Training and Research Hospital, Ear Nose Throat Department, Istanbul, Turkey; þDepartment of Otorhinolaryngology, Tampere University Hospital and the University of Tampere, Tampere, Finland; §Department of Otolaryngology, University of Ottawa, Ottawa, Canada; and kDepartment of Otolaryngology, University of California Los Angeles, California, U.S.A.
Objective: To compare the outcomes of 3 surgical techniques for primary stapes fixation: stapedotomy minus prosthesis (STAMP), circumferential stapes mobilization (CSM), and small fenestra stapedotomy (SFS). Study Design: Retrospective review of 277 primary cases operated for stapes fixation from 1997 to 2007. Setting: Tertiary academic center. Patients: Consecutive adult and pediatric cases operated for conductive hearing loss because of stapes fixation. Interventions: STAMP was performed for otosclerosis limited to the anterior footplate, CSM was conducted for congenital stapes fixation, SFS was performed for more extensive otosclerosis or anatomic contraindications to STAMP/CSM. Main Outcome Measures: Pure-tone audiometry was performed preoperatively and postoperatively (3Y6 wk) and the most recent long-term results (Q12 mo). Results: Ninety-nine ears in 90 patients had audiologic followup data over 12 months. Sixty-seven ears (68%) underwent SFS, 16 (16%) STAMP, and 16 (16%) CSM. There was significant
improvement in average air conduction (AC) thresholds and airbone gap (ABG) for all techniques. Mean ABG for SFS closed from 29 to 7.1 dB (SD, 6.0), for STAMP from 29 to 3.8 dB (SD, 5.8 dB), and for CSM from 34 to 20 dB (SD, 8.2 dB). AC results were better in the STAMP than in the SFS group, especially in high frequencies. Bone conduction improvements were seen in all groups, highest in STAMP (4.3 dB) and CSM (3.8 dB) groups, but the differences between groups were not statistically significant. Conclusion: Satisfactory hearing results were achieved with all the techniques, and STAMP showed better hearing outcomes, especially in high frequencies. CSM is a good option for children and patients in whom it is desirable to avoid a footplate fenestration or prosthesis. CSM and STAMP had significantly higher rates of revision for refixation than SFS. Key Words: Circumferential stapes mobilizationVHearing outcomeVOtosclerosisVSmall fenestra stapedotomyVStapedotomy minus prosthesisVStapes fixation. Otol Neurotol 35:e123Ye129, 2014.
Minimally invasive stapedotomy procedures date back to 1952 and include mobilization, anterior crurotomy, and other stapedioplasty techniques that were designed to preserve some portion of the crura and footplate (1Y6). Initially, it was difficult to reliably control the osteotomies, but with the introduction of the laser and microdrill, the surgical precision has been greatly improved (7,8). Laser stapedotomy minus prosthesis (STAMP) was developed in
1998 for limited anterior otosclerosis in which the anterior crus was divided, and a linear separation in the footplate was created between the anterior one third and posterior two thirds. High-frequency hearing was more favorable than conventional laser stapedotomy (10Y12). Children with bilateral congenital fixation of the stapes or severe tympanosclerosis present special challenges in that they are too young to make their own informed consent, they have a higher incidence of increased perilymph pressure (13), and they have a higher risk of subsequent otitis media. A minimally invasive circumferential footplate mobilization procedure could be beneficial in such cases to improve hearing, yet keep risks reasonably low. The increased surface area of the mobilized footplate
Address correspondence and reprint requests to Dennis S. Poe, M.D., Department of Otolaryngology & Communication Enhancement, Boston Children’s Hospital, 333 Longwood Avenue, Lo-367, Boston, MA 02115; E-mail: [email protected]
e123
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
e124
G. O. ACAR ET AL.
FIG. 1. Diagram of STAMP procedure with the otosclerotic focus involving the anterior one third of the stapes footplate. A laser was used to divide the anterior crus and the footplate transversely across the junction between its anterior one third and the posterior two-thirds portions, mobilizing the remaining posterior footplate.
might be expected to yield advantageous high-frequency hearing results as well. The purpose of this study was to investigate the outcomes of minimally invasive stapes surgical techniques, including circumferential stapes mobilization (CSM) and STAMP. We wanted to test the following hypotheses: 1) preservation of the stapes footplate as a vibrating surface should aid in the high-frequency hearing results compared with conventional small fenestra stapedotomy (SFS) and 2) CSM and STAMP procedures provide acceptable risks and reasonable short- and long-term hearing results compared with SFS.
fenestra. Piston MEEI prosthesis of 0.6-mm diameter was used for reconstruction. The fenestra was sealed around the piston by placing tragal perichondrium covered with gelfoam soaked in saline in the earlier cases, and in recent years, a meatal tissue graft was used, initially covered with gelfoam and, since 2005, without any gelfoam. The laser STAMP procedure has been previously described in detail (11). Using the same approach as for SFS, an assessment of the extent of the otosclerotic focus was made. If the focus was limited to the anterior one third of the footplate and if the oval window niche was sufficiently wide to afford a view of the footplate under the anterior crus, then a STAMP was performed. The laser was used to initially divide the anterior crus; then, the footplate was divided transversely across the junction between its anterior one third and the posterior two thirds. Laser exposure was typically 5 to 7 burns using the same settings as for SFS. Consequently, the energy imparted to the footplate was 1.2 to 1.7 Joules, which was the same for SFS. In most cases, complete mobility of the posterior footplate required division of a small remaining bridge of bone at the superior or inferior annular rim using an oval window pick. Once completely mobilized, the footplate was covered with a tissue seal (Fig. 1). CSM was performed for cases of congenital stapes fixation or ossified tympanosclerosis after CT scanning excluded other abnormalities, including temporal bone anomalies that could be associated with stapes gusher. In adults, this procedure was preferred if there were a past history of otitis media to avoid placing a prosthesis that could present some problems if otitis were to recur in the future. Adults had to be free of any otitis media for greater than 2 years and free of any major chronic ear problems for greater than 5 years to be selected for any of the procedures (CSM, STAMP, or SFS). The operation commenced as for SFS surgery. Wide exposure of the stapes footplate was necessary to do the procedure. The annular margins of the footplate should be visible nearly circumferentially. Oval window straight and angled picks were used to manually create a groove through the bony fixation restoring an annulus circumferentially (Figs. 2 and 3, AYC). The picks were applied with gentle pressure and meticulously drawn
MATERIALS AND METHODS Two hundred seventy-seven consecutive primary operations for stapes fixation were performed by the senior author (D. S. P.) from 1997 to 2007, and the charts were reviewed retrospectively. The study was approved by the institutional review board at Children’s Hospital Boston M08-02-0076 and was conducted in accordance with their guidelines.
Patient Selection and Surgical Methods Patients had clinical evidence of stapes fixation without any history of otologic surgery. For cases of otosclerosis, STAMP and CSM was discussed preoperatively as an option that could be exercised in the event that the otosclerotic focus would be limited to the anterior one third of the footplate, and the anatomy would be favorable. SFS was performed for the remaining cases that had more extensive otosclerosis or anatomic contraindications to STAMP or CSM. Conventional stapedotomy was conducted through a transcanal approach. An argon (HGM) or diode-pumped KTP laser (Iridex Corp., Mountain View, CA, USA) delivered through 200-Km fibers was used to burn a small fenestra in the footplate. Typically, a rosette pattern of 5 to 7 burns was made with settings of 1.2 W, 200 milliseconds (total energy imparted = 1.2 Y 1.7 Joules). In the majority of cases, a Skeeter drill (Medtronics, Jacksonville, FL, USA) with a 0.7-mm diamond burr was used to complete the
FIG. 2. Diagram of the circumferential stapes mobilization procedure. The annular ligament failed to form or has been covered with ossified tympanosclerosis. A circumferential trough has been created in the location of the annular ligament to restore mobility of the footplate.
Otology & Neurotology, Vol. 35, No. 4, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
STAMP, CSM, AND SFS FOR STAPES FIXATION
e125
FIG. 3. A, Intraoperative photo of a left circumferential stapes mobilization procedure. The larger arrow indicates an ossification across the would-be location of the posterior superior annular ligament, and there is no visible trace of a fibrous ligament. The smaller arrow shows a trace of fibrous ligament in the otherwise ossified anterior superior rim. B, Mobilization of the stapes using an oval window pick to create a fine bony trough. C, Postmobilization of the stapes.
over the location of the annulus, slowly and carefully removing bone and creating a narrow circumferential trench. The vestibule was not entered until the trench was uniformly deep around the footplate. Complete mobilization of the footplate was attempted and the footplate covered with a tissue seal followed by assessment of the mobility by viewing the round window reflex as above. Laser techniques were not used for CSM as the annulus was thick in these cases and would have required excessive amounts of energy to accomplish mobilization. Pure tone air conduction (AC) and bone conduction (BC) audiometry with appropriate masking were performed preoperatively and postoperatively (usually 3Y6 wk) and the most recent audiogram that was 12 months or greater. The hearing outcomes data are reported according to AAO-HNS Committee guidelines (14). Pure tone averages (PTAs) were calculated using 4-tone averages (0.5, 1, 2, and 4 kHz). The mean ABG was calculated as the air PTA minus the bone PTA using 4-tone averages (0.5, 1, 2, and 4 kHz). When a negative ABG value was calculated, signifying that BC threshold was falsely greater than the AC threshold, the value obtained for the AC threshold was used for both fields. The results are expressed as mean with standard deviation (SD). Categorical variables were analyzed using W2 tests or Fisher’s exact test, and continuous variables were analyzed using analysis of variance among the 3 treatment groups. Multivariate model was used to evaluate the differences of the 3 treatments at the same time adjusting for the effect of age and sex. All the tests
TABLE 1.
Patients’ characteristics
Small Stapedotomy Circumferential fenestra minus stapes stapedotomy prosthesis mobilization No. of ears Sex Male Female Mean age (standard deviation) Surgical ear Right Left
67 (68%)
16 (16%)
16 (16%)
18 (27%) 49 (73%) 44 (13.0)
6 (37%) 10 (63%) 44 (14.3)
11 (69%) 5 (31%) 18 (14.1)
37 (55%) 30 (45%)
8 (50%) 8 (50%)
8 (50%) 8 (50%)
p
0.007 G0.0001
0.89
were 2 sided with Type I error 0.05. The analyses were conducted in SAS 9.3.
RESULTS Demographics are presented in Table 1. Only 0.6-mm diameter piston prosthesis were included to this study, so the final number of cases were 196, and cases in each subgroups were 149 SFS, 25 STAMP, and 22 CSM. Audiologic 1-year minimum follow-up data were available from 90 patients (99 cases). Thus, final number of
TABLE 2.
Hearing results
Small Stapedotomy Circumferential fenestra minus stapes stapedotomy prosthesis mobilization A/B gap PTA 29.2 (11.0) preoperative, dB mean (standard deviation) A/B gap PTA 11 (6.6) postoperative, dB mean (standard deviation) A/B gap PTA recent, dB Mean 7.1 (6.0) (standard deviation) Air PTA delta, dB Mean 24.6 (12.3) (standard deviation) Bone PTA delta, dB Mean 1.9 (7.0) (standard deviation)
p
29.1 (7.8)
34.0 (8.9)
0.24
3.1 (5.9)
17.7 (12.8)
G0.0001
3.8 (5.8)
19.8 (8.2)
G0.0001
29.9 (10.9)
18.0 (11.0)
0.02
4.3 (3.7)
3.8 (6.3)
0.32
Delta means preoperative value minus recent value. Otology & Neurotology, Vol. 35, No. 4, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
e126
G. O. ACAR ET AL.
FIG. 4. Average air conduction and bone conduction thresholds and air-bone gaps for preoperative, postoperative (3Y6 wk), and most recent (912 mo) audiograms in patients who underwent procedure.
cases in each group was 67 SFS, 16 STAMP, and 16 CSM, respectively. Improvement of hearing results was statistically significant in all groups (Table 2; Fig. 4). Air conduction improvement from preoperative to recent results was most significant in the STAMP group (29.9 dB), then the SFS group (24.6 dB) and least in the CSM group (18.0 dB); differences between groups are statistically significant (analysis of variance, p = 0.02). Overclosure, bone conduction improvement was seen in all groups. The largest improvement was in the STAMP group (4.3 dB), although it was not statistically significantly better than in other groups. ABG improvement was statistically largest in the STAMP group (25.3 dB, p = 0.01). In most recent audiogram, ABG was below 10 dB in 88% of STAMP group, in 81% of SFS group, and in 17% of CSM group, see Figure 5. The chorda tympani were preserved at a rate of 98% in the study. PTA BC thresholds worsened greater than 5 dB in 5 cases in the SFS group, in 1 case in the CSM group, and in none in the STAMP group. There was 1 case of delayed onset anacusis with persistent disequilibrium that occurred in a STAMP patient, 1 (4%) of 25, which was 1 (0.36%) of 277 for the entire series.
The revision surgery rate for recurrence of conductive hearing loss was 8.3% (2/24 cases) for the STAMP procedure (excluding the anacusis case from the denominator), 0.67% (1/149) for SFS, and 4.54% (1/22) for CSM. In the STAMP cases, conductive hearing loss recurred at 3 and 5 years in the 2 cases, and both were revised shortly afterward. In the CSM case, who was an adult, there was initial improvement after removal of ossified obliterative
FIG. 5. Most recent hearing results, air-bone gaps, in different groups.
Otology & Neurotology, Vol. 35, No. 4, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
STAMP, CSM, AND SFS FOR STAPES FIXATION tympanosclerosis, but refixation occurred within 3 months. A total stapedectomy was subsequently successfully accomplished. There were 3 cases in which STAMP was initially attempted but could not be completed, and the operation was converted to SFS. In 2 of these cases, there was inability to fully separate the footplate at either the superior or inferior annulus. In the remaining case, the posterior crus separated from the footplate during the division of the anterior crus. There were 2 cases (2/24; 8%) in which CSM was intended, but the footplate exposure was inadequate to perform the procedure, and it was terminated, followed by recommendation of a hearing aid. All patients were asked postoperatively whether they experienced any symptoms of disequilibrium or head movementYinduced vertigo at all, and affirmative responses were reported as postoperative vertigo. There were no cases of severe vertigo, and the only case of persistent disequilibrium occurred in the anacusis STAMP case above. The rate of postoperative vertigo was significantly lower in CSM patients (0 cases) than in STAMP (5/16, 31%) or SFS patients (25/67, 38%) (p = 0.005). The median duration of postoperative vertigo was 2 days in the STAMP group, and 7 days in the SFS group (Table 2). Three months postoperatively, 1 (0.67%) of 149 SFS patient developed vertigo with straining and with positive or negative insufflation of the tympanic membrane without nystagmus. Middle ear exploration for fistula was negative, but there were considerable adhesions around the prosthesis and footplate that were lysed with subsequent significant relief of the symptoms.
DISCUSSION Enduring mobilization has been reported in the temporal bone histology from a patient who had undergone bilateral stapes mobilization for tympanosclerosis (15). Another temporal bone from a patient who had a successful stapes mobilization showed that the fracture in the footplate had occurred posterior to the anteriorly situated otosclerotic focus and that the enchondral bone of the footplate had healed with a fibrous union. The long-term success of many mobilizations with the anterior crurotomy technique may be due to a similar mechanism, but the fracture location was difficult to control, and the duration of success was variable (16,17). Malfunctioning or displacement of the prosthesis has been consistently identified as the most common reason for failure (980%) of small fenestra stapedotomy (SFS) and stapedectomy (18,19) and as reported by Fisch et al. (20) and Lesinski (21). Association with incus erosion was noted in most cases (80%) (21). Avoidance of a prosthesis is especially desirable in children who have a higher risk of developing otitis media in the future than in adults. CSM required a lesser opening of the vestibule than STAMP or SFS with a minimal circumferential trough to reconstruct an annulus. The rate of postoperative vertigo was less than in either STAMP or SFS, which
e127
were comparable between them. These procedures preserve most of the anatomy of the stapes and most of the footplate and annular ligament, which may provide some advantage to hearing results (22). The most common reason for failure in STAMP and CSM was refixation. The success of the laser STAMP procedure depends on the proper selection of cases in which the otosclerosis is limited to the anterior footplate. The degree of conductive hearing loss is not reliably predictive of the size of the otosclerotic focus. Merchant et al. (18) noted that limited otosclerosis does not translate into more limited conductive hearing loss because the anterior otosclerotic focus can displace the footplate posteriorly to impact on the posterior stapediovestibular joint (SVJ) and cause hearing loss out of proportion to the size of the otosclerotic focus. As expected from their results, the preoperative audiograms of our laser STAMP patients showed a trend that did not rise to statistical significance for less severe conductive hearing loss on average when compared with SFS patients. Furthermore, the footplate laser cut requires wide exposure of the oval window niche that must be judged at the time of surgery. Therefore, determining candidacy for a STAMP remains an intraoperative decision (10,11). The important factors for long-term success of partial stapedectomy procedures are as follows: 1) making the laser cut through healthy footplate enchondral bone that is likely to heal with a fibrous union and not through an otosclerotic focus, which would have a high risk of refixation; 2) lack of expansion of the otosclerotic focus or regrowth of footplate bone that could refix the footplate; and 3) lack of bony ankylosis or otosclerosis involving the posterior SVJ. In the temporal bone study by Merchant et al., 38% of temporal bones had neither otosclerosis nor ankylosis of the posterior SVJ, suggesting that a laser STAMP, anterior crurotomy, or a similar partial stapedectomy procedure on this subset of bones would be expected to have a lasting successful outcome (18). CSM is a technically challenging procedure that requires patient, slow removal of bone circumferentially around the perimeter of the footplate. It is useful in cases of congenital fixation of the stapes, which would be expected to heal with a fibrous union in most cases. It can also be used for cases of tympanosclerosis with bony fixation if the history of otitis media is reasonably remote and unlikely to reactivate, which would risk regrowth of the tympanosclerosis with refixation. CSM is a surgical option for younger children with congenital fixation in whom it is preferred to avoid a prosthesis until they are beyond the otitis media prone years. Complete mobilization of the stapes footplate is desirable to achieve optimal results but partial mobilization did significantly improve the severity of conductive hearing loss. Five-year follow-up results of STAMP reported by Silverstein showed a refixation rate of 9%, which was similar to the 8% in the present study (12). Successful SFS was accomplished in these 2 cases. When refixation occurs, conventional stapedotomy can be performed nearly as if it were a primary stapes operation as the central footplate remains unviolated. SFS after STAMP has shown Otology & Neurotology, Vol. 35, No. 4, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
e128
G. O. ACAR ET AL.
results comparable to primary stapedotomy (9Y12). Similarly, CSM does not involve the central portion of the footplate and leaves open the option for SFS if indicated at a later date. The recurrent surgery rates for STAMP (8%) and CSM (4.5%) were higher than for SFS (1.3%), but these less invasive procedures offer advantages that make these higher rates acceptable. Particularly with CSM, the operation may be offered to patients younger than would ordinarily be considered for SFS, especially if there has been any past history of otitis media. Hearing Results STAMP results were similar or even better than previously reported (9Y12). The observed incidence of anacusis was 4.0%, and it should be noted that the true incidence may lie within the 95% confidence interval from essentially 0% up to 11.8%. This unfortunate case underscores that complications can occur whenever the vestibule is opened, even in minimally invasive procedures. Profound hearing loss has also been previously reported with STAMP, 3.6% (2/55), which is similar to the present data (10). It has been speculated that patients with a limited extent of otosclerotic disease for which STAMP is an appropriate approach may have better preoperative hearing results, and this was seen in Silverstein’s studies. In our patients, however, preoperative hearing results were equal in the STAMP and SFS groups. To our knowledge, there are no reports of CSM in the literature with which to compare results. AC PTA and AC thresholds at 4 and 8 kHz significantly improved postoperatively but did not significantly improve further by more than 12 months. Postoperative results in SFS group were comparable to other series in the literature (6,8,10,19). Hearing improvements are typically better in the low to mid frequencies than in the high frequencies (4Y8 kHz), and this was also noted in the present study. Some worsening of the 4 kHz BC is common after SFS and tends to improve over time but not always to preoperative levels. It is not clear if this effect represents sensorineural injury or a mechanical disadvantage of the reduced vibrating surface area of the piston compared with an intact footplate. AC at 500 Hz and 2 kHz improved between the postoperative and 12 months or greater results, indicating that the healing process likely causes some mechanical advantage to the hearing, possibly from fibrosis tightening the fixation to the incus. ABG gap closure was similar between STAMP and SFS, but CSM results were worse because of the partially mobilized cases. Significant improvement was achieved in these partially mobilized cases such that it was effective in reducing or eliminating the need for hearing aids for these individuals, who were not otherwise good candidates for SFS. Average overclosure, bone conduction improvements was better in STAMP and CSM patients than in SFS patients. Overall, for 12 months, AC thresholds for STAMP were better than the other groups. Low-frequency AC thresholds were worse in the CSM than the other groups
reflecting that complete circumferential mobilization is necessary to yield hearing comparable to STAMP and SFS. AC thresholds at 4 and 8 kHz were better in the CSM and STAMP groups than in the SFS group postoperatively and more than 12 months’ follow-up. These results supported our hypothesis that the increased vibratory surface area of the footplate may be advantageous over a piston reconstruction for high-frequency hearing. For 12 months’ follow-up, the AC results were affected by the cases of refixation in STAMP and CSM. STAMP and CSM are technically more challenging procedures than SFS because they require operating on the footplate while the superstructure is still in place. Having previously practiced the procedures in the temporal bone lab, the learning curve for becoming comfortable with the new techniques was between 5 to 10 cases for each of these operations. Patients should be selected for favorable exposure of the stapes footplate. In the case of STAMP, otosclerosis should be limited to the anterior one third of the footplate, and the risk of refixation would be to be reduced if patients are older than 40 years. CSM can be done on a thick annulus, but care must be taken to bevel the trough, making it wider laterally, to prevent limitation of the mobilized footplate by friction with the thick walls of the surrounding otic capsule. The less invasive techniques have the advantage that they preserve nearly all of the functional anatomic structures. The stapedius tendon and most of the footplate are preserved. Late failures of STAMP and CSM may still be treatable with conventional stapedotomy techniques and with results similar to primary cases. Failed stapedotomy cases often have incus necrosis that will require reconstructive techniques that may not yield results as favorable as in primary cases. Limitations of the Study The conclusions from this study are significantly limited by selection bias including the ages of patients, pathology, and procedure. The types of procedures to be performed were determined beforehand or intraoperatively. By intention, the median age of the CSM cases was younger than for the other procedures, and most of the CSM cases were performed for congenital fixation and not for otosclerosis. However, the audiograms would be expected to be reasonably reproducible and not subject to significant age-related differences in the accuracy of responses to behavioral testing as all of the patients were 6 years or older. Therefore, one would not anticipate that there would be a measurable age-related bias when comparing audiograms from these different age groups. In fact, the multivariate analysis adjusted for age did not alter the results. Stapes fixation of all types results in conductive hearing loss that is diagnostically indistinguishable based on etiology. The purpose of this study was to present the feasibility of the two other options, STAMP or CSM, to give surgeons additional choices to address fixation due to any of these causes; hence, comparisons between groups were made despite the differences in pathology.
Otology & Neurotology, Vol. 35, No. 4, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
STAMP, CSM, AND SFS FOR STAPES FIXATION Cases with otosclerosis who underwent SFS had either larger disease foci or unfavorable anatomy that precluded STAMP. Despite these intended differences, the study was able to demonstrate that the newer procedures are feasible and, with these results, are acceptable alternatives to SFS. CSM expands the indications for stapes surgery without adding undue risks. There may be an added benefit to hearing in the higher frequencies with STAMP and CSM, but these data cannot be considered conclusive because of the selection bias. The outcomes of SFS versus STAMP could only be truly compared by a randomized comparison trial, which treated similarly sized anterior otosclerotic foci. CONCLUSION Techniques are available for less invasive methods to mobilize the stapes in cases of otosclerosis and other types of fixation. Avoidance of a prosthesis has the advantages of eliminating incus necrosis and other piston-related complications. CSM may be performed in younger children and in patients who have had a history of otitis media. Refixation rates are sufficiently low with these techniques that they are reasonable alternatives to conventional stapedotomy and stapedectomy procedures. Additional experience in the hands of other surgeons and in larger studies with long term follow-up is needed. Acknowledgments: The authors thank Ahmet Dirican, M.D, professor, Biostatistic Department of Cerrahpasa School of Medicine of Istanbul University; and Hasan YananlN, M.D., associate professor, Pharmacology Department of Marmara University, for the preliminary statistical analyses of this project. The authors also thank Tali Rasooly, research assistant at Boston Children’s Hospital, for the help in the preparation of this article.
REFERENCES 1. Rosen S. Mobilization of the stapes to restore hearing in otosclerosis. N Y J Med 1953;53:2650Y3. 2. Rosen S. Rosen mobilization and stapedectomy: long time results with special reference to high frequencies. Arch Otolaryngol 1969;89:241Y4.
e129
3. Fowler EP Jr. Anterior crurotomy and mobilization of the ankylozed stapes footplate: introduction to motion picture demonstration. Acta Otolaryngol 1956;46:319Y22. 4. Fowler EP Jr. Anterior crurotomy with footplate fracture. Arch Otolaryngol 1960;71:296Y304. 5. Shea JJ Jr. Fenestration of the oval window. Ann Otol Rhinol Laryngol 1958;67:932Y51. 6. Fisch U. Stapedotomy versus stapedotomy. Am J Otol 1982;4:112Y7. 7. Palva T. Argon laser in otosclerosis surgery. Acta Otolaryngol 1979;104:153Y7. 8. Perkins RC. Laser stapedotomy for otosclerosis. Laryngoscope 1980;90:228Y40. 9. Silverstein H. Laser stapedotomy minus prosthesis (laser STAMP): minimally invasive procedure. Am J Otol 1998;19:277Y87. 10. Silverstein H, Hoffmann KK, Thompson JH, Rosenberg SI, Sleeper JP. Hearing outcome of laser stapedotomy minus prosthesis (STAMP) versus conventional laser stapedotomy. Otol Neurotol 2004;25:106Y11. 11. Poe DS. Laser-assisted endoscopic stapedectomy: a prospective study. Laryngoscope 2000;110(5 Pt 2):1Y37. 12. Silverstein H, Jackson LE, Conlon WS, et al. Laser stapedotomy minus prosthesis (laser STAMP): absence of refixation. Otol Neurotol 2002;23:152Y7. 13. De la Cruz A, Angeli S, Slattery WH. Stapedectomy in children. Otolaryngol Head Neck Surg 1999;120:487Y92. 14. American Academy of OtolaryngologyYHead & Neck Surgery Foundation, Inc. Committee on Hearing and Equilibrium guidelines for the evaluation of results if treatment of conductive hearing loss. Otolaryngol Head Neck Surg 1995;113:186Y7. 15. Thiel G, Mills R. Rosen mobilization of the stapes: does it have a place modern otology? J Laryngol Otol 2006;120:1067Y71. 16. Jaisignhani VJ, Sxhachern PA, Paparella MM. Stapes mobilization in otosclerosis. Ear Nose Throat J 2001;80:586Y90. 17. Myers EN, Ishiyama E, Heisse JW. Histology of a successful stapes mobilization. Ann Otol Rhinol Laryngol 1970;79:321Y30. 18. Merchant SN, Incesulu A, Glynn RJ, Nadol JB. Histologic studies of the posterior stapediovestibularjoint in otosclerosis. Otol Neurotol 2001;22:305Y10. 19. Silverstein H, Hester TO, Rosenberg SI, Deems D. Preservation of the stapedius tendon in laser stapes surgery. Laryngoscope 1998;108:1453Y8. 20. Fisch U, Acar GO, Huber MH. Malleostapedotomy in revision surgery for otosclerosis. Otol Neurotol 2001;22:776Y85. 21. Lesinski SG. Causes of conductive hearing loss after stapedectomy or stapedotomy: a prospective study of 279 consecutive surgical revisions. Otol Neurotol 2002;23:281Y8. 22. Causse JB, Gherini S, Lopez A, Juberthie L, Olivier JC, Bastianelli G. Impedance transfer: acoustic impedance of the annular ligament and stapedial tendon reconstruction in otosclerosis surgery. Am J Otol 1993;14:613Y7.
Otology & Neurotology, Vol. 35, No. 4, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
Comparison of Stapedotomy Minus Prosthesis, Circumferential Stapes Mobilization, and Small Fenestra Stapedotomy for Stapes Fixation *†Gu¨l Ozbilen Acar, *‡Ilkka Kiveka¨s, *§Bassem M. Hanna, *Lin Huang, *kQuinton Gopen, and *Dennis S. Poe *Department of Otolaryngology & Communication Enhancement, Boston Children’s Hospital and Harvard Medical School, Boston, Massachusetts, U.S.A.; ÞIstanbul Goztepe Training and Research Hospital, Ear Nose Throat Department, Istanbul, Turkey; þDepartment of Otorhinolaryngology, Tampere University Hospital and the University of Tampere, Tampere, Finland; §Department of Otolaryngology, University of Ottawa, Ottawa, Canada; and kDepartment of Otolaryngology, University of California Los Angeles, California, U.S.A.
Objective: To compare the outcomes of 3 surgical techniques for primary stapes fixation: stapedotomy minus prosthesis (STAMP), circumferential stapes mobilization (CSM), and small fenestra stapedotomy (SFS). Study Design: Retrospective review of 277 primary cases operated for stapes fixation from 1997 to 2007. Setting: Tertiary academic center. Patients: Consecutive adult and pediatric cases operated for conductive hearing loss because of stapes fixation. Interventions: STAMP was performed for otosclerosis limited to the anterior footplate, CSM was conducted for congenital stapes fixation, SFS was performed for more extensive otosclerosis or anatomic contraindications to STAMP/CSM. Main Outcome Measures: Pure-tone audiometry was performed preoperatively and postoperatively (3Y6 wk) and the most recent long-term results (Q12 mo). Results: Ninety-nine ears in 90 patients had audiologic followup data over 12 months. Sixty-seven ears (68%) underwent SFS, 16 (16%) STAMP, and 16 (16%) CSM. There was significant
improvement in average air conduction (AC) thresholds and airbone gap (ABG) for all techniques. Mean ABG for SFS closed from 29 to 7.1 dB (SD, 6.0), for STAMP from 29 to 3.8 dB (SD, 5.8 dB), and for CSM from 34 to 20 dB (SD, 8.2 dB). AC results were better in the STAMP than in the SFS group, especially in high frequencies. Bone conduction improvements were seen in all groups, highest in STAMP (4.3 dB) and CSM (3.8 dB) groups, but the differences between groups were not statistically significant. Conclusion: Satisfactory hearing results were achieved with all the techniques, and STAMP showed better hearing outcomes, especially in high frequencies. CSM is a good option for children and patients in whom it is desirable to avoid a footplate fenestration or prosthesis. CSM and STAMP had significantly higher rates of revision for refixation than SFS. Key Words: Circumferential stapes mobilizationVHearing outcomeVOtosclerosisVSmall fenestra stapedotomyVStapedotomy minus prosthesisVStapes fixation. Otol Neurotol 35:e123Ye129, 2014.
Minimally invasive stapedotomy procedures date back to 1952 and include mobilization, anterior crurotomy, and other stapedioplasty techniques that were designed to preserve some portion of the crura and footplate (1Y6). Initially, it was difficult to reliably control the osteotomies, but with the introduction of the laser and microdrill, the surgical precision has been greatly improved (7,8). Laser stapedotomy minus prosthesis (STAMP) was developed in
1998 for limited anterior otosclerosis in which the anterior crus was divided, and a linear separation in the footplate was created between the anterior one third and posterior two thirds. High-frequency hearing was more favorable than conventional laser stapedotomy (10Y12). Children with bilateral congenital fixation of the stapes or severe tympanosclerosis present special challenges in that they are too young to make their own informed consent, they have a higher incidence of increased perilymph pressure (13), and they have a higher risk of subsequent otitis media. A minimally invasive circumferential footplate mobilization procedure could be beneficial in such cases to improve hearing, yet keep risks reasonably low. The increased surface area of the mobilized footplate
Address correspondence and reprint requests to Dennis S. Poe, M.D., Department of Otolaryngology & Communication Enhancement, Boston Children’s Hospital, 333 Longwood Avenue, Lo-367, Boston, MA 02115; E-mail: [email protected]
e123
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
e124
G. O. ACAR ET AL.
FIG. 1. Diagram of STAMP procedure with the otosclerotic focus involving the anterior one third of the stapes footplate. A laser was used to divide the anterior crus and the footplate transversely across the junction between its anterior one third and the posterior two-thirds portions, mobilizing the remaining posterior footplate.
might be expected to yield advantageous high-frequency hearing results as well. The purpose of this study was to investigate the outcomes of minimally invasive stapes surgical techniques, including circumferential stapes mobilization (CSM) and STAMP. We wanted to test the following hypotheses: 1) preservation of the stapes footplate as a vibrating surface should aid in the high-frequency hearing results compared with conventional small fenestra stapedotomy (SFS) and 2) CSM and STAMP procedures provide acceptable risks and reasonable short- and long-term hearing results compared with SFS.
fenestra. Piston MEEI prosthesis of 0.6-mm diameter was used for reconstruction. The fenestra was sealed around the piston by placing tragal perichondrium covered with gelfoam soaked in saline in the earlier cases, and in recent years, a meatal tissue graft was used, initially covered with gelfoam and, since 2005, without any gelfoam. The laser STAMP procedure has been previously described in detail (11). Using the same approach as for SFS, an assessment of the extent of the otosclerotic focus was made. If the focus was limited to the anterior one third of the footplate and if the oval window niche was sufficiently wide to afford a view of the footplate under the anterior crus, then a STAMP was performed. The laser was used to initially divide the anterior crus; then, the footplate was divided transversely across the junction between its anterior one third and the posterior two thirds. Laser exposure was typically 5 to 7 burns using the same settings as for SFS. Consequently, the energy imparted to the footplate was 1.2 to 1.7 Joules, which was the same for SFS. In most cases, complete mobility of the posterior footplate required division of a small remaining bridge of bone at the superior or inferior annular rim using an oval window pick. Once completely mobilized, the footplate was covered with a tissue seal (Fig. 1). CSM was performed for cases of congenital stapes fixation or ossified tympanosclerosis after CT scanning excluded other abnormalities, including temporal bone anomalies that could be associated with stapes gusher. In adults, this procedure was preferred if there were a past history of otitis media to avoid placing a prosthesis that could present some problems if otitis were to recur in the future. Adults had to be free of any otitis media for greater than 2 years and free of any major chronic ear problems for greater than 5 years to be selected for any of the procedures (CSM, STAMP, or SFS). The operation commenced as for SFS surgery. Wide exposure of the stapes footplate was necessary to do the procedure. The annular margins of the footplate should be visible nearly circumferentially. Oval window straight and angled picks were used to manually create a groove through the bony fixation restoring an annulus circumferentially (Figs. 2 and 3, AYC). The picks were applied with gentle pressure and meticulously drawn
MATERIALS AND METHODS Two hundred seventy-seven consecutive primary operations for stapes fixation were performed by the senior author (D. S. P.) from 1997 to 2007, and the charts were reviewed retrospectively. The study was approved by the institutional review board at Children’s Hospital Boston M08-02-0076 and was conducted in accordance with their guidelines.
Patient Selection and Surgical Methods Patients had clinical evidence of stapes fixation without any history of otologic surgery. For cases of otosclerosis, STAMP and CSM was discussed preoperatively as an option that could be exercised in the event that the otosclerotic focus would be limited to the anterior one third of the footplate, and the anatomy would be favorable. SFS was performed for the remaining cases that had more extensive otosclerosis or anatomic contraindications to STAMP or CSM. Conventional stapedotomy was conducted through a transcanal approach. An argon (HGM) or diode-pumped KTP laser (Iridex Corp., Mountain View, CA, USA) delivered through 200-Km fibers was used to burn a small fenestra in the footplate. Typically, a rosette pattern of 5 to 7 burns was made with settings of 1.2 W, 200 milliseconds (total energy imparted = 1.2 Y 1.7 Joules). In the majority of cases, a Skeeter drill (Medtronics, Jacksonville, FL, USA) with a 0.7-mm diamond burr was used to complete the
FIG. 2. Diagram of the circumferential stapes mobilization procedure. The annular ligament failed to form or has been covered with ossified tympanosclerosis. A circumferential trough has been created in the location of the annular ligament to restore mobility of the footplate.
Otology & Neurotology, Vol. 35, No. 4, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
STAMP, CSM, AND SFS FOR STAPES FIXATION
e125
FIG. 3. A, Intraoperative photo of a left circumferential stapes mobilization procedure. The larger arrow indicates an ossification across the would-be location of the posterior superior annular ligament, and there is no visible trace of a fibrous ligament. The smaller arrow shows a trace of fibrous ligament in the otherwise ossified anterior superior rim. B, Mobilization of the stapes using an oval window pick to create a fine bony trough. C, Postmobilization of the stapes.
over the location of the annulus, slowly and carefully removing bone and creating a narrow circumferential trench. The vestibule was not entered until the trench was uniformly deep around the footplate. Complete mobilization of the footplate was attempted and the footplate covered with a tissue seal followed by assessment of the mobility by viewing the round window reflex as above. Laser techniques were not used for CSM as the annulus was thick in these cases and would have required excessive amounts of energy to accomplish mobilization. Pure tone air conduction (AC) and bone conduction (BC) audiometry with appropriate masking were performed preoperatively and postoperatively (usually 3Y6 wk) and the most recent audiogram that was 12 months or greater. The hearing outcomes data are reported according to AAO-HNS Committee guidelines (14). Pure tone averages (PTAs) were calculated using 4-tone averages (0.5, 1, 2, and 4 kHz). The mean ABG was calculated as the air PTA minus the bone PTA using 4-tone averages (0.5, 1, 2, and 4 kHz). When a negative ABG value was calculated, signifying that BC threshold was falsely greater than the AC threshold, the value obtained for the AC threshold was used for both fields. The results are expressed as mean with standard deviation (SD). Categorical variables were analyzed using W2 tests or Fisher’s exact test, and continuous variables were analyzed using analysis of variance among the 3 treatment groups. Multivariate model was used to evaluate the differences of the 3 treatments at the same time adjusting for the effect of age and sex. All the tests
TABLE 1.
Patients’ characteristics
Small Stapedotomy Circumferential fenestra minus stapes stapedotomy prosthesis mobilization No. of ears Sex Male Female Mean age (standard deviation) Surgical ear Right Left
67 (68%)
16 (16%)
16 (16%)
18 (27%) 49 (73%) 44 (13.0)
6 (37%) 10 (63%) 44 (14.3)
11 (69%) 5 (31%) 18 (14.1)
37 (55%) 30 (45%)
8 (50%) 8 (50%)
8 (50%) 8 (50%)
p
0.007 G0.0001
0.89
were 2 sided with Type I error 0.05. The analyses were conducted in SAS 9.3.
RESULTS Demographics are presented in Table 1. Only 0.6-mm diameter piston prosthesis were included to this study, so the final number of cases were 196, and cases in each subgroups were 149 SFS, 25 STAMP, and 22 CSM. Audiologic 1-year minimum follow-up data were available from 90 patients (99 cases). Thus, final number of
TABLE 2.
Hearing results
Small Stapedotomy Circumferential fenestra minus stapes stapedotomy prosthesis mobilization A/B gap PTA 29.2 (11.0) preoperative, dB mean (standard deviation) A/B gap PTA 11 (6.6) postoperative, dB mean (standard deviation) A/B gap PTA recent, dB Mean 7.1 (6.0) (standard deviation) Air PTA delta, dB Mean 24.6 (12.3) (standard deviation) Bone PTA delta, dB Mean 1.9 (7.0) (standard deviation)
p
29.1 (7.8)
34.0 (8.9)
0.24
3.1 (5.9)
17.7 (12.8)
G0.0001
3.8 (5.8)
19.8 (8.2)
G0.0001
29.9 (10.9)
18.0 (11.0)
0.02
4.3 (3.7)
3.8 (6.3)
0.32
Delta means preoperative value minus recent value. Otology & Neurotology, Vol. 35, No. 4, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
e126
G. O. ACAR ET AL.
FIG. 4. Average air conduction and bone conduction thresholds and air-bone gaps for preoperative, postoperative (3Y6 wk), and most recent (912 mo) audiograms in patients who underwent procedure.
cases in each group was 67 SFS, 16 STAMP, and 16 CSM, respectively. Improvement of hearing results was statistically significant in all groups (Table 2; Fig. 4). Air conduction improvement from preoperative to recent results was most significant in the STAMP group (29.9 dB), then the SFS group (24.6 dB) and least in the CSM group (18.0 dB); differences between groups are statistically significant (analysis of variance, p = 0.02). Overclosure, bone conduction improvement was seen in all groups. The largest improvement was in the STAMP group (4.3 dB), although it was not statistically significantly better than in other groups. ABG improvement was statistically largest in the STAMP group (25.3 dB, p = 0.01). In most recent audiogram, ABG was below 10 dB in 88% of STAMP group, in 81% of SFS group, and in 17% of CSM group, see Figure 5. The chorda tympani were preserved at a rate of 98% in the study. PTA BC thresholds worsened greater than 5 dB in 5 cases in the SFS group, in 1 case in the CSM group, and in none in the STAMP group. There was 1 case of delayed onset anacusis with persistent disequilibrium that occurred in a STAMP patient, 1 (4%) of 25, which was 1 (0.36%) of 277 for the entire series.
The revision surgery rate for recurrence of conductive hearing loss was 8.3% (2/24 cases) for the STAMP procedure (excluding the anacusis case from the denominator), 0.67% (1/149) for SFS, and 4.54% (1/22) for CSM. In the STAMP cases, conductive hearing loss recurred at 3 and 5 years in the 2 cases, and both were revised shortly afterward. In the CSM case, who was an adult, there was initial improvement after removal of ossified obliterative
FIG. 5. Most recent hearing results, air-bone gaps, in different groups.
Otology & Neurotology, Vol. 35, No. 4, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
STAMP, CSM, AND SFS FOR STAPES FIXATION tympanosclerosis, but refixation occurred within 3 months. A total stapedectomy was subsequently successfully accomplished. There were 3 cases in which STAMP was initially attempted but could not be completed, and the operation was converted to SFS. In 2 of these cases, there was inability to fully separate the footplate at either the superior or inferior annulus. In the remaining case, the posterior crus separated from the footplate during the division of the anterior crus. There were 2 cases (2/24; 8%) in which CSM was intended, but the footplate exposure was inadequate to perform the procedure, and it was terminated, followed by recommendation of a hearing aid. All patients were asked postoperatively whether they experienced any symptoms of disequilibrium or head movementYinduced vertigo at all, and affirmative responses were reported as postoperative vertigo. There were no cases of severe vertigo, and the only case of persistent disequilibrium occurred in the anacusis STAMP case above. The rate of postoperative vertigo was significantly lower in CSM patients (0 cases) than in STAMP (5/16, 31%) or SFS patients (25/67, 38%) (p = 0.005). The median duration of postoperative vertigo was 2 days in the STAMP group, and 7 days in the SFS group (Table 2). Three months postoperatively, 1 (0.67%) of 149 SFS patient developed vertigo with straining and with positive or negative insufflation of the tympanic membrane without nystagmus. Middle ear exploration for fistula was negative, but there were considerable adhesions around the prosthesis and footplate that were lysed with subsequent significant relief of the symptoms.
DISCUSSION Enduring mobilization has been reported in the temporal bone histology from a patient who had undergone bilateral stapes mobilization for tympanosclerosis (15). Another temporal bone from a patient who had a successful stapes mobilization showed that the fracture in the footplate had occurred posterior to the anteriorly situated otosclerotic focus and that the enchondral bone of the footplate had healed with a fibrous union. The long-term success of many mobilizations with the anterior crurotomy technique may be due to a similar mechanism, but the fracture location was difficult to control, and the duration of success was variable (16,17). Malfunctioning or displacement of the prosthesis has been consistently identified as the most common reason for failure (980%) of small fenestra stapedotomy (SFS) and stapedectomy (18,19) and as reported by Fisch et al. (20) and Lesinski (21). Association with incus erosion was noted in most cases (80%) (21). Avoidance of a prosthesis is especially desirable in children who have a higher risk of developing otitis media in the future than in adults. CSM required a lesser opening of the vestibule than STAMP or SFS with a minimal circumferential trough to reconstruct an annulus. The rate of postoperative vertigo was less than in either STAMP or SFS, which
e127
were comparable between them. These procedures preserve most of the anatomy of the stapes and most of the footplate and annular ligament, which may provide some advantage to hearing results (22). The most common reason for failure in STAMP and CSM was refixation. The success of the laser STAMP procedure depends on the proper selection of cases in which the otosclerosis is limited to the anterior footplate. The degree of conductive hearing loss is not reliably predictive of the size of the otosclerotic focus. Merchant et al. (18) noted that limited otosclerosis does not translate into more limited conductive hearing loss because the anterior otosclerotic focus can displace the footplate posteriorly to impact on the posterior stapediovestibular joint (SVJ) and cause hearing loss out of proportion to the size of the otosclerotic focus. As expected from their results, the preoperative audiograms of our laser STAMP patients showed a trend that did not rise to statistical significance for less severe conductive hearing loss on average when compared with SFS patients. Furthermore, the footplate laser cut requires wide exposure of the oval window niche that must be judged at the time of surgery. Therefore, determining candidacy for a STAMP remains an intraoperative decision (10,11). The important factors for long-term success of partial stapedectomy procedures are as follows: 1) making the laser cut through healthy footplate enchondral bone that is likely to heal with a fibrous union and not through an otosclerotic focus, which would have a high risk of refixation; 2) lack of expansion of the otosclerotic focus or regrowth of footplate bone that could refix the footplate; and 3) lack of bony ankylosis or otosclerosis involving the posterior SVJ. In the temporal bone study by Merchant et al., 38% of temporal bones had neither otosclerosis nor ankylosis of the posterior SVJ, suggesting that a laser STAMP, anterior crurotomy, or a similar partial stapedectomy procedure on this subset of bones would be expected to have a lasting successful outcome (18). CSM is a technically challenging procedure that requires patient, slow removal of bone circumferentially around the perimeter of the footplate. It is useful in cases of congenital fixation of the stapes, which would be expected to heal with a fibrous union in most cases. It can also be used for cases of tympanosclerosis with bony fixation if the history of otitis media is reasonably remote and unlikely to reactivate, which would risk regrowth of the tympanosclerosis with refixation. CSM is a surgical option for younger children with congenital fixation in whom it is preferred to avoid a prosthesis until they are beyond the otitis media prone years. Complete mobilization of the stapes footplate is desirable to achieve optimal results but partial mobilization did significantly improve the severity of conductive hearing loss. Five-year follow-up results of STAMP reported by Silverstein showed a refixation rate of 9%, which was similar to the 8% in the present study (12). Successful SFS was accomplished in these 2 cases. When refixation occurs, conventional stapedotomy can be performed nearly as if it were a primary stapes operation as the central footplate remains unviolated. SFS after STAMP has shown Otology & Neurotology, Vol. 35, No. 4, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
e128
G. O. ACAR ET AL.
results comparable to primary stapedotomy (9Y12). Similarly, CSM does not involve the central portion of the footplate and leaves open the option for SFS if indicated at a later date. The recurrent surgery rates for STAMP (8%) and CSM (4.5%) were higher than for SFS (1.3%), but these less invasive procedures offer advantages that make these higher rates acceptable. Particularly with CSM, the operation may be offered to patients younger than would ordinarily be considered for SFS, especially if there has been any past history of otitis media. Hearing Results STAMP results were similar or even better than previously reported (9Y12). The observed incidence of anacusis was 4.0%, and it should be noted that the true incidence may lie within the 95% confidence interval from essentially 0% up to 11.8%. This unfortunate case underscores that complications can occur whenever the vestibule is opened, even in minimally invasive procedures. Profound hearing loss has also been previously reported with STAMP, 3.6% (2/55), which is similar to the present data (10). It has been speculated that patients with a limited extent of otosclerotic disease for which STAMP is an appropriate approach may have better preoperative hearing results, and this was seen in Silverstein’s studies. In our patients, however, preoperative hearing results were equal in the STAMP and SFS groups. To our knowledge, there are no reports of CSM in the literature with which to compare results. AC PTA and AC thresholds at 4 and 8 kHz significantly improved postoperatively but did not significantly improve further by more than 12 months. Postoperative results in SFS group were comparable to other series in the literature (6,8,10,19). Hearing improvements are typically better in the low to mid frequencies than in the high frequencies (4Y8 kHz), and this was also noted in the present study. Some worsening of the 4 kHz BC is common after SFS and tends to improve over time but not always to preoperative levels. It is not clear if this effect represents sensorineural injury or a mechanical disadvantage of the reduced vibrating surface area of the piston compared with an intact footplate. AC at 500 Hz and 2 kHz improved between the postoperative and 12 months or greater results, indicating that the healing process likely causes some mechanical advantage to the hearing, possibly from fibrosis tightening the fixation to the incus. ABG gap closure was similar between STAMP and SFS, but CSM results were worse because of the partially mobilized cases. Significant improvement was achieved in these partially mobilized cases such that it was effective in reducing or eliminating the need for hearing aids for these individuals, who were not otherwise good candidates for SFS. Average overclosure, bone conduction improvements was better in STAMP and CSM patients than in SFS patients. Overall, for 12 months, AC thresholds for STAMP were better than the other groups. Low-frequency AC thresholds were worse in the CSM than the other groups
reflecting that complete circumferential mobilization is necessary to yield hearing comparable to STAMP and SFS. AC thresholds at 4 and 8 kHz were better in the CSM and STAMP groups than in the SFS group postoperatively and more than 12 months’ follow-up. These results supported our hypothesis that the increased vibratory surface area of the footplate may be advantageous over a piston reconstruction for high-frequency hearing. For 12 months’ follow-up, the AC results were affected by the cases of refixation in STAMP and CSM. STAMP and CSM are technically more challenging procedures than SFS because they require operating on the footplate while the superstructure is still in place. Having previously practiced the procedures in the temporal bone lab, the learning curve for becoming comfortable with the new techniques was between 5 to 10 cases for each of these operations. Patients should be selected for favorable exposure of the stapes footplate. In the case of STAMP, otosclerosis should be limited to the anterior one third of the footplate, and the risk of refixation would be to be reduced if patients are older than 40 years. CSM can be done on a thick annulus, but care must be taken to bevel the trough, making it wider laterally, to prevent limitation of the mobilized footplate by friction with the thick walls of the surrounding otic capsule. The less invasive techniques have the advantage that they preserve nearly all of the functional anatomic structures. The stapedius tendon and most of the footplate are preserved. Late failures of STAMP and CSM may still be treatable with conventional stapedotomy techniques and with results similar to primary cases. Failed stapedotomy cases often have incus necrosis that will require reconstructive techniques that may not yield results as favorable as in primary cases. Limitations of the Study The conclusions from this study are significantly limited by selection bias including the ages of patients, pathology, and procedure. The types of procedures to be performed were determined beforehand or intraoperatively. By intention, the median age of the CSM cases was younger than for the other procedures, and most of the CSM cases were performed for congenital fixation and not for otosclerosis. However, the audiograms would be expected to be reasonably reproducible and not subject to significant age-related differences in the accuracy of responses to behavioral testing as all of the patients were 6 years or older. Therefore, one would not anticipate that there would be a measurable age-related bias when comparing audiograms from these different age groups. In fact, the multivariate analysis adjusted for age did not alter the results. Stapes fixation of all types results in conductive hearing loss that is diagnostically indistinguishable based on etiology. The purpose of this study was to present the feasibility of the two other options, STAMP or CSM, to give surgeons additional choices to address fixation due to any of these causes; hence, comparisons between groups were made despite the differences in pathology.
Otology & Neurotology, Vol. 35, No. 4, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
STAMP, CSM, AND SFS FOR STAPES FIXATION Cases with otosclerosis who underwent SFS had either larger disease foci or unfavorable anatomy that precluded STAMP. Despite these intended differences, the study was able to demonstrate that the newer procedures are feasible and, with these results, are acceptable alternatives to SFS. CSM expands the indications for stapes surgery without adding undue risks. There may be an added benefit to hearing in the higher frequencies with STAMP and CSM, but these data cannot be considered conclusive because of the selection bias. The outcomes of SFS versus STAMP could only be truly compared by a randomized comparison trial, which treated similarly sized anterior otosclerotic foci. CONCLUSION Techniques are available for less invasive methods to mobilize the stapes in cases of otosclerosis and other types of fixation. Avoidance of a prosthesis has the advantages of eliminating incus necrosis and other piston-related complications. CSM may be performed in younger children and in patients who have had a history of otitis media. Refixation rates are sufficiently low with these techniques that they are reasonable alternatives to conventional stapedotomy and stapedectomy procedures. Additional experience in the hands of other surgeons and in larger studies with long term follow-up is needed. Acknowledgments: The authors thank Ahmet Dirican, M.D, professor, Biostatistic Department of Cerrahpasa School of Medicine of Istanbul University; and Hasan YananlN, M.D., associate professor, Pharmacology Department of Marmara University, for the preliminary statistical analyses of this project. The authors also thank Tali Rasooly, research assistant at Boston Children’s Hospital, for the help in the preparation of this article.
REFERENCES 1. Rosen S. Mobilization of the stapes to restore hearing in otosclerosis. N Y J Med 1953;53:2650Y3. 2. Rosen S. Rosen mobilization and stapedectomy: long time results with special reference to high frequencies. Arch Otolaryngol 1969;89:241Y4.
e129
3. Fowler EP Jr. Anterior crurotomy and mobilization of the ankylozed stapes footplate: introduction to motion picture demonstration. Acta Otolaryngol 1956;46:319Y22. 4. Fowler EP Jr. Anterior crurotomy with footplate fracture. Arch Otolaryngol 1960;71:296Y304. 5. Shea JJ Jr. Fenestration of the oval window. Ann Otol Rhinol Laryngol 1958;67:932Y51. 6. Fisch U. Stapedotomy versus stapedotomy. Am J Otol 1982;4:112Y7. 7. Palva T. Argon laser in otosclerosis surgery. Acta Otolaryngol 1979;104:153Y7. 8. Perkins RC. Laser stapedotomy for otosclerosis. Laryngoscope 1980;90:228Y40. 9. Silverstein H. Laser stapedotomy minus prosthesis (laser STAMP): minimally invasive procedure. Am J Otol 1998;19:277Y87. 10. Silverstein H, Hoffmann KK, Thompson JH, Rosenberg SI, Sleeper JP. Hearing outcome of laser stapedotomy minus prosthesis (STAMP) versus conventional laser stapedotomy. Otol Neurotol 2004;25:106Y11. 11. Poe DS. Laser-assisted endoscopic stapedectomy: a prospective study. Laryngoscope 2000;110(5 Pt 2):1Y37. 12. Silverstein H, Jackson LE, Conlon WS, et al. Laser stapedotomy minus prosthesis (laser STAMP): absence of refixation. Otol Neurotol 2002;23:152Y7. 13. De la Cruz A, Angeli S, Slattery WH. Stapedectomy in children. Otolaryngol Head Neck Surg 1999;120:487Y92. 14. American Academy of OtolaryngologyYHead & Neck Surgery Foundation, Inc. Committee on Hearing and Equilibrium guidelines for the evaluation of results if treatment of conductive hearing loss. Otolaryngol Head Neck Surg 1995;113:186Y7. 15. Thiel G, Mills R. Rosen mobilization of the stapes: does it have a place modern otology? J Laryngol Otol 2006;120:1067Y71. 16. Jaisignhani VJ, Sxhachern PA, Paparella MM. Stapes mobilization in otosclerosis. Ear Nose Throat J 2001;80:586Y90. 17. Myers EN, Ishiyama E, Heisse JW. Histology of a successful stapes mobilization. Ann Otol Rhinol Laryngol 1970;79:321Y30. 18. Merchant SN, Incesulu A, Glynn RJ, Nadol JB. Histologic studies of the posterior stapediovestibularjoint in otosclerosis. Otol Neurotol 2001;22:305Y10. 19. Silverstein H, Hester TO, Rosenberg SI, Deems D. Preservation of the stapedius tendon in laser stapes surgery. Laryngoscope 1998;108:1453Y8. 20. Fisch U, Acar GO, Huber MH. Malleostapedotomy in revision surgery for otosclerosis. Otol Neurotol 2001;22:776Y85. 21. Lesinski SG. Causes of conductive hearing loss after stapedectomy or stapedotomy: a prospective study of 279 consecutive surgical revisions. Otol Neurotol 2002;23:281Y8. 22. Causse JB, Gherini S, Lopez A, Juberthie L, Olivier JC, Bastianelli G. Impedance transfer: acoustic impedance of the annular ligament and stapedial tendon reconstruction in otosclerosis surgery. Am J Otol 1993;14:613Y7.
Otology & Neurotology, Vol. 35, No. 4, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
