Abstract
Objectives. Surgery for otosclerosis has a highly satisfactory hearing outcome, for both the patient and the otologic surgeon. However, subsequent conductive hearing loss, dizziness/vertigo, or sound distortion could necessitate revision surgery. The aim of the present study is to evaluate the surgical findings and hearing outcomes of 84 revision stapes surgeries.
Study Design. Case series with chart review.
Setting. Tertiary referral center.
Subjects and Methods. At our institution, 84 revision cases were performed between 1998 and 2009. Conductive hearing loss was the revision indication in 69 cases, severe dizziness/vertigo in 8 patients, sound distortion in 5 cases, and progressive hearing loss with dizziness in 2 patients. Operative findings were noted in every case and evaluated separately.
Results. Surgical intervention revealed problems related to prosthesis in 51 cases, fibrotic bands in 26 cases, adhesions in 13 cases, incus necrosis in 15 cases, perilymphatic fistula in 3 cases, intact footplate in 5 cases, incus-malleus fixation in 2 cases, and reobliteration in 2 cases. Mean follow-up period was 19 months (range, 12-53 months). Successful hearing results (air-bone gap <10 dB) were reached in 58% of the cases, and satisfactory hearing results (air-bone gap <20 dB) were reached in 71% of the cases.
Conclusions. The ideal patient for revision stapes surgery is one who benefits from the initial surgery but complained of conductive hearing loss. In the present study, improvement in pure-tone average was 13.2 dB, and the mean air-bone gap was 9.6 dB.
Surgery of otosclerosis is one of the most challenging operations in otology. Satisfactory hearing results could be obtained in primary and revision cases. The hearing outcome of stapes surgery frequently satisfies the surgeon and the patient. Unfortunately, up to 20% of patients may require revisions in the long run.1-3 Most of these patients suffer from delayed or persistent conductive hearing loss, followed by dizziness, vertigo, sensorineural hearing loss, and distortion of sound. Surgeons could face a variety of challenges during the surgery and should be prepared to overcome each and every problem. Surgeons require a detailed description of the initial surgery if the surgery was performed elsewhere. However, revision surgeries may sometimes be even more problematic; for example, sensorineural hearing loss was reported as 0.4% to 20% and total deafness as 1.2% to 14%.1-8
Patients who initially had successful surgery may later present with delayed conductive hearing loss. The result of revision surgery in this group of patients is highly successful. However, some patients may need more than one revision surgery. When the patient needs more than one surgery, the risk of sensorineural hearing loss (SNHL) increases. If dizziness accompanies progressive hearing loss in the postoperative course, the risk of developing profound hearing loss rises. Lasers have improved hearing outcome in revisions, reaching a success rate of 91% in some selected cases.3,7,9-11 Unfortunately, every center does not have laser facilities. Therefore, the otologic surgeon is forced to perform revision stapes surgery without laser assistance but still may achieve satisfactory results. In the present study, we reported the operative findings, type of prosthesis, and hearing outcomes of 84 consecutive revisions and briefly emphasize the factors for achieving a good outcome in revision surgeries.
Subjects and Methods
Eighty-four consecutive revision stapes surgeries were performed at Izmir Bozyaka Teaching and Research Hospital ENT Clinic, in Izmir, Turkey, by 3 senior surgeons from 1998 to 2009, and their charts were retrospectively analyzed. Forty-seven women (55.9%) and 37 men (44%) comprised the study group. Mean age was 45.4 years (range, 16-71 years).
Preoperative Evaluation
The indications for revision surgery were as follows:
Conductive hearing loss (CHL) with air-bone gap (ABG) ≥20 dB (69 cases, 82.14%)
Persistent CHL (26 cases, 30.95%)
CHL after an initial good hearing result (43 cases, 51.19%)
Severe dizziness/vertigo (8 cases, 9.52%)
Distortion of sound (5 cases, 5.96%)
Progressive SNHL and dizziness (2 cases 2.38%)
Thirty-one cases (37%) of primary surgeries were performed at our institution, and the rest were performed elsewhere. Postoperative audiometry was performed at the third postoperative month and the first postoperative year, and then the test was recommended annually. Mean follow-up of the patients was 19 months (12-53 months). Best postoperative hearing thresholds were accepted for the evaluation. Postoperative ABG values were obtained by subtracting postoperative bone conduction pure-tone average (PTA) levels from postoperative air conduction PTAs at frequencies of 0.5, 1, 2, and 3 KHz, as recommended by American Academy of Otolaryngology—Head and Neck Surgery (AAO–HNS) guidelines. 12
Surgical Intervention
All operations were performed under local anesthesia. The transcanal approach was used without a speculum holder. The tympanomeatal flap was elevated with the Rosen incision and the middle ear was explored. Any adhesions or fibrotic bands were removed to gain access to the prosthesis and oval window area. Malleus and incus mobility was checked by careful palpation of the ossicles. Surgical findings were noted intraoperatively.
If the prosthesis was displaced and migrated onto the solid otic capsule margin, it was removed carefully in order not to endanger the inner ear. The oval window area was inspected with higher magnification for lateralization of a neomembrane. If any soft tissue thickening existed, it was meticulously thinned or removed. The vestibule was entered with the aid of a sharp pick or low-speed diamond burr (Skeeter Ultra-Lite, Xomed-Treeace, North Jacksonville, Florida), and a new small stapedotomy was created. A new piston was introduced on the long process of the incus, when available. When incus necrosis was encountered, different types of prosthesis were used. Small amounts of adipose tissue were used as a sealing material around the prosthesis. The tympanomeatal flap was replaced and covered with a gelatin sponge. This study was approved by the local review board of the Izmir Egitim Hastanesi Lokal Etik Kurulu (2010-43) institution.
Results
Of 84 consecutive revision stapes surgeries, 78 had primary revision surgery, 4 had secondary revision surgery, and 2 had tertiary revision surgeries on the same ear. We observed displacement of the prosthesis in 34 cases (40.4%), long prosthesis in 8 cases (9.5%), short prosthesis in 4 cases (4.8%), loose prosthesis in 3 cases (3.6%), and incus-prosthesis ankylosis in 2 cases (2.4%).
There were other intraoperative findings. Of the 84 cases observed, in 20 cases (23.8%) we observed major fibrotic bands ( Figure 1 ), which were located between prosthesis and promontorium or between the long process of the incus and the medial wall of the middle ear mucosa. Those bands were either decreasing ossicular mobility or pulling the prosthesis out of position and causing conductive hearing loss. In 6 cases (7.4%), there were minor fibrotic bands; in 9 cases (10.7%), major adhesions were located between the long process of the incus and the manubrium mallei or inferior portion of the prosthesis and oval window niche. Those adhesions were reducing ossicular mobility and causing conductive hearing loss. There were minor adhesions in 4 cases (4.8%), incus necrosis in 15 cases (17.8%), intact footplate in 5 cases (5.9%), perilympathic fistula in 3 cases (3.5%), reparative granuloma in 3 cases (3.5%), incus-malleus fixation in 2 cases (2.4%), and reobliteration in 2 cases (2.4%).
Fibrotic bands between the incus and the middle ear mucosa and between the prosthesis and the promontorium.
Overall successful hearing results (ABG <10 dB) were achieved in 58% of the cases and satisfactory hearing results (ABG <20 dB) in 71% of the cases (primary, secondary, and tertiary surgeries). Overall improvement in PTA was 13.2 dB, and mean ABG was 9.6 dB. For primary revision cases, postoperative PTA was 13.5 dB, and mean ABG was 9.8 dB. For secondary revisions, satisfactory hearing results (ABG <20 dB) could not be obtained. Mean PTA improvement was 9.2 dB, and mean ABG was 26.3 dB. For tertiary revisions, satisfactory hearing results also could not be obtained. Unfortunately, profound hearing loss ensued.
In the persistent CHL group, successful hearing results (ABG <10 dB) were achieved for 17 of 26 patients (65%), and satisfactory hearing results (ABG <20 dB) were achieved for 20 of 26 patients (77%). In the delayed CHL group, successful hearing results (ABG <10 dB) were achieved for 26 of 43 patients (60%), and satisfactory hearing results (ABG <20 dB) were achieved for 32 of 43 patients (74%). In the dizziness group, successful hearing results (ABG <10 dB) were achieved for 3 of 8 patients (38%), and satisfactory hearing results (ABG <20 dB) were achieved in 4 of 8 patients (50%). In the group of patients operated on for distortion of sound, successful hearing results (ABG <10 dB) were achieved for 3 of 5 patients (60%), and satisfactory hearing results (ABG <20 dB) were achieved in 4 of 5 patients (80%). In the progressive SNHL group, 0% or 0 of 2 patients achieved successful or satisfactory hearing results.
Surgical interventions included removal of fibrotic bands and adhesions in 39 cases, incus to oval window prosthesis replacement in 51 cases, malleus to oval window prosthesis placement in 9 cases (5 Shea malleus attachment prostheses [Smith and Nephew, Bartlett, Tennessee] and 4 CliP Piston malleovestibulopexy [MVP] prostheses [Kurz Medical, Dusslingen, Germany]), Gold-Ti Angular piston (Kurz Medical) placement in 2 cases and total ossicular replacement prosthesis (TORP; Gyrus ACMI-ENT, Bartlett, Tennessee) in 2 cases, hydroxyapatite (HA) bone cement (Gyrus ACMI-ENT) and prosthesis in 2 cases, shortening of the prosthesis in 8 cases, sealing of the oval window with connective tissue and fibrin glue and replacement of the prosthesis in 6 cases, and sealing of the oval window and fibrin glue in 2 cases.
Discussion
Would it be exaggerating to say that it is a challenge to make the decision of revision stapes surgery and perform the operation? The surgeon facing the task of revision has extra responsibility compared to the primary surgeon. He or she should be prepared to overcome every problem that might not be able to be ascertained preoperatively.
The number of revision stapes surgeries is ever increasing because in many newly established centers, the primary operation is carried out by surgeons not yet fully experienced in these kinds of surgeries. During 1998-2009, we performed 1843 stapes surgeries at our institution, 84 of which were revisions.
The time to reoperate is critical, and prior to surgery, an observation period of at least 6 weeks is recommended. 5 In our series, 8 cases were operated on for severe dizziness/vertigo. We made the decision for revision in those patients for 2 reasons: first, the duration was more than 6 weeks, and second, their dizziness/vertigo was incapacitating, restricting their daily activities, including not being able to go to work.
In most cases, revision is required as a result of persistent or delayed conductive hearing loss, and the exact cause of this loss generally cannot be ascertained preoperatively.
Problems such as displacement of the prosthesis, fibrotic bands, adhesions, long prosthesis, short prosthesis, and loose prosthesis leading to hearing loss can be diagnosed and should be corrected during revisions. In patients operated on for persistent or delayed conductive hearing loss, prosthesis displacement, major fibrotic bands, incus necrosis, major adhesions, and intact footplate were commonly encountered findings. Another common finding was the existence of a long prosthesis in the group of patients operated on for severe dizziness. Finally, presence of a short prosthesis (4 cases) and also a loose prosthesis (1 case) in the group of patients operated on for distortion of sound was observed.
Among the patients requiring revision surgery, prosthesis displacement was detected in 14.2% to 82% of cases in other series.1-5,7-10 Similar to previous studies, 40.5% of our study patients showed prosthesis displacement. Erosion of the incus was 17.8% in our group and 5% to 43% in other studies.1-5,7-10 Perilymphatic fistula was observed in 3.6% of our patients compared to 1.5% to 12% in other studies.1-5,7-10 Reobliteration and intact footplate were observed in 7 of our patients (8.3%). The finding in other studies for bony regrowth was 6% to 24.3%.1-5,7-10 Major and minor fibrotic bands (31.2%) and adhesions (15.5%) were observed in the middle ear area, for a total of 46.2% of our patients in which this was a common factor for the failure of primary surgery (8.4%-44.6% in other trials).1-5,7-10
There is a wide range of hearing outcomes after revision stapes surgery.1-11 Air-bone gap closure less than 20 dB was between 50% and 92%, and ABG closure to less than 10 dB was 16% to 80.5%.1-11
Unfortunately, both third revision cases, who were operated on for progressive hearing loss accompanied with dizziness/vertigo, developed profound hearing loss at the end of the revision. Improvement of hearing could not be obtained in those tertiary revisions, but their sensation of dizziness and vertigo ceased. Lippy et al 7 reported poor hearing outcome in revision cases with SNHL in which a piston type of prosthesis was used. They reported elimination of vertigo in those patients with a wire gel foam prosthesis.
We encountered 15 cases with incus necrosis in our group of 84 consecutive revisions. A large number of incus necrosis was encountered when the old type wire prosthesis was used (11 cases, 73.3%). Only 2 revision cases with fluoroplastic pistons showed incus necrosis (13.3%). In those incus necrosis cases, we placed a malleus to oval window prosthesis in 9 cases (5 Shea malleus attachment prostheses and 4 CliP Piston MVP prostheses), an angular piston in 2 cases, TORP in 2 cases, and HA bone cement and prosthesis in 2 cases. Malleus to oval window prostheses were not available at the time of those operations, so we preferred to use TORP in those 2 patients. We observed adhesions and displacement of the prosthesis with Shea malleus attachment pistons and also displacement and major fibrotic bands with Gold-Ti Angular pistons. In MVP cases, we observed favorable hearing results with Hausler’s MVP CliP Piston ( Figure 2 ). In 1994, Shea 13 recommended a special fluroplastic malleus attachment prosthesis (Smith and Nephew) designed for the left or right ear, between the malleus handle to the oval window if the incus was eroded. In this particular prosthesis, the loop material (fluoroplastic) does not wrap the malleus handle circumferentially as a traditional incus to oval window fluoroplastic piston does; however, insertion was recommended by closing with forceps and inserting connective tissue between the loop of the prosthesis around the malleus and the overlying drum. 13 In our revisions, we observed adhesions and displacement of this prosthesis. However, we observed more favorable hearing results in our patients with the CliP Piston MVP prosthesis (n = 4) compared to the Shea malleus attachment prosthesis (n = 5). The reason might be the stability of the prosthesis around the malleus in the former, but because the case number is so small, it would be misleading to reach such a conclusion.
Hausler CliP Piston in position around the malleus.
In 84 consecutive revision surgeries, we preferred to use a fluoroplastic prosthesis in 60.7% of the cases. The reasons: this piston has the most reasonable price (cheap), and intraoperatively, you can easily adjust the functional length by simply cutting the inferior edge, or if you do not have the particular piston in your stock, you can easily trim the inferior edge to get your desired length, if needed. Other factors include ease of manipulation and insertion and the fact that only 2 cases of incus necrosis with a fluoroplastic piston in those 84 consecutive revisions were observed.
Fisch et al, 14 who reported 37.5% of partial malleus fixation in 82 consecutive revision surgeries for otosclerosis, recommended superior canaloplasty to detect malleus fixation. In our cases, we found a fixation of 2.4%, which might be due to not using an enlarged endaural approach to reach and check anterior malleal ligament.
Better hearing outcomes were achieved in the persistent CHL group (ABG <20 dB in 77% of cases) and in the delayed CHL group (ABG <20 dB in 74% of cases). These 2 groups constituted our major patient population (69 of 84). Satisfactory hearing outcomes were achieved for these patients, and we encourage revision surgery for this type of patient. Patients with dizziness did benefit by a reduction in their unsteadiness and/or dizziness after revision surgery, but only half of them obtained a hearing improvement in our hands. Our results showed best hearing outcomes for patients operated on for distortion of sound. However, the number of the patients was small (only 5), and the reader should take into consideration this small sample size. We do not recommend revision surgery for patients with progressive SNHL because in our hands, profound hearing loss ensued in this group.
Revision stapes surgery is technically more difficult and requires more experience than the primary surgery.
Some conclusions could be drawn from our study:
Conductive hearing loss was the most commonly encountered indication for revision stapes surgery (82%).
The most frequent surgical finding was displacement of the prosthesis (40.4%).
The best results occurred in conductive hearing loss cases.
Atraumatic surgery is essential for achieving a good outcome. The primary surgeon should be careful not to disturb/hazard the middle ear mucosa to prevent formation of adhesions and fibrotic bands.
Our data indicated a large amount of cases of incus necrosis with old type wire prostheses (73.3%).
Fluoroplastic materials are cheap and stable. Their manipulation and insertion are easy. We observed a low incidence of incus necrosis with such prostheses. We believe that they could be a good alternative in revision surgery.
As the number of revision surgeries increases, hearing outcome decreases.
The functional integration of the loop of the piston to the malleus is the key point for the malleus to oval window prosthesis for long-term stability and hearing outcome.
Although laser-assisted revision stapes surgery has a better hearing outcome in selected cases, it is costly, time-consuming, and not available in every otologic center. At the same time, extra training and precautions in the operating theater are required. Nevertheless, revision stapes surgeries without laser assistance can achieve satisfactory hearing results, when the factors influencing the surgical outcome and limitations of surgical skills are taken into account.
Author Contributions
Disclosures
Footnotes
No sponsorships or competing interests have been disclosed for this article.
This article was presented at the 2010 AAO–HNSF Annual Meeting & OTO EXPO; September 26-29, 2010; Boston, Massachusetts.