Many developments in reconstruction of the ossicular chain have taken place over the last 50 years [
There has been a quest for the ideal middle ear implant with the understanding that the middle ear environment in chronic ear disease is probably the main factor in determining success [
Over 40 years ago, incus repositioning with homograft and autograft incus ossicles was first undertaken [
The 1970s brought additional interest into trying to overcome the deficiencies of the autograft, homograft, and plastic implants of the 1950s and 1960s. Proplast, a combination of two polymers, had a number of advantages that could be utilized for middle ear reconstruction [
A high-density polyethylene sponge, a machine-tooled form known as Plastipore, was then explored as an alternative and found to be a material that could be sculpted and shaped [
More than 20 years ago, Sheehy [
Continued research in the late 1980s sought a more ideal material for the middle ear. Carbon matrix prostheses [
Hydroxyapatite made its debut in the early 1980s; Grote, the first to use hydroxyapatite in the middle ear, found it compatible [
Titanium was introduced by a small number of US otolaryngologists in the late 1990s. In the mid-1990s, European otologists were the first to use titanium middle ear implants in significant numbers of patients [
In 2001, Dornhoffer and Gardner [
In a review focusing on long-term results, Yung [
Despite numerous papers suggesting titanium as the new best implant system, we elected to review our recent results with the porous polyethylene prosthesis (Plastipore) prior to changing our well-established technique that is based on the two-point fixation principle. In the total ossiculoplasty reconstruction, the lateral surface of the prosthesis is covered with native cartilage. This cartilage is placed just medial to the scutum in cases where the canal wall is intact and slightly lateral to the malleus if present. In canal wall down cases the cartilage is level with the facial canal superiorly and the remnant of the bony annulus posteriorly. The medial shaft of the prosthesis is centered over the footplate with a tissue graft interface. Our study focuses on surgical technique rather than on a discussion about which type of prosthesis is optimal. Emphasis on lateral coverage with cartilage is well documented in the literature as are numerous types of prostheses. What has received scant attention is what is happening at the medial side of the prosthesis and its stability at the footplate interface. Therefore, only one type of prosthesis was used in this study. Surgical success can then be associated more directly to the stabilization technique and not to choice of prosthesis type by limiting this as a confounding variable.
With the extensive history of ossiculoplasty over the last five decades in mind, the difficulties in more diseased ears and, in particular, in total ossiculoplasty, this study analyzes hearing results in a group of patients undergoing a method of total ossiculoplasty. The principle of stabilizing the total prosthesis with native tissue both at its medial and lateral end is the driving factor behind the two-point stabilization theory.
There were 50 consecutive total ossiculoplasties performed over a two-and-a-half-year period (2004–2007). The age range was 5 to 79 years with a mean of 43.5 years (s.d. = 22.8 years). There were 22 males and 28 females. Primary surgery was performed in 5 (10%) and revision surgery in 45 (90%) patients. For 19 patients, it was the first revision; for 13 patients, the second revision; for 13 patients, the third or more revision. None of the revisions were second-stage ossiculoplasties. Indications for revision were infection, perforation, failed ossiculoplasty, and greater than 25 dB conductive hearing loss. The preoperative diagnosis was chronic otitis media in 18 (36%) patients. Conductive hearing loss was the preoperative diagnosis in 19 (38%) patients with cholesteatoma diagnosed in 12 (24%) patients and retraction in 1 (2%) patient. Fibrosis in the middle ear cleft was noted in seven patients (14%). Ventilation tube placement was required postoperatively in five patients (10%) with myringotomy performed in one patient. Indication for myringotomy and/or ventilation tube placement was development of serous otitis media postoperatively. In three patients, there was some degree of facial nerve prolapse and one with a very high jugular bulb, but this did not preclude reconstruction. One patient had cleft palate surgery in the past, and there was one patient with a congenital ear (incus and stapes missing). This patient had no history of chronic otitis media or cholesteatoma.
Subjects were tested pre- and postoperatively using standard audiometric procedures in double-walled sound rooms. Air conduction thresholds were measured at 250, 500, 1000, 2000, 3000, 4000, and 8000 Hz. Bone conduction testing was performed at 500, 1000, 2000, 3000, and 4000 Hz. Pure tone average (PTA) results were calculated using 500, 1000, 2000, and 3,000 Hz thresholds. Postoperative air-bone gap (ABG) was calculated by comparing the postoperative air conduction PTA to the postoperative bone conduction PTA. The speech reception threshold (SRT) was defined as the level (dB HL) at which the listener could identify spondee words 50% of the time. Speech discrimination was measured using taped W-22 25-word lists. Lists were generally presented at 30 to 40 dB SL. Masking was used in the nontest ear as needed. Postoperative testing was performed 2–23 months after surgery with a mean of 8.1 months.
The basic approach is the same in primary and revision surgery. Modification of the approach is used in canal wall down procedures. Almost all of our procedures are done under local anesthesia with intravenous sedation. Children under age 12 are operated under general anesthesia. If mucoid fluid is present, a ventilating tube may be placed at the end of the procedure. If the malleus is present, then the tensor tympani tendon is divided with a sharp instrument under direct vision or by palpation, allowing lateralization and enabling easier placement of the reconstruction prosthesis. From an incision above the auricle, loose areolar tissue is harvested. It is pressed in a fascia press and allowed to dry. Tragal cartilage is harvested if the surgery is transcanal. Alternatively, conchal cartilage is harvested if the ossicular reconstruction is being done during a mastoidectomy. The perichondrium is dissected off the cartilage and, if necessary, pressed and used to reinforce the tympanic membrane. The cartilage is sculpted in the shape of a dome to accommodate the tympanic membrane and the size of the posterior superior quadrant in the middle ear. The TOP is cut with a no. 15 blade, on a moistened tongue blade and, if the middle ear space is aerated, it is usually cut in half. In a canal wall down reconstruction or a shallow middle ear cleft, up to two-thirds of the length of the prosthesis will be removed. If there is a remnant of the stapes crura but not enough to support a POP, the crura may be lasered off with an argon laser in order to accommodate the TOP. This is done with 0.1-second exposure at a power of 1-2 watts. If there is an associated perforation, a temporalis fascia or scar graft is placed as an underlay initially.
Figures
Surgical defect with mobile footplate requiring total ossiculoplasty.
Areolar tissue graft placed over footplate and dimpled to receive TOP.
Lateral stabilization achieved with cartilage being placed lateral to TOP and medial to tympanic membrane.
After the tympanomeatal flap has been returned and prior to placing gelfoam, intraoperative audiometry is performed. Those who did not receive intraoperative audiometry (26 patients) generally had a concomitant tympanoplasty or had a history of a canal wall down mastoidectomy.
Intraoperative testing was performed using a Beltone 109 air-conduction audiometer with a TDH 39 headphone. The headphone was inserted into a Maico audiocup that fit around the ear and helped to attenuate the ambient noise that is common in a surgical suite. The sound pressure level measured with a Quest 155 precision sound level meter at the level of the ear was 54 dBA in our operating room with noise from background equipment and 30 dBA without equipment.
A thickened tympanic membrane or perforation would contribute significantly to a conductive hearing loss; therefore no testing was done in these cases, nor was it done in seven children who were operated under general anesthesia. An orthopedic sleeve is used to keep the audiometer cable sterile. Preoperative air-conduction testing at 500 Hz is done in the operating room. This testing allows the surgeon to verify smaller improvements that routine tuning fork testing might not demonstrate. A positive Rinne will confirm closure of the ABG to within 25 dB, but intraoperative testing will verify more specific improvements. In seven cases, inadequate improvement in postoperative hearing prompted adjustment of the prosthesis with subsequent testing confirming improvement. Adjustments were made with a hook to better center the TOP over the central portion of the graft. Depending on the preoperative ABG, a moderate improvement (at least 10 dB) in pure tone thresholds at 500 Hz was expected.
The above testing is performed on patients sedated with the following protocol. In the holding area they are given Versed 2 mg. I.V. with Dramamine 50 mg orally. In the operating room they are given Fentanyl 100 mcg, Propofol 70–100 mg, and Zofran 4 mg all administered intravenously.
The mean postoperative ABG was 15.7 dB (s.d. = 10.8 dB). The mean PTA hearing improvement was 15.7 dB (s.d. = 15.5 dB). Table
Mean pre- and postoperative air conduction (dBHL), bone conduction (dBHL), and word discrimination (%) results (Group A).
250 Hz | 500 Hz | 1000 Hz | 2000 Hz | 3000 Hz | 4000 Hz | 8000 Hz | Discrim | |
---|---|---|---|---|---|---|---|---|
Preoperative air | 62.3 | 59.4 | 57.6 | 53.9 | 58.2 | 63.9 | 71.4 | 92.0 |
(S.D.) | (18.7) | (20.9) | (19.1) | (18.8) | (20.9) | (23.1) | (22.9) | (8.1) |
Preoperative bone | 23.9 | 22.1 | 30.2 | 30.4 | 31.6 | |||
(S.D.) | (14.5) | (14.8) | (15.1) | (16.4) | (18.5) | |||
Postoperative air | 45.6 | 43.2 | 39.2 | 36.6 | 47.5 | 55.4 | 66.4 | 91.7 |
(S.D.) | (22.4) | (23.3) | (22.4) | (21.0) | (23.1) | (25.0) | (25.9) | (16.3) |
Postoperative bone | 24.2 | 21.4 | 28.0 | 30.2 | 30.2 | |||
(S.D.) | (16.3) | (17.4) | (19.0) | (19.6) | (21.2) |
PTA air-bone gap (ABG) following surgery (Group A).
Postoperative ABG | No. of patients | % of patients |
---|---|---|
1–10 dB | 22 | 44 |
11–20 dB | 11 | 22 |
21–30 dB | 10 | 20 |
31–40 dB | 7 | 14 |
Twelve patients in Group A (24%) had had a canal wall down mastoidectomy in the past with a postoperative ABG of 20.2 dB (s.d. = 10.4 dB) and mean hearing improvement of 11.8 dB (s.d. = 12.9 dB). Thirteen patients (26%) had an intact canal wall mastoidectomy in the past with a postoperative ABG of 14.8 dB (s.d. = 9.8 dB) and mean hearing improvement of 18.9 dB (s.d. = 14.9 dB). Neither the 7.1 dB difference in hearing improvement between these two groups (
Intraoperative audiometry was performed in 24 of 50 of the total ossiculoplasties. Intraoperative testing was done only at one frequency, 500 Hz. The mean preoperative air conduction threshold at 500 Hz tested conventionally was 64.5 dB HL (s.d. = 23.2 dB) with intraoperative presurgery testing yielding 64.3 dB HL (s.d. = 18.4 dB). The mean conventional postoperative air-conduction threshold at 500 Hz. was 44.6 dB HL (s.d. = 22.8 dB) with postoperative intraoperative test yielding a mean of 40.1 dB HL (s.d. = 13.2 dB). There was only a mean 0.2 dB difference between preoperative intraoperative and conventional sound room audiometry. Postoperatively, the difference was 4.5 dB. Table
Mean PTA air-bone gaps (ABG) and hearing improvement for patients who had and did not have intraoperative audiometry (IOA) during their surgical procedure (Group A).
ABG | PTA improvement | |
---|---|---|
24 patients with IOA | 14.4 dB | 21.1 dB |
26 patients without IOA | 16.9 dB | 10.6 dB |
Mean difference | 2.5 dB | 10.5 dB* |
*This difference was significant (
The challenge in ossicular reconstruction is well recognized. Certain variables such as middle ear fibrosis, adhesive otitis, and significant Eustachian tube dysfunction are not easily controlled by the otologic surgeon. However, two variables that can be controlled by the surgeon are the type of prosthesis used and the manner in which the prosthesis is used. The last several decades have seen a shift from autologous ossicle use to prosthetics [
The focus of our review is to demonstrate the value of the two-point stabilization in total ossiculoplasty. As pointed out in the prior section on history, patients requiring a total ossiculoplasty generally have advanced disease. These patients typically arrive for reconstruction with a history of a number of procedures including intact canal wall and canal wall down mastoidectomy [
The mean postoperative ABG of 15.7 dB and the finding that the ABG was closed to within 10 dB in 44% of patients are very favorable data for total ossiculoplasty. Table
Comparison of pure-tone average (PTA), air-bone gap (ABG), and hearing improvement results following total ossicular replacement.
Current |
Martin and Harner [ |
Gardner et al. [ |
Fisch et al. [ |
Krueger et al. [ |
Slater et al. [ | |
---|---|---|---|---|---|---|
Mean postoperaive ABG(dB) | 15.7 | 25 | 24.6 | 21.2 | 15.8 | NR |
Mean PTA improvements(dB) | 15.7 | 9 | 15.1 | 16.9 | 22.8 | NR |
% 0–10 dB ABG | 44 | 3 | 7 | 13 | 26.7 | 38 |
% 0–20 dB ABG | 66 | 40 | 44 | 57 | 66.7 | 67 |
PTA calculation | 4 freq | 4 freq | 4 freq | 4 freq | 4 freq | 3 freq |
Mean followup | 8.1 mo | 3 mo–2.5 yr | 1.5 yr | 1 yr | 3 mo | 6 mo |
50 | 30 | 27 | 46 | 15 | 133 |
NR: Not reported, mo: months, yr: years,
Almost two-thirds (66%) of the patients in this study closed their ABG to within 20 dB (Table
In addition, in a number of cases, intraoperative audiometry was very useful in verifying proper implant positioning. Intraoperative audiometry is a concept borrowed from otosclerosis surgery [
Recent literature has confirmed that staging improves results particularly in more advanced chronic ears, especially those requiring total ossiculoplasty [
With different techniques and prostheses being used over the last 3 decades, it is difficult to compare studies. While recognizing that differences between studies in patient selection, technique and prosthesis type limit direct comparison of results, it is still interesting to contrast the audiometric results reported here to those of other studies. Table
The lateral stabilization over the prosthesis with cartilage has been well described [
The major weakness in this study is the retrospective nature, which is inherent in any review such as this. It would be of interest to compare various prostheses such as titanium, plastipore, and hydroxyapatite utilizing the two-point technique. Of course, a prospective matched cohort study with a larger sample size to increase the statistical power of the observations would lend more scientific support to this theory. Ideal comparisons in chronic ear studies with nonmastoid cases, primary and revision cases being compared only to each other, would further eliminate uncontrolled variables. Of interest would be comparisons of areolar tissue to vein, fascia, and perichondrium as the medial support. Areolar tissue was chosen because of its thinness as opposed to perichondrium and its easy availability.
With the above in mind, further studies on large prospective groups with very well-controlled variables, using different prostheses with and without medial support as well as comparing various medial tissue grafts would be valuable. This study does not address long-term results in these difficult cases. Long-term review using this technique would perhaps show the value in two-point stabilization in obtaining more stable, prolonged reconstruction results.
In conclusion, we emphasize the two-point fixation principle in total ossiculoplasty reconstructions. Although this paper has focused on the porous polyethylene (Plastipore) prosthesis, two-point fixation may be achieved with all prostheses. The purpose of this study is not to propose one prosthesis over another. Rather, it is an attempt to overcome the difficult underlying conditions in total ossiculoplasty, particularly in mastoidectomy and revision ears. In total ossiculoplasty, areolar tissue over the footplate assists in the two-point stabilization. Following the above recommendations will maximize hearing improvement even in revision and difficult CWD reconstructions.