Redaelli de Zinis LO. Titanium vs Hydroxyapatite Ossiculoplasty in Canal Wall Down Mastoidectomy. Arch Otolaryngol Head Neck Surg. 2008;134(12):1283-1287. doi:10.1001/archotol.134.12.1283
To report hearing results using a titanium ossicular replacement prosthesis during canal wall down mastoidectomy with tympanoplasty to treat cholesteatoma.
Retrospective medical record review.
Referral university hospital.
Patients with cholesteatoma treated with primary or revision canal wall down mastoidectomy with tympanoplasty in a single stage. Patients with implanted hydroxyapatite prostheses composed a matched control group.
Main Outcome Measures
Medical records were reviewed for type of ossicular condition, type of prosthesis, and hearing threshold at 1-year follow-up.
Results are reported as the 5-frequency average air conduction gain, bone conduction gain, and air-bone gap. The malleus handle was present in 24 patients, and the stapes superstructure in 22 patients. Mean (SD) air conduction gain was 7.6 (14.7) dB (P = .001); it was 8.7 (12.0) dB in the group with titanium prostheses and 6.3 (17.4) dB in the group with hydroxyapatite prostheses (P = .54). Bone conduction gain was 1.1 (4.9) dB (P = .19). No patients experienced postoperative impairment of bone threshold greater than 5 dB. Postoperative air-bone gap was 26.5 (15.3) dB; it was 23.8 (15.7) dB in the titanium group and 29.8 (14.6) dB in the hydroxyapatite group (P = .18). Air-bone gap closure was 40%; it was 46.2% in the titanium group and 33.3% in the hydroxyapatite group (P = .35).
Titanium is a satisfactory material for use in ossicular reconstruction and is comparable to hydroxyapatite, although at present, no definitive conclusion about the superiority of titanium can be drawn.
The goals of surgery to treat cholesteatomatous otitis media are complete eradication of disease, achievement of a dry and self-cleansing ear, creation of anatomical conditions to prevent recurrence, and preservation or improvement of hearing function.1 Preservation or improvement of hearing function is the most difficult to achieve, especially when the ossicular chain is involved. The 2 most important surgical techniques used to treat cholesteatoma are canal wall down (CWD) and canal wall up mastoidectomy with tympanoplasty. Canal wall down mastoidectomy is associated with lower rates of recurrence and fewer surgical procedures. Canal wall up mastoidectomy does not require substantial anatomical modifications and requires less postoperative care. Favorable hearing results remain elusive with both techniques.2 Reports in the literature of various prosthetic materials and designs have failed to produce a consensus about the best practice for ossicular chain reconstruction with hearing preservation.3
The objective of the present study was to compare functional results of ossicular chain reconstruction using either titanium (Ti) or hydroxyapatite (HA) prostheses with CWD mastoidectomy with tympanoplasty. To my knowledge, to date, no published studies have compared these 2 techniques.
An observational retrospective review of medical records was performed for patients who underwent surgical treatment of chronic otitis media with cholesteatoma. Subjects were identified from a database of patients who received care at the Department of Otorhinolaryngology, University of Brescia, Brescia, Italy. Patients were eligible if they had partial or total erosion of at least 1 middle ear ossicle or had cholesteatoma adherent to at least 1 ossicle that required ossicle removal and reconstruction. Only those patients treated with primary or revision CWD mastoidectomy using Kurz Vario Ti prostheses (Kurz GmbH, Dusslingen, Germany) in a single stage were selected for the study. All patients meeting eligibility criteria had a thin fragment of cartilage interposed between the head of the prosthesis and the malleus handle or tympanic membrane. When the Ti prosthesis, which has no notch for the malleus handle, did not fit well under the malleus handle, the malleus handle was removed.
The control group consisted of patients who had undergone CWD mastoidectomy and ossicular reconstruction with HA partial ossicular replacement prostheses or total ossicular replacement prostheses without cholesteatoma recurrence and without prosthesis extrusion 1 year after surgery. There were 4 basic designs for the HA prostheses used, depending on the presence or absence of the malleus handle and of the stapes superstructure. There was a notch for the malleus handle on the head of the prostheses, whereas the head was flat when the malleus handle was absent and a fragment of cartilage was interposed between the head of the prosthesis and the neotympanum. The shaft of the prostheses was always straight. A cup for the head of the stapes was present on the foot of the prostheses, whereas the shaft was thinner and longer when the stapes superstructure was absent. When HA prostheses were used and the malleus handle was present, the notch of the cap of the prostheses was positioned under the malleus handle.
Air and bone conduction thresholds were determined before and 1 year after surgery using a clinical audiometer (Interacoustics AC-40; Interacoustics USA, Eden Prairie, Minnesota). Testing was performed in a sound isolation booth with 5-dB step increases and appropriate masking with narrow-band noise of the contralateral ear using the plateau method. Threshold frequencies of 0.5, 1, 2, and 3 kHz were used. Mean differences in threshold were calculated for air and bone conduction, or air-bone gap (ABG), and for air conduction between preoperative and postoperative tests (gain). Preoperative and postoperative bone threshold variability was also analyzed to document possible iatrogenic cochlear damage.
The present study compares functional results of Ti and HA prostheses on the basis of presence or absence of the malleus handle and the stapes superstructure. Statistical analysis was performed using 3 tests: the Wilcoxon test when comparing the same group before and after surgery on 1 continuous variable, the χ2 test when comparing 2 groups on categorical variables, and the Mann-Whitney test when comparing 2 groups on 1 continuous variable. Results are expressed as mean (SD).
Fifty patients were eligible for the study, including 32 female and 18 male patients with a mean age of 49 years (age range, 17-78 years). Twenty-six patients received Ti prostheses, and 24 received HA prostheses. The malleus handle was present in 24 patients and absent in 26 patients, and the stapes superstructure was present in 22 patients and absent in 28 patients; both were present in 12 patients, and both were absent in 16 patients. The malleus handle was present in 8 of 26 patients (30.8%) who received the Ti prosthesis and in 16 of 24 (66.7%) of those who received the HA prosthesis (P = .01). The stapes superstructure was present in 12 of 26 patients (46.2%)who received the Ti prosthesis and in 10 of 24 (41.7%) those who received the HA prosthesis (P = .74).
No partial or total prosthesis extrusion or cholesteatoma recurrence was observed 1 year after surgery in patients who received the Ti prosthesis. Mean (SD) bone conduction thresholds were 25.8 (13.4) dB preoperatively and 24.7 (12.7) dB postoperatively (P = .19). No patients experienced postoperative bone threshold impairment greater than 5 dB.
Mean (SD) air conduction gain was significant at 7.6 (14.7) dB (P = .001). Postoperative gain was 8.7 (12.0) dB in the Ti group and 6.3 (17.4) dB in the HA group (P = .54). Presence or absence of the stapes superstructure and the malleus handle did not significantly affect air conduction gain. Gain was 11.2 (12.3) dB when the stapes superstructure was present (n = 22) and 4.7 (16.0) dB when it was absent (n = 28) (P = .09), and was 7.8 (16.7) dB when the malleus handle was present (n = 24) and 7.3 (12.9) dB when it was absent (n = 26) (P = .99). There were also no differences between the Ti and HA groups with respect to presence or absence of the stapes superstructure (Table 1). Although postoperative gain was increased in the Ti group when the malleus handle was absent and in the HA group when the malleus handle was present, these differences were not statistically significant (Table 1).
Postoperative ABG closure within 20 dB was observed in 20 of 50 patients (40%) (12 of 26 [46.2%] in the Ti group and 8 of 24 [33.3%] in the HA group) (P = .35). Closure was 54.5% when the stapes superstructure was present (12 of 22) and 28.6% when it was absent (8 of 28) (P = .06), and was 37.5% when the malleus handle was present (9 of 24) and 42.3% when it was absent (11 of 26) (P = .72). Closure rates were higher when the malleus handle was present and the stapes superstructure was absent, although the difference was not significant. The Ti and HA groups did not differ significantly with respect to ABG closure results (Table 2).
The mean (SD) preoperative ABG was 33.0 (12.8) dB; it was 31.6 (13.3) dB in the Ti group and 34.6 (13.3) dB in the HA group (P = .41). The postoperative ABG was 26.5 (15.3) dB; it was 23.8 (15.7) dB in the Ti group and 29.8 (14.6) dB in the HA group (P = .18). A lower ABG was evident in the Ti group at each frequency tested, although the difference was never significant (Table 3).
The mean (SD) postoperative ABG was 21.9 (15.6) dB when the stapes superstructure was present (n = 22) and 30.2 (14.3) dB when it was absent (n = 28) (P = .05), and was 27.0 (16.2) dB when the malleus handle was present (n = 24) and 26.1 (14.7) when it was absent (n = 26) (P = .75). Mean postoperative ABG did not vary significantly between the Ti and HA groups, even after adjusting for presence or absence of the stapes superstructure and the malleus handle (Table 4).
One of the most intriguing and problematic topics in middle-ear surgery is the reconstruction of the conductive mechanism. This has led to the development of several implants including homograft bone and cartilage and biocompatible alloplastic types.4 The potential for disease transmission with homografts has stimulated further research on alloplastic implants.4 Materials under investigation for ossicular chain replacement should be biocompatible to enable osseointegration and biostability and should have appropriate rigidity and weight. To date, no substance meets these qualifications.4
Over the last few decades, use of HA has increased.5 Recently, however, Ti has gained in popularity.6 Titanium more closely matches ossicle weight, has practical advantages for intraoperative manipulations and stability,7,8 and has hypothetical advantages in speech frequencies performance.6,9 In addition, design features of Ti implants such as a slim axis, a small foot, and an open head improve visibility during surgery.7,8
The purpose of this research was to present functional results of Ti prostheses for ossicular chain reconstruction of CWD mastoidectomy performed in a single stage compared with those of the more frequently used HA prostheses that I used in previous cases. To my knowledge, no literature to date compares results for Ti and HA prostheses in open cavities after cholesteatoma removal. Moreover, the inconsistency of data reporting methods in the literature makes comparisons difficult in this field. I, therefore, report the results in multiple ways for ease of comparison with other investigations.
Functional results with Ti prostheses have been compared with those of prostheses made from other materials,10- 12 usually under favorable conditions of tympanoplasty without mastoidectomy or closed procedures.9 Superior hearing results were generally reported for prostheses made of Ti compared with polyethylene11,12 or other nonspecified materials.10 No functional differences were observed with HA prostheses at 2 months in the most consistent group of patients.9
A review of studies published to date involving Ti implants in CWD mastoidectomies12- 19 is useful for comparison of my results (Table 5). The number of patients analyzed by the different investigators varies between 8 and 21; reported ABG closure ranges from 30.8% to 81.2%; extrusion rate varies from 0% to 7.7%; and follow-up was no longer than 18 months (Table 5).
The ABG closure for CWD mastoidectomies with non-Ti prostheses ranged from 24% to 66.3%.20- 22 Gain varies between −0.1 dB and 9.6 dB.22- 26 Results are frequently superior when the stapes superstructure is present.24,27- 29
In my experience with CWD mastoidectomy performed in a single stage, there were no significant differences between Ti and HA prostheses. However, Ti prostheses showed trends toward superiority for gain (8.7 dB vs 6.3 dB), mean ABG (23.8 dB vs 29.8 dB), and percentage of ABG closure (34.6% vs 29.2%).
Among the 26 patients who received Ti prostheses, superior hearing outcomes were observed when the stapes superstructure was present: gain, 12.5 dB; ABG, 19.7 dB; and ABG closure, 58.3%. This is similar to results obtained for other types of prostheses because of the stability that prostheses provide for reconstruction.28 In contrast to reported results for other materials,28,30 I observed improved gain and ABG closure when the malleus handle was absent. It has been demonstrated on the fresh cadaveric human ear model that there may be a mechanical advantage in reconstructing the malleus because the tympanic membrane can then contribute to impedance matching through the catenary lever.31In my experience, the hypothetical advantage of reconstruction is countered by optimizing prosthesis position in close contact with the tympanic membrane with interposition of only a thin fragment of cartilage.
A retrospective study was conducted of functional hearing results after single-stage CWD mastoidectomy with tympanoplasty to treat cholesteatoma, comparing Ti and HA prostheses in different conditions of the middle-ear ossicles residual. To my knowledge, this is the first comparison of Ti and HA prostheses for CWD mastoidectomies. Comparable hearing outcomes were observed with Ti, which has the advantage of better surgical-handling properties. The small sample size limited my ability to demonstrate statistically significant superior results for Ti. Larger studies are needed to further assess functional results of Ti prostheses in CWD mastoidectomy.
Correspondence: Luca O. Redaelli de Zinis, MD, Department of Otorhinolaryngology, University of Brescia, Piazzale Spedali Civili 1, 25123 Brescia, Italy (email@example.com).
Submitted for Publication: November 21, 2007; final revision received February 12, 2008; accepted April 7, 2008.
Author Contributions: Dr Redaelli de Zinis had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Financial Disclosure: None reported.