Abstract
Purpose
The aims of this study were to investigate the prognostic impact of middle ear risk index on the postoperative hearing results in cases with type 2 ossiculoplasty; to compare the middle ear risk index results among primary, staged, and revision cases; and to compare the results of the prostheses used in ossicular reconstruction.
Material and methods
Records of 293 patients who had canal wall up tympanomasteidectomy and type 2 ossiculoplasty due to chronic otitis media between November 1995 and November 2007 were reviewed retrospectively.
Results
The mean preoperative air-bone gap was 32.6 dB, and it decreased to 15.2 dB after a mean follow-up period of 26.8 months postoperatively. The mean change of air-bone gap was 17.4 dB. Postoperative air-bone gap was 20 dB or less in 79% of the cases. The patients with dry perforations were in the low-risk group, and 91% of them had an air-bone gap of 20 dB or less. This value was 86% in the ones with intact malleus. The patients who had primary surgery were found in moderate risk group, whereas staged and revision groups were in the high-risk group. The air-bone gap was 20 dB or less in 84%, 78%, and 59%, respectively, of those groups. The difference between the primary and the revision groups reached a statistical significance.
Conclusions
We had the best ossicular reconstruction results with glass ionomer cement, whereas the worst results were obtained with allograft partial ossicular replacement prostheses. We determined that risk-reducing factors such as dry ear, minimal ossicular chain defect, and intact malleus were important to have successful results. The middle ear risk index is a valuable tool for the surgeon to judge the risks and the probability success of the procedure as well as to make a good patient selection.
1
Introduction
The most frequent ossicular chain defect that the otologic surgeon encounters is the absence of incus or its defect . That is why otologic surgeons use type 2 ossiculoplasty as the most common reconstruction model in the ossicular defects . Type 2 ossiculoplasty includes interposition techniques between the head of the stapes and the malleus handle or the grafted eardrum. Interpositions between the stapedial arch in cases with a missing head or neck of the stapes and the malleus handle or eardrum are also defined as tympanoplasty type 2 .
A number of articles have reported results of various reconstruction techniques and prostheses for the patients with incus defects. The hearing results after ossiculoplasty are not only affected by the quality of the prostheses and the technique used but also by environmental factors (mucosal disease) where the prostheses is placed, presence or absence of malleus, dry or discharging ear, or primary or revision surgery. The prognostic value of pathologic and technical variables influencing the functional outcome of ossiculoplasty has been highlighted in various studies. In the literature, there are some established protocols to standardize the results and the comparisons. First, Austin [ ] developed a classification to standardize the defects of the ossicular chain and the hearing results. Then, Kartush modified the Austin classification and presented the middle ear risk index (MERI) to define those basic data and to stratify cases in different prognostic categories ( Table 1 ).
| Risk factor | Risk value | |
|---|---|---|
| Otorrhea | I, Dry | 0 |
| II, Occasionally wet | 1 | |
| III, Persistently wet | 2 | |
| IV, Wet, cleft palate | 3 | |
| Perforation | Absent | 0 |
| Present | 1 | |
| Cholesteatoma | O, M+ I+ S+ | 0 |
| A, M+ S+ | 1 | |
| B, M+ S− | 2 | |
| C, M− S+ | 3 | |
| D, M− S− | 4 | |
| E, Ossicle head fixation | 2 | |
| F, Stapes fixation | 3 | |
| Middle ear (granulations or effusion) | No | 0 |
| Yes | 1 | |
| Previous surgery | None | 0 |
| Staged | 1 | |
| Revision | 2 | |
⁎ A value is assigned for each risk factor and then the values are added to determine the MERI. M: malleus; I: incus; S: stapes; +: present; −: absent.
In this study, we scored each patient using MERI by Kartush and investigated the relationship between the status of the middle ear and the postoperative hearing results. We have compared the results of revision, staged, and primary cases of which MERIs were determined. We have analyzed the impact of the materials used for reconstruction on the hearing results.
2
Materials and methods
The charts of the patients who had canal wall up tympanomasteidectomy and type 2 ossiculoplasty due to chronic otitis media between November 1995 and November 2007 were reviewed retrospectively. The patients with a traumatic ossicular chain defect; an intact tympanic membrane; and a conduction type hearing loss, a congenital incus defect, and a follow-up less than 6 months were excluded. There were 293 patients who met the inclusion criteria of the study. All patients were operated on by the experienced senior authors who worked in the otology department and obeyed the same clinical principles.
The adults with dry perforations and the ones who had ossiculoplasty in the second stage were operated on under local anesthesia. All other patients were operated on under general anesthesia. All patients had canal wall up tympanomasteidectomy with the postauricular approach except the ones having a second stage. Temporal muscle fascia or tragal cartilage/perichondrium was used with the underlay technique for tympanic membrane reconstruction. The posterosuperior quadrant was supported by a cartilage graft in patients who had a scutum defect (attic cholesteatomas) or adhesive otitis media. The second-stage ossiculoplasty was performed by elevation of the transcanal endomeatal flap to reach the middle ear (6 months to 1 year after the first operation).
The MERI scores were determined for all patients ( Table 1 ). Then, for a detailed analysis, the type of the middle ear disease, the ossicular chain defect, risk categories, the type of the surgical procedure (primary, staged, or revision surgery), and the prostheses used were recorded, and the patients were divided into groups to compare the hearing results.
To compare the type of the middle ear disease and the hearing results, 293 patients were divided into 5 groups as follows: cholesteatoma, polypoid granular, dry scatricial, adhesive otitis, and tympanosclerosis groups. In the cholesteatoma group, the patients had cholesteatomas limited to the attic.
When the patients were grouped according to the ossicular chain defects, there were 3 groups designated as Austin-Kartush group A [(M+ I− S+), (presence of the malleus, absence of the incus, presence of the stapes)] Austin-Kartush group C [(M− I− S+), (absence of the malleus, absence of the incus, presence of the stapes)] and Austin-Kartush group E (fixation of the malleus head).
To compare the risk categories and the hearing results, the patients were divided into 3 groups. The patients with MERIs 1 to 3 were included into mild pathologic group, the ones with MERIs 4 to 6 were grouped into moderate pathologic group, and the ones with MERIs 7 to 12 were grouped into severe pathologic groups.
According to the type of the surgical procedure, the patients were divided into 3 groups as primary, staged, or revision groups. The patients were included into the revision group if they had myringoplasty or type 1 or type 2 ossiculoplasty in our center or in another with unsuccessful results (graft failure, draining ear, extruded prosthesis, or poor hearing).
For ossicular reconstruction, autograft ossicles (malleus or incus), autograft cortical bone, PORP (partial ossicular replacement prosthesis) (plastipore [Sheehy’s PORP prosthesis] or hydroxyapatite [Sheehy’s PORP prosthesis], [Smith and Nephew, Richards Medical Company]), cartilage or glass ionomer cement (Glasspolyalkenoate[ionomer] cement, Voco/meron corresponds to EN 29917/150/9917/1994[CE 0482]) were used. Glass ionomer cement and cortical bone were used to achieve conduction between the stapes head and the incus long process in case of the incus lenticular process defect. No allografts were used after 2003.
The audiogram performed within the preoperative 30 days and the one performed after postoperative sixth month were used for audiometric evaluation. Pure-tone averages (PTAs) were calculated at the frequencies of 500, 1000, 2000, and 3000 Hz.
For the purposes of this study, success was defined according to the guidelines delineated by the Committee on Hearing and Equilibrium of the Academy of Otolaryngology-Head and Neck Surgery for the evaluation of results for the treatment of conductive hearing loss. An air-bone gap (ABG) of 20 dB or less was regarded as a successful hearing result. An ABG of 10 dB or less was regarded as an excellent hearing result.
2.1
Statistical analysis
Comparisons between 2 groups for postoperative ABG and ABG change were performed by Mann-Whitney U test. Differences among more than 2 groups for postoperative ABG and ABG change were evaluated by Kruskal-Wallis variance analysis. When the P value from the Kruskal-Wallis test statistics was statistically significant, the multiple comparison test was used to know which groups differ from which others . SPSS for Windows 11.5 (SPSS Inc, Chicago, IL) was used for statistical evaluation. A P value less than .05 was considered statistically significant.
2
Materials and methods
The charts of the patients who had canal wall up tympanomasteidectomy and type 2 ossiculoplasty due to chronic otitis media between November 1995 and November 2007 were reviewed retrospectively. The patients with a traumatic ossicular chain defect; an intact tympanic membrane; and a conduction type hearing loss, a congenital incus defect, and a follow-up less than 6 months were excluded. There were 293 patients who met the inclusion criteria of the study. All patients were operated on by the experienced senior authors who worked in the otology department and obeyed the same clinical principles.
The adults with dry perforations and the ones who had ossiculoplasty in the second stage were operated on under local anesthesia. All other patients were operated on under general anesthesia. All patients had canal wall up tympanomasteidectomy with the postauricular approach except the ones having a second stage. Temporal muscle fascia or tragal cartilage/perichondrium was used with the underlay technique for tympanic membrane reconstruction. The posterosuperior quadrant was supported by a cartilage graft in patients who had a scutum defect (attic cholesteatomas) or adhesive otitis media. The second-stage ossiculoplasty was performed by elevation of the transcanal endomeatal flap to reach the middle ear (6 months to 1 year after the first operation).
The MERI scores were determined for all patients ( Table 1 ). Then, for a detailed analysis, the type of the middle ear disease, the ossicular chain defect, risk categories, the type of the surgical procedure (primary, staged, or revision surgery), and the prostheses used were recorded, and the patients were divided into groups to compare the hearing results.
To compare the type of the middle ear disease and the hearing results, 293 patients were divided into 5 groups as follows: cholesteatoma, polypoid granular, dry scatricial, adhesive otitis, and tympanosclerosis groups. In the cholesteatoma group, the patients had cholesteatomas limited to the attic.
When the patients were grouped according to the ossicular chain defects, there were 3 groups designated as Austin-Kartush group A [(M+ I− S+), (presence of the malleus, absence of the incus, presence of the stapes)] Austin-Kartush group C [(M− I− S+), (absence of the malleus, absence of the incus, presence of the stapes)] and Austin-Kartush group E (fixation of the malleus head).
To compare the risk categories and the hearing results, the patients were divided into 3 groups. The patients with MERIs 1 to 3 were included into mild pathologic group, the ones with MERIs 4 to 6 were grouped into moderate pathologic group, and the ones with MERIs 7 to 12 were grouped into severe pathologic groups.
According to the type of the surgical procedure, the patients were divided into 3 groups as primary, staged, or revision groups. The patients were included into the revision group if they had myringoplasty or type 1 or type 2 ossiculoplasty in our center or in another with unsuccessful results (graft failure, draining ear, extruded prosthesis, or poor hearing).
For ossicular reconstruction, autograft ossicles (malleus or incus), autograft cortical bone, PORP (partial ossicular replacement prosthesis) (plastipore [Sheehy’s PORP prosthesis] or hydroxyapatite [Sheehy’s PORP prosthesis], [Smith and Nephew, Richards Medical Company]), cartilage or glass ionomer cement (Glasspolyalkenoate[ionomer] cement, Voco/meron corresponds to EN 29917/150/9917/1994[CE 0482]) were used. Glass ionomer cement and cortical bone were used to achieve conduction between the stapes head and the incus long process in case of the incus lenticular process defect. No allografts were used after 2003.
The audiogram performed within the preoperative 30 days and the one performed after postoperative sixth month were used for audiometric evaluation. Pure-tone averages (PTAs) were calculated at the frequencies of 500, 1000, 2000, and 3000 Hz.
For the purposes of this study, success was defined according to the guidelines delineated by the Committee on Hearing and Equilibrium of the Academy of Otolaryngology-Head and Neck Surgery for the evaluation of results for the treatment of conductive hearing loss. An air-bone gap (ABG) of 20 dB or less was regarded as a successful hearing result. An ABG of 10 dB or less was regarded as an excellent hearing result.
2.1
Statistical analysis
Comparisons between 2 groups for postoperative ABG and ABG change were performed by Mann-Whitney U test. Differences among more than 2 groups for postoperative ABG and ABG change were evaluated by Kruskal-Wallis variance analysis. When the P value from the Kruskal-Wallis test statistics was statistically significant, the multiple comparison test was used to know which groups differ from which others . SPSS for Windows 11.5 (SPSS Inc, Chicago, IL) was used for statistical evaluation. A P value less than .05 was considered statistically significant.
3
Results
Of 293 patients, 123 were males and 170 were females. They were between 8 and 67 years (mean age, 31 ± 13 SD). There were 150 left ears and 143 right ears operated on upon. The range of the follow-up was between 6 and 96 months with an average of 26.8 ± 11.6 SD months.
Eighty-one patients had tympanosclerosis, 62 had cholesteatoma, 40 had adhesive otitis, 67 had polypoid granular middle ear disease, and 43 had dry perforations. Autograft ossicles were used in 149, cortical bone was used in 35, allograft ossicles were used in 39, allograft septal cartilage was used in 7, PORP was used in 47, and glass ionomer cement was used in 16 patients for ossicular reconstruction.
Two hundred fifteen patients had primary surgery, 32 had staged surgery, and 46 had revision surgery. Staged procedures were performed for cholesteatoma in 12 and for polypoid granular disease in 7 patients. Thirteen patients with tympanosclerosis who had attic and stapes fixations were staged for a stapedectomy. On the other hand, the stapes could be mobilized in the second-stage surgery. The fixated incus was removed, and a type 2 ossiculoplasty was achieved by positioning the removed incus between the stapes head and the handle of the malleus. Those patients were included in the staged surgery group.
The mean MERI score of all patients was 4.6 ± 2.2 SD, and the preoperative mean ABG of 32.6 dB was reduced to 15.2 dB in the postoperative period. The change of PTA-ABG was 17.4 dB. The closure of ABG to 20 dB or lower was evident in 79%, whereas closure to 10 dB or lower was achieved in 33% of the patients.
The mean MERI scores and the hearing results in relation to the middle ear disease are presented in Table 2 in detail. The mean MERI score was 5.6 in the tympanosclerosis group and the mean preoperative ABG of those patients was 35 dB in the preoperative period, whereas it reduced to 18.2 dB postoperatively. In the group with cholesteatoma, those values were 5.3, 33.9, and 14.8 dB, respectively; in dry perforation group, they were 2.6, 31.1, and 11 dB; in polypoid granular middle ear disease, they were 5.4, 32.9, and 16.1 dB; and in adhesive otitis, 2.8, 27.9, and 12.8 dB. The following were found statistically significant after statistical comparison of the mean postoperative ABGs of the groups: cholesteatoma to dry scatricial, P = .002; dry scatricial to polypoid granular, P = .000; tympanosclerosis to cholesteatoma, P = .015; tympanosclerosis to dry scatricial, P = .000; and tympanosclerosis to adhesive otitis, P = .001. The other comparisons were insignificant.
