- Email: [email protected]
Incidence of Attic Retraction After Staged Intact Canal Wall Tympanoplasty for Middle Ear Cholesteatoma
Auris·Nasus·Larynx (Tokyo) 19, 75-82 (1992)
INCIDENCE OF ATTIC RETRACTION AFTER STAGED INTACT CANAL WALL TYMPANOPLASTY FOR MIDDLE EAR CHOLESTEATOMA Kiyofumi GYO, M.D., Yasuyuki HINOHIRA, M.D., Yoshinari HIRATA, M.D., and Naoaki YANAGIHARA, M.D. Department of Otolaryngology, Ehime University, Shigenobu-cho, Ehime, Japan
Incidence of retraction pocket and recurrent cholesteatoma in the attic after surgery for middle ear cholesteatoma using the staged intact canal wall technique were investigated in 95 ears of 91 patients, all of which had various degrees of bone defect in the tympanic scutum. Surgical procedures employed in the second stage for prevention of attic retraction were classified into three types: Type I, no scutumplasty; Type II, scutumplasty; Type III, scutumplasty plus mastoid obliteration. In 83 ears followed up for more than 1 year after the second stage, such retraction troubles occurred in 2 of 13 ears (15%) in Type I, 8 of 20 ears (40%) in Type II, and 24 of 50 ears ( 48%) in Type III. Incidence of retraction troubles was higher in Types II and III, probably because these procedures were indicated in ears with large scutum defects. Dislocation and atrophy of the graft materials, together with bone resorption around the bone defect were the main reasons for failure in scutumplasty. Dysfunction of the eustachian tube and traction of the eardrum by the scar tissue behind it may have also contributed to attic retraction. Mastoid obliteration with small blocks of hydroxyapatite was more effective in prevention of retraction troubles than that with pedicled temporalis muscle flap. In surgical treatment of chronic otitis media with cholesteatoma, it remains controversial whether the bony wall of the external ear canal should be removed or not. The technique of intact canal wall tympanoplasty (ICWT) has the advantage of avoiding cavity problems; however, residue and recurrence of cholesteatoma occur with some frequency with this procedure. While staging of the procedure is effective to control residual disease, prevention of recurrence is not always possible. We have attempted various procedures to repair the bony defects in the ear canal during ICWT operation. The purpose of the present study is to clarify the Received for publication August 14, 1991 75
76
K. GYOet a/.
incidence of attic retraction and recurrent cholesteatoma after staged ICWT operation for cholesteatoma with respect to surgical procedures, operative findings, and periods after surgery. PATIENTS AND METHODS During the last 10 years, from January 1981 to December 1990, planned two-staged operations by ICWT were attempted in 100 ears of 96 patients suffering from middle ear cholesteatoma with various degrees of bone defect in the tympanic scutum. The patients included 56 men and 40 women, ranging from 5 to 65 years in age, with an average of 31.6± 17.3 years. Cholesteatoma was classified as attic retraction type in 48 ears, PSQ (posterior-superior quadrant) type in 15 ears, adhesive type in 13 ears, and others in 24 ears. Surgical procedures followed techniques described by SHEEHY and PATTERSON (1967). In the first stage (the eradicating stage), following completion of mastoidectomy and posterior tympanotomy, the cholesteatoma was everted into the external ear canal and removed together with the incus after disconnecting the incudo-stapedial joint. A 0.5 mm thick silicone rubber sheet was then introduced into the middle ear through the facial recess and epitympanum to prevent adhesion and to obtain an aerated middle ear cavity. No ossiculoplasty was included in this stage; it was postponed until the second stage. In the present series of patients, various degrees of bone defect were noted in the tympanic scutum. When the defect in the scutum was substantial, it was closed using various graft materials, such as auricular or tragal cartilage, bone chip or a block of hydroxyapatite, in order to prevent invagination of the overlying skin during subsequent stages. Cartilage graft was shaped to be slightly larger than the defect and placed over it from the canal side. Bone or hydroxyapatite graft was trimmed just large enough to wedge into the scutum defect. If the defect was large, a 1 mm thickness of Gore-Tex®, an artificial textile made of polytetrafluoroethylene, was applied over the defect. Following closure of the defects by these procedures, the temporalis fascia was grafted over these materials and any perforation of the tympanic membrane was closed. The second stage (the reconstructing stage) was scheduled for eight to twelve months later. This interval actually averaged 10.2±5.2 months, ranging from 4 to 44 months. Any occurrence of attic retraction was recorded at the preoperative checkup and during surgery. In this stage, the silicone rubber sheet was removed, the remaining pathology eradicated, and the ossicular chain reconstructed. Defects in the ear canal were then repaired with autologous auricular or tragal cartilage, when deemed necessary. Bone and hydroxyapatite used in the first stage were replaced by cartilage, except in three cases in which such grafts were found to be adhered to the surrounding bone. Gore-Tex® was also replaced by cartilage, since long-term biocompatibility of this material in the ear remains unclear. In 50 ears,
ATTIC RETRACTION AFTER TYMPANOPLASTY
77
ill
II HA
Fig. 1. Types of surgery undergone in the second stage for prevention of attic retraction. Type I, no reconstruction of scutum defect except with use of temporalis fascia; Type II, scutumplasty mainly with cartilage; Type III, scutumplasty plus mastoid obliteration. HA, hydroxyapatite ossicle; C, cartilage; PMF, pedicled muscle flap; B, blocks of hydroxyapatite.
the mastoid cavity was obliterated with the pedicled temporalis muscle flap and/or blocks of hydroxyapatite. In 5 ears in which extensive residual or recurrent cholesteatoma were recognized, or the canal wall was extensively absorbed during this stage, the posterior canal wall was drilled down and the mastoid cavity opened to the external ear canal. These five cases were excluded in the following study. Postoperative incidence of retraction pocket and recurrent cholesteatoma were evaluated in respect to the types of surgical procedures, operative findings, and periods after the second stage. Although there were great individual variations in the size and shape of scutum defects and in types of cholesteatoma, these were not included in the analysis of the present study. Surgical procedures employed in the second stage for prevention of attic retraction were classified as follows: Type I, no scutumplasty; Type II, scutumplasty; Type III, scutumplasty plus mastoid obliteration (Fig. 1). The results were evaluated from the latest operating microscopic findings in the patients followed-up for more than 1 year after the second stage. According to Tos and PouLSEN ( 1980), retraction pocket is defined as invagination of the overlying skin into the attic with the bottom visible from the ear canal, and recurrent cholesteatoma as a deep retraction pocket without visible bottom which is usually accompanied with accumulation of debris. In this paper, retraction pocket was further divided by appearance into shallow and deep. Hearing results are not presented in this article, since the main aim of this paper is to describe the anatomical problems arising from the scutum defect in staged ICWT for cholesteatoma. RESULTS Incidences of attic retraction recognized in the second stage are summarized in Table 1 with respect to the materials used in the first stage for closure of the scutum defects. Retraction pocket or recurrent cholesteatoma occurred in 36 of 75 ears
K. GYOetal.
78
Table 1. Incidence of retraction pockets and recurrent cholesteatoma in the second stage operation with respect to the materials used in the first stage for closure of the scutum defects. Retraction troubles
Fascia and
Fascia alone
Cartilage
39
2
13
21 2
0 0 0
2 0
75
2
10
None Retraction pocket Shallow Deep Recurrent chole. Total
Bone
Hydroxyapatite Gore-Tex®
. (number of ears) 7 1
Total
6
55
0
3 2 0
24 3
2
11
100
0
18
Table 2. Incidence of retraction pockets and histological reactions to the silicone sheet recognized in the second stage. In 5 ears, the sheet was covered with recurrent or residual cholesteatoma and such cases were excluded in this table. Encapsulation of silicone sheet
Retraction pocket
Not used
None
None Shallow Deep
0 2 2
8 I 8
Total
4
17
Thin tissue
Thick tissue
(number of ears) 22 7
23
Total
53
8
18
11
3
24
40
34
95
( 48%) in which the defect had been closed with the temporalis fascia alone in the first stage. When the defect had been closed using various graft materials in addition to the fascia, such problems occurred in 9 of 25 ears (36%). Causes of failures to prevent retraction in the latter group were mainly due to dislocation or resorption of the graft materials. In two of the Gore-Tex® implantation cases, retraction occurred by invagination of the eardrum skin creeping below the Gore-Tex®. Histological reactions to the silicone rubber sheet introduced in the middle ear in the first stage were related to the incidence of attic retraction. As shown in Table 2, cases in which the sheet was encapsulated by thick fibrous tissue had a lower incidence of retraction pockets than other groups, especially that of deep retraction pockets; it occurred in 3 of 34 ears (9%) in the thick tissue encapsulated group, 11 of 40 ears (28%) in the thin tissue encapsulated group, and 8 of 17 ears (47%) in the non-encapsulated group. These results indicate that formation of the thick fibrous tissue around the scutum defect served to prevent retraction of the canal skin, although hindering aeration of the mastoid cavity at the same time. Surgical results were evaluated in the patients followed-up for more than one
ATTIC RETRACTION AFTER TYMPANOPLASTY
79
Table 3. (A) Surgical procedures and incidence of retraction troubles after the second stage. (B) Effect of mastoid obliteration with pedicled muscle flap and/or blocks of hydroxyapatite. The data were obtained from patients followed-up for more than one year. (A)
Retraction troubles Uneventful Retraction pocket Shallow Deep Recurrent chole. Total
Type I
11
Type II
Type III
Total
(number of ears) 12 26
2 0 0
5 2
13
20
49
10 10
17 12
4
5
50
83
(B)
Retraction troubles Uneventful Retraction pocket Shallow Deep Recurrent chole. Total
CD Hydroxyapatite
® Pedicled muscle
CD+®
12 (80%)
(number of ears) 8 (35%)
6 (50%)
3 (20%) 0 0
4(17%) 8 (35%) 3 (13%)
3 (25%) 2 (17%) 1 (8%)
15
23
12
year after the second stage. As shown in Table 3A, incidence of retraction troubles, which included shallow and deep retraction pockets and recurrent cholesteatoma, occurred in 2 of 13 ears (15%) in Type I, 8 of 20 ears (40%) in Type II, and 24 of 50 ears ( 48%) in Type III. Incidence of retraction troubles was higher in Types II and III, probably because these procedures were performed in ears with large scutum defects often complicated by severe middle ear pathology. Table 3B shows the effect of mastoid obliteration on prevention of retraction troubles in respect to the graft materials. Compared to blocks of hydroxyapatite, pedicled temporalis muscle was not so effective in prevention of such troubles, probably owing to muscle shrinkage. Table 4 shows the relationships between the incidence of retraction troubles in the second stage and that of follow-up examination more than one year after the second stage. In ears showing deep retraction pockets in the second stage, retraction troubles tended to recur after the second stage in spite of closure of the scutum defect. Incidence of postoperative retraction increased corresponding to the increase in periods after the operation (Table 5). Within one year, the incidence was very low. However, after one year, it increased. This was typically seen in Types II and III, in which scutumplasty was performed in the second stage.
K. GYOetal.
80
Table 4. Relationships between the incidences of retraction troubles in the second stage and those after the second stage. The data were obtained from patients followed-up for more than one year after the second stage. After second stage None
Shallow ret. pocket
Deep ret. pocket
Total
Recurrent cholesteatoma
(number of ears) At second stage None Shallow ret. pocket Deep ret. pocket
31 11 7
8 2 7
6 5
3
46 15 22
Total
49
17
12
5
83
Table 5. Incidence of retraction troubles after the second stage. Denominators indicate the number of ears being followed-up for years after the second stage shown in the left column, and numerators the number of ears in which retraction troubles occurred during these periods. (years)
Type I
Type II
Type III
1/13 1/13 1/8 0/6
(number of ears) 3/24 7/20 5/12 3/8
2/58 16/50 15/25 4/7
6/95 24/83 21/45 7/21
25/58 (43%)
35/95 (37%)
Total
2/13 (15%)
8/24 (33%)
Total (6%) (29%) (47%) (33%)
At present, retraction troubles occurred overall in 35 of 95 ears (37%), in which ICWT was performed in the second stage. The incidence was not related to the age of the patients; they occurred in 9 of 24 ears (38%) in the 15-and-under years in age group and in 26 of 71 ears (37%) in the over-16 age group. However, severe retraction troubles tended to occur in children. Five children aged 5 to 14 years old were re-operated 2 to 7 years after the second stage due to recurrence of cholesteatoma: three by radical mastoidectomy and two by ICWT without mastoid obliteration. Postoperative courses of them are uneventful so far. Incidence of retraction troubles differed by the types of cholesteatoma; they occurred in 12 of 46 ears (26%) in attic retraction type, 6 of 15 ears ( 40%) in PSQ type, 6 of 11 ears (55%) in adhesive type, and 11 of 23 ears (48%) in other types. This probably reflects the severity of middle ear pathology in these ears. DISCUSSION
Repair of a bone defect in the ear canal is important in surgical treatment of cholesteatoma by ICWT, since retraction of the canal skin often occurs through the
ATTIC RETRACTION AFTER TYMPANOPLASTY
81
defect. In the staged ICWT operation, bone defect in the scutum, when large, should be closed in the first stage, in order to prevent invagination of the overlying skin during subsequent stages, which often makes the second procedure difficult. As was indicated in Table 2, a thick fibrous tissue formed around the silicone rubber sheet had some effect on prevention of attic retraction, but this effect cannot be depended upon too greatly. Reconstruction of the canal wall in the second stage is much more important. This is accomplished by using various graft materials. Size, shape, and quality of the materials, together with method of placement, are critical for satisfactory closure of the defect. As pointed out by RoBINSON (1988), closure with fascia or connective tissue alone is often insufficient to prevent retraction because of its later atrophy. At present, autograft cartilage and bone plate taken from the mastoid cortex are commonly used for such a purpose, whether operative procedure is staged or not. Such materials are usually placed over the defect from the canal side or wedged into the scutum defect. However, these procedures are not always effective in prevention of retraction troubles. The main reasons for failure appeared to be (i) migration or (ii) atrophy of the graft material, (iii) progress of bone resorption around the defect, and (iv) dislocation of the graft due to traction by scar tissue behind it. Dysfunction of the eustachian tube and remaining middle ear inflammation may be background causes of these troubles. In order to avoid migration, SAKAI, SHINKAWA, MIYAKE, and Fum (1986) reported an innovative method whereby a bone plate taken from the mastoid cortex is fitted between two grooves made at the scutum. TAKEDA, SAITO, TAKEUCHI, and UCHIDA (1988) described a similar technique using a plate of alumina ceramic. Now, fixture of the graft to the canal wall has been facilitated by the development of fibrin glue. Mastoid obliteration is considered to be a useful method for prevention of attic retraction. Various materials such as muscle, fat, bone chips, and hydroxyapatite (HA) are used for obliteration. However, the former three materials often suffer from postoperative atrophy or absorption (OJALA, SoRRI, SIPILA, and PALVA, 1982). In our cases, as shown in Table 3B, HA was superior to the pedicled muscle flap in avoiding incidence of postoperative retraction troubles. Since HA has good biocompatibility and stability in the human body with identical chemical composition to bone, HA is considered to be a useful material for obliteration of the mastoid cavity. In spite of scutumplasty and mastoid obliteration, retraction pocket and recurrent cholesteatoma occurred in a substantial number of the patients in our series. In some of these cases, the scutum was well-repaired anatomically, but the retraction occurred by invagination of the eardrum below the reconstructed scutum. This indicates that the canal wall defect is not the only cause of attic retraction. Dysfunction of the eustachian tube and traction of the eardrum by scar tissue behind it might also have caused retraction in these cases. Discouraged by these findings, SADE (1982) recommended conservative radical mastoidectomy in
K. GYOetal.
82
surgical treatment for cholesteatoma, especially in ears with poor mastoid pneumatization. In contrast, we still feel that the ICWT technique is an appropriate procedure in surgical treatment for cholesteatoma because of its advantages in avoiding cavity problems and in providing adequate postoperative hearing in most of the patients. As was pointed out by SHELTON and SHEEHY (1990) and MERCKE ( 1986), the ICWT procedure should be planned in two stages, because the frequency of remnant cholesteatoma is so high. Obliteration of the mastoid cavity is helpful to some extent in prevention of retraction troubles in the attic, which present the greatest drawback of ICWT. Although retraction pocket is considered a precursor of cholesteatoma, it maintains a self-cleansing function and does not require re-operation in most cases. Careful and long-term postoperative follow-up is inevitable in surgical management of cholesteatoma, irrespective of the procedures employed. REFERENCES MERCKE, U.: Anatomic findings one year after combined approach tympanoplasty. Am. J. Otol. 7: 15G154, 1986. OJALA, K., SORRI, M., SIPILA, P., and PALVA, A.: Late changes in ear canal volumes after mastoid obliteration. Arch. Otolaryngol. 108: 208-209, 1982. ROBINSON, J. M.: Repair of the outer attic wall in closed cavity mastoidectomy. In Transplant and Implant in Otology (Babighian, G., and Veldman, J. E., eds.), pp. 173-174, Kugler and Ghedini Publications, Amsterdam/Berkeley/Milano, 1988. SADE, J.: Treatment of retraction pockets and cholesteatoma. J. Laryngol. Otol. 96: 685-704, 1982. SAKAI, M., SHINKAWA, A., MIYAKE, H., and Fum, K.: Reconstruction of scutum defects (scutumplasty) for attic cholesteatoma. Am. J. Otol. 7: 188-192, 1986. SHEEHY, J. L., and PATTERSON, M. E.: Intact canal wall tympanoplasty with mastoidectomy. A review of eight years' experience. Laryngoscope 77: 1502-1542, 1967. SHELTON, C., and SHEEHY, J.: Tympanoplasty: review of 400 staged cases. Laryngoscope 100: 679-681, 1990. TAKEDA, T., SAITO, H., TAKEUCHI, S., and UCHIDA, K.: Reconstruction of bony meatus with aluminum ceramics (Bioceram). In Transplant and Implant in Otology (Babighian, G., and Veldman, J. E., eds.), pp. 245-251, Kugler and Ghedini Publications, Amsterdam/Berkeley/Milano, 1988. Tos, M., and POULSEN, G.: Attic retractions following secretary otitis. Acta Otolaryngol. 89: 479-486, 1980.
Request reprints to:
Dr. K. Gyo, Department of Otolaryngology, Ehime University School of Medicine, Shigenobu-cho, Ehime 791-{)2, Japan
INCIDENCE OF ATTIC RETRACTION AFTER STAGED INTACT CANAL WALL TYMPANOPLASTY FOR MIDDLE EAR CHOLESTEATOMA Kiyofumi GYO, M.D., Yasuyuki HINOHIRA, M.D., Yoshinari HIRATA, M.D., and Naoaki YANAGIHARA, M.D. Department of Otolaryngology, Ehime University, Shigenobu-cho, Ehime, Japan
Incidence of retraction pocket and recurrent cholesteatoma in the attic after surgery for middle ear cholesteatoma using the staged intact canal wall technique were investigated in 95 ears of 91 patients, all of which had various degrees of bone defect in the tympanic scutum. Surgical procedures employed in the second stage for prevention of attic retraction were classified into three types: Type I, no scutumplasty; Type II, scutumplasty; Type III, scutumplasty plus mastoid obliteration. In 83 ears followed up for more than 1 year after the second stage, such retraction troubles occurred in 2 of 13 ears (15%) in Type I, 8 of 20 ears (40%) in Type II, and 24 of 50 ears ( 48%) in Type III. Incidence of retraction troubles was higher in Types II and III, probably because these procedures were indicated in ears with large scutum defects. Dislocation and atrophy of the graft materials, together with bone resorption around the bone defect were the main reasons for failure in scutumplasty. Dysfunction of the eustachian tube and traction of the eardrum by the scar tissue behind it may have also contributed to attic retraction. Mastoid obliteration with small blocks of hydroxyapatite was more effective in prevention of retraction troubles than that with pedicled temporalis muscle flap. In surgical treatment of chronic otitis media with cholesteatoma, it remains controversial whether the bony wall of the external ear canal should be removed or not. The technique of intact canal wall tympanoplasty (ICWT) has the advantage of avoiding cavity problems; however, residue and recurrence of cholesteatoma occur with some frequency with this procedure. While staging of the procedure is effective to control residual disease, prevention of recurrence is not always possible. We have attempted various procedures to repair the bony defects in the ear canal during ICWT operation. The purpose of the present study is to clarify the Received for publication August 14, 1991 75
76
K. GYOet a/.
incidence of attic retraction and recurrent cholesteatoma after staged ICWT operation for cholesteatoma with respect to surgical procedures, operative findings, and periods after surgery. PATIENTS AND METHODS During the last 10 years, from January 1981 to December 1990, planned two-staged operations by ICWT were attempted in 100 ears of 96 patients suffering from middle ear cholesteatoma with various degrees of bone defect in the tympanic scutum. The patients included 56 men and 40 women, ranging from 5 to 65 years in age, with an average of 31.6± 17.3 years. Cholesteatoma was classified as attic retraction type in 48 ears, PSQ (posterior-superior quadrant) type in 15 ears, adhesive type in 13 ears, and others in 24 ears. Surgical procedures followed techniques described by SHEEHY and PATTERSON (1967). In the first stage (the eradicating stage), following completion of mastoidectomy and posterior tympanotomy, the cholesteatoma was everted into the external ear canal and removed together with the incus after disconnecting the incudo-stapedial joint. A 0.5 mm thick silicone rubber sheet was then introduced into the middle ear through the facial recess and epitympanum to prevent adhesion and to obtain an aerated middle ear cavity. No ossiculoplasty was included in this stage; it was postponed until the second stage. In the present series of patients, various degrees of bone defect were noted in the tympanic scutum. When the defect in the scutum was substantial, it was closed using various graft materials, such as auricular or tragal cartilage, bone chip or a block of hydroxyapatite, in order to prevent invagination of the overlying skin during subsequent stages. Cartilage graft was shaped to be slightly larger than the defect and placed over it from the canal side. Bone or hydroxyapatite graft was trimmed just large enough to wedge into the scutum defect. If the defect was large, a 1 mm thickness of Gore-Tex®, an artificial textile made of polytetrafluoroethylene, was applied over the defect. Following closure of the defects by these procedures, the temporalis fascia was grafted over these materials and any perforation of the tympanic membrane was closed. The second stage (the reconstructing stage) was scheduled for eight to twelve months later. This interval actually averaged 10.2±5.2 months, ranging from 4 to 44 months. Any occurrence of attic retraction was recorded at the preoperative checkup and during surgery. In this stage, the silicone rubber sheet was removed, the remaining pathology eradicated, and the ossicular chain reconstructed. Defects in the ear canal were then repaired with autologous auricular or tragal cartilage, when deemed necessary. Bone and hydroxyapatite used in the first stage were replaced by cartilage, except in three cases in which such grafts were found to be adhered to the surrounding bone. Gore-Tex® was also replaced by cartilage, since long-term biocompatibility of this material in the ear remains unclear. In 50 ears,
ATTIC RETRACTION AFTER TYMPANOPLASTY
77
ill
II HA
Fig. 1. Types of surgery undergone in the second stage for prevention of attic retraction. Type I, no reconstruction of scutum defect except with use of temporalis fascia; Type II, scutumplasty mainly with cartilage; Type III, scutumplasty plus mastoid obliteration. HA, hydroxyapatite ossicle; C, cartilage; PMF, pedicled muscle flap; B, blocks of hydroxyapatite.
the mastoid cavity was obliterated with the pedicled temporalis muscle flap and/or blocks of hydroxyapatite. In 5 ears in which extensive residual or recurrent cholesteatoma were recognized, or the canal wall was extensively absorbed during this stage, the posterior canal wall was drilled down and the mastoid cavity opened to the external ear canal. These five cases were excluded in the following study. Postoperative incidence of retraction pocket and recurrent cholesteatoma were evaluated in respect to the types of surgical procedures, operative findings, and periods after the second stage. Although there were great individual variations in the size and shape of scutum defects and in types of cholesteatoma, these were not included in the analysis of the present study. Surgical procedures employed in the second stage for prevention of attic retraction were classified as follows: Type I, no scutumplasty; Type II, scutumplasty; Type III, scutumplasty plus mastoid obliteration (Fig. 1). The results were evaluated from the latest operating microscopic findings in the patients followed-up for more than 1 year after the second stage. According to Tos and PouLSEN ( 1980), retraction pocket is defined as invagination of the overlying skin into the attic with the bottom visible from the ear canal, and recurrent cholesteatoma as a deep retraction pocket without visible bottom which is usually accompanied with accumulation of debris. In this paper, retraction pocket was further divided by appearance into shallow and deep. Hearing results are not presented in this article, since the main aim of this paper is to describe the anatomical problems arising from the scutum defect in staged ICWT for cholesteatoma. RESULTS Incidences of attic retraction recognized in the second stage are summarized in Table 1 with respect to the materials used in the first stage for closure of the scutum defects. Retraction pocket or recurrent cholesteatoma occurred in 36 of 75 ears
K. GYOetal.
78
Table 1. Incidence of retraction pockets and recurrent cholesteatoma in the second stage operation with respect to the materials used in the first stage for closure of the scutum defects. Retraction troubles
Fascia and
Fascia alone
Cartilage
39
2
13
21 2
0 0 0
2 0
75
2
10
None Retraction pocket Shallow Deep Recurrent chole. Total
Bone
Hydroxyapatite Gore-Tex®
. (number of ears) 7 1
Total
6
55
0
3 2 0
24 3
2
11
100
0
18
Table 2. Incidence of retraction pockets and histological reactions to the silicone sheet recognized in the second stage. In 5 ears, the sheet was covered with recurrent or residual cholesteatoma and such cases were excluded in this table. Encapsulation of silicone sheet
Retraction pocket
Not used
None
None Shallow Deep
0 2 2
8 I 8
Total
4
17
Thin tissue
Thick tissue
(number of ears) 22 7
23
Total
53
8
18
11
3
24
40
34
95
( 48%) in which the defect had been closed with the temporalis fascia alone in the first stage. When the defect had been closed using various graft materials in addition to the fascia, such problems occurred in 9 of 25 ears (36%). Causes of failures to prevent retraction in the latter group were mainly due to dislocation or resorption of the graft materials. In two of the Gore-Tex® implantation cases, retraction occurred by invagination of the eardrum skin creeping below the Gore-Tex®. Histological reactions to the silicone rubber sheet introduced in the middle ear in the first stage were related to the incidence of attic retraction. As shown in Table 2, cases in which the sheet was encapsulated by thick fibrous tissue had a lower incidence of retraction pockets than other groups, especially that of deep retraction pockets; it occurred in 3 of 34 ears (9%) in the thick tissue encapsulated group, 11 of 40 ears (28%) in the thin tissue encapsulated group, and 8 of 17 ears (47%) in the non-encapsulated group. These results indicate that formation of the thick fibrous tissue around the scutum defect served to prevent retraction of the canal skin, although hindering aeration of the mastoid cavity at the same time. Surgical results were evaluated in the patients followed-up for more than one
ATTIC RETRACTION AFTER TYMPANOPLASTY
79
Table 3. (A) Surgical procedures and incidence of retraction troubles after the second stage. (B) Effect of mastoid obliteration with pedicled muscle flap and/or blocks of hydroxyapatite. The data were obtained from patients followed-up for more than one year. (A)
Retraction troubles Uneventful Retraction pocket Shallow Deep Recurrent chole. Total
Type I
11
Type II
Type III
Total
(number of ears) 12 26
2 0 0
5 2
13
20
49
10 10
17 12
4
5
50
83
(B)
Retraction troubles Uneventful Retraction pocket Shallow Deep Recurrent chole. Total
CD Hydroxyapatite
® Pedicled muscle
CD+®
12 (80%)
(number of ears) 8 (35%)
6 (50%)
3 (20%) 0 0
4(17%) 8 (35%) 3 (13%)
3 (25%) 2 (17%) 1 (8%)
15
23
12
year after the second stage. As shown in Table 3A, incidence of retraction troubles, which included shallow and deep retraction pockets and recurrent cholesteatoma, occurred in 2 of 13 ears (15%) in Type I, 8 of 20 ears (40%) in Type II, and 24 of 50 ears ( 48%) in Type III. Incidence of retraction troubles was higher in Types II and III, probably because these procedures were performed in ears with large scutum defects often complicated by severe middle ear pathology. Table 3B shows the effect of mastoid obliteration on prevention of retraction troubles in respect to the graft materials. Compared to blocks of hydroxyapatite, pedicled temporalis muscle was not so effective in prevention of such troubles, probably owing to muscle shrinkage. Table 4 shows the relationships between the incidence of retraction troubles in the second stage and that of follow-up examination more than one year after the second stage. In ears showing deep retraction pockets in the second stage, retraction troubles tended to recur after the second stage in spite of closure of the scutum defect. Incidence of postoperative retraction increased corresponding to the increase in periods after the operation (Table 5). Within one year, the incidence was very low. However, after one year, it increased. This was typically seen in Types II and III, in which scutumplasty was performed in the second stage.
K. GYOetal.
80
Table 4. Relationships between the incidences of retraction troubles in the second stage and those after the second stage. The data were obtained from patients followed-up for more than one year after the second stage. After second stage None
Shallow ret. pocket
Deep ret. pocket
Total
Recurrent cholesteatoma
(number of ears) At second stage None Shallow ret. pocket Deep ret. pocket
31 11 7
8 2 7
6 5
3
46 15 22
Total
49
17
12
5
83
Table 5. Incidence of retraction troubles after the second stage. Denominators indicate the number of ears being followed-up for years after the second stage shown in the left column, and numerators the number of ears in which retraction troubles occurred during these periods. (years)
Type I
Type II
Type III
1/13 1/13 1/8 0/6
(number of ears) 3/24 7/20 5/12 3/8
2/58 16/50 15/25 4/7
6/95 24/83 21/45 7/21
25/58 (43%)
35/95 (37%)
Total
2/13 (15%)
8/24 (33%)
Total (6%) (29%) (47%) (33%)
At present, retraction troubles occurred overall in 35 of 95 ears (37%), in which ICWT was performed in the second stage. The incidence was not related to the age of the patients; they occurred in 9 of 24 ears (38%) in the 15-and-under years in age group and in 26 of 71 ears (37%) in the over-16 age group. However, severe retraction troubles tended to occur in children. Five children aged 5 to 14 years old were re-operated 2 to 7 years after the second stage due to recurrence of cholesteatoma: three by radical mastoidectomy and two by ICWT without mastoid obliteration. Postoperative courses of them are uneventful so far. Incidence of retraction troubles differed by the types of cholesteatoma; they occurred in 12 of 46 ears (26%) in attic retraction type, 6 of 15 ears ( 40%) in PSQ type, 6 of 11 ears (55%) in adhesive type, and 11 of 23 ears (48%) in other types. This probably reflects the severity of middle ear pathology in these ears. DISCUSSION
Repair of a bone defect in the ear canal is important in surgical treatment of cholesteatoma by ICWT, since retraction of the canal skin often occurs through the
ATTIC RETRACTION AFTER TYMPANOPLASTY
81
defect. In the staged ICWT operation, bone defect in the scutum, when large, should be closed in the first stage, in order to prevent invagination of the overlying skin during subsequent stages, which often makes the second procedure difficult. As was indicated in Table 2, a thick fibrous tissue formed around the silicone rubber sheet had some effect on prevention of attic retraction, but this effect cannot be depended upon too greatly. Reconstruction of the canal wall in the second stage is much more important. This is accomplished by using various graft materials. Size, shape, and quality of the materials, together with method of placement, are critical for satisfactory closure of the defect. As pointed out by RoBINSON (1988), closure with fascia or connective tissue alone is often insufficient to prevent retraction because of its later atrophy. At present, autograft cartilage and bone plate taken from the mastoid cortex are commonly used for such a purpose, whether operative procedure is staged or not. Such materials are usually placed over the defect from the canal side or wedged into the scutum defect. However, these procedures are not always effective in prevention of retraction troubles. The main reasons for failure appeared to be (i) migration or (ii) atrophy of the graft material, (iii) progress of bone resorption around the defect, and (iv) dislocation of the graft due to traction by scar tissue behind it. Dysfunction of the eustachian tube and remaining middle ear inflammation may be background causes of these troubles. In order to avoid migration, SAKAI, SHINKAWA, MIYAKE, and Fum (1986) reported an innovative method whereby a bone plate taken from the mastoid cortex is fitted between two grooves made at the scutum. TAKEDA, SAITO, TAKEUCHI, and UCHIDA (1988) described a similar technique using a plate of alumina ceramic. Now, fixture of the graft to the canal wall has been facilitated by the development of fibrin glue. Mastoid obliteration is considered to be a useful method for prevention of attic retraction. Various materials such as muscle, fat, bone chips, and hydroxyapatite (HA) are used for obliteration. However, the former three materials often suffer from postoperative atrophy or absorption (OJALA, SoRRI, SIPILA, and PALVA, 1982). In our cases, as shown in Table 3B, HA was superior to the pedicled muscle flap in avoiding incidence of postoperative retraction troubles. Since HA has good biocompatibility and stability in the human body with identical chemical composition to bone, HA is considered to be a useful material for obliteration of the mastoid cavity. In spite of scutumplasty and mastoid obliteration, retraction pocket and recurrent cholesteatoma occurred in a substantial number of the patients in our series. In some of these cases, the scutum was well-repaired anatomically, but the retraction occurred by invagination of the eardrum below the reconstructed scutum. This indicates that the canal wall defect is not the only cause of attic retraction. Dysfunction of the eustachian tube and traction of the eardrum by scar tissue behind it might also have caused retraction in these cases. Discouraged by these findings, SADE (1982) recommended conservative radical mastoidectomy in
K. GYOetal.
82
surgical treatment for cholesteatoma, especially in ears with poor mastoid pneumatization. In contrast, we still feel that the ICWT technique is an appropriate procedure in surgical treatment for cholesteatoma because of its advantages in avoiding cavity problems and in providing adequate postoperative hearing in most of the patients. As was pointed out by SHELTON and SHEEHY (1990) and MERCKE ( 1986), the ICWT procedure should be planned in two stages, because the frequency of remnant cholesteatoma is so high. Obliteration of the mastoid cavity is helpful to some extent in prevention of retraction troubles in the attic, which present the greatest drawback of ICWT. Although retraction pocket is considered a precursor of cholesteatoma, it maintains a self-cleansing function and does not require re-operation in most cases. Careful and long-term postoperative follow-up is inevitable in surgical management of cholesteatoma, irrespective of the procedures employed. REFERENCES MERCKE, U.: Anatomic findings one year after combined approach tympanoplasty. Am. J. Otol. 7: 15G154, 1986. OJALA, K., SORRI, M., SIPILA, P., and PALVA, A.: Late changes in ear canal volumes after mastoid obliteration. Arch. Otolaryngol. 108: 208-209, 1982. ROBINSON, J. M.: Repair of the outer attic wall in closed cavity mastoidectomy. In Transplant and Implant in Otology (Babighian, G., and Veldman, J. E., eds.), pp. 173-174, Kugler and Ghedini Publications, Amsterdam/Berkeley/Milano, 1988. SADE, J.: Treatment of retraction pockets and cholesteatoma. J. Laryngol. Otol. 96: 685-704, 1982. SAKAI, M., SHINKAWA, A., MIYAKE, H., and Fum, K.: Reconstruction of scutum defects (scutumplasty) for attic cholesteatoma. Am. J. Otol. 7: 188-192, 1986. SHEEHY, J. L., and PATTERSON, M. E.: Intact canal wall tympanoplasty with mastoidectomy. A review of eight years' experience. Laryngoscope 77: 1502-1542, 1967. SHELTON, C., and SHEEHY, J.: Tympanoplasty: review of 400 staged cases. Laryngoscope 100: 679-681, 1990. TAKEDA, T., SAITO, H., TAKEUCHI, S., and UCHIDA, K.: Reconstruction of bony meatus with aluminum ceramics (Bioceram). In Transplant and Implant in Otology (Babighian, G., and Veldman, J. E., eds.), pp. 245-251, Kugler and Ghedini Publications, Amsterdam/Berkeley/Milano, 1988. Tos, M., and POULSEN, G.: Attic retractions following secretary otitis. Acta Otolaryngol. 89: 479-486, 1980.
Request reprints to:
Dr. K. Gyo, Department of Otolaryngology, Ehime University School of Medicine, Shigenobu-cho, Ehime 791-{)2, Japan