Clin. Otolaryngol.
1992, 17, 6-9
Underlay tympanoplasty with the graft lateral to the malleus handle JAN STAGE & KRISTIAN BAK-PEDERSEN Department qf Otorhinoluryngoli~g.v:v, Rispebjerg Hospitul. Copenhugen. Denmark
Accepted for publication 28 November 1990 STAGE J . & B A K - P E D E R S E N K .
(1992) Clin.Otolaryngol. 17, 6-9
Underlay tympanoplasty with the graft lateral to the malleus handle In conventional underlay tympanoplasty the graft is placed medial to the malleus handle. The present study evaluates the results of a modification of this technique, in which the graft is placed lateral to the malleus handle, which has been dissected from the drum remnants. 39 ears with predominantly large or anterior pars tensa perforations were operated upon in this manner. After a median observation time of 20 months one ear was found to have a small reperforation. All ears had normal tympanomeatal angles, but 12 ears showed a small degree of laterofixation of the graft from the malleus handle. Analysis of the hearing showed good hearing improvement and no adverse effects attributable to the dissection of the ossicular chain. It is concluded that the technique is a good alternative to conventional underlay myringoplasty in ears with perforations involving the area anterior to the malleus handle. Keywords
underlay tympanoplasty
underlay myringoplusty
A very popular method of closing perforations of the tympanic membrane is the underlay technique using temporalis fascia as graft material. Normally, using this method, the graft is placed medial to the malleus handle. In certain pathological conditions of the drum, we have found it advantageous to place the graft lateral to the malleus handle. The purpose of this study was to describe the results of this variant of the underlay technique in a series of patients.
Materials and methods The material comprises 39 ears (38 patients) operated on from June 1984 to January 1989. 10 of the operations were revision operations. The median age of the patients was 46 years (range 7-78 years). The locations of the perforations are shown in Table I . In most cases, the indication for placing the graft lateral to the malleus handle was a perforation extending into the area anterior to the malleus handle (26 ears). Severe atrophy o r tympanosclerosis of the same area was the reason in another 9 ears with initially only posterior or inferior perforations. In Correspondence: Dr K . Bak-Pedersen, Department of Otorhinolaryngology, Bispebjerg Hospital, DK-2400 Copenhagen NV. Denmark.
6
malleus handle
the last 4 cases the drum remnant was dissected from the malleus handle for various reasons: pars flaccida perforation, severe pars Raccida retraction, cholesteatoma of the middle ear, o r in order to preserve a well functioning, earlier ossiculoplasty. The type of operation performed in 26 ears was a myringoplasty or tympanoplasty type I. Type TI tycnpanoplasty with incus interposition was performed in 12 ears, and type 111 with a ceramic prosthesis in one ear. All operations were performed under general anaesthesia. The access in all cases was primarily endaural, but in 21 ears a supplementary retroauricular incision was necessary. Having incised the canal skin posteriorly the skin and drum
Table 1. Location of perforations in 39 ears
No.
010
.
Total or subtotal tensa defects Anterior* Posterior or inferiort
19
49
7
13
18 33
Iota1
39
I00
*Involving the upper anterior quadrant. ?One ear having also a flaccid perforation.
Underlay tympanoplasty with the graft lateral to malleus handle
1
Table 2. Median pre- and post-operative speech reception threshold (SRT) and air-bone gap in tympanoplasty type I and I1
respectively Air-bone gap* (dB HL)
SRT (dB HL) Pre .~
~
Type I ( n = 26) Type I1 ( n = 12) ~
~
Post
Pre
Post
32.5 24.2
12.5 12.5
~~
40 42.5
20 22.5
~
*Mean values for 500, 1000 and 2000 Hz.
point any nccessary ossiculoplasty was performed. Finally, the graft and flaps were placed in situ (Figure I b) and the ear canal filled with Gelfoam innermost and upon this antibiotic-soaked gauze. All patients were operated upon by the authors and followed post-operatively at regular intervals. The results given reflect the status at the latest follow-up.
Results
Figure 1. A, Superior and inferior flaps raised, allowing full inspection of the middle ear. b, Temporalis fascia (F) placed against lateral wall of Eustachian tube (ET), lateral to the malleus handle (MH) and medial to the drum remnants (DR).
remnant were raised. Superior and inferior tympanomeatal flaps were created according to the ‘swinging door’ technique described by Palva et al.’ The inferior flap was elevated to a point near the anterior margin of the perforation and superiorly the drum remnant was carefully dissected from the malleus handle and short process. If necessary the whole pars flaccida area was dissected as well and incorporated in the superior flap (Figure la). In 10 cases the operator suspected epithelial ingrowth under the umbo region, and in these cases the tip of the malleus handle was resected with a malleus nipper. Having dealt with any middle ear abnormality temporalis fascia was then placed as an underlay graft medial to the drum remnant but lateral to the malleus handle. Anteriorly, the graft was pushed forward to achieve a good alignment against the lateral wall of the Eustachian orifice. Subsequently graft and flaps were displaced latcrally while the middle ear was filled with Gelfoam. Also, a t this
The median post-operative observation time was 20 months (range 1.5-58 months). Very few post-operative complications were seen. One car developed a staphylococcal infection and had a temporary reperforation, which closed after conservative treatment. N o cases of vertigo or deafness occurred. One patient felt that a prior existing tinnitus was somewhat worsened after the operation. The post-operative hearing tests of this patient showed a slightly better hearing and no signs of cochlear damage as measured by bone conduction or speech discrimination scores. At the latest follow-up 38 ears had an intact drum, while one ear had a small, dry reperforation anteriorly. This gives a closure rate of 97%. The hearing tests of the 39 ears showed a rise in median speech reception threshold (SRT) from 40 d B HL preoperatively to 20 dB HL post-operatively (P< 0.05). The changes in S R T and air-bone gap for type I and 11 tympanoplasty respectively are listed in Table 2. On pure-tone audiometry a significant rise in air conduction was found for most frequencies (Figure 2). N o significant conduction changes were found in bone, as shown in Figures 3 and 4, which also depict the changes in air conduction for types I and I1 respectively. Normal tympanomeatal angles were found in all ears by otomicroscopic examination. Usually the umbo was less well defined than normal, and in 12 ears the tympanic membrane was judged not to be in contact with the distal half of the malleus handle. The charts of three ears lacked information regarding this problem. To examine whether these small degrees of laterofixation had any influence on post-operative conductive hearing, 24 ears with an intact ossicular chain as
J.Stage and K . Bak- Pedersen
8
O r
T
T
T
T
250
500
1000
2000
20
m
60
80
4000
8000
Frequency ( H z )
Figure 2. 95% confidence limits for pre- and post-operative air Preoperative median values; 0 , postconduction in 39 ears. 0, operative median values.
well as information concerning the relation between tympanic membrane and malleus handle were analysed. In 15 ears with apparently normal malleus-TM relations the median post-operative air-hone gap was 11.7 d B H L compared to 8.1 dB H L in 9 ears having latero-fixation (mean values for 500. 1000 and 2000 Hz).
Discussion Large and anteriorly situated pars tensa perforations have been reported by various authors to be relatively difficult to close successfully, reported closure rates varying between 56 and 91 Our smaller series containing a high proportion of these perforations seems to compare favourably with these results. The described operation is a variant of the 'swinging door' technique described by Palva et al.' Palva and later
O r
6oI 801
I
250
I
500
I
1000
I
2000
I
4000
1
8000
Frequency (Hz)
Figure 3. Median pre- and post-operative air and honeconduction thresholds in 26 ears with tympanoplasty type I. Air conduction: 0, preoperative; 0 , post-operative. Bone conduction: 0, preoperative; m, post-operative.
801
I
I
250
500
I
I
1000 2000 Frequency (Hi!)
I
A
4000
8000
Figure 4. Median pre- and post-operative air and boneconduction thresholds in 12 ears with tympanoplasty type 11. Air conduction: 0, preoperative; 0 , post-operative. Bone conduction: 0 , preoperative; m, post-operative.
Schwaber' both placed the graft lateral to the malleus handle in the repair of large perforations, but neither has published separate results concerning this group of patients. From a theoretical point of view there are a couple of problems inherent in this technique. It requires more dissection of the ossicular chain, which where there is an intact chain might damage the inner ear. In our series we have found no signs of cochlear damage in these ears. In the most sensitive high-tone frequencies is seen first a slight rise in median bone conduction thresholds at 4000 Hz and secondly a lOdB rise in median air conduction at 8000Hz. These results are in accordance with other studies of this problem."' We therefore consider this dissection of the ossicular chain permissible without risk to the cochlea. Another problem is the tendency to slight graft lateralization from the malleus handle, seen in some of the ears. As our technique is essentially an underlay technique and all ears have had normal tympanomeatal angles post-operatively, this tendency to lateralization is of a very limited nature. In accordance with this, our analysis of the postoperative air-bone gap in these ears has not shown any measurable effect on the hearing, and we d o not consider it a serious problem. The advantage of the technique is better access to the upper anterior part of the middle ear, which can be packed with Gelfoam without having to push it through thc relatively narrow space between promontory and malleus handle. Thus a much safer alignment between graft and drum remnant is allowed in this critical area. Also. the anterior part of the middle ear and the umbo region can be clcared of disease with a high degree of certainty. We would like to advocate the described techniquc as a good alternative to the conventional underlay technique in cases with perforation or other significant disease involving the area anterior to the malleus handle.
Underlay tympanoplasty with the graft lateral to malleus handle
References I PALVAT., PALVAA. & KARJA J. (1969) Myringoplasty. Ann. Otol. Rhinol. Laryngol. 78, 1074-1080 2 BOOTH J.B. (1974) Myringoplasty. The lessons of failure. J . Laryngol. Otol. 88, 1223-1236 3 SMYTHG.D.L. (1976) Tympanic reconstruction. Fifteen year report on tympanoplasty. 11. J . Laryngol. O l d . 90, 713-741 4 SADE J., BERCOE., BROWS M., WEISLIERG J . & AVRAHAMS. (1981) Myringoplasty. Short and long-term results in a training program. J . Laryngol. Otol. 95, 653-665
9
5 SCHWABER M . K . (1986) Postauricular undersurface tympanic membrane grafting: some modifications of the ‘swinging door’ technique. Otolaryngol. Head Neck Surg. 95, 182-187 6 KYLEN P., ARLINGER S.. JERVALL L. & HARDERH . (1980) Ossicular manipulation in chronic ear surgery. Arch. Ofolaryngol. 106, 598-601 L. & KUYLENSTIERNA R . (1985) A clinical comparison of 7 MENDEL. the results of two different methods of closing tympanic membrane perforations. J . Laryngol. O l d . 99, 339-342
1992, 17, 6-9
Underlay tympanoplasty with the graft lateral to the malleus handle JAN STAGE & KRISTIAN BAK-PEDERSEN Department qf Otorhinoluryngoli~g.v:v, Rispebjerg Hospitul. Copenhugen. Denmark
Accepted for publication 28 November 1990 STAGE J . & B A K - P E D E R S E N K .
(1992) Clin.Otolaryngol. 17, 6-9
Underlay tympanoplasty with the graft lateral to the malleus handle In conventional underlay tympanoplasty the graft is placed medial to the malleus handle. The present study evaluates the results of a modification of this technique, in which the graft is placed lateral to the malleus handle, which has been dissected from the drum remnants. 39 ears with predominantly large or anterior pars tensa perforations were operated upon in this manner. After a median observation time of 20 months one ear was found to have a small reperforation. All ears had normal tympanomeatal angles, but 12 ears showed a small degree of laterofixation of the graft from the malleus handle. Analysis of the hearing showed good hearing improvement and no adverse effects attributable to the dissection of the ossicular chain. It is concluded that the technique is a good alternative to conventional underlay myringoplasty in ears with perforations involving the area anterior to the malleus handle. Keywords
underlay tympanoplasty
underlay myringoplusty
A very popular method of closing perforations of the tympanic membrane is the underlay technique using temporalis fascia as graft material. Normally, using this method, the graft is placed medial to the malleus handle. In certain pathological conditions of the drum, we have found it advantageous to place the graft lateral to the malleus handle. The purpose of this study was to describe the results of this variant of the underlay technique in a series of patients.
Materials and methods The material comprises 39 ears (38 patients) operated on from June 1984 to January 1989. 10 of the operations were revision operations. The median age of the patients was 46 years (range 7-78 years). The locations of the perforations are shown in Table I . In most cases, the indication for placing the graft lateral to the malleus handle was a perforation extending into the area anterior to the malleus handle (26 ears). Severe atrophy o r tympanosclerosis of the same area was the reason in another 9 ears with initially only posterior or inferior perforations. In Correspondence: Dr K . Bak-Pedersen, Department of Otorhinolaryngology, Bispebjerg Hospital, DK-2400 Copenhagen NV. Denmark.
6
malleus handle
the last 4 cases the drum remnant was dissected from the malleus handle for various reasons: pars flaccida perforation, severe pars Raccida retraction, cholesteatoma of the middle ear, o r in order to preserve a well functioning, earlier ossiculoplasty. The type of operation performed in 26 ears was a myringoplasty or tympanoplasty type I. Type TI tycnpanoplasty with incus interposition was performed in 12 ears, and type 111 with a ceramic prosthesis in one ear. All operations were performed under general anaesthesia. The access in all cases was primarily endaural, but in 21 ears a supplementary retroauricular incision was necessary. Having incised the canal skin posteriorly the skin and drum
Table 1. Location of perforations in 39 ears
No.
010
.
Total or subtotal tensa defects Anterior* Posterior or inferiort
19
49
7
13
18 33
Iota1
39
I00
*Involving the upper anterior quadrant. ?One ear having also a flaccid perforation.
Underlay tympanoplasty with the graft lateral to malleus handle
1
Table 2. Median pre- and post-operative speech reception threshold (SRT) and air-bone gap in tympanoplasty type I and I1
respectively Air-bone gap* (dB HL)
SRT (dB HL) Pre .~
~
Type I ( n = 26) Type I1 ( n = 12) ~
~
Post
Pre
Post
32.5 24.2
12.5 12.5
~~
40 42.5
20 22.5
~
*Mean values for 500, 1000 and 2000 Hz.
point any nccessary ossiculoplasty was performed. Finally, the graft and flaps were placed in situ (Figure I b) and the ear canal filled with Gelfoam innermost and upon this antibiotic-soaked gauze. All patients were operated upon by the authors and followed post-operatively at regular intervals. The results given reflect the status at the latest follow-up.
Results
Figure 1. A, Superior and inferior flaps raised, allowing full inspection of the middle ear. b, Temporalis fascia (F) placed against lateral wall of Eustachian tube (ET), lateral to the malleus handle (MH) and medial to the drum remnants (DR).
remnant were raised. Superior and inferior tympanomeatal flaps were created according to the ‘swinging door’ technique described by Palva et al.’ The inferior flap was elevated to a point near the anterior margin of the perforation and superiorly the drum remnant was carefully dissected from the malleus handle and short process. If necessary the whole pars flaccida area was dissected as well and incorporated in the superior flap (Figure la). In 10 cases the operator suspected epithelial ingrowth under the umbo region, and in these cases the tip of the malleus handle was resected with a malleus nipper. Having dealt with any middle ear abnormality temporalis fascia was then placed as an underlay graft medial to the drum remnant but lateral to the malleus handle. Anteriorly, the graft was pushed forward to achieve a good alignment against the lateral wall of the Eustachian orifice. Subsequently graft and flaps were displaced latcrally while the middle ear was filled with Gelfoam. Also, a t this
The median post-operative observation time was 20 months (range 1.5-58 months). Very few post-operative complications were seen. One car developed a staphylococcal infection and had a temporary reperforation, which closed after conservative treatment. N o cases of vertigo or deafness occurred. One patient felt that a prior existing tinnitus was somewhat worsened after the operation. The post-operative hearing tests of this patient showed a slightly better hearing and no signs of cochlear damage as measured by bone conduction or speech discrimination scores. At the latest follow-up 38 ears had an intact drum, while one ear had a small, dry reperforation anteriorly. This gives a closure rate of 97%. The hearing tests of the 39 ears showed a rise in median speech reception threshold (SRT) from 40 d B HL preoperatively to 20 dB HL post-operatively (P< 0.05). The changes in S R T and air-bone gap for type I and 11 tympanoplasty respectively are listed in Table 2. On pure-tone audiometry a significant rise in air conduction was found for most frequencies (Figure 2). N o significant conduction changes were found in bone, as shown in Figures 3 and 4, which also depict the changes in air conduction for types I and I1 respectively. Normal tympanomeatal angles were found in all ears by otomicroscopic examination. Usually the umbo was less well defined than normal, and in 12 ears the tympanic membrane was judged not to be in contact with the distal half of the malleus handle. The charts of three ears lacked information regarding this problem. To examine whether these small degrees of laterofixation had any influence on post-operative conductive hearing, 24 ears with an intact ossicular chain as
J.Stage and K . Bak- Pedersen
8
O r
T
T
T
T
250
500
1000
2000
20
m
60
80
4000
8000
Frequency ( H z )
Figure 2. 95% confidence limits for pre- and post-operative air Preoperative median values; 0 , postconduction in 39 ears. 0, operative median values.
well as information concerning the relation between tympanic membrane and malleus handle were analysed. In 15 ears with apparently normal malleus-TM relations the median post-operative air-hone gap was 11.7 d B H L compared to 8.1 dB H L in 9 ears having latero-fixation (mean values for 500. 1000 and 2000 Hz).
Discussion Large and anteriorly situated pars tensa perforations have been reported by various authors to be relatively difficult to close successfully, reported closure rates varying between 56 and 91 Our smaller series containing a high proportion of these perforations seems to compare favourably with these results. The described operation is a variant of the 'swinging door' technique described by Palva et al.' Palva and later
O r
6oI 801
I
250
I
500
I
1000
I
2000
I
4000
1
8000
Frequency (Hz)
Figure 3. Median pre- and post-operative air and honeconduction thresholds in 26 ears with tympanoplasty type I. Air conduction: 0, preoperative; 0 , post-operative. Bone conduction: 0, preoperative; m, post-operative.
801
I
I
250
500
I
I
1000 2000 Frequency (Hi!)
I
A
4000
8000
Figure 4. Median pre- and post-operative air and boneconduction thresholds in 12 ears with tympanoplasty type 11. Air conduction: 0, preoperative; 0 , post-operative. Bone conduction: 0 , preoperative; m, post-operative.
Schwaber' both placed the graft lateral to the malleus handle in the repair of large perforations, but neither has published separate results concerning this group of patients. From a theoretical point of view there are a couple of problems inherent in this technique. It requires more dissection of the ossicular chain, which where there is an intact chain might damage the inner ear. In our series we have found no signs of cochlear damage in these ears. In the most sensitive high-tone frequencies is seen first a slight rise in median bone conduction thresholds at 4000 Hz and secondly a lOdB rise in median air conduction at 8000Hz. These results are in accordance with other studies of this problem."' We therefore consider this dissection of the ossicular chain permissible without risk to the cochlea. Another problem is the tendency to slight graft lateralization from the malleus handle, seen in some of the ears. As our technique is essentially an underlay technique and all ears have had normal tympanomeatal angles post-operatively, this tendency to lateralization is of a very limited nature. In accordance with this, our analysis of the postoperative air-bone gap in these ears has not shown any measurable effect on the hearing, and we d o not consider it a serious problem. The advantage of the technique is better access to the upper anterior part of the middle ear, which can be packed with Gelfoam without having to push it through thc relatively narrow space between promontory and malleus handle. Thus a much safer alignment between graft and drum remnant is allowed in this critical area. Also. the anterior part of the middle ear and the umbo region can be clcared of disease with a high degree of certainty. We would like to advocate the described techniquc as a good alternative to the conventional underlay technique in cases with perforation or other significant disease involving the area anterior to the malleus handle.
Underlay tympanoplasty with the graft lateral to malleus handle
References I PALVAT., PALVAA. & KARJA J. (1969) Myringoplasty. Ann. Otol. Rhinol. Laryngol. 78, 1074-1080 2 BOOTH J.B. (1974) Myringoplasty. The lessons of failure. J . Laryngol. Otol. 88, 1223-1236 3 SMYTHG.D.L. (1976) Tympanic reconstruction. Fifteen year report on tympanoplasty. 11. J . Laryngol. O l d . 90, 713-741 4 SADE J., BERCOE., BROWS M., WEISLIERG J . & AVRAHAMS. (1981) Myringoplasty. Short and long-term results in a training program. J . Laryngol. Otol. 95, 653-665
9
5 SCHWABER M . K . (1986) Postauricular undersurface tympanic membrane grafting: some modifications of the ‘swinging door’ technique. Otolaryngol. Head Neck Surg. 95, 182-187 6 KYLEN P., ARLINGER S.. JERVALL L. & HARDERH . (1980) Ossicular manipulation in chronic ear surgery. Arch. Ofolaryngol. 106, 598-601 L. & KUYLENSTIERNA R . (1985) A clinical comparison of 7 MENDEL. the results of two different methods of closing tympanic membrane perforations. J . Laryngol. O l d . 99, 339-342
