Jesic SD, Dimitrijevic MV, Nesic VS, Jotic AD, Slijepcevic NA. Temporalis Fascia Graft Perforation and Retraction After Tympanoplasty for Chronic Tubotympanic Otitis and Attic Retraction PocketsFactors Associated With Recurrence. Arch Otolaryngol Head Neck Surg. 2011;137(2):139–143. doi:10.1001/archoto.2010.242
To correlate the recurrence of temporalis fascia graft perforation and retraction in adults and children after tympanoplasty for chronic tubotympanic otitis and deep attic retraction pockets with age, pathologic process, mucosal lesions, mucociliary transport time, chronic sinusitis, and lateral attic wall reconstruction.
Tertiary referral center.
Two hundred seventy-four adult ears and 41 child ears were operated on for chronic tubotympanic otitis, 50 adult ears were operated on for traumatic tympanic membrane rupture, and 56 adult ears were operated on for attic retraction pockets.
Underlay tympanoplasty with or without mastoidectomy and lateral attic wall reconstruction for attic retraction pockets. Mucociliary transport time was investigated using saccharin testing.
Main Outcome Measures
Recurrent perforation, recurrent attic retraction, and temporalis fascia graft retraction were evaluated.
Higher incidences of recurrent perforation were found in adults operated on for tubotympanic otitis vs traumatic tympanic membrane rupture (P = .02) and in children (P = .004), especially those 9 years and younger (P = .02). A risk factor in adults was chronic sinusitis (risk ratio, 35.0; 95% confidence interval, 32.1-38.2; P = .004). Temporalis fascia graft retraction correlated with slower mucociliary transport time in adults (P < .001). A lower incidence of recurrent attic retraction was associated with lateral attic wall reconstruction (P < .001).
Recurrent temporalis fascia graft perforation after tympanoplasty for tubotympanic otitis may be related to age and coexisting chronic sinusitis. Temporalis fascia graft retraction correlates with slower mucociliary transport time in the eustachian tube. Lateral attic wall reconstruction minimizes recurrent attic retraction in adults.
Modern tympanoplasty is successful in radical eradication of pathologic processes owing to microscopic and endoscopic control, preoperative computed tomography, and nuclear magnetic resonance imaging. However, inadequate aeration is unresolved in tympanoplasty. Negative tympanic cavity pressure can lead to ossicular osteolysis, extrusion of the prosthesis, or recurrent perforation and retraction of the graft. Attention has been directed to the choice of grafting material, which should be mechanically stable and have necessary acoustic performance for good sound transmission. Autologous materials such as temporalis fascia, cartilage, and perichondrium provide these characteristics, and each has specific advantages and disadvantages. Studies1- 3 have shown that cartilage and temporalis fascia are associated with minimal tissue reaction and absorption; temporalis fascia shows particularly poorer dimensional stability and significant shrinkage after rehydrating. In contrast, some investigations have demonstrated loss of acoustic energy using thick-shield cartilage grafting, while others have shown the opposite.4 The choice of temporalis fascia or cartilage should not be one of surgeon preference but should be derived based on knowledge of different prognostic aspects such as demographic factors, disease-associated issues, and technical reasons (perioperative bleeding, the use of a graft that is too small, etc).5,6
The objectives of this study were to evaluate the recurrence of temporalis fascia graft perforation and retraction in adults and children after tympanoplasty for chronic tubotympanic otitis with mild to large central perforations and of deep attic retraction pockets and tympanic membrane rupture in adults. These findings were correlated with age, pathologic process, mucosal lesions, mucociliary (MC) transport time in the eustachian tube, coexistence of chronic sinusitis, follow-up duration, status of the contralateral ear in children, and surgical procedure performed, particularly lateral attic wall reconstruction (LAR).
A retrospective study was performed of 863 middle ears of patients who underwent temporalis fascia graft underlay tympanoplasty in the Clinic of Otorhinolaryngology and Maxillofacial Surgery, Clinical Center of Serbia, Belgrade, from October 1, 2000, to October 31, 2006. The study was performed in accord with the ethical standards in the 1964 Declaration of Helsinki, and all patients or parents of children signed an informed consent form before study inclusion. Adults in the study also underwent saccharin testing for estimation of MC transport time and were fully informed about the procedure and gave their consent.
Excluded from the study were 326 ears that were operated on for cholesteatoma and 58 ears that were operated on for chronic otitis with total perforation. Two hundred seventy-four adult ears and 41 child ears were followed up after tympanoplasty for chronic tubotympanic otitis with mild to large central perforations, 50 adult ears for traumatic tympanic membrane rupture, and 56 adult ears for attic retraction pockets of grades 3 and 4 according to the classification by Tos.7 Lost to follow-up were 58 ears (33 adult and 8 child ears operated on for tubotympanic otitis, 6 adult ears for traumatic tympanic membrane rupture, and 11 adult ears for attic retraction pockets). The mean age of adults was 41 years. Among children, 28 were 9 years or younger and 13 were older than 9 years.
In patients with chronic tubotympanic otitis, the disease was unilateral, with mild to large central perforations occupying 50% to 75% of the pars tensa. Attic retraction pockets were classified according to the system by Tos, which comprises 5 depth grades of attic retraction pockets. Grade 1 represents slight retraction toward the neck of the malleus with air present between the Shrapnell membrane and the neck of the malleus. Grade 2 indicates retraction that is extended to the neck of the malleus without air behind the Shrapnell membrane. In grade 3, retraction is beyond the bony annulus, and the base of retraction is visible. Grade 4 consists of bone destruction of the scutum with the membrane being adherent to the head of the malleus and the body of the incus and with the base of retraction being still visible under the microscope. Grade 5 is assigned if the bottom of the pocket is not visible under the microscope but is clearly viewed with an endoscope.
Eustachian tube function was investigated under the operative microscope with Valsalva and Toynbee maneuvers. Mucociliary transport time was investigated before surgery by saccharin testing in 30 adults with chronic tubotympanic otitis, in 16 adults with attic retraction pockets, and in 9 adults with traumatic tympanic membrane rupture.
Saccharin testing was performed in the following manner. During preoperative testing, sterile saccharine sodium salt solution, 5% (10 μL of 1,1-dioxo-1,2-benzothiazol-3-1) was instilled through the existing eardrum perforation. The interval in seconds was measured from the application of the solution until the perception of sweet taste in the patient's mouth and pharynx.8,9
For patients who had attic retraction pockets, we performed preoperative otomicroscopy, tonal liminar audiometry, and tympanometry. We also performed preoperative computed tomography and intraoperative exploration of the middle ear mucosa and auditory ossicles.
Fifty ears with traumatic tympanic membrane rupture were operated on using myringoplasty technique with underlay temporalis fascia graft. Ears with chronic tubotympanic otitis (274 adults and 41 children) were operated on using intact canal wall tympanoplasty and mastoidectomy with underlay temporalis fascia graft. Fifty-six adult ears with deep attic retraction pockets were treated using intact canal wall tympanoplasty with underlay temporalis fascia graft. Among 56 ears, we used only temporalis fascia in 13, ventilation tube insertion in the pars tensa with temporalis fascia for attic retraction in 10, and demineralized bone matrix (Osteovit; B. Braun Melsungen GmbH, Melsungen, Germany) or tragal cartilage for bone defect reconstruction together with underlay temporalis fascia in 33 ears that underwent LAR.
A retroauricular approach was used in all cases. For fixation of the graft, we used absorbable gelatin (Gelfoam [Pfizer Inc, New York, New York] or Spongostan [Johnson and Johnson Ltd, New Brunswick, New Jersey]). A sterile gauze tape was kept in the external acoustic canal for 14 days after surgery.
Every 3 months (mean follow-up period, 36 months), we performed otomicroscopic inspection of the tympanic graft and noted recurrence of perforation or retraction. Incidence of tympanic graft postoperative change was correlated with age, pathologic process, mucosal lesions, MC transport time, coexistence of chronic sinusitis, follow-up duration, and malleus manubrium handle presence. Descriptive and statistical analyses were performed using commercially available software (SPSS version 8.0; SPSS Inc, Chicago, Illinois). The Pearson χ2 test, Fisher exact test, and logistic regression analysis were used.
Adult patients operated on for chronic tubotympanic otitis had recurrent perforation more often vs adults operated on for traumatic tympanic membrane rupture or those with normal middle ear structures (13.9% [38 of 274] vs 2.0% [1 of 50]) (P = .02). The incidence of recurrent perforation was higher among children than among adults (P = .004) (Table 1). Graft retraction toward the middle tympanic wall was more frequent in children (22.0% [9 of 41]) than in adults (7.7% [21 of 274]) (P = .001).
Most graft perforations in adults were detected 3 to 6 months after surgery (P = .02). The incidences of early perforation were 15.8% (6 of 38) immediately after surgery, 31.6% (12 of 38) after 3 to 6 months, and 52.6% (20 of 38) at 12 months (P = .02) (Table 2). Most patients (29 of 38) with chronic tubotympanic otitis and recurrent perforation had chronic sinusitis (P = .005). Primarily among patients with early perforation, possible causes for graft perforation were technical reasons such as perioperative bleeding or the use of a graft that is too small.
Mucosal lesions were not severe; edema occurred in 55.1% (20 patients) with recurrent perforation. In 44.9% (18 patients), we observed polypoid lesions and granulations. We found no correlation between type of mucosal lesion in chronic tubotympanic otitis and recurrent perforation (P = .52). In patients with edematous mucosa, the mean MC transport time was 15 minutes; in patients with more intensive lesions, it was 24 minutes. We detected no recurrent perforation in these groups but found that graft retraction toward the promontory was more frequent in the group with longer MC transport time (P < .001, Fisher exact test) (Table 3).
Log regression analysis demonstrated an association of chronic sinusitis with recurrent perforation in adults. The risk for recurrent perforation was 35 times higher (risk ratio, 35.0; 95% confidence interval, 32.1-38.2; P = .004) in the group of patients with chronic sinusitis who underwent tympanoplasty for chronic tubotympanic otitis than in the other groups (Table 4).
Of 41 child ears that were operated on for chronic tubotympanic otitis, 28 children were 9 years or younger, while 13 children were older than 9 years. In the group of children who were 9 years or younger, the incidence of recurrent perforation was 46.4% (13 of 28), whereas it was 15.4% (2 of 13) in the group of children who were older than 9 years (P < .001, Fisher exact test) (Table 5). In the group of younger children, there was no statistically significant association between the type of tympanic membrane change (perforation, retraction, or other) in the operated-on ear vs the contralateral ear (P = .28). This was not the case in the group of older children, among whom the absence of tympanic membrane change in the contralateral ear was statistically significant (P = .001).
In adult patients with attic retraction pockets, the mean MC transport time was 44 minutes, which was significantly longer than that in patients with chronic tubotympanic otitis, indicating poor eustachian tube function. Three types of procedures were performed in these patients. We used only temporalis fascia graft with or without a ventilation tube at the beginning of the study, whereas we introduced LAR with bovine demineralized bone matrix (Osteovit) or with tragal cartilage in the last 2 years of the study. The recurrent attic retraction rate was lower in patients who underwent LAR (P < .001) (Table 6). In 42.4% (14 of 33) of ears that underwent LAR, pars tensa retraction toward the medial wall was detected. Although the results varied slightly for the use of a graft alone without LAR, there was no statistically significant difference in the incidence of pars tensa retraction associated with the type of surgery (LAR vs graft only) (P = .86).
Recurrent perforation and retraction are the most common outcome measures of temporalis fascia graft stability after tympanoplasty. The literature contains comprehensive discussions of prognostic factors such as age, eustachian tube function, state of the middle ear mucosa, perforation size, integrity of the ossicular chain, the presence of a cholesteatoma, status of the contralateral ear, type of surgery, and choice of grafting material used.
In our study, the overall success rate for closure of mild tympanic membrane perforation was 86.1% after wall-up tympanoplasty using temporalis fascia graft in adults with chronic tubotympanic otitis. Fishman et al10 reported overall success rates of 84% using dermis allograft and 97% using temporalis fascia graft, both for closure of large perforations. Kazikdas et al11 reported a 75% success rate with temporalis fascia graft for cases with large perforations. Perforation size may be predictive of outcome, with report of a 56% success rate among large perforations.12 The success rate can deteriorate with time. Halik and Smyth13 noted a decline to 81% after 11 years. Success rates may vary because of shorter follow-up periods or because of failure to address the appearance of attic retraction pockets.
In our study, most graft perforations appeared 3 to 12 months after surgery. Kartush et al14 noted that the critical period for graft perforation is within the first 6 months. As our study shows, the most frequent causes of immediate graft perforation are technical in nature, such as perioperative bleeding or the use of a graft that is too small. During the later postoperative period, chronic sinusitis was the most significant factor associated with graft perforation. It is well known that this and similar conditions have an important effect on middle ear ventilation. Endoscopy has shown that the nasopharyngeal orificium in the eustachian tube is blocked by secretion in 73% of children and by hypertrophy of the peritubal lymphoid tissue in 17%; mucosal edema of the orificium has been detected in 27% of adults.15 Some investigations showed that preoperative endoscopic treatment of chronic sinusitis resulted in better aeration after tympanoplasty.16
Ossicular chain integrity (especially of the malleus manubrium), mucosal lesions, and aeration of the middle ear improve the stability of the tympanic graft.17 In this study, the presence of mucosal lesions did not predict graft failure in chronic tubotympanic otitis; however, MC transport time in the eustachian tube could be of importance. The presence of normal healthy mucosa in patients with traumatic tympanic membrane rupture could be responsible for their high graft success rate of 98.0% in our study. Some authors point out the prognostic value of the presence of mucosal lesions but not the type of ear suppuration before surgery.18 Angeli et al5 demonstrated that their tympanic graft success rate was not significantly affected by the presence of mucosal lesions.
Our success rates were significantly higher in patients with traumatic tympanic membrane rupture who had healthy mucosa (98.0%) and normal MC transport time (86.1%) than in patients with chronic tubotympanic otitis who had slower MC transport time. In adult patients with chronic tubotympanic otitis, MC transport time was unassociated with perforation recurrence rate. In contrast, slower MC transport time was a significant factor associated with temporalis fascia graft retraction after tympanoplasty in adults. Retraction of the graft cannot be explained solely by simple tubal obstruction and causative dysfunction. Tubal dysfunction is a causative factor of graft perforation in 90% of cases, but graft perforation occurred in 16% of cases with a normal tubal opening.19 Among several eustachian tube functions, MC transport time is the most mysterious. In previous investigations, the normal MC transport time associated with healthy middle ear mucosa is 7 minutes, whereas it is much longer with inflamed mucosa.8 As the present study shows, longer MC transport time has an effect on tympanic membrane retraction. We detected a 42.4% incidence of pars tensa retraction among patients operated on for an attic retraction pocket. Sade20 reported a 40% incidence of pars tensa retraction associated with an attic retraction pocket. This is much higher than the incidence herein of graft retraction after tympanoplasty for chronic tubotympanic otitis (7.7% [21 of 274]) vs for attic retraction pockets (42.9% [24 of 56]) (P = .001). In the present study, the LAR technique decreased the incidence of recurrent graft retraction of the attic but not pars tensa retraction of the tympanic membrane. Among patients in whom LAR was not performed, the incidence of recurrent graft retraction of the attic was much higher, exceeding 80% during the follow-up period. Other authors demonstrated the benefit of LAR in preventing attic retraction pocket recurrence.21 An attic retraction pocket operation without reconstruction of a much deteriorated lateral attic wall gives poor anatomical results. Ventilation tube insertion is controversial. It is more effective in treating pars tensa retraction than attic retraction pockets.22 Some authors propose obliteration of the attic instead of its reconstruction.23 According to these authors, aeration of the mesotympanum was established in 75% of cases in which obliteration of the attic was performed.
The success rate of tympanic graft closure herein was significantly lower in children. Recurrent perforation occurred in 31.7% of children with chronic tubotympanic otitis vs in 13.9% of adults, which is in accordance with data from the literature.24 Possible reasons for more recurrent perforation among children could be the presence of upper respiratory tract infections, otitis media with effusion, or adenoid infection. Age at the time of surgery, status of the contralateral ear, and middle ear gas volume are the most important prognostic factors for tympanic membrane graft closure. Middle ear gas volume of less than 3 mL is a poor prognostic factor for tympanic membrane graft closure in children.20 According to other authors, there is no adequate method for preoperative assessment of eustachian tube function.25
In our study, the incidence of perforation recurrence was statistically significantly higher among children 9 years or younger. All children in this age group demonstrated middle ear pathologic findings in the contralateral ear, suggesting tubal dysfunction, in contrast to findings among children older than 9 years. The difference was statistically significant. However, in a study by Angeli et al,5 age was not predictive of perforation closure by fascia grafting.
Temporalis fascia graft is an adequate material for closure of tympanic membrane perforation among patients with chronic tubotympanic otitis, especially among adults with fewer mucosal lesions and less MC transport damage. According to our study results, closure success rates are affected primarily by the coexistence of chronic sinusitis, suggesting that it should be treated before attempting tympanoplasty. Occurrence of graft retraction correlates with longer MC transport time and with tubal dysfunction. Temporalis fascia graft closure success rates are significantly lower in children and are affected by poor ventilation associated with middle ear pathologic findings in the contralateral ear. Finally, the choice of surgical technique affects attic retraction recurrence in adults with deep attic retraction pockets and poor eustachian tube function. Lateral attic wall reconstruction reduces the incidence of attic retraction recurrence, although it does not prevent pars tensa retraction.
Correspondence: Snezana D. Jesic, MD, PhD, Clinic of Otorhinolaryngology and Maxillofacial Surgery, Clinical Center of Serbia, School of Medicine, University of Belgrade, Pasterova 2, 11000 Belgrade, Serbia (firstname.lastname@example.org).
Submitted for Publication: May 31, 2010; final revision received August 22, 2010; accepted September 7, 2010.
Author Contributions: All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Jesic and Nesic. Acquisition of data: Jesic, Jotic, and Slijepcevic. Analysis and interpretation of data: Jesic, Dimitrijevic, and Nesic. Drafting of the manuscript: Nesic, Jotic, and Slijepcevic. Critical revision of the manuscript for important intellectual content: Jesic and Dimitrijevic. Statistical analysis: Nesic. Administrative, technical, and material support: Dimitrijevic, Jotic, and Slijepcevic. Study supervision: Jesic.
Financial Disclosure: None reported.
Previous Presentation: This study was presented in part as a poster at the First Meeting of the European Academy of Otorhinolaryngology–Head and Neck Surgery; June 28, 2009; Mannheim, Germany.