Introduction
Chronic suppurative otitis media (CSOM) is a major cause of long-standing hearing impairment in Sub-Saharan African countries.Reference Guntinal-Lichius, Wittenkindt, Baier and Manni 1 The main worries for CSOM patients are the associated conductive hearing loss and suffering from chronically draining ear, both of which carry significant social stigma in most societies. These concerns are greater for individuals with bilateral ear disease.
The exact prevalence of bilateral CSOM in most populations is unknown. In 2012, Kabir et al. reported that 23.64 per cent of CSOM patients in Dhaka had bilateral disease.Reference Kabir, Joarder, Ekramuddaula, Uddin, Islam and Habib 2 It was previously reported that one-third of Nigerians with significant hearing loss due to CSOM had bilateral disease.Reference Olusesi 3 In 2008, Gianoli and Soileau reported that 13 out of 25 CSOM patients awaiting tympanomastoid surgery had bilateral disease.Reference Gianoli and Soileau 4
The traditional approach of staging bilateral ear surgery is premised on the theoretical risk of iatrogenic sensorineural hearing loss (SNHL).Reference Mane, Patio, Mohite and Varute 5 For bilateral CSOM patients, tympanic membranes have traditionally been grafted sequential at two different sittings.Reference Raghuwanshi and Asati 6 However, for bilateral CSOM patients undergoing tympanoplasty, it is significantly more expensive to undergo surgery on two ears at different sittings. Therefore, in a resource-constrained economy with a high prevalence of bilateral CSOM, the best way of providing good quality economical ear care has to be found.
Bilateral same-day type 1 tympanoplasty using different graft materials has been reported.Reference Mane, Patio, Mohite and Varute 5 , Reference Raghuwanshi and Asati 6 A study comparing same-day bilateral middle-ear surgery for otitis media with bilateral sequential (i.e. staged) middle-ear surgery concluded that the former provides good hearing outcomes with lower costs and shorter surgery times, with a low incidence of complications.Reference Kim, Lee, Park, Kwon, Kim and Lee 7 However, no previous study has compared outcomes between bilateral same-day and sequential (different-day) cartilage tympanoplasty.
Materials and methods
A prospective, randomised observational study was conducted at National Hospital Abuja, Nigeria, between May 2005 and April 2014 in accordance with the declaration of Helsinki of 1975, as revised in 2008. All participants received counselling and gave informed consent to participate in this study.
Patients from the hearing loss clinic were enrolled subject to the following inclusion criteria: (1) a diagnosis of bilateral tubotympanic CSOM; (2) absence of cholesteatoma on otomicroscopy and/or computed tomography of the ear; (3) an air–bone gap (ABG) of less than 40 dB; and (4) otorrhoea from either ear last reported more than 12 weeks previously.
A total of 43 consecutive patients met the inclusion criteria, of whom 38 consented to participate in the study and were allocated to undergo either bilateral sequential same-day tympanoplasty or bilateral sequential different-day tympanoplasty (staged 3 months apart). The choice of surgical approach was determined by the perforation size and location, as well as the presence or absence of an overhanging bony canal wall. The same-day tympanoplasty group were informed of the possibility of reduced hearing in the first 10 days following surgery due to aural packing. Perforations were graded into small (affecting less than 30 per cent of the pars tensa), medium-sized (affecting 30–60 per cent) and large (affecting more than 60 per cent).
All patients underwent cartilage tympanoplasty using the underlay grafting technique under general endotracheal anaesthesia. Grafts were harvested from the conchal cartilage for a post-auricular approach or the tragal cartilage for an endaural or permeatal approach. Aural packs were removed and the first graft inspection took place on post-operative day 10. A second graft inspection and post-operative pure tone audiometry were performed at six weeks and patients were followed up for at least six months.
Post-operative hearing improvement was defined as either complete ABG closure or an ABG improvement of at least 10 dB at speech frequencies (0.5, 1, 2 and 4 kHz), assessed by comparing pre- and post-operative pure tone audiograms.
Parameters and outcomes were compared between both patient groups. Parameters included CSOM duration, intra-operative findings, overall duration of surgery, post-operative graft appearance at six weeks, post-operative complications and re-do rate. The overall cost of surgery for each patient was also calculated at six months post-operatively as the sum of the direct costs of pre-operative visits, surgery and post-operative visits, including fees payable for hospital admission, surgery, pre-operative investigations, post-operative follow up and laboratory investigations.
Results
A total of 38 participants underwent either bilateral sequential same-day tympanoplasty (18 patients, 36 ears) or bilateral sequential different-day tympanoplasty (20 patients, 40 ears). Two patients in the same-day tympanoplasty group and three in the different-day tympanoplasty group failed to turn up for surgery after receiving counselling and being scheduled for surgery.
The mean (± standard deviation (SD)) age was 33.7 ± 12.67 years for patients in the same-day tympanoplasty group and 37.4 ± 12.69 years for those in the different-day tympanoplasty group. The male-to-female ratio was 11:7 in the same-day tympanoplasty group and 5:15 in the different-day tympanoplasty group (that is, 1:3). The mean (± SD) duration of CSOM symptoms was 8.7 ± 5.16 years in the same-day tympanoplasty group and 9.14 ± 6.33 years in the different-day tympanoplasty group.
Most perforations were large, affecting more than 60 per cent of the pars tensa (22 out of 36 ears in the same-day tympanoplasty group, 27 out of 40 ears in the different-day tympanoplasty group; p = 0.36, odds ratio = 0.75). There was a weak positive correlation between symptom duration and perforation size (r = 0.379)
Most grafts were harvested from the conchal cartilage (same-day tympanoplasty group, 28 out of 36 ears; different-day tympanoplasty group, 31 out of 40 ears) and the others were harvested from the tragal cartilage (same-day tympanoplasty group, 8 out of 36 ears; different-day tympanoplasty group, 9 out of 40 ears; p = 0.59, odds ratio = 1.016).
Table I shows the different surgical approaches used to treat all bilateral CSOM patients in this study. A hospital stay of three days was needed for each procedure, that is, a total of six days for different-day tympanoplasty (two admissions) and three days for same-day tympanoplasty. The mean (± SD) duration of follow up for all patients was 2.2 years ± 19.43 months (range, 8 months to 7 years).
Surgical approach by perforation size for both patient groups
Data are n (%). *Pars tensa perforation of less than 30 per cent. †Pars tensa perforation of between 30 and 60 per cent. ‡Pars tensa perforation of more than 60 per cent. BSST = bilateral sequential same-day tympanoplasty; BSDT = bilateral sequential different-day tympanoplasty
The overall graft take rate was 88 per cent (32 out of 36 ears in the same-day tympanoplasty group, 35 out of 40 ears in the different-day tympanoplasty group; p = 0.96, odds ratio = 0.984). Complications for the whole cohort included residual perforation in 7 per cent of ears (2 out of 36 ears in the same-day tympanoplasty group and 3 out of 40 ears in the different-day tympanoplasty group; p = 0.55, odds ratio = 0.72), graft oedema in 1 per cent (1 out of 36 ears in the same-day tympanoplasty group, 0 out of 40 ears in the different-day tympanoplasty group) and graft infection in 5 per cent (1 out of 36 ears in the same-day tympanoplasty group, 3 out of 40 ears in the different-day tympanoplasty group; p = 0.34, odds ratio = 0.35). Intra-operative finding of mucous exudate in the middle ear correlated with post-operative graft infection (r = 0.82). The overall surgery duration was also longer for different-day tympanoplasty (mean ± SD, 4 hours and 50 minutes ± 35 minutes) than for same-day tympanoplasty (mean ± SD, 3 hours and 35 minutes ± 18 minutes).
Post-operative hearing improvement, represented by an improvement in air-conduction pure tone average (PTA), was observed in 32 out of 36 ears in the same-day tympanoplasty group (mean ± SD, 15 ± 3.2 dB) and 35 out of 40 ears in the different-day tympanoplasty group (mean ± SD, 17.5 ± 1.75 dB). There was no significant difference in hearing improvement between groups (p = 0.56, odds ratio = 1.14).
The revision rate was 8 per cent (3 out of 36 ears) in the same-day tympanoplasty group and 10 per cent (4 out of 40) in the different-day tympanoplasty group (p = 0.55, odds ratio = 0.81). No SNHL was observed in either group.
Table II shows intra-operative findings in both patient groups. The odds of having abnormal middle-ear granulation and middle-ear mucoid exudate intra-operatively was higher in the same-day tympanoplasty group than in the different-day tympanoplasty group (odds ratio = 1.72 and 1.62, respectively).
Intra-operative findings for both patient groups
n = 76. BSST = bilateral sequential same-day tympanoplasty; BSDT = bilateral sequential different-day tympanoplasty; OR = odds ratio; CI = confidence interval
The cost of treatment was significantly higher for different-day tympanoplasty than for same-day tympanoplasty. Therefore, an estimated total cost of more than US $600 per patient could be saved by treating patients with bilateral sequential same-day rather than different-day bilateral tympanoplasty. Table III shows the adjusted cost of treatment for each group at the six-month follow up.
Adjusted direct treatment costs for both patient groups
Data are Naira (US dollar). *Second ear surgery performed three months after the first. †The admission cost includes cost of laboratory procedures during the admission period. ‡Note that the last two follow ups for the first ear overlap with pre-operative visits for the second ear. BSST = bilateral sequential same-day tympanoplasty; BSDT = bilateral sequential different-day tympanoplasty
Discussion
Since tympanoplasty was first described, various graft materials have been used.Reference Genc 8 Cartilage is used as a graft material to replace the missing portion of the tympanic membrane or ossicles, reinforce retraction pockets, or reconstruct the eroded canal wall in cholesteatoma surgery. Cartilage tympanoplasty is reported to be effective for both small and large perforations, leading to long-term continuous hearing improvement and ABG closure in patients with large perforations.Reference Wu, Wang, Huang, Lee and Huang 9
Untreated or poorly treated middle-ear disease can still result in serious complications in the post-antibiotic era.Reference Heah, Soon and Yuen 10 Patients with untreated CSOM experience disturbing otorrhoea and hearing loss and are advised not to play water sports; these consequences are greater in those with bilateral ear disease. Bilateral tympanoplasty is the best surgical option for such patients.
Conventional wisdom among otosurgeons was previously to be extremely cautious about performing bilateral ear surgery based on the resultant risk of SNHL. In 1977, Kylén et al. reported a post-operative temporary threshold shift of 5–40 dB at 4 and 8 KHz, attributed to drilling into the temporal bone during ear surgery to treat chronic otitis media.Reference Kylén, Arlinger and Bergholtz 11 These authors postulated that the noise caused by drilling during ear surgery carries a risk of post-operative high tone SNHL. However, technological improvements, including the development of so-called noiseless drills, have necessitated a review of this hypothesis.
Rai et al. recently concluded that bilateral same-day type 1 tympanoplasty is a feasible treatment option in modern otology for most CSOM patients (e.g. excluding high risk surgical patients and those with ossicular dissolution or CSOM-associated with cholesteatoma).Reference Rai, Singh, Sahu, Singh and Arora 12 That study compared unilateral and bilateral type 1 tympanoplasty, utilising inlay temporalis fascia as the graft material. Similar studies have utilised the fascia lata, while others used the temporalis fascia either alone or with the perichondrium.Reference Mane, Patio, Mohite and Varute 5 , Reference Raghuwanshi and Asati 6 The present study is the first to explore the use of cartilage grafts in all patients and to compare the outcomes of bilateral sequential same-day and different-day (three months apart) tympanoplasty.
A significant proportion of participants in both patient groups had had CSOM symptoms for over eight years. There is a worrisome, although not uncommon, trend in most developing countries (where poverty and access to otosurgeons are challenging) for patients to have experienced recurrent otorrhoea with associated hearing loss for at least eight years before seeking specialist care. However, although CSOM duration correlated with perforation size, it did not appear to influence the surgical outcome. Notably, perforation size correlated with use of a post-auricular approach in both groups.
The post-auricular approach was used in most patients. This route provides easy access to the anterior annular rim, which is crucial for preventing anterior blunting and graft medialisation, and influenced the choice of conchal cartilage for grafting in most patients. Of all available graft materials, autologous cartilage graft is believed to provide the best resistance to infection, with a low resorption rate.
In the present study, the graft take rate at six weeks was 88.0 per cent for same-day tympanoplasty and 87.5 per cent for different-day tympanoplasty. These rates are not significantly different from the values published by Rai et al.,Reference Rai, Singh, Sahu, Singh and Arora 12 but are lower than the rate of 96 per cent reported by Mane et al.Reference Mane, Patio, Mohite and Varute 5 The latter study included 14 patients (far fewer than in the present series) and utilised fascia lata and/or temporalis fascia as graft material, both of which might explain the difference in outcome.
In the present study, residual perforation and graft infection were the commonest complications (with rates of 6.5 per cent and 5.2 per cent, respectively). The correlation between intra-operative findings of mucous exudate in the middle ear (although the patient reported no obvious otorrhoea) and post-operative graft infection was not unexpected: most non-cholesteatomatous chronic otitis media patients with previous failed tympanoplasty are reported to have radiographic and intra-operative evidence of middle-ear and/or mastoid disease.Reference Ruble and Pensak 13 Those requiring re-do surgery underwent aerating mastoidectomy and re-do tympanoplasty, with good outcomes.
Most patients with graft take in the present series had functional hearing improvement: 89.0 per cent in the same-day tympanoplasty group and 87.5 per cent in the different-day tympanoplasty group. This finding might be explained by resilience of the cartilage against resorption and its good long-term survival. Unlike fascia, cartilage does not undergo atrophy; hence, graft take is generally associated with functional hearing improvement.
In the present study, revision rates were similar for same-day and different-day tympanoplasty. This might be because a post-auricular approach was used for most patients in both groups, unless complete visualisation of the annulus was possible on otomicroscopy. In a patient series with follow up of 5–7 years, Albera et al. reported that the surgical approach is the principal prognostic factor influencing the anatomical outcome of myringoplasty.Reference Albera, Ferrero, Lucilla and Canale 14
As in previous studies, no patient developed SNHL in either group despite the use of a micro-drill to remove anterior and/or posterior bony overhangs in most ears.Reference Mane, Patio, Mohite and Varute 5 – Reference Kim, Lee, Park, Kwon, Kim and Lee 7 However, Kazikdas et al. recently reported that SNHL can be a temporary outcome of over-under tympanoplasty, which requires significant manipulation of the ossicles.Reference Kazikdas, Onal and Yildrim 15 The present study used underlay cartilage grafts with either the cartilage shield or palisade cartilage technique for all patients, and this outcome was not observed.
-
• This study compared the outcomes of bilateral same-day and staged cartilage tympanoplasty
-
• There was no significant difference in anatomical and functional outcomes between tympanoplasty types
-
• Staged tympanoplasty had higher surgical care costs
-
• Same-day bilateral tympanoplasty is recommended for most bilateral chronic suppurative otitis media patients
Two significant differences between the two patient groups were the overall surgery duration and the cost of care. However, neither of these factors influenced the tympanoplasty outcome in either group. Suzuki et al. recently reported that prolonged anaesthesia is independently associated with early local wound complications after tympanoplasty for chronic otitis media.Reference Suzuki, Yasunaga, Matsui, Fushimi and Yamasoba 16 However, a limitation of the current study was the short follow up: it is possible that a longer follow up may have produced different results.
Surgery is usually a low priority in global health planning in most developing countries, and surgical interventions have traditionally been perceived as expensive.Reference Löfgren, Mulowooza, Nordin, Wladis and Forsberg 17 In these settings, most people are assumed to make out-of-pocket payments for surgical procedures. Considering the low SNHL risk and high cost of surgical care for staged (i.e. different-day) tympanoplasty, same-day bilateral tympanoplasty is recommended for most patients, with staged surgery reserved for specific bilateral CSOM patients (i.e. high risk surgical patients and those with ossicular dissolution or CSOM-associated with cholesteatoma).
Acknowledgements
The authors acknowledge Drs N Undie, E Opaluwah, U C Ukwuije, Y Oyeyipo and I Odiba for their referral of all patients in this study.
