Audiological function in a group of adults following myringoplasty: an exploratory study in South Africa

Introduction Chronic suppurative otitis media is a global middle ear disease with quality of life as economic implications, which are worse felt in low and middle income (LAMI) countries; thus the need for myringoplasty. This study aimed to explore audiological function in a group of adults following myringoplasty in South Africa, with an exploration of the possible influence of factors such as HIV/AIDS and type of surgical technique on hearing outcomes. Methods Within a retrospective chart review research design, 41 participant files for a six-year period from two academic hospitals in Johannesburg, South Africa, were reviewed. Data were analysed using both descriptive and inferential statistics. Results Participant files comprised of 16 males and 25 females between 18-63 years. Findings revealed that clinically, overall hearing improved post-operatively, as indicated by improved tympanometry findings, pure tone air-conduction and speech reception thresholds. Descriptively, the predictors of improved hearing outcomes post-operatively appeared to be HIV negative status and butterfly cartilage inlay surgery as a surgical technique adopted. Although clinically, hearing outcomes improved post-operatively at all air-conduction frequencies tested; these clinical improvements were only statistically significant at specific frequencies. Conclusion Current findings provide useful initial evidence on the benefits of myringoplasty from the South African context; particularly because of the HIV/AIDS prevalence and its potential influence on middle ear disease and its management. Prospective efficacy studies with bigger sample sizes are recommended, with early identification strategies for middle ear disease to reduce the need for myringoplasty seriously considered bearing in mind the resource constraints.


Introduction
Globally, hearing loss is the most common sensory deficit across the human race [1]. Eighty percent of deaf and hearing-impaired people live in low and middle-income countries where services are either totally absent or very limited [2]. WHO [1], further highlights that the major preventable causes of hearing impairment in these countries include middle ear infections. The burden of otitis media occurs overwhelmingly in these countries with almost nine times more cases reported if compared to developed countries [3]. Despite being the most prevalent disabling condition globally and one of the major contributors to the global burden of disease, hearing loss has historically been ignored on global health care agendas [4]. According to the WHO [5], it is "easily overlooked and underestimated" because it is not as "dramatic" as other health care conditions. It is therefore not surprising that hearing loss has been referred to as a silent epidemic. Hearing health care surveys confirm a global paucity of ear and hearing care services due to the limited numbers of available hearing health care professionals [2,6]. A recent survey of countries in sub-Saharan Africa indicated that many of these countries do not have any audiology or otolaryngology services [2]. This shortage of hearing health care professionals is primarily due to a reported lack of government funding, professional and public awareness, and, most significantly, available training programs [6]. Only two African countries, for example, indicated having any training programs in audiology and many countries also indicated that they had no otolaryngology training programs [2].
Factors influencing health care in developing countries are numerous.
South Africa has the largest number of people living with HIV/AIDS [7]. Disorders of the auditory and vestibular system, including otitis media, are often associated with HIV/AIDS [7,8].
Otitis media as a global middle ear disease is reported to be as high as 11% in some African countries, with severe economic implications [9]. In South Africa, the high prevalence of otitis media has been attributed to factors such as overcrowding, inadequate housing, poor hygiene, not breastfeeding, poor nutrition, and high rates of nasopharyngeal colonisation with potentially pathogenic bacteria [10]. As with most infectious diseases, the burden of acute otitis media varies substantially across countries [11]. Studies have shown that the main differences in the manifestation of the burden of acute otitis media resides in the frequency of suppurative complications such as mastoiditis and meningitis and in the consequences such as hearing loss due to chronic suppurative otitis media as well as its management [12][13][14]. Management varies and presents a range of efficacy [15,16]. Due to South Africa facing a number of challenges that complicate the progressive realization of access to health care [17], it can be suggested that not all patients are able to access treatment early and promptly thus leading to untreated otitis media which can result in permanent hearing loss.
South Africa has the highest income inequality globally and the gap between public and private health care, with regards to affordability and quality of service remains a great concern [17].  [19,20]. Where spontaneous closure has not occurred, three principal indications for surgical intervention in the form of myringoplasty have been documented: (1) recurrent otorrhea, (2) desire to swim without wearing water proof in the ear, and (3) need to improve the conductive hearing loss resulting from a non-healing perforation [20]. Myringoplasty can improve hearing in many cases.
Although a 90% success rate of surgery is frequently quoted for myringoplasty, the routine nature of the procedure, as well as the effect of many influencing factors on the success or failure of this procedure remains unresolved [20]. It has been suggested that factors such as the age of the patient; site of the perforation, size of the perforation, length of time that the ear has been dry for prior to surgery, the presence of infection at the time of surgery, as well as the status of the opposite ear may all be influencing factors affecting outcomes of myringoplasty [20,21]. Several studies have reported that the most popular techniques for closure of a tympanic membrane perforation include either the temporalis fascia underlay approach or the butterfly cartilage inlay approach [21][22][23][24][25][26]; and these are the techniques investigated in this study as they were found to be the most common types of techniques performed at tertiary hospitals in Gauteng.
Clinical studies comparing outcomes of tympanic membrane reconstruction with fascia and cartilage show contradictory results [27], due to confounding variables such as revision surgery, variances in size and location of the perforation, a draining ear at the time of surgery, bilateral disease, ossicular discontinuity or cholesteatoma [25]. Hearing outcomes also vary considerably, with some studies indicating no significant difference in hearing outcomes post operatively while others do [26][27][28], with some studies from district hospitals in developing countries and remote indigenous communities demonstrating promising results [29][30][31][32]. However, some of these studies have not included audiological outcomes [31] or were limited by small sample sizes [32], while others have reported disappointing outcomes that are significantly poorer than those achieved at tertiary institutes in developed nations [33].

Research design
A retrospective chart review design was adopted [37], where medical, surgical and audiological records were reviewed.

Participants and sampling
Through non-probability purposive sampling technique where participant files meeting specific inclusion criteria were included in the study [38]; a total of 41 participant files were included in the study (Table 1) where available [39]. 2) Air-conduction and bone-conduction thresholds: air-bone gaps (ABGs) at 500Hz, 1000Hz, 2000Hz and 4000Hz were calculated. According to Valente [39], normal hearing thresholds for adults are recorded at 25dB or less. Air-bone gaps are considered to be significant when greater than or equal to 10dB [39].

3) Speech reception threshold (SRT): Speech reception threshold
results for each ear were documented pre and post operatively.

Data analysis and statistical procedures
Data were analysed using both descriptive and inferential statistics

Ethical considerations
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional guidelines on human experimentation. Ethical approval from the University's Medical Ethics Committee (Protocol number M121012) was obtained.

Results
The

Type of hearing loss
Descriptively, the type of hearing loss was analysed pre and post operatively ( Table 1). Results revealed that the total number of ears with hearing loss (including conductive hearing loss and mixed hearing loss) pre operatively was 49 and this decreased postoperatively to 36. As depicted in Table 1, the total number of hearing loss decreased post-operatively suggesting that overall hearing (airconduction and bone-conduction thresholds) improved post operatively in the total sample. This was particularly noted in participants with purely conductive hearing loss (CHL); without any sensory (inner ear) involvement. Five participants who had presented with mixed hearing loss pre-operatively had their conductive components of the hearing loss eliminated, leaving them with sensory loss postoperatively.

Degree of hearing loss
As depicted in Table 1

HIV status
Results in Table 2 show statistically significant differences at 250, 500 and 1000Hz at air-conduction frequencies in HIV negative participants but only at 250Hz in HIV positive participants. A paired t-test comparison in air-conduction outcomes in HIV positive participants to HIV negative participants showed no evidence of statistically significant differences across all air-conduction thresholds as shown in Table 2.
Results presented in Table 3 show no statistically significant differences in mean change in hearing function in both HIV negative and positive participants at all frequencies tested by bone-conduction.
Similarly, a comparison of mean differences in bone-conduction outcomes in HIV positive participants to HIV negative participants show no evidence of statistically significant differences across all frequencies as shown in Table 3.

Type of surgical technique
Results in Table 4 show statistically significant differences with regard to air-conduction mean change in hearing function in participants who had butterfly cartilage inlay surgery at 250, 500, 1000 and 2000Hz.
On the other hand, no evidence of statistically significant difference in air-conduction mean change in hearing function was found in the participants who underwent fascia underlay surgical technique at all frequencies (Table 4). Table 5 show no statistically significant differences with respect to mean change in hearing function in participants who had butterfly cartilage inlay and fascia underlay surgeries at all frequencies. post-operatively that improves air-conduction thresholds [20].

Discussion
In the current study, the mean hearing gain for air-conduction and bone-conduction thresholds showed that air-conduction thresholds remained the same or improved and bone-conduction thresholds deteriorated overall. The improvements found were both clinically and statistically significant changes; which strengthens current findings. Geber et al. [28] also found that participants who underwent butterfly cartilage inlay surgery had better speech reception thresholds when compared to those who underwent fascia underlay surgery.
HIV status was found to be a possible variable that may have an When assessing the influence of butterfly cartilage inlay surgery on hearing outcomes post-operatively; statistically significant differences with regard to air-conduction mean change in hearing function were observed at 250Hz, 500Hz, 1000Hz and 2000Hz where (p < 0.05), therefore, suggesting that butterfly cartilage inlay surgery had an influence on air-conduction thresholds at specific frequencies. This correlates well with findings by Chhapola and Matta [26] where they claimed that cartilage perichondrium seems be an ideal graft material for tympanic membrane in terms of postoperative healing and improved acoustic properties as it can easily withstand negative middle ear pressure. Furthermore, they found statistical significant differences in all air-conduction thresholds [26]; as in the current study.
On the other hand, no evidence of statistically significant difference in air-conduction mean change in hearing function was found in the ears that underwent fascia underlay surgery at all air-conduction thresholds; suggesting that this type of surgical technique had limited influence on hearing outcomes post-operatively; although this is a preliminary finding since the sample size was small for this type of surgery. Although clinically, there was a difference in the postoperative findings between the two surgeries, these were not statistically significant at all air-conduction thresholds. This suggests that, statistically, the type of surgical technique has no influence on the hearing outcomes post-operatively. These findings highlight the importance of clinical evaluation and indicate that more detailed clinical conclusions must be drawn from a descriptive clinical evaluation rather than just on statistical analysis. Gerber et al. [28] compared hearing outcomes in participants who underwent fascia underlay and butterfly cartilage inlay and found no significant differences in air and bone-conduction thresholds. Mauri et al. [23] found that the audiometric results following inlay cartilage tympanoplasty or underlay tympanoplasty were similar; although inlay butterfly cartilage tympanoplasty presented with benefits of not requiring general anesthesia, was less expensive, and more comfortable to the patient -benefits which are important in a developing country where resources are limited. A single study conducted within the South African context by Becker and Lubbe [24] looked at the type of graft used (cartilage or temporalis fascia) as one of the prognostic factors that may influence outcomes of surgery and found that none of the prognostic factors assessed were statistically significant (p >0.05).
It must be noted that many of the studies reviewed did not use the same criteria to assess hearing outcomes when compared to that of

Conclusion
The current study which aimed to explore the changes in hearing