Tympanic membrane perforation is relatively a common finding, and it is caused by various factors including trauma and otitis media. Perforated eardrum results in conductive hearing loss, and this range is reported not to exceed 50 dB [1,2,3]. But the effects of tympanic membrane perforations on middle-ear sound transmission are not well characterized, largely because ears with perforations typically have additional pathological changes [1,2,3].

Delivery of sound to the middle ear is known to be based on the sound pressure difference between the external and middle ear, and corresponding models have been suggested [1,2,3]. According to corresponding models, the size of perforation and the volume of middle ear and mastoid are related to the degree of hearing loss, and the significance has been reported in other study [4]. However, the impedance audiometry has been used to estimate the volume of middle ear and mastoid pneumatization, which may be limitation of those studies.

In this study, the authors measured the size of perforation, the volume of middle ear and mastoid pneumatization on temporal bone computerized tomography (TBCT), and the location of the perforation in the patients with tympanic membrane perforation and intact ossicular chain. The purpose of this study was to investigate how these factors affect the hearing.

The mean ABG was 19.9 dB. There was no significant relation between the size of tympanic membrane perforation and the volume of middle ear and mastoid (P=0.862, Pearson correlation), or between the location of tympanic membrane perforation and the volume of middle ear and mastoid (P=0.412, Pearson correlation). There was no significant difference in the size of perforation according to the location of tympanic membrane perforation (P=0.161, Pearson correlation). The univariate general linear model, three-way ANOVA, and scheffe post hoc test, showed that the size of tympanic membrane perforation and the volume of middle ear and mastoid pneumatization significantly affected the mean ABG. There was no effect of the perforation location on the mean ABG (Table 1). The volume of middle ear and mastoid pneumatization (F=6.004) seemed to be more effective than the size of tympanic membrane perforation (F=4.269).

In normal ears, the difference in sound pressure that develops in the border of tympanum causes the movements of eardrum and ossicles [3]. Voss et al. [1,2,3] reported that hearing loss in case of tympanic membrane perforation is caused by decreased negative pressure between the external and middle ear when the eardrum was the boundary, and that this difference in negative pressure is determined by the impedance of tympanic membrane perforation and of middle ear, when the movements of ossicles were hypothesized as normal. Therefore, a simple model was suggested based on the auditory structure to predict hearing loss due to tympanic membrane perforation. Mehta et al. [4] used this model for their study and demonstrated that hearing loss in case of tympanic membrane perforation is greater with larger size of perforation and with smaller middle ear and mastoid volume. However, it was not be related to the location of tympanic membrane perforation.

Our results also showed that the mean ABG significantly increased as the size of perforation increased (P=0.018) and the volume of middle ear and mastoid decreased (P=0.005), which is well agreed with the other previous studies [4,6,7,8,9]. There was no difference in the mean ABG according to the location of perforation (P=0.924). There are two major differences between the previous reports by others and this study. Firstly, in the previous studies above, the impedance audiometry was used to estimate the volume of middle ear and mastoid pneumatization [4]. However, the volume exceeding 7 mL could not be measured due to limitation of the instrument, thus in that cases the data were processed as 7 mL, which would be one of the limitations of that study [4]. In addition, although there have been reports on the feasibility of the impedance audiometry for the measurement of the middle ear and mastoid volume [10,11], but when this method was used in case of small volume of middle ear and mastoid, the error increased [4]. In contrast, in this study, the volumes of middle ear and mastoid pneumatization were obtained by multiplication of the total area of pneumatization in the middle ear and mastoid by the distance between the cross sections on TBCT. One limitation of this study is that mastoid pneumatization above the superior semicircular canals could not be included since TBCT did not cover these area in some of the patients. Secondly, previous studies used Student t-test to prove significance [4], whereas in this study, univariate general linear model, three-way ANOVA was used.

Mastoid pneumatization begins from 33rd week after birth, and develops until the age of 15 for boys, and 10 for girls [12,13,14]. However, the degree of pneumatization between individuals is known to show a big difference [15,16], and that the volumes of normal ears are in the range of 2-20 mL. The volume of mastoid pneumatization in Korean adults was 7,095.2 mm³ in average, but showed to have big standard deviation of 3,087.1 mm³, which demonstrated individual diversity [17]. The difference on mastoid pneumatization may come from pathological changes in the middle ear and mastoid due to the otitis media that developed after birth [16,18,19,20].

In conclusion, hearing loss in case of tympanic membrane perforation was related to the size of perforation and the degree of middle ear and mastoid pneumatization when the mucosa inside the middle ear cavity and ossicles were hypothesized as normal. This shows that not only the size of perforation but also the degree of middle ear and mastoid pneumatization need to be considered when we treat the patients with tympanic membrane perforation. Therefore, small pneumatization of middle ear and mastoid may be the causative reason as long as middle ear and ossicles are normal when ABG is relatively larger than that expected from the size of perforation. In these cases, successful closures of ABG after tympanoplasty can be expected without manipulation of ossicles. Further studies may be needed on middle ear and mastoid pneumatization through reconstruction of 3-dimensional images of TBCT.