Key points:
Introduction
Endoscopic myringoplasty has gained popularity in recent years, as its use can help to achieve a clear surgical vision,1 less postoperative pain, and a high success rate both in anatomy and function.2 Full-thickness tragal cartilage (FTTC) is the grafting material of choice for endoscopic myringoplasty.3,4 FTTC is easy to harvest and suitable for one-handed operations under the endoscope due to its hardness. However, there is a considerable difference in the thickness, stiffness, and elasticity between FTTC and the natural tympanic membrane. Many otologists doubt its postoperative hearing performance. Performing myringoplasty under the microscope with a temporalis fascia (TF) graft is the classic surgical approach. However, in endoscopic myringoplasty, partial-thickness tragal cartilage (PTTC) is used as a graft, which is closer to the natural tympanic membrane. Here, we aimed to compare the hearing performance between the two myringoplasties, microscopic and endoscopic, and between FTTC and PTTC graft in endoscopic myringoplasty. We also compared the postoperative audiological performance between them.
Materials and Methods
Subjects
This retrospective cohort study was conducted at the Department of Otolaryngology of a tertiary hospital in China between January 2017 and February 2020. Adult patients (>18 years) with chronic otitis media, persistent tympanic membrane perforation, an intact ossicular chain on computed tomography scan, and ossicular chain mobility during the intraoperative assessment were included. Hearing performance in microscopic and endoscopic myringoplasties, each using a different graft, was compared. For the FTTC group, cartilage with no thinning and perichondrium on one side was used; the myringoplasty was performed under the endoscope. For the TF group, traditional myringoplasty was performed under the microscope using TF as a graft. For the PTTC group, ultra-thin tragal cartilage was used and the myringoplasty was performed under the endoscope. Postoperative tympanum closure rate and hearing performance were evaluated 3 months after the operation.
The study protocol was approved by the Ethics Committee of the Hospital.
Definitions of variables
The mean air-bone gap (ABG) was the average ABG value at 500 Hz, 1000 Hz, 2000 Hz, and 4000 Hz. The mean ABG was divided into four categories: 1) ≤10 dB, 2) 11–20 dB, 3) 21–30 dB, 4) ≥31 dB. Three factors affecting the ABG were included in this study: time (before and after surgery), tympanic perforation size (large [≥50%] and small [<50%]), and exposure of malleus handle (exposed and unexposed).
Statistical analysis
Continuous variables were presented as mean ± standard deviation for normally distributed variables and as median (interquartile range) for variables that followed a non-parametric distribution. Categorical variables were compared using the chi-square test. Quantitative continuous variables were compared using the unpaired Student’st -test or the Mann-Whitney U test for normally and non-normally distributed variables, respectively. The factors affecting ABG were compared using the general linear model repeated measures test.
Propensity score matching (PSM) was performed to balance the baseline characteristics between the TF group and the FTTC group. Age, gender, size of perforation, exposure of malleus handle, and ABG values at frequencies from 250 Hz to 4000 Hz before surgery were matched. Before PSM, there were 40 patients in the microscopic group and 111 patients in the endoscopic group. However, after the cohorts were propensity score-matched to create a 1:1 matched set using a caliper width of 0.02; a total of 38 patients were included in each group. After the two groups were matched, the distribution of the ABG values for frequencies of 250 Hz, 500 Hz, 1000 Hz, 2000 Hz, and 4000 Hz did not obey normal distribution; therefore, a non-parametric test, to compare two-related-samples, was used for analysis.
P -values < .05 was considered statistically significant. IBM SPSS for Windows version 22.0 (IBM Corp., Armonk, NY, USA) was used for analyses.
Myringoplasty procedure
After infiltration of the canal with local anesthesia, the edges of the perforation were freshened. A broad tympanomeatal flap was elevated from 11 to 6 o’clock (left ear) and the middle ear cavity was entered; care was taken to preserve the chorda tympani nerve. The ossicular chain was palpated and its mobility was confirmed. The tympanic mucosa, anterior space of malleus, anterior and posterior tympanic isthmus, and eustachian tube orifice were explored. FTTC was harvested with perichondrium on one side, and a ”V” shaped notch was made. The epithelium on the malleus handle was removed, and the FTTC was inserted between the malleus handle and the tympanum, with the perichondrium facing outward. The ”V” shape was embedded in the short process of the malleus. The graft was supported with gelatin sponges soaked in normal saline. A zero-degree endoscope (diameter, 3 mm; length, 140 mm) was used under an endoscope monitoring system (Storz TC200, Germany).
In the TF group, a classic microscopic myringoplasty was performed through postauricular and transcanal approaches utilizing the TF for reconstruction. The graft was placed under the malleus handle.
In the PTTC group, endoscopic myringoplasty was performed utilizing ultra-thin tragus cartilage with a perichondrium layer as the graft, which was placed under the malleus handle.5
Results
Myringoplasty with FTTC graft: Factors affecting the mean ABG and its trend
The average thickness of the FTTC was 0.850 ± 0.117 mm (range, 0.642–1.083 mm). The mean operation time was 52.44 ± 14.04 minutes, and the average postoperative dry ear time was 3.457 ± 1.78 weeks. The tympanic membrane integrity was restored in 91% of the cases. Figure 1 presents key images before, during, and after surgery. A mean ABG of 12.17 ± 6.58 dB was observed postoperatively (≤10 dB, 45%; 11–20 dB, 45.9%; 21–30 dB, 7.2%; ≥31 dB, 1.8%; Figure 2A.)
Further, the impact of the factors affecting ABG was explored. The factors assessed before and after surgical intervention were perforation size and malleus handle exposure status. The mean ABG changed significantly after surgical intervention (F(1,108) = 38.707, P < .001). The tympanic perforation size and malleus handle exposure status significantly affected the mean preoperative ABG (F(1,108) = 5.969, P = .016; F(1,108) = 7.281, P = .008, respectively). Further, no interaction was found between surgical intervention and perforation size (F(1,108) = 1.022, P = .314) and between surgical intervention and malleus handle exposure status (F(1,108) = 0.053, P = .818). To summarize these findings, the perforation size and malleus handle exposure status did not affect the ABG trend (change in the ABG values before and after surgery). The distribution of mean ABG under different factors is presented in Figure 2B.
Comparison of audiological performance betweenTF and FTTC groups