3.3 Comparison of audiological performance between the FTTC and PTTC groups
First, baseline characteristics between the two groups were compared, and no statistically significant differences were observed (Table 2). The mean ABG values in the PTTC and FTTC groups were 12.5 (8.75–16.25) dB and 11.25 (6.25–15) dB, respectively, with no significant difference between the two groups (Z = -0.958, P = .338, Figure 3B.). At frequencies 500 Hz, 1000 Hz, 2000 Hz, and 4000 Hz, there was no significant difference in ABG values between the two groups (P> .05) (Table 3). The postoperative ABG in the FTTC group at 250 Hz was 15 (10–20) dB, which was better than that in the PTTC group (15 [15–20] dB, Z = -2.277, P = .023).
Discussion
Hearing improvement was achieved irrespective of the preoperative tympanic perforation size and malleus handle exposure status in the FTTC group. The postoperative ABG was comparable in the TF and PTTC groups. Moreover, the audiological improvement at 250 Hz was better in the FTTC group than in the PTTC group.
Endoscopic myringoplasty has received special attention in recent years. The grafts used in myringoplasty are diverse, such as temporalis muscle fascia,6 tragus perichondrium,7,8tragus cartilage and perichondrium complex,9 and porcine small intestine submucosa.10 Minimal invasion and easy harvest, less postoperative pain, quicker postoperative recovery in anatomy and function, and strong ability to resist inflammation are the requirements of an ideal graft; since it meets the requirements, tragus cartilage is the material of choice. Moreover, it does not require an extra incision behind the ear, thereby avoiding long-term numbness 11 after the operation.
Endoscopic myringoplasty is a flexible and variable procedure as it facilitates multiple choices of grafts, varied graft placement, and different surgical approaches; however, on the downside, diversity may lead to discrepancies in postoperative hearing performance. There are three main cartilaginous grafts: perichondrium, full-thickness cartilage with perichondrium complex, and thinned cartilage with perichondrial complex. The commonly employed surgical approaches include the ”put through”12 method, which does not involve the elevation of the tympanomeatal flap, and the method involving the elevation of the tympanomeatal flap. The location of graft placement also varies, such as incarceration on the tympanic membrane (”butterfly” method),13 placed between the tympanic membrane and the malleus handle, or medial to the malleus handle. In this study, a broad tympanomeatal flap was lifted, and the FTTC was placed between the malleus handle and the tympanomeatal flap. In both traditional microscopic tympanoplasty with the TF and the FTTC surgery, a similar hearing performance was achieved after the surgery. In a meta-analysis14 where different research groups adopted different surgical and graft placement strategies, a sub-analysis of prospective studies showed a lower mean postoperative ABG in patients using TF graft. The heterogeneity may have affected the stability of the results and led to the difference.
For all frequencies except 250 Hz, a similar audiological performance was observed in the FTTC and PTTC groups; FTTC had a better performance at 250 Hz. Similar comparisons were drawn in another study,4 where no significant differences were observed in hearing improvement between the PTTC and the FTTC groups at 250 Hz, 500–2000 Hz, and 8000 Hz. Overall, our findings were in agreement, but PTTC had a better performance at 4000 Hz in the other study, whereas FTTC had a better performance at 250 Hz in our study. In the same study, the tensor tympani tendon was cut to make space in cases where the handle was foreshortened or medially displaced. This manipulation may have led to poor performance in the FTTC group at 250 Hz. The tensor tympani tendon cut did not affect hearing at 0.5 Hz–4 kHz after tympanoplasty,15 but tensor tympani tendon contraction is related to low-frequency hearing loss.16 In addition, barred cartilage, which might result in decreased tension of drum, was placed under the malleus, which may be responsible for the poor performance at 4000 Hz in the FTTC group in their study. In the future, the effect of tensor tympani tendon cut on low-frequency hearing can be explored.
Although reports on myringoplasty with FTTC graft have demonstrated good hearing performance, only a few articles have provided a convincing theoretical explanation. A study on biomechanical analysis17 suggested the use of grafts with a thickness of 0.1–0.2 mm for medium-to-large perforations and <1.0 mm for small perforations, where the graft is a substitute for both mechanical stability and audibility. However, in this study, the thickness of the FTTC was 0.850 ± 0.117 mm, and good hearing results were obtained at each frequency. The middle ear can only provide about 20 dB of gain on average18 and most of the gain comes from the area ratio of the tympanic membrane to the stapes. The air-containing middle ear cavity can reduce the sound pressure on the round window membrane.19 Figure 1F shows that the graft fits well with the malleus handle; the new cartilage drum could provide a sufficient drum area and an aired cavity, so the graft would fit well and a good hearing performance with FTTC graft could be achieved. Moreover, a better hearing performance at 250 Hz was achieved in the FTTC group than that in the PTTC group. The possible reason for this could be that FTTC changes the stiffness of the tympanic membrane, which may be beneficial in low-frequency conductions. However, it does not mean that the FTTC graft can replace the PTTC graft, which is convenient for postoperative examination and manipulation.
Postoperative hearing performance in the FTTC group was independent of the preoperative tympanic perforation size and the exposure status of the malleus handle. According to our previous research, factors such as the state of mucosa and location of tympanic perforation do not affect the trends of ABG.5Another study also reported similar results.20 In a previous study, a linear correlation was observed between the preoperative tympanum perforation size and postoperative ABG (P = .0017) at frequencies of 0.5 to 4 kHz, indicating an increased postoperative ABG in larger perforations; however, here, we specified only the P -values and did not provide the correlation coefficient or other parameters that reflect the relationship between the perforation size and postoperative ABG.
There are some limitations of this study. First, this was a retrospective analysis performed at a single center. Second, the sample size of the TF group was small, and the baseline characteristics between the TF and FTTC group were imbalanced. However, PSM was performed to resolve the issue of imbalanced baselines and sample sizes. Third, the follow-up duration of this study was short. The patients at this center came from all over the country. Since improvement in hearing and dry ear was observed in the early days after the surgery and a longer follow-up meant increased loss to follow-up, we chose 3 months as the endpoint of follow-up.