Introduction
Tympanic membrane (TM) perforation caused by middle ear infections or traumatic damage is a common disease in the clinical practice of Otolaryngology. Persistent perforation of TM may result in repeated infection of the middle ear and hearing loss, thus requires surgical repair to improve the patients’ quality of life. Microscopic type I tympanoplasty/myringoplasty has been the most widely used technique and achieves good results for repair of TM perforation.Reference Lee, Lee, Seo and Kim1 However, microscopic type I tympanoplasty/myringoplasty has several disadvantages despite its effectiveness. Firstly, in certain cases, the complete exposure of the tympanic annulus may be challenging due to anatomical variations of the external auditory canal, necessitating canalplasty for some patients.Reference Tseng, Lai, Wu, Yuan and Ding2 Additionally, the use of post-auricular incision may lead to post-operative ear numbness and cosmetic concerns.Reference Chen, Jiang, Yang, Wang and Li3
Transcanal endoscopic ear surgery (TEES) was introduced in the 1990s as an innovative alternative to traditional microscopic ear surgery.Reference Schwam and Cosetti4 Compared to microscopic techniques, otoendoscopy provides a broader field of view, especially for the anterior aspect of the ear, through a less invasive approach via a narrower ear canal.Reference Furukawa, Watanabe, Ito, Kubota and Kakehata5, Reference Marchioni, Gazzini, De Rossi, Di Maro, Sacchetto and Carner6 Consequently, TEES has gained increased popularity in recent years. Generally, endoscopic TM perforation repair can be performed using two distinct approaches: endoscopic type I tympanoplasty/myringoplasty with tympanomeatal flap elevation (ETMFE), or without tympanomeatal flap elevation (ETMFNE).Reference Chen, Jiang, Yang, Wang and Li3, Reference Fermi, Maccarrone, Villari, Palermo, Alicandri-Ciufelli and Ghirelli7 Recently, several studies have attempted to compare the efficacy of these two techniques. However, conclusive evidence supporting the superiority of one technique over the other, particularly for the repair of large tympanic membrane perforations, continues to be elusive.Reference Chen, Jiang, Yang, Wang and Li3, Reference Tseng, Lai, Wu, Yuan and Ding8–Reference Chen, Kuo, Weng and Hsieh14 Therefore, the present study aims to conduct a systematic review and meta-analysis of existing literature comparing post-operative outcomes between the two approaches in patients with TM perforations. We believe this study will offer valuable insights into the differences between ETMFE and ETMFNE, thus informing clinical decision-making in TM perforation repair.
Methods
This systematic review and meta-analysis adheres to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines and is registered with PROSPERO under registry number CRD42023475060.
Search strategy and selection criteria
Two researchers independently searched the following databases: PubMed, Embase and Cochrane Library. The search terms were: ((endoscopy) OR (endoscope)) AND ((tympanoplasty) OR (myringoplasty)). Databases were searched from inception to 27 October 2023, and no restrictions regarding the language of publication were applied. Additionally, we manually searched the reference lists of the retrieved articles to make sure that relevant studies were not missed.
The inclusion criteria for this meta-analysis were: (1) studies compared the efficacy of ETMFE and ETMFNE for TM repair; (2) randomised controlled trials (RCTs), retrospective or prospective cohort studies; (3) details of operation method were described and the main outcome measures was reported. The exclusion criteria were: (1) studies including patients with additional needs for epitympanotomy, mastoidotomy, or ossicular chain reconstruction for ossicular chain disease and cholesteatoma; (2) relevant data could not be fully extracted; (3) there is no comparability between the ETMFE and ETMFNE groups; (4) reviews, comments, thesis, or conference abstracts.
Main outcome measures
Main outcome measures were prespecified in this meta-analysis: graft success rate, air–bone gap (ABG) changes, operation time.
Quality assessment of included articles
The level of evidence each included study was evaluated according to the Oxford Center for Evidence-Based Medicine (OCEBM).15 Bias for the randomised and nonrandomised studies was assessed using the Jadad ScaleReference Jadad, Moore, Carroll, Jenkinson, Reynolds and Gavaghan16 and the Methodological Index for Non-Randomized Studies (MINORS)Reference Slim, Nini, Forestier, Kwiatkowski, Panis and Chipponi17 respectively.
Data collection
Relevant data were extracted from all eligible studies, including first author, year of publication, sample size of each study arm, mean age and gender distribution of the study group, size of tympanic membrane, graft material, main outcome measures and follow-up period.
Statistical analyses
Statistical analyses were performed using Review Manager software, version 5.4 from the Cochrane Collaboration (London, UK). Meta-analyses were conducted regarding the proportion of patients with healed perforation, ABG changes in decibels and operation time. The risk ratio (RR) and 95 per cent confidence interval (CI) were estimated for the numbers of patients with healed perforation. The mean difference (MD) and 95 per cent CI were estimated for the ABG changes and operation time. Cochrane’s I2 index was used to assess heterogeneity, and if the I2 test suggested a high heterogeneity (I2 ˃ 50 per cent), the RRs and MDs were pooled by the random-effects model. Otherwise, the fixed-effects model was used. Publication bias was assessed by Deek’s funnel plot asymmetry test. A p < 0.05 was considered to indicate a statistically significant difference for all analyses.
Results
Study selection
The initial database search yielded 1220 potentially relevant articles. Following the removal of duplicates, 849 records remained. After a thorough review of titles and abstracts, 840 records were excluded. An additional record was found through other sources. Subsequently, 10 articles underwent a comprehensive full-text review. Among these, two studies were excluded because the sizes of tympanic membrane perforations between the two groups were not comparable, and one study was excluded due to relevant data not being fully extracted. Ultimately, seven studies were considered eligible for inclusion in the final meta-analysis. The process of article selection is depicted in Figure 1. Table 1 offers a summary of the characteristics of the included studies.
A flowchart summarizing the step-by-step article selection process for the study.
Summary of the characteristics of the included studies
ABG = air–bone gap; ETMFE = endoscopic type I tympanoplasty/myringoplasty with tympanomeatal flap elevation; ETMFNE = endoscopic type I tympanoplasty/myringoplasty without tympanomeatal flap elevation; GSR = graft success rate; TCPC = tragal cartilage–perichondrium composite; TF = temporalis fascia; TP = tragal perichondrium.
Quality assessment
For nonrandomised studies, two reviewers independently assessed comparative studies using the MINORS tool, which consists of 12 items. Each item was scored from 0 to 2, with a maximum cumulative score of 24. All four included studies received a high score of 20 points. Randomised studies were evaluated using the Jadad scale, which allocates points based on criteria such as the adequacy of randomisation, blinding and comprehensive reporting of patients’ data, utilising a five-point scale. Two of the three randomised studies achieved a perfect score of 5/5, while the remaining study scored 3/5 (Table 2).
Study site and quality assessment of included studies
Results of the meta-analysis
All seven studies included in this study reported data on graft success rate. Given the absence of statistical evidence of heterogeneity across these studies (I2 = 0; p = 0.72), a fixed-effect model was employed for the meta-analysis. The calculation of the RR indicated that there was no significant difference in graft success rates between the ETMFNE group and the ETMFE group (RR = 1.01; 95 per cent CI: 0.98–1.04; p = 0.54; Figure 2). Of the seven studies included, six provided data on graft success rates for small and medium perforations (smaller than 50 per cent of the TM), while four included data for large perforations (larger than 50 per cent of the TM). The subgroup analysis, stratified by perforation size, indicated no significant difference in graft success rates between the ETMFNE and ETMFE groups for both small to medium (RR = 1.00; 95 per cent CI: 0.97–1.03; p = 1.00) and large perforations (RR = 1.04; 95 per cent CI: 0.96–1.13; p = 0.36). Additionally, the comparison between these subgroups did not yield a significant difference (p = 0.40), as depicted in Figure 2.
Forest plot illustrating the comparative analysis of graft success rates in the meta-analysis.
Five studies provided data on the post-operative ABG gain. We employed a random-effect model for pooling the trials due to observed heterogeneity between the studies (I2 = 81 per cent). The results indicated that there was no difference in the ABG improvement between ETMFNE and ETMFE group (MD = –0.08; 95 per cent CI: –2.02–1.85; p = 0.93). Further subgroup analysis, stratified by the presence of large perforations, also showed no significant differences between the groups in both the small to medium perforation subgroup (MD = –0.12; 95 per cent CI: –3.25–3.01; p = 0.94) and the large perforation subgroup (MD = 0.17; 95 per cent CI: –1.29–1.64; p = 0.81). Comparisons between subgroups did not reveal significant differences (p = 0.87; Figure 3).
Forest plot illustrating ABG improvement of the meta-analysis. ABG indicates air-bone gap.
Additionally, six studies provided information on the operation time. In contrast to the ETMFNE group, the result of meta-analysis revealed a significant increase in operation time in the ETMFE group (MD = 19.50; 95 per cent CI: 10.75–28.25; p < 0.0001), with high heterogeneity among included studies (I2 = 98 per cent; Figure 4).
Forest plot summarizing the surgical operation times in the meta-analysis.
Complication
Six studies provided data on complications in either the ETMFNE or the ETMFE group. Atchariyasathian et al. reported two cases of temporary dysgeusia in the ETMFE group. Notably, none of the patients in the ETMFNE group experienced changes in taste in all the studies. Lou et al. and Wang et al. documented middle ear infection in three patients in ETMFE group and one patient in ETMFNE group. Additionally, Tseng et al. reported that one patient in the ETMFNE group and two patients in the ETMFE group developed post-operative fungal infection.
Publication bias
Potential publication bias in our meta-analysis was assessed both visually through the inspection of the funnel plot and statistically using Egger’s test. Although the funnel plot showed slight asymmetry (Figure 5), the Egger’s test indicated no significant publication bias (t = −0.79; p = 0.467).
A funnel plot evaluating the risk of publication bias in the included studies.
Discussion
Principal findings
TEES employs the endoscope to navigate the external auditory canal, bypassing anatomical constraints and offering a high-resolution, magnified view of the entire TM. This has been particularly beneficial in cases with narrow or tortuous ear canals.Reference Akyigit, Sakallıoglu and Karlidag18, Reference Schwam and Cosetti19 Endoscopic repair of TM perforation can be performed via two general techniques, ETMFNE or ETMFE. ETMFE involves the elevation of a tympanomeatal flap to enter the middle ear. This technique requires a high level of skill and expertise in endoscopic surgery, which can be challenging for many surgeons to acquire. ETMFE is also associated with risk of damage to the chorda tympani, bleeding and longer operation times, especially for inexperienced surgeons.Reference Doğan and Bayraktar20 In contrast to ETMFE, the ETMFNE procedure is characterised by its minimal invasiveness, with the graft being delivered through the perforation, thus avoiding the necessity of raising the tympanomeatal flap. Although numerous studies on this subject have been published in recent years, there remains no consensus on which procedure leads to superior anatomical and functional outcomes. In this meta-analysis, we synthesised the available data on anatomic and auditory results to examine the comparative efficacy of ETMFNE and ETMFE. The results showed that the graft success rate of ETMFNE was 94.3 per cent (399/423), and that of ETMFE was 95.0 per cent (395/416). Regarding the auditory outcome, our analysis showed that both techniques enhanced ABG closure, with a mean difference of 0.087 dB between the two groups, which implies that there is no substantial clinical difference in hearing results between the two techniques. Consequently, the findings from this meta-analysis indicate that there is no significant difference in the perforation closure success rate and hearing outcomes between the two procedural techniques.
Despite the absence of definitive documentation, a common belief exists that ETMFNE is particularly fitting for minor to moderate perforations, and ETMFE is deemed appropriate for all perforation sizes.Reference Nicholas, Jeong, Nguyen and Lambert21 However, emerging studies have demonstrated promising outcomes for ETMFNE even in cases of larger perforations.Reference Chen, Jiang, Yang, Wang and Li3, Reference Atchariyasathian, Suwannajak, Plodpai and Pitathawatchai12–Reference Chen, Kuo, Weng and Hsieh14 The subgroup analysis results indicated no significant difference in graft success rates and hearing improvement between the ETMFNE and ETMFE groups when stratified by perforation sizes exceeding and not exceeding 50 per cent of the TM. At the same time, a notable finding was the significantly shorter operation time observed in the ETMFNE group compared to the ETMFE group. The short operation time may not only minimise patients’ exposure to the inherent risks associated with anesthesia but also substantially decrease healthcare costs.
Nicholas et al. conducted a meta-analysis comparing myringoplasty to type I tympanoplasty in the repair of the tympanic membrane, including both microscopic and endoscopic surgical approaches.Reference Nicholas, Jeong, Nguyen and Lambert21 Different from their study, our meta-analysis exclusively focused on the comparison of ETMFE to ETMFNE. We included three additional studies published recently, one study that was previously overlooked and excluded one study due to incomparable study groups. Despite these differences, our results are consistent with those of Jungbauer et al., further validating the similarity in anatomic success and ABG outcomes for both surgical techniques.
Limitations of the study
It is crucial to recognise the limitations inherent in our study. Firstly, despite including seven studies in the analysis, the relatively small number of studies may limit the comprehensive interpretation and generalisability of the results. Secondly, the presence of non-RCTs and retrospective studies in the analysis could introduce bias, which might affect the reliability of the results. Lastly, the follow-up periods in the included studies varied from 6 to 12 months post-operatively. Data from a more extended follow-up duration may provide unique insights into the enduring efficacy of the surgical interventions.
• The results of previous meta-analyses showed that the anatomic success rate and air–bone gap (ABG) improvements are similar for tympanoplasty and myringoplasty.
• Transcanal endoscopic ear surgery has gained increased popularity in recent years. Endoscopic repair of tympanic membrane (TM) perforation can be performed via two general techniques, endoscopic type I tympanoplasty/myringoplasty with tympanomeatal flap elevation (ETMFE) or without tympanomeatal flap elevation (ETMFNE).
• This systematic review and meta-analysis has compared the efficacy of ETMFNE and ETMFE for tympanic membrane repair.
• Seven studies were included in this meta-analysis.
• The result of this meta-analysis suggests that both techniques demonstrate similar graft success rates and post-operative auditory improvements, ETMFNE offers the advantage of shorter operation times, particularly in appropriately selected cases.
Conclusion
The findings of this meta-analysis provide valuable insights into the comparative effectiveness of ETMFNE and ETMFE in TM perforation repair. While both techniques demonstrate similar graft success rates and post-operative auditory improvements, ETMFNE offers the advantage of shorter operation times, particularly in appropriately selected cases. However, this finding should be interpreted with caution due to the limitations mentioned. Future research with larger sample sizes, longer follow-up durations and robust study designs will be crucial to confirm these results and further inform clinical decision-making in TM perforation repair.
