4. Discussion
The two most widely used graft materials in tympanoplasty are temporal fascia (TF) and auricular cartilage and and the popularity of the cartilage graft has been increasing in recently [1, 2]. PCG and ICG are popular graft types that have long been used for TM repair in cartilage tympanoplasty. Dornhoffer [7] reported an analysis including 1000 cases of tympanoplasty using ICG and PCG (conchal and tragal cartilage) in 2003 and calculated the perforation rate as 4.2% (n = 9) during an average of 12 months follow-up with 215 patients who were operated for pure TM perforation. This study included revision cases, children, and adults. The average preoperative ABG was found to be 21.7 ± 13.5 dB, and 11.9±9.3 dB at the end of the follow-up (p< 0.05). In this study, atelectasis occurred as a complication in during follow up [7]. The technique of PCG (conchal cartilage) and the indications have been identified for the first time in the literature by Heerman et al. [8]. The PCG is divided into groups according to their thickness as a slice or strip and also combined with fascia [3]. Khan and Parab [9] observed graft failure in 5 (2.3%) cases after 4 years of follow-up in 223 cases of primary tympanoplasty with PCG (conchal cartilage). In the same study, perforation was not detected in patients with small perforations (perforation< 50% of TM), while all 5 graft failures were detected in 154 (3.2%) patients with large perforations (perforation> 50% of TM). The mean of preoperative ABG was 30.6±4.7 dB and 7.1±3.0 dB after four years of follow-up. In this study, the perichondrium of the cartilage slices was preserved and no retraction was detected postoperatively [9]. Velepic et al. [10] identified graft defects in 14 (25%) of 51 patients (9 children and 42 adults) who underwent tympanoplasty with PCG (conchal cartilage) and also detected perforation in 2 patients (3.5%) in 11 years of follow-up. In this study, the indication was reported as total tympanic membrane (TM) perforation in 38 patients (68%), and retraction and cholesteatoma were reported in other patients. The mean preoperative ABG was found to be 27.2±10.2 and 10.7±7.8 dB after 11 years. PCG has applied with perichondrium on both sides of the palisades and cartilage resorption was detected in 14 (25.0%) patients in this study [10]. Kazikdas et al. [11]. reported a high success rate as 95.7% for PCG. In this study, the preoperative PTA was 31.4±10.7, and 22.4±12.0 dB at the end of 18.7 months of follow-up, and the mean decrease in ABG was recorded as 8.3 dB. Perichondrium was preserved only on the lateral side of the cartilage in this study and retraction was not reported [11]. A similar study reported the graft success rates as 97% (n=34) and 78% (n=19) in ICG (conchal cartilage) and PCG (tragal cartilage) graft groups respectively [12]. According to audiological results, the average ABG decreased to 11.9 dB and 15.2 dB, respectively. Similarly, PCG was prepared with one-side perichondrium, and retraction was not mentioned [12]. Vashishth et al. [13] compared PCG with the temporal fascia graft for TM perforation and reported 90% graft success for PCG after one-year follow-up. The mean of PTA was determined as 29±6.21 dB preoperatively and 7.33±3.88 dB after surgery. An average ABG decrease of 21.6±6.7 dB was recorded and retraction was not detected [13]. Jeffery et al. [14] calculated the weighted average graft take rate for PCG in type-1 tympanoplasty as 96% and the average postoperative ABG closure as 20.9 dB in a systematic meta-analysis. It was also discussed in this review that while postoperative complications are usually listed the only perforation but subgroups are not detailed, and follow up periods approximately 12 months [14]. In this study, the mean of postoperative PTA was 24.5±9.7 dB and ABG closure was 7.5±7.0 dB in the PCG group and results were similar to the literature. Also, we found that retraction developed between cartilage slices in two of the four failure case in the PCG group.
Island cartilage graft (ICG) was used for similar indications and has been described as a composite graft on the perichondrium containing different numbers of independent cartilage islands in different shapes [3, 6]. Dornhoffer [15] presented the results of 22 type-1 tympanoplasty cases with ICG for atelectatic TM and recurrent perforation and graft failure was detected in 3 patients (13.6%) at the end of an average of 11 months of follow-up. The mean ABG was defined before the operation as 21.1±8.4 dB and 6.8±6.1 dB after the operation. Also, it was reported that the cartilage graft was intact, even though otitis media with effusion occurred in 3 patients during follow-up [15]. Kirazlı et al. [16] presented an average of 17 months of follow-up outcomes of 15 type-1 tympanoplasty cases performed with ICG (tragal cartilage). The indications for tympanoplasty in this study are similar to the presented study but graft take rates has not mentioned. In the audiological examination, the mean ABG before surgery was 28.1±8.8 dB and 16.2±6.2 dB at the end of follow-up [16]. Önal et al. [17] detected the success of graft as 93.2% in 44 type-1 tympanoplasty cases with ICG and ABG gain as 12.7 dB after 21.8 months follow-up. In this study, while emphasizing the resistance of ICG to retraction, it was argued that it could adversely affect hearing results due to the thickness and rigidity of cartilage. Consistent with the initial study, Önal et al. [18] reported the success of the graft as 93.2% in 39 patients using ICG grafts after an average of 20.9 months. Also, graft failure is not detailed in this study. De Seta et al. [19] reported any recurrent perforation with ICG after a one-year follow-up period. ABG was 15 dB and below in 61% of patients in this study after one year of follow-up. However, in this study, retraction was observed in 2 patients (5.5%) at the end of one year. The number of studies with ICG that offer a detailed audiological examination preoperatively and postoperatively, detailed graft success definition, adequate follow-up period, and homogeneous study groups were limited in the literature. In presented study, the success of tympanoplasty with ICG was 86.1%. Preoperative PTA was 33.5±8.4 dB, and postoperative PTA was 28.8±3.0 dB (p< 0.036). One of the graft failures was due to retraction, and four were marginal perforation due to possible surgical technique. In addition, unfavorable audiological results were found in the ICG group compared to PCG and WsCCG groups.
WsCCG tympanoplasty was defined by in 2005 Shin et al. [20] as for the prevention of retraction pockets and the prevention of partial ossicular replacement prosthesis excretion. At the end of an average follow-up duration of 50 months, graft failure was observed in six of 47 patients (12.7%) and retraction was detected in three (6.4%) patients. Mean 6 dB ABG gain was observed at the end of the follow-up. This study emphasized the unsuitable consequences of a thick cartilage plate due to thickness, stiffness, and mass that could compromise the acoustic results. WsCCG re-designed to provide flexibility and avoid stiffness on the perichondrium instead of a single plate-shaped graft or palisade to achieve better audiological results by the authors. [5, 20]. WsCCG (conchal cartilage) was modified and used for type-1 tympanoplasty and the graft success was detected as 97.7% (43/44). After a mean of 15 months of follow-up, ABG improvement was found 6.6±11.0 dB. WsCCG was used in this study for the repair of total or subtotal perforations, and anterior-inferior recurrent perforation was detected in only one patient, no retraction was observed [5]. In the presented study, graft success of WsCCG was 97.2% and ABG closure was detected as a mean of 9.1±6.0 dB after an average of 32.8 months follow up period. ABG closure was detected higher in the WsCCG group than the PCG and ICG groups. Also, only one graft failure detected at the anterior margin of the annulus, and retraction was not observed after a long follow up period compared to the literature.
The results are consistent with the referential study, and the success of the graft is similar to the other study in the literature. However, the fact that this technique has not been experienced by different surgeons and this increases the risk of bias. Also, in this study, a set of homogeneous study groups was aimed and the adhesive COM or retraction pockets were excluded at the beginning of the study. But a total of three retractions were relapsed. This was another limitation of the study.