This study aimed to retrospectively evaluate the prognosis and outcome of tympanic membrane perforations with a particular focus on the fate of the perforation edge flaps.Chart records of 329 patients with a single ear traumatic tympanic membrane perforation were retrieved and analyzed. Of these patients, 70 were left to heal spontaneously, 93 received gelatin sponge patching treatment and 114 were subjected to otoendoscopic eardrum repair before gelatin sponge patching. The complete perforation closure rate at 3 months was 94.29% (66/70), 98.92% (92/93) and 98.24% (112/114) in the 3 groups, respectively, with no statistically significant difference (p = 0.608). The mean closure time was 28.2 ± 3.6 days in the spontaneous healing group, which was significantly longer than that in the sponge patching group (11.1 ± 2.1 days, p = 0.0017) and in the eardrum repair + sponge patching group (12.5 ± 1.9 days, p = 0.0032), while there was no significant difference between the 2 gelatin sponge patching groups (p = 1.86). The hearing ability improved in the 3 groups (6.4 ± 0.83, 7.2 ± 1.65 and 9.6 ± 2.37 dB, respectively), with no statistically significant difference (p >0.05). In the gelatin sponge patching group, new tympanic membrane tissue of the eardrum flap edge proliferated, and the contour of the eardrum flap was not obvious. In the eardrum flap repair group, the eardrum flap retracted to the perforation edge. In conclusion, the eardrum flap of the perforation edge does not have any obvious effect on the perforation closure so that it is unnecessary to conduct an intervention procedure on the flap in the clinical treatment.

In the group treated with the gelatin sponge patch technique, the patients sought medical treatment at different time intervals since the injury. Accordingly, the outcome of the treatment varied in terms of the perforation closure rates achieved in different patients in this group. The respective perforation closure rates were 100%, 100%, 96%, 94%, and 89% in accordance with the time interval at which the patients were treated since the injury. The results were not significantly different when compared by statistical analysis (p > 0.05); the mean closure times in each of the different sets of cases in this group were calculated and the following values were reported: 7.1 ± 2.3, 8.2 ± 1.6, 8.7 ± 1.2, 9.2 ± 3.1, and 10.7 ± 3.9 days. On the other hand, in the edge approximation plus gelfoam patching group, the perforation closure rates were 100%, 97%, 96%, 97%, and 94%, respectively. This was in accordance with the time elapsed since the injury for the patients who visited the hospital on different days. Statistical analysis confirmed that the perforation closure rates for the different cases of this group did not have any significant difference (p > 0.05); the mean closure times were 7.6 ± 1.9, 7.9 ± 2.2, 9.2 ± 2.8, 8.5 ± 3.6, and 11.2 ± 4.1 days, respectively, indicating that differences were not significant even in terms of mean closure rates for the different cases of this group (p > 0.05).