Abstract: Objective To investigate the learning curve and short-term clinical efficacy of otoendoscopic ossicular prosthesis reconstruction in patients with difficult conductive hearing loss. Methods The clinical data of 52 patients(52 ears)with difficult conductive hearing loss who underwent otoendoscopic ossicular prosthesis reconstruction were retrospectively analyzed. The patients included 15 males and 37 females, with a mean age of 44.7 years(range, 18-72 years); 27 patients had left-ear involvement and 25 had right-ear involvement. All procedures were performed by the same surgeon. The preoperative mean air-conduction(AC)threshold was(56.31±18.12)dB HL, the mean bone-conduction(BC)threshold was(23.23±13.59)dB HL, and the mean air-bone gap(ABG)was(33.08±9.09)dB HL. The learning curve for ossicular prosthesis reconstruction time was analyzed using the cumulative sum(CUSUM)method. A cubic polynomial model was used for curve fitting(R2=0.912, P<0.001), and the inflection point of the learning curve was identified at the 25th case. The patients were subsequently divided into a learning group(n=25)and a proficiency group(n=27). Hearing was reassessed 3 months after surgery to evaluate the overall clinical efficacy and compare outcomes between the two groups. Results All procedures were completed entirely under otoendoscopic visualization. No cases of postoperative facial palsy, cerebrospinal fluid otorrhea, persistent vertigo, infection, or total deafness occurred. Postoperative complications included taste disturbance in 3 cases, tympanic membrane graft retraction in 1 case, collapse of the cartilage graft flap of the external auditory canal in 2 cases, transient tinnitus in 2 cases, and exposed bone of the external auditory canal in 1 case. The mean BC threshold increased by 18 dB HL and 10 dB HL in 2 cases, respectively. There were no statistically significant differences between the learning and proficiency groups in sex, age, affected side, or the type of ossicular prosthesis used(all P>0.05). In all patients, the postoperative mean AC threshold was(45.04±18.75)dB HL, significantly lower than the preoperative value of(56.31±18.12)dB HL(t=6.436, P<0.001). The postoperative mean ABG was(22.63±9.08)dB HL, significantly smaller than the preoperative value of(33.08±9.09)dB HL(t=6.748, P<0.001). The postoperative mean BC threshold was(22.23±13.85)dB HL, with no statistically significant difference from the preoperative value of(23.23±13.59)dB HL(t=1.244, P=0.219). The learning curve for ossicular prosthesis reconstruction time reached its peak at the 25th case. The reconstruction time was significantly shorter in the proficiency group than in the learning group ［(13.85±4.03)min vs.(25.44±3.78)min, P<0.001］. No statistically significant differences were observed between the two groups in postoperative audiological outcomes(all P>0.05). Conclusion Otoendoscopic ossicular prosthesis reconstruction is a safe and effective treatment for difficult conductive hearing loss caused by complex middle ear lesions in the short term. For surgeons with a certain level of experience in microscopic otologic surgery, a relatively proficient stage may be reached after approximately 25 cases, as reflected primarily by a reduction in reconstruction time without a decline in short-term hearing outcomes.

Key words: Otoendoscopy, Difficult conductive hearing loss, Ossicular prosthesis reconstruction, Learning curve