Learning Objectives: How to report on term results of tympanoplasty and mastoidectomy.
Introduction: (1) Exact etiology of middle ear cholesteatoma remains unknown and its recurrence is unavoidable during the long-term follow up. We showed recurrence rate using Kaplan-Meier analysis because follow-up patients decreased with the time. (2) We analyzed the long-term outcomes of perforated COM using multivariate analysis to examine the prognostic factors and to determine whether mastoidectomy is useful for tympanoplasty in patients with perforated COM.
Subjects: (1) Between 1987 and 2002, 345 patients with cholesteatoma were operated on by the same surgeon. They were 140 attic cholesteatomas (40.6%) and 90 pars tensa cholesteatoma, and 115 other types (33.3%). Canal wall down tympanoplasty (CWDT) was performed in 113 patients (32.8%), canal wall reconstruction (CWR) after CWDT in 70 patients (20.3%) and intact canal wall up tympanoplasty (ICWT) in 162 patients (47.0%). (2) Between 1989 and 2002, 213 patients with perforated COM underwent tympanoplasty with mastoidectomy (34 ears, 16.0%) and without mastoidectomy (179 ears, 84.0%), and were followed up for more than 5 years.
Results: (1) The mean follow-up period was 6.3 years. Using the standard calculation method, the 5-year recurrence rate in patients with CWDT and with ICWT/CWR were 3.5% and 12.1%, respectively. Using Kaplan-Meier analysis, they were 3.9 and 16.7%, respectively. (2) Successful hearing outcomes (A-B gap: 20 dB or smaller) was 174/213 (81.7%). Using multivariate logistic regression analysis, normal ossicular chain was the only factor to long-term successful hearing outcomes. Graft success rate was 204/213 (95.8%). There were no significant predictors of long-term successful graft outcomes.
Conclusion: (1) Because the follow-up rate decreased with year, Kaplan-Meier analysis shows more correct recurrence rate than the standard calculation method. (2) Mastoidectomy was not a significant factor predicting long-term outcomes.