Cholesteatoma Recidivism in Pediatric Versus Adult Patients: Disease Extent and Prognostic Factors

OBJECTIVE: To determine whether pediatric age independently predicts cholesteatoma recidivism beyond anatomical disease extent using the EAONO/JOS and SAMEO-ATO classifications. METHODS: A retrospective cohort study of 128 primary cholesteatoma surgeries (128 ears in 126 patients; median follow-up, 5.8 years) was conducted. Etiology, stage, topographical extension (STAM), and surgical category (SAMEO-ATO) were recorded. A data-driven threshold (< 20 years) was identified via maximally selected rank statistics. Predictors were determined by Group Lasso selection and Firth's penalized Cox regression. RESULTS: Recidivism was identified in 21 of 43 pediatric and 16 of 85 adult ears. The 5-year cumulative risk was 53.9% (95% CI, 34.2%-67.8%) in children versus 19.1% (95% CI, 9.5%-27.7%) in adults (log-rank p < 0.001), driven by early residual disease (p = 0.004) rather than true recurrence (p = 0.058). The sinus tympani (S2) was a site of recidivism in 38.5% of pediatric versus 5.6% of adult cases (p = 0.016). Multivariable analysis identified four independent predictors: pediatric age (hazard ratio [HR] 2.24; 95% CI, 1.09-4.63), supratubal recess involvement (HR 3.41; 95% CI, 1.56-7.45), mastoid extension (HR 2.43; 95% CI, 1.01-5.84), and absence of mastoidectomy relative to canal wall-down technique (HR 8.67; 95% CI, 1.95-38.41). CONCLUSION: Pediatric age, supratubal recess involvement, mastoid extension, and mastoidectomy omission relative to canal wall-down technique independently predict cholesteatoma recidivism. The predominance of residual over recurrent disease, with a median detection interval of approximately 14 months, supports active 12-18 month postoperative surveillance and long-term clinical follow-up on a risk-stratified basis.