RESUMEN
OBJECTIVE: To investigate the clinical relevance of the classification systems used for temporal bone osteoradionecrosis (ORN) and to define a treatment protocol for temporal bone ORN. STUDY DESIGN: Retrospective case series. SETTING: Amsterdam, department of otorhinolaryngology and head and neck surgery. SUBJECTS AND METHODS: Classification of temporal bone ORN was performed through use of clinical data and radiologic imaging. Outcomes of conservative and surgical treatment were investigated and compared for different grades of ORN. RESULTS: Of the 49 ears included in this study, 35 were primarily treated conservatively. At start of conservative treatment, 23 were classified as a localized and 8 as a diffuse form of ORN; 4 could not be classified. There was a significant difference in clinical outcome between the localized and diffuse forms of ORN (χ(2) = 5.862, P = .015), and mastoid air cell destruction on preoperative computed tomography scan was found to be a significant predictor for a negative outcome of conservative treatment (χ(2) = 4.34, P = .037). Fourteen ears with diffuse ORN were primarily treated surgically, and 11 were secondarily treated surgically following a period of conservative treatment. Twenty-two patients were treated with subtotal petrosectomy, of which 20 were cured. Three patients were treated with canal wall down mastoidectomy, and 2 had recurrence of disease. CONCLUSION: Ramsden's classification system is clinically relevant in predicting conservative treatment outcomes. Mastoid air cell destruction on computed tomography differentiates between the localized and diffuse forms of ORN. Given our results and experience with treating temporal bone ORN, we propose a treatment protocol.