Objective: To determine how classification of petrous bone cholesteatomas (PBCs) using the 5-point Sanna classification can predict major structural involvement, facial nerve outcomes, hearing outcomes, postoperative complications, and disease recurrence.
Study design: Retrospective case series.
Setting: Tertiary referral center in Bergamo, Italy.
Patients: Eighty-one sequential patients with radiologic and surgically confirmed diagnoses of PBC treated at a single tertiary referral center during a 20-year period.
Main outcome measures: Major structural involvement, facial nerve outcomes, hearing outcomes, postoperative cerebrospinal fluid leak, and disease recurrence were evaluated on the basis of Sanna classification.
Results: Using the Sanna classification, 70% (57) were supralabyrinthine, 12% (10) infralabyrinthine, 7% (6) infralabyrinthine-apical, 5% (4) apical, and 5% (4) massive. Massive classification was statistically significantly associated with cochlear involvement (p = 0.009) and internal auditory canal involvement (p = 0.02). The infralabyrinthine-apical class was associated with carotid canal involvement (p = 0.03). Facial nerve interruption was observed in 35% of patients and most frequently in the apical group (75%). Neither hearing nor facial nerve outcomes were associated with Sanna classification. House-Brackmann score improved or was maintained postoperatively in 89% of patients.
Conclusion: The Sanna classification provides anatomic detail on location of PBCs and is predictive of IAC, cochlear, and carotid artery involvement. However, classification systems for this rare condition continue to pose a challenge in being able to accurately predict facial nerve and hearing outcomes in surgical obliteration of PBC.