Temporal Bone Malignancies and Modifications of Surgical Methods

Aim: To evaluate surgical outcome in temporal bone cancers and analyze the change in surgical practices.

Materials and Methods: This is a study of surgeries performed for temporal bone cancers in a tertiary cancer center, which was conducted retrospectively over ∼13 years, from July 1994–August 2002, and prospectively over ∼6 years, from September 2002–July 2008. Sixty-two cases were analyzed, 40 men and 22 women, with a median age of 49.5 years (range, 18–75years). Nearly 45% of patients underwent a salvage procedure: prior surgery in 20 cases and prior radiation or chemotherapy and radiation in 8 cases. Subtotal temporal bone resection (STBR) was performed in 48.5%, lateral temporal bone resection (LTBR) in 44.5%, sleeve resection and total temporal bone resection (TTBR) in 3%. In 18% of the STBR cases, the procedure was modified by initial LTBR followed by drill out of involved areas. Facial nerves were preserved in patients with operable disease extending to the hypotympanum and stylomastoid foramen with intact facial nerve or anterior soft tissue extension, where the cuts of en bloc resection were within diseased areas. The endpoints of recurrence (LR) and overall survival (OS) were studied using Cox regression and Kaplan-Meier survival analyses.

Results: The stages at presentation were: stage I in 13%, stage II in 17.5%, stage III in 14.5%, and stage IV in 55%(modified Pittsburgh staging). The most common subsite involved was the external auditory canal with extension to the middle ear in 52%. Squamous cell carcinoma accounted for 84% of cases. The estimated 2-year disease-free survival and overall survival were 77% and 73.5%, respectively. Recurrences occurred in 18 cases, with 15 local and locoregional and 3 distant failures. Increased disease recurrence was found with petrous apex involvement (P = 0.03), extent of surgery—lateral vs. subtotal vs. total temporal bone resection—(P = 0.01), and presence of perineural invasion (P = 0.02). On multivariate analysis only perineural invasion was significant (P = 0.02). Factors influencing OS were petrous apex involvement (P = 0.04), soft tissue involvement (P = 0.01), and prior treatment (P = 0.01). Modification of the surgical technique—modified STBR vs. STBR—did not worsen LR or OS (73 and 82% vs. 42 and 58%, respectively)(P = 0.40). Presence of parotid involvement strongly correlated with nodal involvement on pathology (P = 0.00).

Factors like facial nerve involvement, margins, nodal involvement, and perineural invasion showed an association in OS on univariate analysis but did not reach statistical significance due to the small number of cases.

Conclusion: Cancer of the temporal bone is a locoregional disease with few distant failures. Modifications of described surgical techniques may be used in some cases to preserve function without compromising oncologic safety. Perineural invasion is associated with poor prognosis, and parotid involvement merits nodal dissection.