Utility of neck dissections in squamous cell carcinomas of the temporal bone

Presentation: P769
Topic: Skin Cancers
Type: Poster
Authors: Benjamin R Johnson, MD; Claudia I Cabrera Aviles, MD; Armine Kocharyan, MD; Nauman Manzoor, MD; Akina Tamaki, MD
Institution(s): University Hospitals Cleveland Medical Center / Case Western Reserve University


There is a lack of a standardized treatment algorithm for the management of the neck in squamous cell carcinoma (SCC) involving the temporal bone. There is significant heterogeneity of disease given that primary malignancy can arise from multiple adjacent epithelial subsites including the external auditory canal (EAC), auricle, and periauricular skin. Selective neck dissection in patients undergoing temporal bone resection for SCC varies based on institution given the limited data available regarding risk for occult cervical lymph node metastasis (CLNM); occult disease defined as having no abnormal lymph nodes based on preoperative exam or imaging.


 Determine reliability of preoperative clinical exam and imaging in detecting nodal disease in patients with SCC of the temporal bone. Identify the overall rate of nodal disease and occult nodal disease in SCC of the temporal bone in addition to its impact on overall survival.  


Patients who underwent a lateral temporal bone resection or subtotal temporal bone resection for a diagnosis of SCC involving the EAC, auricle, or periauricular skin from January 2008 to January 2010 were reviewed retrospectively. Preoperatively, patients noted to have cervical lymph node involvement based on clinical exam or imaging were identified. A portion of the patients underwent selective neck dissection, including all patients with clinical or radiographic concern for lymph node involvement. Analysis on whether either of these two variables predicted CLNM when compared to operative pathology was performed.  The rate of overall and occult lymph node metastasis was assessed. A Cox regression analysis was also performed comparing overall survival for patients with CLNM versus those without CLNM.


31 patients underwent temporal bone resection for a diagnosis of SCC and 28 patients underwent concomitant selective neck dissection. 2 patients underwent previous selective neck dissections and SCC resection with subsequent local recurrence. Neither had pathologic CLNM. 1 patient did not undergo a neck dissection but had no evidence of recurrence at 3 years of follow up. The overall rate of pathologic CLNM was 5/28 (17.9%) and the rate of occult CLNM, was 2/24 (8.33%). The odd ratios of a patient having pathologic CLNM with abnormal preoperative imaging was 14.2 (p=0.038). There was found to be no statistically significant correlation between a preoperative clinical exam concerning for abnormal cervical lymph nodes and rate of CLNM based on pathology (p>0.9). Overall survival analysis revealed that patients with CLNM had a death hazard ratio of 5.23 (p<0.001, CI 1.94-14.1) when compared to patients without CLNM. Additionally, mean survival time in days was 555.44 for patients with CLNM versus 1187.86 without CLNM (p=0.043).


The presence of cervical lymph node metastasis in patients with SCC of the temporal bone is a poor prognostic factor with a negative impact on overall survival, however, the occult rate of cervical lymph node metastasis is under 10%. This suggests that in the absence of radiographic suspicion for cervical lymph node metastasis selective neck dissection may not be warranted.