Radiological detection of extracapsular spread in head and neck squamous cell carcinoma (HNSCC) cervical metastases

Extracapsular spread of cervical lymph nodes deteriorates the prognosis of patients with head and neck squamous cell carcinoma. Postoperative radiochemotherapy is superior to postoperative radiotherapy alone in patients with histologically proven extracapsular spread. If extracapsular spread can be detected preoperatively, patients may favor primary radiochemotherapy instead of primary surgery plus postoperative radiochemotherapy.

Computed tomography (CT) scans of nodal positive head and neck squamous cell carcinoma patients treated between 2008 and 2010 with comprehensive neck dissection as part of first line surgical treatment were retrospectively scanned for extracapsular spread by two blinded radiologists. If a positive lymph node was identified by the pathologist, CT scans were assessed for extracapsular spread retrospectively. CT criteria for Extracapsular spread were apparent fat and soft tissue infiltration or infiltration of sternocleidomastoid muscle, internal jugular vein or carotid artery. Radiologic judgment was compared with histological evidence of extracapsular spread and specificity and sensitivity of CT detection was calculated.

Forty-nine patients with histologically proven positive lymph nodes (pN+) were included. Extracapsular spread was histologically proven in 17 cases; the number of all affected lymph nodes was not listed. Radiologist 1 found extracapsular spread in CT scans of 15/49 patients and radiologist 2 in 16/49 patients (Cohen's kappa=0.86; p<0.01). Sensitivity of radiologic extracapsular spread detection was 73% (95% confidential index (CI): 44.0-89.7%) and specificity 91% (75.0-98.0%).

Extracapsular spread depicted on computed tomography using strict criteria has high specificity.