Comparison of 16-slice MSCT and MRI in the assessment of squamous cell carcinoma of the oral cavity

The goal of this retrospective study was to assess the accuracy of 16-slice multislice CT (MSCT) and MRI in staging of patients with primary squamous cell carcinoma (SCC) of the oral cavity. Fifty-two patients with histologically proven primary SCC were examined with contrast enhanced MSCT and MRI at 1.5 T with a combined head and surface neck coil. Image modalities were evaluated in a blinded fashion by two radiologists and an oral-maxillofacial surgeon in consensus concerning tumour depiction, local tumour infiltration and cervical lymph node metastases. Results of the radiological assessment were correlated with the intraoperative and histopathological findings in all patients. 36 of 52 primary tumours (69.2%) were depicted by MSCT while 44 were localized by MRI (84.6%). Regarding muscle infiltration MRI versus MSCT had a sensitivity of 81.8% versus 72.7%, but a low specificity and an accuracy of 63.4% versus 61% and 67.3% versus 63.5%, respectively, were found. There was a trend towards a better detection of bony infiltration by MRI than MSCT with a sensitivity of 100% versus 71.4%, a specificity of 93.3% versus 95.5% and an accuracy 94.2% versus 92.3%, respectively. Detection of cervical lymph node involvement was similar for MRI and MSCT with a sensitivity of 84.2% and 78.9%, a specificity of 63.6% and 75.7% and an accuracy of 71.1% and 76.9%, respectively. For N-staging both methods failed to detect small metastasis. For T-staging MRI was superior to MSCT, because there was a tendency to underestimate the tumour size by MSCT more often (19.4% versus 6.8% by MRI). Therefore, pre-operative MRI is recommended as the basic imaging modality of choice for treatment planning of oral SCC. MSCT is a valid alternative imaging method especially in cases with low patient compliance. (c) 2005 Elsevier Ireland Ltd. All rights reserved.