PURPOSE: To define MR accuracy in the evaluation of the T stages of tumors of the tongue and floor of the mouth. METHODS: Fifty-two patients affected by squamous cell carcinoma were studied with a superconductive system at 1.5 T. The study was performed with spin-echo T1- and T2-weighted images before contrast and short spin-echo T1-weighted and gradient-echo sequences after gadopentetate dimeglumine administration. The results obtained with MR were compared with pathologic data. RESULTS: Good correlation of T stages (TNM system) was obtained in 45 of 52 cases. MR did not show four superficial lesions. In one case, infiltration of the cortical bone of the mandible was not demonstrated (MR, T2; pathologic, T4), and in another the lesion was classified as T2 instead of T3, as it was pathologically. One lesion was classified as T4 on MR because of infiltration of the alveolar ridge but was classified as T2 at surgery. In 46 patients who underwent surgery, the accuracy of MR was excellent for predicting the relationship of tumor to midline and oral floor musculature. The results obtained with gadopentetate dimeglumine were better than those obtained in noncontrast studies in 32 (62%) of 53 cases. CONCLUSION: MR showed high accuracy in the study of tumors of the tongue and floor of the mouth.
PURPOSE: To define MR accuracy in the evaluation of the T stages of tumors of the tongue and floor of the mouth. METHODS: Fifty-two patients affected by squamous cell carcinoma were studied with a superconductive system at 1.5 T. The study was performed with spin-echo T1- and T2-weighted images before contrast and short spin-echo T1-weighted and gradient-echo sequences after gadopentetate dimeglumine administration. The results obtained with MR were compared with pathologic data. RESULTS: Good correlation of T stages (TNM system) was obtained in 45 of 52 cases. MR did not show four superficial lesions. In one case, infiltration of the cortical bone of the mandible was not demonstrated (MR, T2; pathologic, T4), and in another the lesion was classified as T2 instead of T3, as it was pathologically. One lesion was classified as T4 on MR because of infiltration of the alveolar ridge but was classified as T2 at surgery. In 46 patients who underwent surgery, the accuracy of MR was excellent for predicting the relationship of tumor to midline and oral floor musculature. The results obtained with gadopentetate dimeglumine were better than those obtained in noncontrast studies in 32 (62%) of 53 cases. CONCLUSION: MR showed high accuracy in the study of tumors of the tongue and floor of the mouth.