Nobuo Kashiwagi1, Takamichi Murakami2, Katsuyuki Nakanishi3, Osamu Maenishi4, Kaoru Okajima5, Hiroto Takahashi2, Mio Sakai3, Yasuhiko Tomita6, Kyoichi Terao7, Noriyuki Tomiyama8. 1. Department of Radiology, Kinki University School of Medicine, Osaka kashiwaginobuo@yahoo.co.jp. 2. Department of Radiology, Kinki University School of Medicine, Osaka. 3. Department of Diagnostic Radiology, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka. 4. Department of Pathology, Kinki University School of Medicine, Osaka. 5. Department of Radiology, Nara Hospital, Kinki University Faculty of Medicine, Osaka. 6. Department of Pathology, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka. 7. Department of Otolaryngology, Kinki University School of Medicine, Osaka. 8. Department of Radiology, Osaka University Graduate School of Medicine, Osaka, Japan.
Abstract
BACKGROUND: Although pleomorphic adenomas account for over 90% of all benign submandibular gland tumors, the imaging features of submandibular pleomorphic adenomas have not been reported in a large number of cases. PURPOSE: To assess the conventional magnetic resonance imaging (MRI) findings for predicting the submandibular pleomorphic adenoma. MATERIAL AND METHODS: MR studies of 42 pleomorphic adenomas and 28 other types of tumor were reviewed. MR images were assessed for the presence of hyperintense areas on T2-weighted images (first sign), a well-defined margin (second sign), and presence of crescent-shaped compression of the ipsilateral normal submandibular gland (third sign). RESULTS: For identifying submandibular pleomorphic adenoma, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 97.6%, 50.0%, 74.5%, 93.3%, and 78.6% for the first sign, 95.2%, 46.4% 72.7%, 86.7%, and 75.7% for the second sign, and 23.8%, 100%, 86.7%, 46.7%, and 54.3% for the third sign, respectively. Combining the first and second findings achieved to 85.7% specificity and 90.9% accuracy. CONCLUSION: Although non-specific, submandibular pleomorphic adenomas usually have hyperintense areas on T2-weighted images and well-defined margins. In addition, crescent-shaped compression of the ipsilateral normal gland seems to represent a highly specific sign.
BACKGROUND: Although pleomorphic adenomas account for over 90% of all benign submandibular gland tumors, the imaging features of submandibular pleomorphic adenomas have not been reported in a large number of cases. PURPOSE: To assess the conventional magnetic resonance imaging (MRI) findings for predicting the submandibular pleomorphic adenoma. MATERIAL AND METHODS: MR studies of 42 pleomorphic adenomas and 28 other types of tumor were reviewed. MR images were assessed for the presence of hyperintense areas on T2-weighted images (first sign), a well-defined margin (second sign), and presence of crescent-shaped compression of the ipsilateral normal submandibular gland (third sign). RESULTS: For identifying submandibular pleomorphic adenoma, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 97.6%, 50.0%, 74.5%, 93.3%, and 78.6% for the first sign, 95.2%, 46.4% 72.7%, 86.7%, and 75.7% for the second sign, and 23.8%, 100%, 86.7%, 46.7%, and 54.3% for the third sign, respectively. Combining the first and second findings achieved to 85.7% specificity and 90.9% accuracy. CONCLUSION: Although non-specific, submandibular pleomorphic adenomas usually have hyperintense areas on T2-weighted images and well-defined margins. In addition, crescent-shaped compression of the ipsilateral normal gland seems to represent a highly specific sign.