Ryo Yajima1, Aki Kido2, Ryo Kuwahara3, Yusaku Moribata1, Yoshitsugu Chigusa4, Yuki Himoto1, Yasuhisa Kurata1, Yuka Matsumoto1, Satoshi Otani1, Naoko Nishio1, Sachiko Minamiguchi5, Masaki Mandai4, Yuji Nakamoto1. 1. Department of Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate School of Medicine, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto, 606-8507, Japan. 2. Department of Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate School of Medicine, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto, 606-8507, Japan. akikido@kuhp.kyoto-u.ac.jp. 3. Department of Radiology, National Hospital Organization Kyoto Medical Center, 1-1 Mukaihatacho, Fushimi-ku, Kyoto, 612-8555, Japan. 4. Department of Gynecology and Obstetrics, Kyoto University Graduate School of Medicine, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto, 606-8507, Japan. 5. Department of Diagnostic Pathology, Kyoto University Graduate School of Medicine, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto, 606-8507, Japan.
Abstract
PURPOSE: To evaluate the diagnostic performance of MRI findings for differentiating uterine leiomyoma with intraligamentous growth, or broad ligament fibroid, from subserosal leiomyoma. METHODS: This study included 37 patients with surgically confirmed uterine smooth muscle tumors (36 leiomyomas and one smooth muscle tumor of uncertain malignant potential) with intraligamentous growth (IL) and size-matched control of 37 patients with subserosal leiomyoma (SS). Two radiologists independently evaluated eight preoperative MRI findings: tumor shape, degeneration, attachment to uterus, ovary elevation, ureter displacement, bladder deformation, rectal displacement, and separation of round ligament (RL) and uterine artery (UA). The diagnostic values of these findings and interobserver agreement were assessed. Receiver-operating characteristic (ROC) analysis of the number of positive MRI findings for diagnosing IL was performed. Clinical outcomes including surgical method, operation time, intraoperative blood loss, perioperative complications, and postoperative hospital stay of the two groups were compared. RESULTS: Significant differences in tumor shape, attachment to uterus, ovary elevation, ureter displacement, and separation of RL and UA were found between IL and SS. Four of these findings, excluding ureter displacement, showed moderate to substantial interobserver agreement. When two or more of these four findings were positive, sensitivity, specificity, and area under the ROC curve were 91%, 77%, 0.90 in reader 1 and 82%, 89%, 0.91 in reader 2. The operation time was significantly longer for IL than for SS. CONCLUSION: Tumor shape, attachment to uterus, ovary elevation, and separation of RL and UA are useful MRI findings for differentiating intraligamentous leiomyoma from subserosal leiomyoma.
PURPOSE: To evaluate the diagnostic performance of MRI findings for differentiating uterine leiomyoma with intraligamentous growth, or broad ligament fibroid, from subserosal leiomyoma. METHODS: This study included 37 patients with surgically confirmed uterine smooth muscle tumors (36 leiomyomas and one smooth muscle tumor of uncertain malignant potential) with intraligamentous growth (IL) and size-matched control of 37 patients with subserosal leiomyoma (SS). Two radiologists independently evaluated eight preoperative MRI findings: tumor shape, degeneration, attachment to uterus, ovary elevation, ureter displacement, bladder deformation, rectal displacement, and separation of round ligament (RL) and uterine artery (UA). The diagnostic values of these findings and interobserver agreement were assessed. Receiver-operating characteristic (ROC) analysis of the number of positive MRI findings for diagnosing IL was performed. Clinical outcomes including surgical method, operation time, intraoperative blood loss, perioperative complications, and postoperative hospital stay of the two groups were compared. RESULTS: Significant differences in tumor shape, attachment to uterus, ovary elevation, ureter displacement, and separation of RL and UA were found between IL and SS. Four of these findings, excluding ureter displacement, showed moderate to substantial interobserver agreement. When two or more of these four findings were positive, sensitivity, specificity, and area under the ROC curve were 91%, 77%, 0.90 in reader 1 and 82%, 89%, 0.91 in reader 2. The operation time was significantly longer for IL than for SS. CONCLUSION:Tumor shape, attachment to uterus, ovary elevation, and separation of RL and UA are useful MRI findings for differentiating intraligamentous leiomyoma from subserosal leiomyoma.