OC07.02: A trial of preoperative diagnosis of uterine sarcoma using tumor signal intensity at magnetic resonance imaging (MRI)

Abstract

Uterine sarcoma is a rare and poor prognostic disease and its preoperative diagnosis from uterine tumor is difficult, since there are many cases of myoma uteri. We focused on tumor signal intensity at MRI T2 weighted imaging for the preoperative diagnosis of uterine sarcoma. The purpose of this study is to determine the tumor signal intensity as an accurate preoperative evaluating method for uterine sarcoma. MRI was performed in 16 cases of uterine sarcoma in the last 5 years at our hospitals (sarcoma group). In contrast, 22 cases of leiomyoma received MRI in 2013 (myoma group). We selected uterine muscle, urinary bladder, subcutaneous fat and other leiomyoma of uterus as control structure. We calculated tumor to other structures signal intensity ratio (SIR) and compared between two groups. In the structure that has significant difference, the cutoff value to predict sarcoma was calculated. The sarcoma group had significantly higher tumor to mural layer SIR when compared with the leiomyoma group (1.95 ± 0.634 vs. 0.879 ± 0.592, p < 0.01). The same results were confirmed on tumor to urinary bladder SIR (0.720 ± 0.181 vs. 0.399 ± 0.329, p < 0.01) and tumor to fat SIR (0.810 ± 0.205 vs. 0.456 ± 0.336, p < 0.01). Except for other leiomyoma of uterus, every SIR has significant difference between the two groups. The best cutoff value to diagnose sarcoma is over 1.37 in tumor to uterine muscle SIR (sensitivity: 84%, specificity: 93%, area under the curve (AUC): 0.902), 0.451 in tumor to urinary bladder SIR (sensitivity: 77%, specificity: 93%, AUC 0.851) and 0.512 in tumor to fat SIR (sensitivity: 75%, specificity: 100%, AUC 0.833). It is suggested that preoperative diagnosis of uterine sarcoma is possible by evaluation of its signal intensity on T2-weighted images at MRI.