Preoperative diagnosis of atypical pelvic leiomyoma and sarcoma: the potential role of diffusion-weighted imaging

摘要

The objective of our study was to determine the utility of diffusion-weighted magnetic resonance (DWMR) to differentiate the atypical uterine leiomyomas and sarcomas, establishing a cut-off value of the apparent diffusion coefficient (ADC) to rule out the malignancy. We performed a diagnostic accuracy retrospective study including 10 patients with pelvic sarcomas and 17 patients with leiomyomas. Atypical morphological features in magnetic resonance (MR) studies occurred in 58.8% of the patients, leading to a significant number of indeterminate diagnoses. In contrast, ADC values were consistent for leiomyomas, sarcomas, primary tumours, recurrences, intrauterine and in the extrauterine pelvic locations. The ADC cut-off value was set in 1 (x10(-3) mm(2)/s). Thus, the ADC values equal or superior to 1 x 10(-3) mm(2)/s were always associated with a leiomyoma. The structural MR accuracy was 66.7%, reaching 100% when using DWMR with dichotomised ADC values. Diffusion-weighted imaging with the quantitative measurement of ADC may be considered a useful preoperative test for the differentiation of atypical leiomyomas from sarcomas. Impact statement What is already known on this subject? Papers reporting the utility of a diffusion-weighted MR for the diagnosis of uterine sarcomas are scarce and consist of a small series. However, the published results are consistent with our study, with the decreased ADCs in the case of malignancy. What do the results of this study add? The main differential characteristic of our study is that we selected only the atypical leiomyomas: they share sonographic and MR features with sarcomas, which often leads to an inaccurate diagnosis. This is also the first paper reporting on the role of DWMR with ADC for these types of tumours in extrauterine pelvic locations. We demonstrated a consistent relationship between dichotomised ADC values in leiomyomas/sarcomas for these particular cases and in recurrent tumours, with no overlap between both the groups, as a difference with the previous reports.