Introduction/Background The objective was to evaluate the feasibility and the performance of MR and ultrasound fusion imaging (MUFI) for characterizing adnexal masses.
Methodology A prospective monocentric study was conducted in the gynaecologic oncologic department of Creteil hospital. Between 01st January 2014 and 31 July 2018, we ‘consecutively’ included all women referred for characterization of indeterminate or suspicious pelvic mass who accepted to be enrolled in our study (n=109). We excluded all patients who did not undergo ultrasound or MR imaging (n=3). Feasibility was evaluated on this first population. A subgroup of patients who underwent surgery was identified to evaluate MIFU performance (n=26). Clinical and radiological features were reported in our database. Two independent readers retrospectively and blindly rated ultrasonography according to IOTA simples rules and MR imaging according to ADNEXMR score. Reference standard was final pathology at surgery (n=106). Sensitivity, specificity, positive and negative predictive values and likelihood ratio of each imaging were calculated for predicting malignant ovarian masses.
Results Final population included 106 patients (mean age (SD)=57 (18) years old). Mean time duration between first consultation and MUFI was 15 days (7). The subgroup of patients who underwent surgery consisted in 12 patients with primitive ovarian cancer, 2 with ovarian metastases, 3 patients with borderline and 9 with benign tumours. Sensitivity and specificity were respectively of 94% and 44% for ultrasound, 94% and 89% for MR imaging and 100% and 89% for MUFI for the diagnosis of cancer. In MUFI, high colour score, number of vegetations, carcinosis were the most pertinent malignant features whereas the absence of vascularisation was the most pertinent benign feature.
Conclusion MUFI is feasible and reproducible. Moreover, MUFI may help improve diagnostic performances of the combination of simple ultrasound and MR imaging. Further research is needed to confirm our results and define proper indications.
Disclosure Nothing to disclose.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.