[Magnetic resonance in the evaluation of Mullerian duct anomalies]. / La risonanza magnetica nella valutazione delle anomalie dei dotti di Müller.

Müllerian duct alterations in development or fusion in the embryo cause congenital uterine anomalies which may be responsible for decreased fertility or problems in carrying out a normal pregnancy. In this study, the MR findings in uterine agenesis (1 case), unicornuate (2 cases), didelphys (3 cases), bicornuate (3 cases), arcuate (6 cases) and septate uterus (8 cases) are described, together with the optimal section planes for their demonstration. The examinations were performed with an 0.5-T superconductive magnet, the spin-echo technique and mostly T2-weighted sequences. The anomalies were grouped according to Buttram and Gibbons classification, which is the most used in clinics. In particular, the bicornuate uterus was distinguished from the septate uterus, the latter associated with the highest spontaneous abortion rates, on the basis of external fundal outline appearance. In such anomalies, the muscular or fibrotic nature of any intracavitary septum was assessed based on septal thickness more than on signal intensity at this level. MR diagnostic accuracy in 23 patients with Müllerian anomalies, compared with surgical, hysteroscopic, laparotomic and laparoscopic findings, was 100%. Nevertheless, if Müllerian duct anomalies responsible for gynecologic-obstetric problems are known or suspected, MRI should always be used, on the basis of a close gynecologist-radiologist collaboration, for classification agreement and the evaluation of any intracavitary septum morpho-biometric appearance and possibly nature, to discuss treatment options.

[Carcinoma of the uterine cervical canal. Staging and biometric assessment with magnetic resonance]. / Il carcinoma del canale cervicale dell'utero. Stadiazione e valutazione biometrica con risonanza magnetica.

In uterine cervical canal carcinoma, the current clinical FIGO criteria often fail not only to differentiate stage IA2 from stage IB but also to demonstrate possible parametrial involvement. Moreover, the analysis of tumor volume and of the depth of neoplastic stromal invasion is not very reliable. The authors investigated MR accuracy in the definition of such variables: to this purpose, 24 patients with histologically confirmed endocervical adenocarcinoma were submitted to MRI, which was performed with an 0.5-T superconductive magnet. Sagittal and oblique transverse or sometimes coronal SE images, oriented so as to be perpendicular to longitudinal cervical major axis were obtained with T2 weighting (TR 1800 ms, TE 25-90 ms). MR data were correlated with pathologic findings. MR accuracy in demonstrating parametrial involvement was 92%, its sensitivity was 86% and specificity 97%. Volumetric MR data showed a high correlation (r = 0.970) with those derived from pathologic findings. In 92% of cases stromal invasion exceeded 5 mm. MRI, thanks to its high accuracy, should be included in diagnostic pretreatment protocols, even though FIGO criteria do not require it yet, especially in the presence of an endocervical adenocarcinoma. Moreover, the accurate definition of tumor volume can allow less extensive surgery with the same survival rates and fewer complications, which are frequently observed after radical hysterectomy.

[Magnetic resonance in the evaluation of parametrial involvement in carcinoma of the cervix uteri]. / La risonanza magnetica nella valutazione del coinvolgimento parametriale nel carcinoma della cervice uterina.

Cervical carcinoma is one of the most frequent gynecologic malignancies. Its prognosis depends on both tumor volume at diagnosis and its stage. Staging accuracy is important not only for prognosis but also for optimal treatment planning. According to FIGO criteria, carcinomas without parametrial involvement (stage I and limited stage IIA disease) can be surgically treated. For more advanced stages, treatment, in most cases, consists of radiation therapy or chemotherapy alone. The authors evaluated MR accuracy in the diagnosis of parametrial involvement; to this purpose, 32 patients with histologically proven lesions were referred for MR imaging, which was performed with a 0.5 T superconductive magnet. Transverse and sagittal SE images were obtained with T2 weighting (TR 1800 ms, TE 30-100 ms); transverse and sometimes sagittal images were obtained with T1 weighting (TR 450/300 ms, TE 20/30). T1-weighted images distinguished neoplasm from cervical stroma or dense parametrial connective tissue in 40% of cases only. T2-weighted images, instead, demonstrated the difference in all cases, showing tumor as a hyperintense area in 90% of patients. Neoplastic involvement of pericervical connective tissue was diagnosed, with those sequences, on the basis of focal disruptions of the outer hypointense fibrous cervical stroma; findings were correlated with those from a previous clinical staging and in 26/32 patients with pathologic findings. MR accuracy in demonstrating parametrial involvement was 88%, sensitivity was 77% and specificity was 94%. Clinical staging accuracy in the evaluation of this parameter was 66%. In 6 cases with no surgical findings, MR confirmed extensive parametrial and vesical or rectal neoplastic involvement, as diagnosed at clinics. MR imaging, thanks to its multiplanar and multiparametric imaging capabilities is a very reliable technique in the preoperative staging of cervical carcinoma. Moreover, since clinical staging can sometimes underestimate pericervical connective spread, the higher accuracy of MR imaging can help avoid useless interventional procedures.