THE USE OF MAGNETIC-RESONANCE-IMAGING IN THE ASSESSMENT OF MALIGNANT VAGINAL FISTULAS

The objective of this study was to determine whether magnetic resonanc e imaging (MRI) could reliably demonstrate fistulas and any associated mass and to see whether these findings were beneficial in the managem ent of the fistula. Twelve consecutive patients presenting with suspec ted vaginal fistulas were examined prospectively with MRI, using a com bination of sequences, for the presence, extent and configuration of f istulas and any associated mass. Comparison was made with CT when avai lable. All patients underwent examination under anesthesia (EUA) and t he findings compared. Of the 12 women presenting, seven had vesico-vag inal fistulas (VVF) and seven had recto-vaginal fistulas (RVF). Four w omen had both types of fistulas. The underlying pathology was cervical cancer (seven cases), colonic cancer (three cases), breast cancer (on e case) and ovarian cancer (one case). Vaginal fistulas were unequivoc ally seen on MRI in eight of 10 cases with fistulas. In the two cases with a difference between the MRI and EUA findings, the MRI was interp reted as showing more than was found at EUA, In the seven women with V VF, MRI detected five of the cases. In the seven women with RVF, MRI d etected all seven cases. Magnetic resonance imaging was correct in det ermining the presence of recurrent disease in the pelvis when an assoc iated mass was seen (seven cases). Computer-assisted tomography was co mpared in 10 cases and in six eases, the results were comparable and i n four cases, more information was obtained from the MRI. Magnetic res onance imaging appears to be accurate in detecting and defining comple x gynecologic fistulas and should be considered the investigation of c hoice to aid the planning of restorative, salvage or palliative surger y.