Purpose: Our goal was to determine the efficacy of MRI with rectal adm
inistration of perflubron in the evaluation of perirectal and pelvic f
istulae, with an emphasis on development of an improved method for fis
tula visualization. Method: Fourteen pelvic and/or perirectal fistulae
were evaluated in nine patients with MRI after the rectal administrat
ion of 200 mi of perflubron. Images were reviewed by two independent r
eaders and evaluated for the presence or absence of fistulae. Qualitat
ive assessment was also performed, rating the examinations for the abi
lity to visualize the fistulae. Comparison was made with the clinical
examination and/or operative findings, as well as available preperflub
ron imaging studies (CT = 10; fluoroscopy = 1; preperflubron MRI = 4).
Results: MRI with rectal perflubron demonstrated all 14 fistulae (100
%), while preperflubron imaging demonstrated only 7 of 14 (50%) fistul
ae (p less than or equal to 0.008, Wilcoxon signed rank test). In addi
tion, MRI with perflubron resulted in significantly improved qualitati
ve visualization of fistulae (overall very good rating versus poor rat
ing for preperflubron imaging, p less than or equal to 0.006, Wilcoxon
signed rank test). Conclusion: MRI with rectal perflubron is a superi
or modality for imaging of fistulae, compared with conventional method
s, since it allows a noninvasive means of demonstrating the presence o
f a fistula as well as offers improved qualitative assessment of fistu
lous disease.