Purpose: The aim of the study was to evaluate the value of dynamic magnetic resonance imaging (MRI) to classify suspect lesions into benign and malignant in patients previously treated for cancer of the cervix.
Materials and methods: Eleven patients with 14 suspect lesions after treatment of cervical carcinoma were examined by dynamic contrast-enhanced MRI. The imaging findings were compared to the giant cross-section specimen as the reference standard. Computed tissue-specific enhancement parameters were obtained (pharmacokinetic mapping) and displayed as color-coded images. The regions of interest were retrospectively defined according to the pharmacokinetic images over the most suspect areas. Therein, the threshold values were determined that achieved the greatest overall accuracy and a low rate of false-positive results.
Results: Analysis of the lesions on T2-weighted images revealed sensitivity of 88%, specificity of 40%, and accuracy of 71%. Analysis of the dynamic MR data showed significantly shorter (p < 0.01) and stronger (p < 0.001) contrast media enhancement of malignant (n = 9) than benign lesions (n = 5).
Conclusion: Pharmacokinetic mapping appears to yield important information for dividing suspect lesions into malignant and benign following treatment of cervical cancer.