Cervical carcinoma: computed tomography and magnetic resonance imaging for preoperative staging

To assess the accuracy of computed tomography (CT) and magnetic resonance imaging (MRI) in the evaluation of invasive cervical cancer.

Seventy-nine women with untreated cervical cancer underwent pre-treatment MRI (n = 71) and/or CT (n = 37) within 4 weeks of surgical evaluation. Twenty-nine women had both MRI and CT. Images were evaluated for tumor detection, size, stromal invasion, local extension, and nodal metastases.

Tumor size was evaluated accurately by MRI, with a correlation coefficient of 0.93. Magnetic resonance estimates of tumor size were within 0.5 cm of the surgical sample in 64 of 69 women (93%). Magnetic resonance was 88% accurate evaluating the presence of stromal invasion and 78% accurate for depth of stromal invasion. Computed tomography could not evaluate tumor size or stromal invasion because it could not distinguish cancer from the surrounding normal cervical tissue. In evaluating stage of disease, MRI had an accuracy of 90%, compared with 65% for CT (P < .005). Magnetic resonance imaging was more accurate than CT (94 versus 76%, P < .005) in assessing parametrial invasion. Both modalities were comparable in evaluating lymph node metastases (86% each). In determining operative candidates (stage I and minimal IIA), MRI was 94% accurate, compared with 76% for CT (P < .005).

Compared with CT, MRI offered significantly improved evaluation of tumor size, stromal invasion, and local and regional extent of disease in pre-treatment imaging for cervical cancer.