Computed tomography in the pretreatment assessment of carcinoma of the cervix.

Fifty-nine patients with primary or recurrent carcinoma of the cervix were evaluated by computed tomography as part of their presurgical evaluation. The computed tomography staging results were compared with the surgical staging. Computed tomography staging was accurate in 71% (42 of 59), whereas clinical staging was accurate in 66% (39 of 59). In assessing paraaortic nodes by CT, there were 10 true-positive, 20 true-negative, 1 false-positive, and 2 false-negative results (sensitivity, 83%; specificity, 95%), for an overall accuracy of 91%. For pelvic nodes, there were 10 true-positive, 11 true-negative, 3 false-positive, and 6 false-negative results (sensitivity, 62.5% specificity, 78%), for an overall accuracy of 70%. Excretory urograms and barium enemas provided no information not obtained by computed tomography and are probably unnecessary if computed tomography is used as a routine staging examination. At present, computed tomography should not replace clinical assessment of extent of the disease. Its chief advantage over clinical staging is its ability to detect metastases beyond the true pelvis.