The use of computed tomography in the evaluation of mediastinal masses

A retrospective analysis was conducted to ascertain whether computed tomography had increased diagnostic accuracy while decreasing the number of tests needed in the preoperative assessment of patients with mediastinal masses. A total of 42 patients were entered into the study: Fifteen patients were evaluated before the advent of computed tomography (No CT) and 27 patients had computed tomography during their evaluation (CT). The No CT group comprised 10 male and five female patients (2:1 ratio); the age range was 8 months to 61 years. The CT group included 15 male and 12 female patients (1.25:1.0 ratio), the age range being 21 to 70 years. In each group, both invasive and noninvasive studies were done. Although the CT group had 40 noninvasive tests, 27 were computed tomographic scans. The additional 13 noninvasive tests and the five invasive tests added no significant diagnostic information. In the No CT group, preoperative evaluation as to the cystic or solid nature of the mass was correct only four of 13 times (31%). In the CT group, 22 of 25 patients had accurate assessment as to the cystic or solid nature of the lesions (88%). In addition, extension of the mass into other structures, consistent with malignancy, was correctly diagnosed preoperatively in nine of the patients in the CT group. Two had extension of the mass at operation not preoperatively diagnosed (82% accuracy). None of the No CT group was given an assessment of possible mass extension preoperatively. The results suggest that mediastinal masses can be evaluated by computed tomography with a high degree of accuracy for predicting the nature, size, location, and involvement of other organs by the mass. The use of other tests before resection generally yields little additional information.