Computed tomography in lung cancer.

Computed-assisted tomography produces a cross-sectional image of the body using x-ray absorption measurements. Density differences are much more apparent than on conventional radiographs, although spatial resolution is not so fine. Lung cancer, can be evaluated accurately with regard to size, location, and whether regional or mediastinal lymph nodes are enlarged. Although enlarged granulomatous nodes generally tend to be dense, histologic specificity is not available. Computed tomography is recommended as a screening technique in lung cancer staging for patients whose routine radiographs are normal but whose primary lesions fulfill criteria for mediastinoscopy. If findings on computed tomography are normal, thoracotomy should then be performed; if enlarged nodes are detected, then a biopsy is needed to exclude operability. This strategy presumes that computed tomography has high sensitivity, although it remains to be proven by further experience whether the false-negative rate of computed tomography is an acceptable 10 to 15 percent. Computed tomography has been shown to be of considerable value in optimizing radiation therapy of lung cancer, and in diagnosing pleural complications of the cancer or its treatment.