Staging of lung cancer.

arl in the history of thoracic surgery, attempts were made to classify patients as to the extent of disease, in order to avoid unnecessary surgery. Evarts A. Graham,1 the surgeon who first successfully removed a carcinomatous lung in 1933, listed in 1944 the following signs of inoperabiity: (1) the presence of a bloody pleural effusion; (2) paralysis of the ipsilateral hemidiaphragm; (3) paralysis of the left vocal cord; (4) severe pain in the thoracic wall or down the arm; (5) bronchoscopic evidence of extension of the tumor into the trachea; (6) the presence of distant metastases; and (7) advanced age is not an absolute contraindication. More recently, the American Joint Committee for Cancer Staging has advocated staging of cancer of the lung to permit more precise definition of the extent of the disease. Staging is essential in the evaluation of treatment and also in the selection of proper treatment for individual patients. The classification covers the size and location of the tumor, the extent of the metastases to the lymph nodes, and the presence and extent of distant metastases.2 Tables 1 and 2 summarize the classification. Table 1 defines the size and location of primary tumors (TO to T4), the extent of nodal metastases (NO to N2), and whether distant metastases are present. This leads to the staging classification shown in Table 2. This system of classification is important for the evaluation of the results of therapy and for defining appropriate therapy for the individual patient. For the clinician, another important question is how to assess the extent of the tumor in the individual patient. A chest x-ray film and bronchoscopic examination usually suffice to provide the T classification of the tumor. The assessment of N or M classifications is more difficult. At the Veterans Administration Hospita! in San Diego, Calif, we have completed a study of 100 consecutive patients with cancer of the lung, exclusive of oat cell carcinoma.3 These patients all had bone, liver, and brain scans and scans with radioactive gallium performed. At the time of this presentation to the American College of Chest Physicians, only preliminary conclusions were available, since analysis had not been completed. Final analysis confirms the preliminary one. There were some surprising conclusions. We found that if the neurologic history and physical examination gave no evidence of abnormality, the brain scan did not add significant information. If the findings from physical examination and studies of hepatic function were normal, a liver scan was not helpful. In fact, it gave both false-positive and falsenegative results. Bone scanning gave many falsepositive results (old fractures, arthritis, etc) but did not find occult metastases that were not indicated by symptoms of pain. Thus, in the asymptomatic patient with normal findings on physical examination, scans of the brain, liver, and bones are not helpful.4 Scanning with radioactive gallium did prove useful in evaluating the need for mediastinoscopy or anterior thoracostomy, as follows: (1) if the primary