As was indicated at the beginning of this review, a concensus does not exist regarding many aspects of the use of CT in evaluating bronchogenic carcinoma. When and how CT is used, therefore, becomes a function of the beliefs of the physician caring for the patient. The radiologist must be familiar with this philosophy to be able to advise when CT will be of value. Despite all of the variables considered on the preceding pages, there are some facts. 1. (1) Normal mediastinal lymph nodes may be larger than 1 cm in maximal transverse diameter; the majority are not. 2. (2) An enlarged node (independent of definition) need not harbor metastases. Histologic proof is necessary, especially if this information will preclude surgery. 3. (3) CT less frequently offers usable information in small peripheral cancers. The use of CT in peripheral cancers is very much dependent on the surgeon's philosophy. 4. (4) Important information for patient care is more frequently obtained in patients with central lesions or peripheral lesions associated with abnormal hili or mediastinums. This is also closely related to surgical philosophy. 5. (5) Prediction of either chest wall or mediastinal invasion is treacherous and should only be diagnosed when the findings are certain. © 1990.