Faculty Bibliography

Forty-nine patients with chest wall lesions were evaluated by computed tomography (CT) and conventional radiography. Computed tomography was found to be indispensable for detecting and precisely localizing these lesions. It revealed unsuspected bone destruction and lung, pleural, and mediastinal involvement, as well as invasion of the spinal canal. In more than two thirds of the patients, CT provided additional information of clinical importance in management and, in one third, treatment was altered or the surgical approach modified because of the CT findings. Computed tomography is an essential diagnostic modality in evaluating chest wall lesions

The computed tomographic appearance of collapse without endobronchial obstruction is reviewed. These 57 cases were classified by the etiology of collapse. The largest group consisted of 29 patients with passive atelectasis, i.e., collapse secondary to fluid, air, or both in the pleural space. Twenty-three of 29 proved secondary to malignant pleural disease. Computed tomography accurately predicted a malignant etiology in 22 of 23 cases. The second largest group of patients had lobar collapse secondary to cicatrization from chronic inflammation. In all cases the underlying etiology was tuberculosis. Radiation caused adhesive atelectasis in six patients secondary to a lack of production of surfactant. In each case a sharp line of demarcation could be defined between normal and abnormal collapsed pulmonary parenchyma. Three cases of unchecked tumor growth caused a peripheral form of collapse (replacement atelectasis). This form of collapse was characterized by an absence of endobronchial obstruction and extensive tumor not delineated by the normal boundaries of the pulmonary lobes

The computed tomographic (CT) appearance of lobar collapse has yet to be defined. In an attempt to determine the characteristic appearance of collapse 95 cases were reviewed retrospectively in a wide variety of clinical settings over a 3 year period ending January 1983. In this report 38 cases of lobar collapse secondary to endobronchial occlusion are analyzed; the appearance of collapse without endobronchial obstruction forms the basis of a subsequent report. Computed tomography was accurate in determining the site of bronchial occlusion in all cases. In 36 of 38 cases collapse was caused by endobronchial tumors, including bronchogenic carcinoma, bronchial carcinoids, endobronchial metastases, and lymphoma. Differentiation between these tumors was not feasible with CT. Most cases of collapse were caused by central tumor. In those cases in which a bolus of contrast material was used differentiation between tumor mass and collapsed pulmonary parenchyma was possible. Two of 38 cases were found to have benign bronchial occlusion. In one case a mucous plug obstructing the left lower lobe bronchus was accurately defined. In another case a bronchial stricture occluded the right lower lobe bronchus. This represented the only false positive case in this series. It is concluded that CT is an accurate means for establishing the diagnosis of endobronchial obstruction. In most cases the diagnosis of neoplasia was possible, provided a bolus of contrast material was used to define tumor mass. The potential role of CT in evaluating patients with lobar collapse is discussed

Thirty patients with lung involvement with Pneumocystis carinii and other opportunistic organisms, many of whom also had Kaposi sarcoma, were seen from December 1980 through March 1982. Clinical manifestations consisted of a prodrome of weeks to months with weight loss, fever, and malaise. When clinical pneumonia became apparent, four distinct radiographic patterns were identified. Pneumocystis carinii was uniformly present, and the most common pattern encountered was a relatively symmetric, homogeneous perihilar pneumonia that progressed to diffuse consolidation. Asymmetric and focal infiltrates were seen in patients who proved to have concomitant opportunistic infection, most commonly fungal in all but two cases. A third pattern of nodular and linear densities with or without adenopathy was seen in patients without pneumonia who had biopsy-positive Kaposi sarcoma involving the lung parenchyma. A fourth pattern represented a combination of any of the first three, and these patients had multiple infections as well as Kaposi sarcoma in the lung. Any significant change in the radiograph indicating progression of disease while on therapy prompted a rebiopsy, and in five of 10 patients other infections and/or Kaposi sarcoma were identified