Patients with pulmonary sarcoidosis are less likely to smoke than persons of a similar age in the general population. This could be because smoking reduces the likelihood of developing sarcoidosis, or alternatively smoking could reduce the severity of the disease process so that smoking patients are underrepresented among patients with clinically overt disease. To evaluate these possibilities 64 patients with sarcoidosis of recent onset were studied at presentation and after a one year follow up period, clinical, functional, radiographic, and bronchoalveolar lavage fluid criteria being used to identify factors that might account for the lower incidence of sarcoidosis in smokers and to determine whether the disease is less severe in smoking patients. Smoking was less common in the patients with sarcoidosis (30%) than in the control subjects (46%). The study did not support the conclusion that sarcoidosis is less severe in smokers, as clinical, radiographic, and functional abnormalities were similar in smokers and non-smokers at initial evaluation and after a one year follow up period. Nevertheless, smoking did influence various indices used to assess disease "activity." Cigarette smoking was associated with a significant increase in the serum angiotensin converting enzyme activity (SACE), and patients with very high SACE and pulmonary gallium-67 uptake were smokers. Furthermore, more CD8+ (but not CD4+) lymphocytes were recovered by lavage from smoking than from non-smoking patients, giving a lower CD4 :CD8 ratio in smokers. Fewer alveolar macrophages were recovered by lavage from smokers with sarcoidosis than from normal subjects with a similar smoking history. These findings support the possibility that smokers, particularly those with a prominent accumulation of alveolar macrophages in the lower respiratory tract, may be less likely to develop sarcoidosis.

pulmonale in pulmonary sarcoidosis. Right-heart overload happens relatively rarely in sarcoidosis, even with fibroemphysematous changes. Of 21 cases that we studied, six (28%) had clinical and/or electrocardiographic features of cor pulmonale. The cause of cor pulmonale often evoked is an invasion of the walls of pulmonary vessels by sarcoid granulomas or their compression by the fibrotic process. Pathological studies in one patient showed compression of large pulmonary arteries associated with specific sarcoid lesions in small and medium-sized arteries.The development of chronic cor pulmonale due to pulmonary hypertension has rarely been described in pulmonary sarcoidosis, even in chronic fibroemphysematous forms. The mechanisms often evoked to explain the occurrence of pulmonary hypertension are based on histological findings, the perivascular occurrence of granulomas with the fibrosis which sometimes results, and the possibility of thrombosis or invasion of arterial and vein walls by granulomas. Material and methodsTwenty-one cases of chronic sarcoidosis were studied.The diagnosis was made using clinical, radiological, pathological (bronchial biopsy), and immunological (Kveim test, tuberculin anergy) findings. Such a diagnosis was confirmed in life by these methods even in the two cases (2 and 5) where no definite evidence of sarcoidosis was found at necropsy many years later.The radiographic classification was based on postero-anterior and lateral views, occasionally confirmed by tomography.Evidence of right-sided heart involvement was based upon clinical examination and six electrocardiographic criteria (Schaub, 1966): 1 QRS vector angle in the frontal plane (angle a)>+900.2 Negative T wave in S2 and S3 or S, and ca. +900.3 S wave in V, and V >07 mV.4 R wave in V,+S wave in V5>1 mV.5 QRS direction to the right and posterior:rS of V1 to V4, sometimes rS or Qs of V1 to V6 with positive or flattened T waves.6 Incomplete right bundle-branch block: rS R' in V1 with R' in V1>10 mV. Radiological measurement of the size of the right heart chambers and the diameter of the pulmonary arteries was in most cases impossible because of the perihilar pulmonary opacities which obscured the cardiac and arterial contours. Thus radiological criteria were not taken into account.

Some properties of the alveolar epithelial barrier during transalveolar transport of water and solutes were studied in normal humans and patients with sarcoidosis by means of the transalveolar capillary concentration gradients of various solutes. A total of 9 normal control subjects (Group A) and 60 sarcoidosis patients, 52 with an evolving disease (Group B) and 8 recovered (Group C), underwent bronchoalveolar lavage (BAL). The second aliquot of fluid was used to measure urea, glucose, potassium, and albumin, which were also investigated in plasma. Urea was used to determine the volume of alveolar epithelial lining fluid (AELF volume). Results are expressed as the ratio of solute concentration in AELF over that in plasma (A/P ratio). In Group A there were clear concentration gradients of glucose, potassium, and albumin between the AELF and plasma, as the A/P ratios of glucose, potassium, and albumin were 0.02 +/- 0.006 (mean +/- SEM), 3.2 +/- 0.34, and 0.04 +/- 0.008, respectively. In Group B the A/P ratios of glucose (0.21 +/- 0.02, p less than 0.001) and albumin (0.17 +/- 0.02, p less than 0.001) were significantly increased but that of potassium remained unchanged (2.9 +/- 0.2). The A/P ratios of these various solutes were independent of chest x-ray typing. The albumin but not the glucose A/P ratio was correlated with the percentage of lymphocytes recovered from BAL (p less than 0.02); however, there was no correlation between the albumin A/P ratio and the CD4+/CD8+ T lymphocyte ratio. In group C there was a striking contrast between the albumin A/P ratio, which was normal, and the glucose A/P ratio, which was clearly elevated despite a normal lymphocyte count.(ABSTRACT TRUNCATED AT 250 WORDS)

Thirty-three cases of sarcoidosis (4.4% of 746 patients) showed an alveolar radiological pattern. A study of pulmonary function was carried out in 25 patients and compared with that of 46 patients with the interstitial radiological type of sarcoidosis. Twenty-two cases have been followed up from one to six years after the initial examination. The radiographic lesions were most often bilateral and included nodules greater than 15 mm with ill-defined margins or diffuse, infiltrative, non-retractile opacities with fluffy margins. Bilateral mediastinal lymph nodes were present in 27 patients. In 20 patients an associated reticulation was found on radiography. In four patients an open lung biopsy was done. The granulomatous nodules were identical to those found in other forms of sarcoidosis, although they were more confluent in the affected areas. Clinical and functional findings did not differ from those in the more common forms of sarcoidosis. Alveolar sarcoidosis has a sudden course. The alveolar radiological patterns always disappeared, with or without steroid treatment, while reticular patterns persisted in four patients. Rapid radiological changes were observed. Some functional abnormalities persisted in cases that were followed. It is concluded that alveolar sarcoidosis is a distinct acute form of sarcoidosis.Sarcoid granulomas are found in the alveolar septa and less frequently in the walls of bronchi, pulmonary arteries, and veins.' The radiological findings in pulmonary sarcoidosis may typically be of one of three types of diffuse interstitial opacities2: reticular, reticulomicronodular, or nodular (nodules less than 5 mm in diameter). Less The TLCO was scaled according to age and height (indicated as TLCO Ht). The static ventilatory mechanics were estimated through the relation between the volume measured at the mouth and the transpulmonary pressure during slow expiration. The following indices were calculated: static expiratory compliance (CL) meas-

To examine the possibility that mast cells have a central role in the pathogenesis of hypersensitivity pneumonitis, 20 patients with this disease were studied with the aim of seeking evidence for mast cell degranulation.The number of mast cells recovered by bronchoalveolar lavage from patients with hypersensitivity pneumonitis was more than 1000 times greater than those recovered from normal individuals. Furthermore, discontinuation of antigen exposure resulted in an increase in the number of mast cells observed, consistent with the possibility that antigen exposure had induced mast cell degranulation. Cessation of antigen exposure also resulted in a rapid decrease in the number of neutrophils and eosinophils recovered by lavage, followed by an increase in the number of T8 + T lymphocytes present. In each case the time course of the changes was consistent with the possibility that mast cell degranulation had been important in regulating the number of the immune and inflammatory cells present in the lung. Histamine was present in lavage fluid supernatant from patients with hypersensitivity pneumonitis. The amount of histamine present was, however, closely correlated with the number of mast cells present and not with the interval since last antigen exposure. Delay in separating cells from lavage fluid supernatant resulted in an increase in histamine content. These results suggest that the free histamine in lavage fluid resulted from the degranulation of mast cells induced by the lavage procedure as histamine released in vivo has a short half life. We suggest that hypersensitivity pneumonitis results from a "late phase reaction" initiated by antigen induced mast cell degranulation.