The prevalent theory in the pathogenesis of emphysema proposes that increased numbers of activated neutrophils and/or alveolar macrophages produce large amounts of proteases, an activity that cannot be regulated by alpha 1-antiproteases, resulting in lung destruction. However, the cells in the lung parenchyma of smokers have not been properly identified. We characterized and quantitated the inflammatory cell load in the lungs of smokers and correlated these findings with the degree of lung destruction. Twenty-one patients, six nonsmokers and 15 smokers, undergoing lung resection were studied. Lungs or lobes were fixed and stained for light microscopy and neutrophil identification and immunohistochemically stained for identification of lymphocytes and macrophages. By point counting, we determined the extent of emphysema by the volume density of the lung parenchyma (Vvalv), and the different cell numbers per cubic millimeter in all lungs. In nonsmokers Vvalv was greater than in smokers. The number of neutrophils/mm3 of lung correlated directly with the Vvalv, (r = 0.71, p < 0.01), whereas the number of alveolar macrophages (r = -0.70) and T-lymphocytes (r = -0.78) correlated negatively with the Vvalv. The number of T-lymphocytes correlated negatively with the number of neutrophils (r = -0.58) and positively with the numbers of alveolar macrophages (r = 0.77). Our data suggest that as long as the inflammatory reaction is predominantly of neutrophils there is no destruction of the lung. However, the extent of lung destruction becomes evident, and its extent is directly related to the number of alveolar macrophages and T-lymphocytes/mm3. We conclude that the T-lymphocyte might be importantly implicated in the pathogenesis of emphysema in smokers.