Epidemiology of pulmonary lesions in nontextile and cotton textile workers: a retrospective autopsy analysis.

A collection of 565 unselected inflation-fixed lungs was divided into three groups: (1) normal (209 lungs); (2) centrilobular emphysema (231) lungs); and (3) "other" (125 lungs), the last including examples of fibrosis, tuberculosis, cancer, and other forms of emphysema. Clinical hospital records were reviewed to ascertain smoking history [no smoking (105 lungs); greater than 0.5 pack cigarettes per day (427 lungs); or pipe/cigar (33 lungs)] and occupation [nontextile (521 lungs), or textile (44 lungs)]. Lungs were subjected to morphometric determination of the extent of centrilobular emphysema, mucus gland hyperplasia in large bronchi, and goblet cell metaplasia in bronchioles. Extent of tissue pigmentation in normal lungs was also measured. Associations between morphologic data and background factors were examined by covariance analysis. As in many previous studies, data show highly significant cigarette smoking effects on all factors measured. Significant pipe smoker effects were also found, and when the cigarette group was excluded, a significant association was found between cotton dust exposure and both mucus gland hyperplasia and goblet cell metaplasia, but not emphysema. The results suggest that centrilobular emphysema is not associated occupationally in the textile industry, although bronchitis and bronchiolitis probably are. If byssinotic symptoms and physiologic impairment are as prevalent as some have reported, they must be primarily related to airway lesions. It might follow that they should be reversible. Textile workers with irreversible impairment and morphologic emphysema who are also smokers probably have little or no justification for attributing this to their occupation.