The relationship of ventilatory function to cardiovascular events was studied in 12,511 men and women, enrolled in 1976-1978 in a prospective population study. Until the end of 1983, 388 subjects died because of a cardiovascular disease, 133 died within 30 days of developing myocardial infarction (fatal myocardial infarction), while 238 had a non-fatal myocardial infarction. Cox proportional hazards models were employed for the analysis. In the models including tobacco smoking, cholesterol level, blood pressure, diabetes mellitus and body-mass index as covariates, forced expiratory volume in one second in percentage of predicted (FEV1% pred), forced vital capacity in percentage of predicted (FVC% pred), and the ratio between FEV1 and FVC (FEV1/FVC) were significantly related to the risk of cardiovascular death, e.g. compared with subjects with FEV1% pred or FVC% pred greater than or equal to 80 the risk of death from cardiovascular disease among subjects with FEV1% pred or FVC% pred less than 60 was approximately twice as high. There was a nonsignificant trend towards an increased risk of fatal myocardial infarction with reduction of FVC% pred, but in general the regression models did not show consistent relationship between lung function impairment and risk of myocardial infarction. In the regression models, which only included age and sex as covariates, reduced FVC% pred and FEV1% pred were significantly related to both cardiovascular mortality and to fatal myocardial infarction, but not to the non-fatal infarction, whereas the FEV1/FVC ratio was not related to fatal or to non-fatal myocardial infarction.(ABSTRACT TRUNCATED AT 250 WORDS)