Objective: To investigate the relation of baseline antioxidant, fruit, vegetable and fish intake with 20 y chronic obstructive pulmonary disease (COPD) mortality in middle-aged men from three European countries. Design: Prospective study (1970 -1990). Setting: Five population-based cohorts of middle-aged men from Finland, Italy and The Netherlands. Subjects: A total of 2917 men aged 50 -69 y at baseline. Methods: Baseline information on diet was collected using the cross-check dietary history method. After 20 y of follow-up the underlying cause of death of those who died was established centrally. Survival analyses were performed using the Cox Proportional Hazards Model. Results: After adjustment for age, smoking and country, we observed an inverse trend (P-trend < 0.05) of 20 y COPD mortality across tertiles of fruit and vitamin E intake. No trend was observed for vegetables, fish, vitamin C and b-carotene. When modelled continuously, a 100 g increase in fruit intake was associated with a 24% lower COPD mortality risk (RR ¼ 0.76, 95% CI ¼ 0.60 -0.92). For vitamin E intake (per 5 mg) the RR was 0.77 (95% CI ¼ 0.55 -1.06), after adjustment for age, smoking and country. Additional adjustment for body mass index, total energy intake and alcohol consumption reduced the RR to 0.86 (95% CI ¼ 0.69 -1.07, P ¼ 0.12) for fruit and 0.93 (95% CI ¼ 0.65 -1.33) for vitamin E. Conclusions: Our results suggest a protective effect of fruit and possibly vitamin E intake against COPD. No effect was observed for intake of vitamin C, b-carotene, vegetables and fish.

Ozone exposure has been related to adverse respiratory effects, in particular to lung function decrements. Antioxidant vitamins are free-radical scavengers and could have a protective effect against photo-oxidant exposure. To evaluate whether acute effects of ozone on lung functions could be attenuated by antioxidant vitamin supplementation, we conducted a randomized trial using a double-blind crossover design. Street workers (n = 47) of Mexico City were randomly assigned to take daily a supplement (75 mg vitamin E, 650 mg vitamin C, 15 mg beta carotene) or a placebo and were followed from March to August 1996. Pulmonary function tests were done twice a week at the end of the workday. During the follow-up, the mean 1-h maximum ozone level was 123 ppb (SD = 40). During the first phase, ozone levels were inversely associated with FVC (beta = -1.60 ml/ppb), FEV1 (beta = -2.11 ml/ppb), and FEF25-75 (beta = -4.92 ml/ppb) (p < 0.05) in the placebo group but not in the supplement group. The difference between the two groups was significant for FVC, FEV1, and FEF25-75 (p < 0.01). During the second phase, similar results were observed, but the lung function decrements in the placebo group were smaller, suggesting that the supplementation may have had a residual protective effect on the lung. These results need to be confirmed in larger supplementation studies.