Background: Flavonoids may contribute to preserve lung function, but epidemiological evidence is limited to high-income countries.

Objective: To investigate the association of dietary flavonoids and ventilatory function in adults from the BOLD study.

Methods: We analysed data from 3,879 adults aged ?40 years from Albania, Kyrgyzstan, Morocco, Pakistan, and Sri Lanka, who underwent post-bronchodilator spirometry. Dietary intakes of total, and seven flavonoid subclasses (flavan-3-ols, flavanones, flavones, flavonols, anthocyanidins, isoflavones, and proanthocyanidins) were also estimated. The association between FVC, FEV1/FVC and flavonoids (highest vs. lowest quintile) was examined within each site, using adjusted linear regression models. Effect estimates from all sites were then combined using meta?analysis.

Results: The total median flavonoid intake was 384.8 mg/d (IQR 262.2 ? 527.0). Compared to those in the lowest quintile, individuals in the highest quintile of total flavonoid intake had a higher FVC (49.1ml; 95% CI -51.9, 84.5; 150.0; I2=57.2%) but this association did not reach statistical significance. Among subclasses, Isoflavone intake was positively associated with FVC (69.1ml; 95% CI 0.97, 137.3; I2=49.0%). In stratified analyses, flavanone and isoflavone intakes were associated with higher FVC among current smokers (222.5ml; 95% CI 46.4, 398.7; I2=0%), and never smokers (52.1ml; 95% CI 2.4, 101.8; I2=0%), respectively. Differences in FEV1/FVC were not associated with flavonoids.

Conclusion:  In this multinational survey, we found some evidence of a cross-sectional association between specific subclasses of flavonoids and FVC, but not FEV1/FVC.