Epidemiological studies have shown that higher intake of flavonoid is inversely associated with CHD risk. However, which flavonoid subclass could reduce CHD risk has remained controversial. The present meta-analysis of prospective cohort studies aimed to quantitatively assess the associations between flavonoid subclasses and CHD risk. A systematic literature search was implemented from PubMed and Web of Science databases up to March 2021, and eligible studies were identified. Multivariate-adjust relative risks (RR) with corresponding 95 % CI were pooled by using a random-effects model. A restricted cubic spline regression model was performed for non-linear dose–response analysis. A total of 19 independent prospective cohort studies with 894 471 participants and 34 707 events were included. The results showed that dietary intakes of anthocyanins (RR = 0·90; 95 % CI: 0·83, 0·98), proanthocyanidins (RR = 0·78; 95 % CI: 0·65, 0·94), flavonols (RR = 0·88; 95 % CI: 0·79, 0·98), flavones (RR = 0·94; 95 % CI: 0·89, 0·99) and isoflavones (RR = 0·90; 95 % CI: 0·83, 0·98) were negatively associated with CHD risk. Dose–response analysis showed that increment of 50 mg/d anthocyanins, 100 mg/d proanthocyanidins, 25 mg/d flavonols, 5 mg/d flavones and 0·5 mg/d isoflavones were associated with 5 % reduction in CHD risk, respectively. Sensitivity and subgroup analyses were used to further support these associations. The present results indicate that dietary intakes of fruits and vegetables abundant five flavonoid subclasses, namely anthocyanins, proanthocyanidins, flavonols, flavones and isoflavones, are associated with a lower risk of CHD.