Fifteen cohort studies were included in the review (n=381,903; 7,415 incidents of CHD, ranging between 101 and 1,356 events). The extent of population overlap between studies was not clear.
Comparison of participants in the upper third versus the lower third at baseline showed small to moderately statistically significantly greater risk of CHD in the lower third for patients receiving vitamin C with RR 0.84 (95% CI: 0.73, 0.95; 14 study arms) and vitamin E with RR of 0.76 (95% CI: 0.63, 0.89; nine study arms). No statistically significant relationship was found between Beta-carotene and CHD, 0.78 (95% CI: 0.53, 1.04; three study arms).
There was evidence of statistical heterogeneity among studies for intake of vitamin C (I2=63, p=0.0005), vitamin E (I2=42, p=0.08) and Beta-carotene (I2=64, p=0.06). Sensitivity analyses did not significantly alter the results for vitamin C and E. A dose relationship was reported for vitamin E, with an increase of 30IU per day potentially lowering the risk of CHD by four per cent. No statistically significant dose-response relationship was found for vitamin C or Beta-carotene.
There was no evidence of publication bias for any antioxidant using funnel plot analysis.