Fourteen RCTs (38,941 participants) were included. Study size ranged from 240 to 12,064 participants. Twelve trials were double-blind placebo-controlled and two were open control, compared to usual care. Tests found no evidence of publication bias.
Compared to control, folic acid decreased homocysteine levels in all trials. The pooled net reduction was 2.9μmol/L (95% CI 2.4 to 3.4, I2=91%).
Compared to control, folic acid had no effect on the primary clinical outcomes (I2=38%), risk of cardiovascular disease (I2=0%), coronary heart disease (I2=31%), stroke (I2=25%) and all-cause mortality (I2=0%). Tests showed no evidence of publication bias. Sensitivity analyses that removed one trial at a time did not significantly alter the results.
Subgroup analyses according to country status of folic acid fortification showed no significant difference in baseline homocysteine, net homocysteine decrease and the primary clinical outcome.
In sensitivity analyses grouped according to baseline homocysteine levels that compared folic acid to control there was a significant difference (p=0.030) in risk ratios for the primary outcome between those above and below the mean (12μmol/L). Those with a higher baseline homocysteine had a higher risk of cardiovascular disease (RR 1.06, 95% CI 1.00 to 1.13) compared with those with a lower than average homocysteine level (RR 0.94, 95% C I 0.86 to 1.03). When results were analysed stratified by baseline homocysteine levels, comparing folic acid to control, on average the pooled risk ratio for primary clinical outcomes increased risk by 3.9% (95% CI -3.0 to 11.3) for each 5μmol/L increase in baseline homocysteine (I2=8%).