B-Complex Vitamins (B1, B2, B3/Niacin, B5, B6, B7/Biotin, B9/Folate, B12) for Homocysteine

Finding: Mean plasma homocysteine decreased 2.4 µmol/L in the active arm and rose 0.8 µmol/L in placebo. Despite this clear surrogate effect, the primary composite endpoint occurred in 18.8% of active vs 19.8% of placebo (RR 0.95, 95% CI 0.84-1.07, p=0.41). MI was unchanged; cardiovascular death was unchanged; stroke was modestly reduced (RR 0.75, 95% CI 0.59-0.97). Hospitalization for unstable angina was significantly increased (RR 1.24, 95% CI 1.04-1.49). Net conclusion: homocysteine lowering does not reduce major cardiovascular events in patients with vascular disease.

Finding: Folic acid + B12 reduced plasma homocysteine by 27%. Despite this, primary composite occurred in 18.0% of folic acid + B12 + B6, 16.5% of folic acid + B12, 17.1% of B6 alone, and 16.2% of placebo. Combined folic acid + B12 + B6 produced a non-significant trend toward HARM (RR 1.22, 95% CI 1.00-1.50, p=0.05). No benefit observed in any active arm; signal of increased risk with the triple combination.

Finding: Mean homocysteine reduction 2 µmol/L greater in high-dose vs low-dose arm. Recurrent ischemic stroke risk was equivalent between arms (HR 1.0, 95% CI 0.8-1.3) at 2 years; combined endpoint of stroke, CHD event, or death also showed no difference (HR 1.0, 95% CI 0.8-1.1). Conclusion: moderate homocysteine reduction had no effect on vascular event rates after stroke. Note: background U.S. folate fortification began during the trial (1998), which may have blunted differential effects.

Finding: Allocation to folic acid + B12 reduced homocysteine by 3.8 µmol/L (28%). Primary major vascular event occurred in 25.5% of B-vitamin arm vs 24.8% of placebo (risk ratio 1.04, 95% CI 0.97-1.12, p=0.28). No effect on major coronary events, stroke, revascularisation, vascular mortality, or all-cause mortality. Largest single homocysteine-lowering trial; findings definitive for null effect on hard CV outcomes in secondary prevention.

Finding: Plasma homocysteine reduced 18.5% in active vs placebo. Primary composite occurred in 14.9% of active vs 14.3% of placebo (RR 1.03, 95% CI 0.90-1.19, p=0.65). No effect on individual outcomes including MI (RR 1.04), stroke (RR 1.14), revascularisation (RR 1.00), or CV death (RR 0.92). Despite robust homocysteine reduction over >7 years in high-risk women, no clinical benefit observed.

Finding: Pooled across 15 RCTs (n=71,422): no effect of homocysteine-lowering interventions on non-fatal/fatal MI (RR 1.02, 95% CI 0.95-1.10; high-quality evidence), all-cause mortality (RR 1.01, 95% CI 0.96-1.06), or cardiovascular mortality. Borderline reduction in stroke (RR 0.90, 95% CI 0.82-0.99) appearing only in trials conducted in regions WITHOUT folate fortification and dominated by Chinese trial CSPPT. Authors conclude no convincing evidence supporting homocysteine-lowering for CV event prevention.