Vitamin K for Cardiovascular Disease
Despite encouraging observational data and a plausible mechanism, vitamin K (including K2/MK-7) has not been shown in rigorous trials to prevent heart disease or slow arterial and valve calcification, so it is not a justified supplement for cardiovascular protection. Anyone on warfarin or another vitamin K antagonist should not change their vitamin K intake without medical supervision.
Why this grade7-layer evidence engine
The evidence shows a classic split between observation and intervention. The Rotterdam Study (PMID 15514282, n=4,807) linked higher dietary K2 intake to lower coronary heart disease mortality (RR 0.43, 95% CI 0.24-0.77), but observational diet cohorts cannot rule out reverse causation or that K2-rich eaters are simply healthier overall.
Randomized trials, which carry far more weight, do not confirm a benefit. The largest and best-designed trial, AVADEC (PMID 35465686, n=365, MK-7 720 ug plus vitamin D over 24 months), found no slowing of aortic valve calcification (between-group difference 17 AU, P=0.64) and no effect on coronary calcification. Positive signals are confined to surrogate markers and are weak: a 2023 meta-analysis (PMID 37252246, 14 RCTs, n=1,533) found only a marginal reduction in coronary calcium score (MD -17.37 AU, upper CI essentially at null, P=0.04), Knapen 2015 (PMID 25694037) improved arterial stiffness but was industry-funded, and a 2024 meta-analysis (PMID 38282652, 17 RCTs) found everything but HOMA-IR unchanged. The earlier positive Brandenburg trial (PMID 28533322) was small, open-label, and superseded by AVADEC.
Regulators and clinicians reinforce the low grade. The FDA, EFSA, and WHO authorize vitamin K only for blood clotting, bone maintenance, and newborn bleeding prophylaxis, with no cardiovascular claim, and the NHS advises a food-first approach. Mayo Clinic states clinical trials show no evidence it slows arterial calcium, Harvard calls the research very limited, and the American Heart Association highlighted AVADEC's negative result. No trial has demonstrated reduced heart attack, stroke, or cardiovascular death.
Scoring transparency
All scores computed by a 7-layer evidence engine — fully auditable▸View the full decision path (audit trail)
- compute_raw_score — 加權公式: L2×0.30 + L3×0.25 + L5×0.25 + L11×0.10 + L1×0.10 = 0.385
- tier_from_score — 依分數區間映射至 tier letter
- apply_hec_rules — 無高階證據可裁決
- tier_strict_requirement_check — Tier 條件達標，未降階
- detect_disputes — 偵測到 0 個 hard + 1 個 soft dispute
- decide_status — 依 tier + dispute 結果決定 status