Vitamin K1 vs K2: different vitamins despite the shared letter

Both forms of vitamin K act on the same enzyme family — gamma-glutamyl carboxylases that activate vitamin-K-dependent proteins. But K1 and K2 differ enough in tissue distribution and half-life that the consumer market increasingly treats them as separate vitamins. Here is what the chemistry and the trial evidence actually show.

What K1 and K2 are

Vitamin K1 (phylloquinone) is the form synthesised by plants. Leafy greens — kale, spinach, collards, broccoli — are the dominant dietary source, with one cup of cooked kale providing about 1,000 mcg. K1 has a phytyl side chain. Vitamin K2 (menaquinones) is the bacterial form found in animal-derived and fermented foods. The menaquinones are designated MK-n by the number of isoprenoid units in the side chain. MK-4 is found in egg yolk, butter, and some organ meats; MK-7 through MK-13 are produced by gut and food-fermentation bacteria, with MK-7 most abundant in natto (Japanese fermented soy).

Half-life, tissue distribution, and hepatic vs. extrahepatic action

K1 has a plasma half-life of about 90 minutes and is preferentially taken up by hepatocytes, where it activates clotting factors II, VII, IX, and X. K2 menaquinones — especially the longer-chain MK-7 — have plasma half-lives of 60–72 hours and reach extrahepatic tissues (vessel wall, bone) more efficiently. This pharmacokinetic difference is the rationale for the K2 cardiovascular and bone hypothesis: hepatic clotting needs are easily met by background K1 intake, but extrahepatic matrix Gla-protein and osteocalcin carboxylation depend on sustained tissue exposure that K2 provides more efficiently (PMID: 36511772).¹

What the clinical trials show by indication

For warfarin reversal, K1 (phytonadione) is the only form used clinically because of its rapid onset on hepatic clotting factors. K2 should not be used for warfarin reversal. For bone health, the most rigorous data come from Japanese MK-4 menatetrenone trials at 45 mg/day, which modestly reduced vertebral fracture risk in postmenopausal osteoporosis but did not consistently meet primary endpoints in the larger OF Study (PMID: 19330493).² For arterial calcification, MK-7 at 180 mcg/day improved carotid arterial stiffening over 3 years in postmenopausal women (PMID: 25694037).³ For neonatal vitamin K deficiency bleeding, all guidelines specify K1 — the standard 1 mg IM dose at birth — and K2 is not interchangeable (PMID: 36014527).⁴

Dietary intake estimates

Adequate intake (AI) for K1 is 90 mcg/day for adult women and 120 mcg/day for adult men in the US Dietary Reference Intakes. Average US intakes are typically 80–110 mcg/day. There is no formal AI for K2 because dietary K2 intake varies enormously by population: Japanese diets containing natto provide 50–100 mcg/day of MK-7, while typical Western diets provide perhaps 10–25 mcg/day of total menaquinones (PMID: 36514345).⁵ A 2022 USDA analysis added MK-4 through MK-10 to the National Nutrient Database for the first time, reflecting growing recognition of K2 as a distinct dietary input (USDA Food Data Central, 2022).

Practical decisions on supplementation

A person consuming several servings per day of leafy greens generally has adequate K1 status; supplementation is unnecessary outside specific clinical settings (cystic fibrosis, certain malabsorption syndromes). The K2 case is different: typical Western dietary K2 intake is well below trial doses, and supplementing 90–180 mcg/day of MK-7 reliably improves carboxylated MGP biomarkers. Whether that biomarker shift translates to clinical outcomes in advanced calcified disease remains uncertain (PMID: 35460244).⁶ For postmenopausal bone or vascular concerns in patients who do not eat natto, 90–180 mcg/day of MK-7 is the most defensible regimen.

Warfarin warning that the supplement market keeps under-emphasising

Both K1 and K2 antagonise warfarin's anticoagulant effect. Patients on warfarin should keep K intake consistent rather than fluctuating, and should not start either form of K supplement without anticoagulation oversight. The 2024 ACC/AHA dyslipidaemia and lipid update reiterates this caution and notes that DOACs (apixaban, rivaroxaban) do not have this interaction (PMID: 39836398).⁷ A 2023 European review estimated that uncoordinated K supplementation accounts for 5–10 percent of unexpected INR fluctuations in long-term warfarin patients (PMID: 37852135).⁸