Vitamin K is considered to be involved in the pathological mechanisms of coronary artery calcification (CAC). Correlation between CAC and plasma vitamin K levels was studied. A total of 103 patients, with at least one coronary risk factor, were studied. CAC was measured using 64-slice multislice computed tomography (MSCT) and divided into three groups: none (CAC score = 0; n 25), mild to moderate (0 < CAC score < 400; n 52) and severe (CAC score > 400; n 26). Phylloquinone (PK) and menaquinone (MK)-4 and MK-7 were measured by HPLC-tandem MS. Mean age of patients was 64 (sd 13) years, of which 57 % were male. Median CAC score was 57·2. Median levels of PK, MK-4 and MK-7 were 1·33, 0 and 6·99 ng/ml, showing that MK-7 was the dominant vitamin K in this population. MK-7 showed a significant inverse correlation with uncarboxylated osteocalcin (ucOC, P = 0·014), protein induced by vitamin K absence of antagonist-2 (PIVKA-2, P = 0·013), intact parathyroid hormone (P = 0·007) and bone-specific alkaline phosphatase (P = 0·018). CAC showed an inverse correlation with total circulating uncarboxylated matrix Gla protein (t-ucMGP, P = 0·018) and Hb (P = 0·05), and a positive correlation with age (P < 0·001), creatinine, collagen type 1 cross-linked N-terminal telopeptide (NTX, P = 0·03), pulse wave velocity (P < 0·001) and osteoprotegerin (P < 0·001). However, CAC did not have a significant correlation with plasma levels of PK, MK-4 or MK-7. In conclusion, plasma MK-7, MK-4 or PK level did not show significant correlation with CAC despite the association between plasma vitamin K levels and vitamin K-dependent proteins such as ucOC or PIVKA-2.