Scientific Opinion on the substantiation of a health claim related to
vitamin K2 and contribution to the normal function of the heart and blood
vessels (ID 125, further assessment) pursuant to Article 13(1) of Regulation
(EC) No 1924/2006[sup]1[/sup]
EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA)2, 3
European Food Safety Authority (EFSA), Parma, Italy
ABSTRACT
Following a request from the European Commission, pursuant to Article 13 of Regulation (EC) No 1924/2006,
the Panel on Dietetic Products, Nutrition and Allergies was asked to provide a scientific opinion on a health
claim related to vitamin K2 and contribution to the normal function of the heart and blood vessels. The food
constituent that is the subject of the claim, vitamin K2, is sufficiently characterised. The claimed effect,
contribution to the normal function of the heart and blood vessels, is a beneficial physiological effect. The
proposed target population is the general population. In weighing the evidence, the Panel took into account the
absence of human intervention studies from which conclusions could be drawn for the scientific substantiation of
the claim, the inconsistency of the results reported in two cross-sectional studies regarding arterial calcification in
women, that the results of two prospective cohort studies are in conflict regarding the risk of coronary heart
disease associated with vitamin K2 intakes, that high intakes of vitamin K2 were associated with a significantly
lower degree of aortic calcification in one prospective cohort study after adjustment for confounders, and that the
evidence provided for a proposed mechanism is weak. On the basis of the data presented, the Panel concludes
that a cause and effect relationship has not been established between the dietary intake of vitamin K2 and
contribution to the normal function of the heart and blood vessels.
© European Food Safety Authority, 2012
Słowa kluczowe:
Vitamin K
blood vessels
calcification
health claims
heart
menaquinones
3. Naukowe uzasadnienia wpływu na zdrowie człowieka - Funkcjonowanie serca i naczyń krwionośnych
-
Funkcjonowanie serca i naczyń krwionośnych
In its earlier opinion (for ID 124, 125 and 2880) (EFSA Panel on Dietetic Products Nutrition and Allergies (NDA), 2009), the Panel considered three cross-sectional studies (Beulens et al., 2009; Jie et al., 1995; Maas et al., 2007) and one prospective cohort study (Geleijnse et al., 2004) which investigated the relationship between vitamin K intake and arterial calcification or the elastic properties of the arteries, which may interfere with normal vascular structure and function. One of the studies did not report on vitamin K2 intakes specifically and will not be considered further in this evaluation, as requested in the application for further assessment (Jie et al., 1995). In addition, one animal study (Luo et al., 1997), which reported on a Matrix Gla Protein (MGP) knockout mouse model, and one human study (Munroe et al., 1999) reporting on a mutation of the gene encoding human MGP, were considered for the scientific substantiation of the claim.
In the framework of the further assessment, one prospective cohort study on the association between vitamin K2 intakes and the risk of coronary heart disease (CHD) (Gast et al., 2009), as well as three unpublished human intervention studies which investigated the effect of vitamin K2 on blood concentrations of MGP (Braam and Vermeer, 2011, unpublished; Moschonis et al., 2011, unpublished; Vermeer, 2008, unpublished) and two animal studies on the mechanisms by which vitamin K2 could exert the claimed effect (Schurgers et al., 2007; Spronk et al., 2003), were provided.
This evaluation is based on the scientific references provided in the present and the previous submissions which addressed the relationship between vitamin K2 intake and changes in vascular/heart function (e.g. CHD) or changes in vascular/heart structure leading to changes in vascular/heart function (e.g. arterial calcification). Scientific references on the mechanisms by which vitamin K2 could exert the claimed effect in the target population were also considered.
Two prospective cohort studies investigated the association between vitamin K2 intakes and incidence of CHD (Gast et al., 2009; Geleijnse et al., 2004) and the degree of aortic calcification (Geleijnse et al., 2004).
A Dutch prospective cohort study (Geleijnse et al., 2004) was conducted in 7,983 men and women aged 55 years and over, who completed at baseline a semi-quantitative food-frequency questionnaire to assess foods and beverages consumed more than once a month during the preceding year. At baseline, calcified deposits were detected in the abdominal aorta parallel and anterior to the lumbar spine on a lateral radiographic film, and the severity of this calcification was graded as “absent or mild” (≤1 cm calcification), “moderate” (>1 and <5 cm) or “severe” (≥5 cm). Risk of non fatal myocardial infarction, of incident CHD and of CHD mortality during follow-up was examined. After exclusion of institutionalised subjects, subjects with no or unreliable dietary data, and subjects with a history of myocardial infarction, 4,807 individuals, with a mean follow-up of 7.2 years, were included in the analysis for coronary events. After exclusion of subjects with no or unreliable baseline
radiographic films, 4,473 subjects were included in the analysis on abdominal aortic calcification. The risk ratios (RR) of coronary events in energy-adjusted tertiles of intake of vitamin K2 were studied in a Cox regression model and the association (odd ratios, OR) between the tertiles of intake and moderate or severe aortic calcification was assessed in a multivariate logistic regression model. Analyses were performed with adjustment for age, sex and total energy intake (model 1), and with additional adjustment for Body Mass Index (BMI), diabetes mellitus, smoking status, pack/year of cigarette smoking, category of education, and intakes of alcohol, calcium, flavonols, and saturated and polyunsaturated fatty acids (model 2). In this population, 64.3 % (n=2,874), 30.4 % (n=1,359) and 5.4 % (n=240) of subjects had “absent or mild”, “moderate” or “severe” calcification, respectively. Compared with the first tertile of vitamin K2 intake, high intakes of vitamin K2 (third tertile only) were associated with a lower incidence of CHD only in model 2 (RR: 0.59, 95 % CI: 0.40-0.86), and with a lower CHD mortality (models 1 and 2), but not with a lower risk of non fatal myocardial infarction (models 1 and 2). Compared with the first tertile of vitamin K2 intake, high intakes of vitamin K2 (third tertile only) were associated with a lower risk of severe aortic calcification (models 1 and 2; OR: 0.48, 95 % CI: 0.32-0.71 for model 2). No significant association was observed with moderate calcification. Results did not change with further adjustment for intakes of vitamin E, vitamin C, beta-carotene and fibre. The Panel notes that higher intakes of vitamin K2 were associated with a significantly lower degree of aortic calcification and lower incidence of CHD in this study, after adjustment for confounders.
Another Dutch prospective cohort study (Gast et al., 2009) was conducted in 17,357 women aged 49- 70 years, with a mean follow-up of 8.1 years, and investigated the relationship between vitamin K2 intake and risk of CHD. After exclusion of women with dietary energy intake below 500 kcal/day or above 6,000 kcal/day, with no baseline general questionnaires or no retrievable data on vital status, or with either a reported history of CHD or cerebrovascular events or a reported use of vitamin K antagonists, 16,057 post-menopausal women completed a food frequency questionnaire at baseline to estimate intakes of vitamin K2 during the year preceding enrolment in the study. Cox proportional hazards regression models were used to estimate the hazard ratio (HR) for coronary events, in a univariate analysis (model 1), after adjustment for age at baseline (model 2), after additional adjustment for BMI, smoking status, presence of diabetes, hypertension and hypercholesterolaemia, and energy-adjusted alcohol intake (model 3), after further adjustment for dietary factors, i.e. energy and energy-adjusted intakes of saturated and polyunsaturated fatty acids (model 4), or after further adjustment for calcium intakes, which were found to be strongly correlated with vitamin K2 intakes (model 5). The HR for risk of CHD per 10 µg increase in vitamin K2 intake was significantly reduced in models 1 and 2, but not in model 3. The HR was 0.92 (95 % CI: 0.83 to 1.01) in model 5, and was 0.91 (95 % CI: 0.85 to 1.00) when calcium was excluded from the model (model 4). The Panel notes that higher intakes of vitamin K2 were not significantly associated with a reduced risk of CHD in this study, after adjustment for confounders.
The Panel notes that the results of two prospective cohort studies (Gast et al., 2009; Geleijnse et al., 2004) are in conflict regarding the risk of CHD associated to vitamin K2 intakes, and that high intakes of vitamin K2 were associated with a significantly lower degree of aortic calcification after adjustment for confounders in one prospective cohort study (Geleijnse et al., 2004).
Two cross-sectional studies (Beulens et al., 2009; Maas et al., 2007) investigated the relationship between vitamin K intake and arterial calcification in women in random samples of the Dutch cohort described by Gast et al. (2009).
The study by Beulens et al. (2009) was undertaken on post-menopausal women (n=564, after exclusion of eight subjects with missing information on coronary calcification and one subject with missing information on vitamin K intakes) who did not use contraceptives or were not on hormone replacement therapy and who underwent (seven to eleven years after enrolment) a multislice multi- detector computed tomography, from which coronary arterial calcium was quantified using the
Agatston calcium score. Presence of calcium in the coronary arteries was defined as a score >0. Energy-adjusted menaquinone intakes (estimated during the year preceding enrolment in the study) were categorised in quartiles. Prevalence ratios of coronary calcification for these quartiles were estimated using Poisson regression, and were adjusted for age (model 1), as well as smoking status, presence of diabetes and hypertension, categories of education, HDL- and LDL-cholesterol (model 2), and, in addition, alcohol consumption and energy-adjusted intakes of protein, calcium and fibre (model 3). Sixty-two percent of the women had coronary calcification. The prevalence ratio of coronary calcification was significantly lower only for the highest quartile of menaquinone intake, compared to the lowest, in models 1 and 3 (respectively 0.82, 95 % CI: 0.68-0.99; and 0.80, 95 % CI: 0.65-0.98). The Panel notes that vitamin K2 intake was inversely related to the presence of coronary calcification, after adjustment for confounders.
The study by Maas et al. (2007) was conducted in women for whom the frozen blood samples and the baseline mammogram, on which calcium deposits along the breast arteries were searched for in one or both breasts, were retrieved. After exclusion of 18 subjects with reported use of vitamin K antagonists and of 29 subjects with no blood sample or baseline general questionnaires, 1,689 women were considered in the analysis. Mean K2 intakes were calculated for each quartile, and for women with or without breast arterial calcification. Adjustment for age, smoking, diabetes, energy-adjusted intakes of saturated, monounsaturated and polyunsaturated fatty acids and protein, and for intakes of calcium was done with univariate analysis of variance (ANOVA) (general linear model). The prevalence of breast arterial calcification was not statistically significantly different (OR 0.7, 95 % CI: 0.5-1.1) between the highest (9 %) and the lowest (13 %) quartile of vitamin K2 intake. After adjustment for confounders, mean vitamin K2 intake was not statistically significantly different between participants with or without breast arterial calcification. The Panel notes that this study showed no association between vitamin K2 intake and breast arterial calcification, after adjustment for confounders.
The Panel notes that one cross-sectional study reported that vitamin K2 intake was inversely related to the presence of coronary calcification (Beulens et al., 2009), while the other (Maas et al., 2007) showed no association between vitamin K2 intake and breast arterial calcification, after adjustment for confounders.
The proposed mechanism by which vitamin K2 could exert the claimed effect is by contributing to the vitamin K-dependent activation (carboxylation) of MGP, a matrix Gla-protein which has been identified in vascular tissue. Increased levels of carboxylated MGP would reduce vascular calcification and decrease the risk of vascular (including coronary) events.
There is some evidence for a role of MGP in preventing calcification of soft tissues. In an MGP knock-out mouse model, spontaneous calcification of soft tissues (mostly arteries) occurred (Luo et al., 1997). Also in the Keutel syndrome, which results from a mutation of the gene encoding the human MGP, patients display several of the same features as the knockout mice, including abnormal calcification of ear and nose cartilage, and of the respiratory tract (Munroe et al., 1999). However, whether changes in vitamin K2 intakes may induce changes in MGP carboxylation, which in turn could affect vascular function (e.g. calcification) or the risk of vascular events, has not been established.
Three additional randomised, placebo-controlled, parallel intervention studies investigated the effects of vitamin K2 supplementation on blood concentrations of (uncarboxylated/carboxylated) MGP (Braam and Vermeer, 2011, unpublished; Moschonis et al., 2011, unpublished; Vermeer, 2008, unpublished). The Panel notes that these studies do not provide any information about how, and to which extent, changes in (uncarboxylated/carboxylated) MGP may affect vascular function, and considers that no conclusions can be drawn from these studies for the scientific substantiation of the claim.
As regards the animal studies, one study (Spronk et al., 2003) investigated the effects of diets containing warfarin and vitamin K1, vitamin K2 (MK-4, 1.5 mg per g of food), or both, on calcification of the aorta and carotid arteries in male rats. The Panel considers that no conclusions can be drawn from a study using a fixed combination for the substantiation of a claim on vitamin K2 alone.
The second study (Schurgers et al., 2007) investigated the effects of diets with different contents of vitamin K1 or vitamin K2 on arterial calcium content and stiffness, total plasma MGP, MGP deposits in vascular tissues, and the content of vitamin K1 and K2 in the aorta, in male rats with preformed arterial calcification. Thirty rats received a six-week diet containing warfarin and 1.5 mg vitamin K1 per g of food (“W&K”) to induce arterial calcification, while the controls (n=18) received vitamin K1 without warfarin. Six rats from the control group were killed to measure the baseline arterial calcium content, and after the six weeks of treatment, six additional control rats and six treated rats were killed to monitor the effect of treatment, while the remaining control rats continued their diet for six additional weeks. The remaining treated rats were divided into four groups (n=6 each), which either continued for six additional weeks the same “W&K” diet with warfarin and vitamin K1, or received diets without warfarin for six weeks, but either with vitamin K1 (“normal” and “high K1 groups”), or with 100 µg vitamin K2 per g of food (“high K2 group” receiving MK-4). Antibodies were raised against, respectively, total, carboxylated and uncarboxylated MGP. Calcium content was assessed in the abdominal aorta and left carotid artery, and the right carotid artery was used for monitoring distensibility and compliance. Immunohistochemistry was performed on the aortic arch and thoracic aorta. The Panel notes that although the outcomes assessed in this study were related to the claimed effect, no information was provided to establish the validity of this animal model for the scientific substantiation of the claim in humans.
The Panel considers that the evidence provided in these animal studies is not sufficient to predict the occurrence of an effect of the dietary intake of vitamin K2 on contribution to the normal function of the heart and blood vessels in humans.
In weighing the evidence, the Panel took into account the absence of human intervention studies from which conclusions could be drawn for the scientific substantiation of the claim, the inconsistency of the results reported in two cross-sectional studies regarding arterial calcification in women, that the results of two prospective cohort studies are in conflict regarding the risk of CHD associated with vitamin K2 intakes, that high intakes of vitamin K2 were associated with a significantly lower degree of aortic calcification in one prospective cohort study after adjustment for confounders, and that the evidence provided for a proposed mechanism is weak.
The Panel concludes that a cause and effect relationship has not been established between the dietary intake of vitamin K2 and contribution to the normal function of the heart and blood vessels.