Abstract

Introduction

Vitamin K is a fat soluble vitamin found in two natural forms: phylloquinone (vitamin K1) and menaquinones (collectively known as vitamin K2), which differ from phylloquinone in length and saturation of the side chain [1*] (Figure 1). Phylloquinone, found predominantly in dark green leafy vegetables and vegetable oils is the primary dietary form of vitamin K in Western diets [2,3]. Menaquinones, which are found in animal based foods such as dairy and meats, as well as in fermented foods, are thought to contribute less than phylloquinone to overall vitamin K intakes in Western diets [4–6]. The current recommended Adequate Intake for vitamin K set by the United States’ Institute of Medicine is 120 micrograms and 90 micrograms per day for adult males and females respectively [3]. Vitamin K intake has shown great variation among age groups and geographic location [7]. Among those at greater risk for low vitamin K status are older adults [8,9].

Open in a separate window

Figure 1

Forms of Vitamin K.

The only known function of vitamin K is as an enzymatic co-factor for the post-translational carboxylation of certain proteins (called vitamin K-dependent proteins). Carboxylation confers function to these proteins. While the most common vitamin K-dependent proteins are clotting proteins, vitamin K dependent proteins have been discovered in several extra-hepatic tissues and have important physiological functions, for example in soft-tissue calcification and bone metabolism [1*,10–12]. Additionally, vitamin K has been shown to have anti-inflammatory effects, through mechanisms that appear to be independent of its role as an enzymatic co-factor [13–15]. Vitamin K and vitamin K-dependent proteins have been linked to several age-related diseases in observational and intervention studies. The purpose of this review is to summarize key population-based studies and randomized trials focused on vitamin K nutritional status and age-related health outcomes in community-dwelling adults, namely inflammation, cardiovascular disease and osteoarthritis (Table 1). Vitamin K has been implicated in age-related bone loss and studied extensively in that regard. This body of literature was reviewed extensively in 2014 [16], so studies focused on bone loss and fractures are not included here. Additionally, kidney function declines with age and a unique role for vitamin K in kidney disease has been proposed, which has been thoroughly reviewed [17]. Therefore studies focused on kidney disease are not included here.

Vitamin K and inflammation

Aging is characterized by a chronic low-grade pro-inflammatory state [18]. Age related increases in C-reactive protein (CRP), interleukin- (IL-6), and tumor necrosis factor-α (TNF-α) resulting in the low-grade inflammation appear to contribute to the onset and progression of chronic aging diseases including cardiovascular disease, osteoarthritis and other chronic diseases [19–22].

In vitro and animal experiments have found vitamin K suppresses production of pro-inflammatory cytokines [13–15]. At this time, however, the relevance of vitamin K nutritional status to inflammation in humans remains unclear. In a cross-sectional analysis of the Framingham Offspring (n= 1,381; mean age = 59±8 years), higher vitamin K intake was associated with lower inflammation overall and with lower concentrations of several individual pro-inflammatory biomarkers [11]. In a secondary analysis of the PREDIMED trial, a Mediterranean diet intervention being conducted in Spain, (n= 510; mean age = 67.2±6 years) participants who increased their dietary phylloquinone intake the most (≥70 mcg/d) had the greatest reductions in IL-6, and TNF-α over 1 year. In this analysis, the intervention and control groups were combined, which is not consistent with the RCT design [23]. In both studies [11,23] vitamin K intake was estimated using self-report measures, which are inherently limited (as recently reviewed [24**]). Phylloquinone is found in generally healthy foods, so it cannot be discounted that the reported associations reflect a healthy diet rather than phylloquinone intake specifically.

Nutritional biomarkers are considered more objective measures of nutrient status and reflect nutrient intake, metabolism, and absorption [25]. There are multiple biomarkers indicative of vitamin K status, but no single biomarker is considered the ‘gold-standard’ [24**]. Circulating phylloquinone, a global indicator of vitamin K status, has been evaluated in relation to inflammation. In the Framingham Offspring higher plasma phylloquinone was also associated with lower inflammatory-burden cross-sectionally, consistent with the findings of vitamin K intake [11]. Higher serum phylloquinone was also associated with lower inflammation cross-sectionally in a multi-ethnic cohort of adults without clinically apparent CVD (the Multi-ethnic Study of Atherosclerosis, MESA, n= 662; mean age=62±10 years) [26].

Vitamin K status can also be estimated by measuring the uncarboxylated fractions of certain vitamin K dependent proteins in circulation, as recently reviewed [24**]. Osteocalcin (OC) is a vitamin K dependent protein synthesized in bone and higher circulating uncarboxylated OC (ucOC) reflects low vitamin K status [24**]. In the Framingham Offspring, ucOC, however, was overall not associated with inflammation [11]. That plasma phylloquinone, but not ucOC, was associated with an inflammatory burden, suggests vitamin Ks role in inflammation is independent of its role as an enzyme factor. Since this has not been well studied, it is premature to draw conclusions about the mechanism underlying the apparent anti-inflammatory effects of vitamin K. Additionally, these studies are cross-sectional [11,26], so causal relationship between vitamin K status and inflammation is uncertain.

Randomized trials can address causality and inflammatory measures have been evaluated as secondary outcomes in two trials conducted in older adults. However, neither phylloquinone [27] nor menaquinone-7 [28] supplementation reduced circulating inflammatory biomarkers in older men and women over 3 years. Both of these studies enrolled generally healthy older adults, who are less likely to have substantial increases in inflammation, so any ability to see an effect of vitamin K (or any nutrient) in reducing inflammation may have been blunted. It is plausible the anti-inflammatory effects of vitamin K may be more relevant to groups with a higher inflammatory burden.

Vitamin K, arterial calcification, and cardiovascular disease

Cardiovascular disease (CVD) is the leading cause of mortality in adults ≥ 65 years old [29]. Coronary artery calcification (CAC) is indicative of subclinical CVD and predicts clinical cardiovascular events and all-cause mortality [30–33]. A role for vitamin K in CVD has been proposed, based on the presence of vitamin K-dependent proteins, such as matrix Gla protein (MGP), in vascular tissue [34,35*]. When MGP is carboxylated, which requires vitamin K, it inhibits calcification in arterial and other soft tissues.

The association between vitamin K status, CAC, and CVD has been evaluated, with equivocal results, as reviewed in 2012 [36*]. Results of more recent studies are also inconclusive. In a case-cohort analysis of the MESA, low serum phylloquinone was associated with a 34% higher odds of CAC progression over 3 years, but statistical significance was not reached [OR(95%CI) 1.34(0.94–1.90)] [37]. In secondary analysis, low serum phylloquinone was associated with a 2-fold higher odds of CAC progression in persons treated for hypertension [OR (95% CI): 2.37 (1.38 – 4.09)] but was not associated with CAC progression in persons not treated for hypertension. Although this was not hypothesized a priori, it was replicated in a post hoc analysis of a phylloquinone supplementation trial [37]. The circulating concentration of phylloquinone considered sufficient is not yet defined clinically, so in this study low plasma phylloquinone was defined as < 1.0nmol/L, which is the concentration achieved when Adequate Intakes are met [8,38]. In a sub-study of the Dutch Prospect cohort, post-menopausal women (n=508 mean±SD age 57±5 years), with higher plasma phylloquinone, defined as >0.7 nmol/L, had a higher prevalence of CAC [39]. However in this study, CAC was defined as absent or present at a single time point measured 7 – 11 years after plasma phylloquinone was measured, so it is not known if CAC was actually prevalent at the time phylloquinone status was assessed. Furthermore, this analysis did not account for triglycerides, which may have confounded the findings. Phylloquinone is transported on triglyceride-rich lipoproteins [40], so adjustment for triglycerides is imperative in studies utilizing circulating phylloquinone as a biomarker, especially in relation to CVD since elevated triglycerides are a CVD risk factor [41].

Assays that measure uncarboxylated MGP in plasma have been developed [42] and are now commercially available. The dephosphorylated uncarboxylated MGP ((dp)ucMGP) responds to changes in vitamin K intake [43,44] and is thought to be a functional indicator of vitamin K status in tissues that use MGP. Higher plasma (dp)ucMGP reflects lower vitamin K status. (Dp)ucMGP has been evaluated in relation to cardiovascular outcomes in clinical observational studies. (There are other circulating forms of MGP, some of which have been evaluated in relation to CVD, but do not reflect vitamin K status [42], so are not considered here.) In 195 post-menopausal women analyzed cross-sectionally, higher (dp)ucMGP was associated with more CAC, although statistical significance was borderline (p=0.065) [45]. In a secondary analysis of the randomized controlled trial that found phylloquinone supplementation reduced CAC progression in older men and women, (dp)ucMGP was reduced by phylloquinone supplementation, but the change in (dp)ucMGP did not correlate with change in CAC [43]. In type II diabetics in the Dutch European Prospective Investigation into Cancer and Nutrition (EPIC) cohort, each standard deviation increase in (dp)ucMGP (reflecting lower vitamin K status) was associated with a 21% higher risk for CVD over 11 years of follow-up (HR(95%CI) 1.21 (1.06–1.38) [46]. However, a case-cohort analysis of the same cohort, did not find (dp)ucMGP to be associated with coronary heart disease or stroke [47]. In contrast, a longitudinal study conducted in 577 community dwelling older men and women in the Logitudinal Aging Study Amersterdam (LASA) who were free of CVD at baseline found that after a mean follow-up of 5.6+1.2 years, individuals with the highest circulating (dp)ucMGP had a 2-fold significantly higher risk of CVD [HR(95%CI) 2.69(1.09–6.62)] [48].

A growing body of evidence suggests arterial calcification is also implicated in arterial stiffness [49–51]. Arterial stiffness increases with age and is also an independent risk factor for CVD [52–54]. Examination of data from the National Health and Nutrition Examination Surveys 2007–2008, 2009–2010 of older adults (n=5296, age >50 yrs) showed that inadequate phylloquinone intake was a significant independent predictor of high arterial stiffness [55], although it is plausible this association is due, in part, to participants reporting low phylloquinone intake also consuming a less healthy diet. A cross sectional analysis of the Czech post-MONICA study found individuals in the highest dp-ucMGP quartile had a higher risk of elevated aortic stiffness [OR(95% CI): 1.73 (1.17–2.5)] [56]. Similar findings were obtained in a cross-sectional analysis of a family based study in Switzerland, in which (dp)ucMGP was positively associated with arterial stiffness after adjusting for various confounders [57]. Menaquinone-7 supplementation improved some parameters of arterial stiffness over 3 years in post-menopausal women (n=244; mean±age = 60±3) in a randomized trial designed test the effect of menaquinone supplementation on bone strength, but measured arterial stiffness as a secondary outcome [28]. It is not known if these findings are generalizable to men or other groups.

There are two systematic reviews examining the association between vitamin K status and CVD. The first one, published in 2010, suggested a beneficial effect of menaquinone intake, but not phylloquinone intake, in lowering CVD risk [58]. This is primarily due to three studies conducted in the Netherlands that all reported inverse associations between menaquinone intake and CVD [59–61]. Whether or not these findings generalize to other countries or nationalities is not known. Menaquinone intake was assessed using food frequency questionnaires, which carry inherent limitations [62]. At the time these studies were conducted food composition databases for menaquinones were limited. (They are currently being expanded) [24**]. It is therefore premature to draw conclusions regarding the relative importance of phylloquinone and menaquinones with respect to CVD based on the available studies. A systematic review of vitamin K and CVD published in 2015 sought to include only vitamin K supplementation trials of at least 3 months in duration and conducted in healthy adults or adults at high risk for CVD [63]. Ultimately only one trial that tested the effect of menaquinone-7 supplementation on blood pressure and lipid levels over 12 weeks 60 men and women 45–60 years old was included in the review [63,64]. Overall, the authors concluded there is insufficient evidence to conclude vitamin K affects CVD [63]. Of note, this review did not include the only randomized trial designed to test the effect of phylloquinone supplementation on CAC progression in older adults (n=388, mean±SD; age = 68±6 yrs) because the intervention and control groups both received calcium and vitamin D (to assure all participants were replete in those nutrients); hence the trial lacked a pure placebo group [27,63]. This trial found older adults who adhered to the 3-year phylloquinone supplementation intervention had significantly less CAC progression compared to adherent participants in the control group, suggesting a protective effect of phylloquinone against subclinical CVD. In the intent-to-treat analysis, however, phylloquinone supplementation did not significantly affect CAC progression [27]. In a subsequent post-hoc analysis of this trial it was found that phylloquinone supplementation reduced CAC progression in participants treated for hypertension but did not affect CAC progression in participants not treated for hypertension [37]. While this suggests vitamin K may be particularly beneficial to hypertension therapeutic regimens, these findings need to be confirmed in studies designed to study treated hypertensives.

Although some observational studies suggest improving vitamin K status may reduce subclinical and clinical CVD, data are conflicting [27,37,39,59,61]. Additional prospective studies are necessary to determine whether increasing vitamin K intake decreases risk for cardiovascular events and subclinical CVD and whether this is modulated by (dp)ucMGP.

Vitamin K and osteoarthritis

Osteoarthritis is the leading cause of lower-extremity disability in older adults, and there is currently no therapy known to reduce osteoarthritis progression. Osteoarthritis is characterized by pathological changes in all joint tissues, including cartilage and bone. Vitamin K has been implicated in osteoarthritis because vitamin K dependent proteins are found in cartilage and bone [65–67]. MGP is among the most studied vitamin K dependent protein expressed in human cartilage. Uncarboxylated MGP (the nonfunctional form) is elevated in human arthritic cartilage, while carboxylated (functional) MGP is more abundant in healthy cartilage [65]. Gla rich protein (GRP), which is another vitamin K dependent implicated in calcification, has recently identified in human articular cartilage [68]. Similar to MGP, GRP from arthritic cartilage is primarily uncarboxylated, whereas GRP from healthy cartilage is primarily carboxylated [69].

In a cross-sectional analysis of older men and women from Japan, low vitamin K intake was associated with a higher prevalence of radiographic knee osteoarthritis [70]. In a cross-sectional analysis of the Framingham Offspring, low plasma phylloquinone, was associated with higher knee and hand osteoarthritis prevalence [71]. In the Multicenter Osteoarthritis (MOST) Study (mean±SD age = 62±8 yrs), participants with subclinical vitamin K deficiency (defined as plasma phylloquinone < 0.5 nmol/L) were 1.5 – 2 times more likely to develop radiographic knee osteoarthritis and cartilage damage over 30 months (risk ratio (95%CI) 1.56(1.08–2.25) and 2.39(1.05–5.40) respectively, compared to those without subclinical deficiency) [72]. In the Health Aging and Body Composition (Health ABC) Study (mean±SD age = 74±3 yrs), older community-dwelling adults with very low plasma phylloquinone (defined as below the assay limit of detection, <0.2 nmol/L) had a 1.7-and 2.6-fold higher odds of worsening cartilage damage and meniscal damage over 3 years. This analysis also suggested participants with very low phylloquinone were more likely to have bone attrition, subarticular cyst, and osteophyte progression, but statistical significance was not reached [OR (95% CI): 1.9(0.9–3.6); 1.5(0.8–2.7); 1.5(0.8–2.8) respectively] [73]. In this same study, individuals with elevated (dp)ucMGP (reflecting low vitamin K status) were more likely to have osteophytes, bone marrow lesions, subarticular cysts, and meniscus damage cross-sectionally, but this was not associated with progression of any structural abnormalities in the longitudinal analysis [73]. This may suggest vitamin Ks role in these pathologies is independent of its function as an enzymatic co-factor in the carboxylation of MGP. Neogi et al assessed the effect of phylloquinone supplementation on radiographic hand osteoarthritis using x-rays obtained at the end of study and did not find any effect of 3 years phylloquinone supplementation [74]. No baseline measures of hand osteoarthritis were available in this study. When participants with baseline plasma phylloquinone <1.0nM were analyzed separately, there was a trend towards less joint space narrowing (a measure of radiographic osteoarthritis) in the phylloquinone supplemented group, suggesting persons with low vitamin K status are more likely to benefit from vitamin K supplementation, but this finding needs to be confirmed in trials designed specifically to test the effect of vitamin K on osteoarthritis development and progression. In this same cohort, MGP genotype was found to be associated with hand OA, but circulating total MGP concentrations were not [75].

While collective data suggest a protective role for vitamin K in osteoarthritis, several questions remain. Low circulating phylloquinone was associated with more OA in three cohorts, but low circulating phylloquinone has not been consistently defined [72,73,75]. It is not clear what level is ‘sufficient’ in terms of joint health. Vitamin Ks role in osteoarthritis was proposed because vitamin K dependent proteins are present in cartilage and bone. However, alternate mechanisms may exist. Osteoarthritis has been characterized as having an inflammatory component with measureable cytokines and inflammatory mediators present in the synovium of the joint resulting in joint pain, swelling, and stiffness [22]. Vitamin K appears to have anti-inflammatory effects [13–15], suggesting an alternate pathway through which vitamin K may affect joint health – an area of research that merits attention. Clinical trials are needed to evaluate the efficacy of vitamin K supplementation in OA development and progression.

Conclusion

There is accumulating evidence to support a protective role for vitamin K in chronic aging conditions and diseases, such as inflammation, cardiovascular disease, and osteoarthritis, but there are also inconsistencies in the studies conducted to date. It is therefore premature to make recommendations regarding vitamin K’s efficacy in improving inflammation, and cardiovascular and joint health. Clinical trials designed to test the effect of vitamin K supplementation on these chronic age-related conditions are needed. As the aging population continues to grow, the prevalence of these diseases will rise dramatically. Identifying and understanding nutritional factors that impact the progression or treatment of these diseases is imperative to address the health and function of older adults worldwide.

Acknowledgments

Footnotes

Contributor Information

Stephanie G. Harshman, Jean Mayer Human Nutrition Research Center on Aging, Tufts University 711 Washington Street, Boston, MA 02111, Phone number: 617-556-3151, Fax number: 617 556 3149.

M. Kyla Shea, Jean Mayer Human Nutrition Research Center on Aging, Tufts University 711 Washington Street, Boston, MA 02111, Phone number: 617-556-3073, fax number: 617 556 3344.

References