Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.

Association of Vitamin K and Chondrocalcinosis of the Knee: The MOST and Framingham Osteoarthritis Studies.

Misra3,  Devyani, Booth8,  Sarah L., LaValley3,  Micheal, Toltsykh6,  Irina, Nevitt6,  Micheal, Lewis5,  C. Elizabeth, Torner7,  James

Boston Univ Schl of Med, Boston, MA
Boston University School of Medicine, Boston, MA
BUSM, Boston, MA
BWH, Harvard University, Boston, MA
UAB, Birmingham, AL
UCSF, San Francisco, CA
U-Iowa, Iowa City, IA
USDA HNRCA, Tufts University, Boston, MA


Vitamin K (VK) plays a key role in regulation of calcification in soft tissue. A knockout mice model demonstrated that in absence of a VK–dependent protein (MGP), there was excessive calcification of cartilage. Because chondrocalcinosis (CC) has a low prevalence in the general population, we studied the association of VK with CC in two cohorts: The Multicenter Osteoarthritis (MOST) Study and the Framingham Osteoarthritis Study, to obtain sufficient numbers for this study.


Participants were from the MOST Study (cohort of persons with or at high risk for knee osteoarthritis (OA) and the Framingham Osteoarthritis Study (community-based cohort unselected for OA) who had knee x-rays obtained and serum VK (phylloquinone) measured. X-rays for both studies were read for presence of CC at baseline and follow up visits by the same academically based MSK radiologist. A person was classified as having CC if it was present in either knee at either of two time points. We examined the association of VK with prevalent CC in both cohorts combined, with VK assessed as 1) dichotomous variable, deficient (serum phylloquinone <0.5 nM) vs not deficient and 2) quartiles, using Poisson regression to calculate prevalence ratios (PR), adjusting for age, sex, BMI, BMD, 25(OH)-D, warfarin use, race, cohort and history of knee injury. We also performed local regression (LOESS), which avoids assuming a parametric form, plotting phylloquinone concentrations (log-transformed) against prevalence of CC. Effect measure modification by cohort was tested separately.


Among 1795 participants (1180 from MOST, 615 from Framingham; mean age 59±9.2 yrs, 59% women, mean BMI 29±5.6 kg/m2, mean serum phylloquinone concetration 1.7±1.8 nM, range 0–22nM), there were 126 subjects with prevalent CC. Those who were VK deficient had 40% higher prevalence of CC compared with those who were not VK deficient, although this was not statistically significant (PR 1.4, 95% CI 0.8–2.4, p=0.3), and did not change with exclusion of warfarin users. Although all VK deficient individuals (n=173) were in the lowest quartile of VK, the analysis using serum VK quartiles did not demonstrate a dose-response relation, but rather suggested a potential U-shaped relation (adjusted PR: 1.0 (ref), 0.8, 0.9, 0.9 for quartiles 1 through 4, respectively). However, the local regression plot did not suggest any linear or other association (Fig. 1). There was no effect measure modification by cohort.


Although there seems to be a plausible biologic rationale for an association of deficient VK with CC, we were unable to find such an association in our study using serum phylloquinone. Because this association remains unclear, further studies are warranted to explore potential effects of VK on extra-skeletal mineralization, using different measures of VK.

Fig. 1. Local regression (LOESS) plot of association of vitamin K with prevalent chondrocalcinosis.

To cite this abstract, please use the following information:
Misra, Devyani, Booth, Sarah L., LaValley, Micheal, Toltsykh, Irina, Nevitt, Micheal, Lewis, C. Elizabeth, et al; Association of Vitamin K and Chondrocalcinosis of the Knee: The MOST and Framingham Osteoarthritis Studies. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :872
DOI: 10.1002/art.28640

Abstract Supplement

Meeting Menu