Effect of Vitamin K on Bone Mineral Density and Fractures in Adults : An updated systematic review and meta-analysis of randomised controlled trials

Purpose: Vitamin K may affect bone mineral density and fracture incidence. Since publication of a previous systematic review the integrity of some of the previous evidence has been questioned and further trials have been published. Our objective was to assess the effectiveness of oral vitamin K supplementation for increasing bone mineral density and reducing fractures in adults. Methods: MEDLINE, EMBASE, CENTRAL, CINAHL, clinicaltrials.gov, WHO-ICTRP were searched for eligible trials. Randomised controlled trials assessing oral vitamin K supplementation that assessed bone mineral density or fractures in adult populations were included. A total of 36 studies were identified. Two independent reviewers extracted data using a piloted extraction form. Results: For post-menopausal or osteoporotic patients meta-analysis showed that the odds of any clinical fracture were lower for vitamin K compared to controls (OR:0.72, 95%CI: 0.55 to 0.95). Restricting the analysis to low risk of bias trials reduced the OR to 0.76 (95%CI: 0.58 to 1.01). There was no difference in vertebral fractures between the groups (OR:0.96, 95%CI: 0.83 to 1.11). In the bone mineral density meta-analysis, percentage change from baseline at the lumbar spine was higher at 1 year (MD:0.93, 95% CI: -0.02 to 1.89) and 2 years (MD:1.63%, 95%CI: 0.10 to 3.16) for vitamin K compared to controls; however, removing trials at high risk of bias tended to result in smaller differences that were not statistically significant. At 6 months it was higher in the hip (MD:0.42%, 95%CI: 0.01 to 0.83) and femur (MD:0.29%, 95%CI: 0.17 to 0.42). There was no significant difference at other anatomical sites. Conclusions: For post-menopausal or osteoporotic patients there is no evidence that vitamin K affects bone mineral density or vertebral fractures, it may reduce clinical fractures however the evidence is insufficient to confirm this. There are too few trials to draw conclusions for other patient groups.