Article Text


Boning up on vitamin K
  1. Department of Medicine, Box 157
  2. Level 5, Addenbrooke's Hospital
  3. Hills Rd, Cambridge CB2 2QQ,UK
  4. jec1001{at}

Statistics from

See article on page 473

The nature and significance of the relationship between vitamin K status and bone health has been debated for some years. Vitamin K is required for the gamma carboxylation of glutamic acid residues in three bone proteins, namely osteocalcin, matrix Gla protein, and protein S. Although the roles of these proteins have not been clearly defined, there is evidence that osteocalcin, which is produced by cells of the osteocyte/osteoblast lineage and has three vitamin K dependent gamma carboxyglutamic acid (Gla) residues, may be involved in the mineralisation of bone matrix1; in addition, it may function as a negative regulator of bone formation, deletion of the osteocalcin gene in mice resulting in increased bone mass.2 Approximately 30% of synthesised osteocalcin is released into the circulation and serum levels of the protein are widely used as an indicator of bone formation.3 Vitamin K deficiency is associated with a reduction in circulating osteocalcin concentrations and a decrease in the gamma carboxylated fraction, the latter being associated with reduced binding of the protein to bone mineral.4 Despite this theoretical basis for adverse effects of vitamin K deficiency in the skeleton, however, direct evidence is lacking. Thus although low vitamin K levels and undercarboxylation of osteocalcin have been described in patients with osteoporosis,5 6 these findings may reflect poor nutritional status rather than a specific effect of vitamin K deficiency on bone. Similarly, there is currently no direct evidence that therapy with warfarin, which inhibits the secretion by osteoblasts of osteocalcin,7 increases the risk of osteoporosis, and although increased fracture risk has been reported in patients receiving oral anticoagulants, this finding has not been universal.

Malabsorption of fat soluble vitamins is a recognised complication of Crohn's disease and, in particular, an increased prevalence of vitamin D deficiency has been reported in many studies.8 In rare cases this results in osteomalacia but more importantly it contributes to the increased prevalence of osteoporosis associated with Crohn's disease as a consequence of secondary hyperparathyroidism and the resulting increase in bone turnover. In contrast, vitamin K status has been less well studied although in one study a high prevalence of deficiency of the vitamin was found in patients with ileal Crohn's disease.9 The study by Schoon et al in this issue of Gut, performed in a cohort of patients with small intestinal Crohn's disease, provides some evidence, albeit indirect, that there may be an inverse association between vitamin K status and bone mineral density of the spine (see page 473).10 Thus undercarboxylation of osteocalcin was higher and binding of circulating osteocalcin to hydroxyapatite lower in patients with Crohn's disease compared with controls, indicating reduced vitamin K status. Furthermore, serum levels of undercarboxylated osteocalcin were inversely related to bone mineral density in the spine, although this relationship was not observed for proximal femur or total body bone mineral density. Interestingly, although there was a significant correlation between serum levels of the major circulating metabolite of vitamin D, 25-hydroxyvitamin D, and serum vitamin K, no correlation was observed between vitamin D status and bone mineral density at any of the sites assessed.

The results of this study would thus be consistent with a role for vitamin K deficiency in the pathogenesis of osteoporosis associated with Crohn's disease. Nevertheless, further studies are required in view of the relatively small sample size in this study (n=32) and lack of a statistically significant correlation between serum vitamin K levels and bone mineral density. In addition, the reduction in bone mineral density was relatively modest in this cohort, with mean z scores higher than −0.5 at all sites, and in this respect it would be of interest to investigate the relationship between vitamin K status and bone mass across a wider spectrum of bone disease. Failure to demonstrate a correlation between the serum “free” (undercarboxylated) osteocalcin and bone mineral density in the femoral neck and total body may reflect inadequate statistical power or alternatively might indicate differential effects on cancellous and cortical bone.

The pathogenesis of osteoporosis in patients with Crohn's disease is multifactorial, glucocorticoid therapy, hypogonadism, vitamin D deficiency, and malnutrition all being potential contributory factors.8 In the study of Schoon et al, patients were in remission at the time of assessment and were treated with no more than 5 mg of oral prednisolone daily. Nevertheless, bone mineral density values assessed at one point in time reflect multiple past and present influences and thus the effects of glucocorticoids and disease activity could not be excluded completely, as the authors claimed. The question of whether vitamin K supplementation prevents or reduces bone loss in patients with Crohn's disease (or in other populations) has not been established and this will be an important area for future research if an association between vitamin K status and bone mineral density is confirmed in subsequent studies.

See article on page 473


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles