Vitamin K is thought to be important for bone health because it activates several proteins involved in bone formation. Poor dietary intake of vitamin K (mainly found in dark green leafy vegetables) is associated with bone loss and fractures. Giving supplements of the main dietary form of vitamin K (called K1) or another common form which our bodies make from K1(called MK4), to improve bone health have given mixed results. This confusion is thought to have arisen because these studies involved people who already had enough vitamin K or did not have osteoporosis. We want to test the hypothesis that treatment with bisphosphonates combined with vitamin K, in vitamin K deplete elderly women with osteoporosis, may offer additional benefit on skeletal metabolism and reduction of fracture risk. We want to test this by measuring vitamin K status in post-menopausal women with osteoporosis who are on the recommended treatment with a bisphosphonate and calcium/vitamin D supplements. Those with low vitamin K will then be recruited to study the effect of supplementation with either K1 or MK4.

Vitamin K is important for skeletal health. Vitamin K is essential for the carboxylation of several Gla proteins in bone which are implicated in bone formation and mineralization. These include osteocalcin (OC) and matrix Gla protein (MGP). Carboxylation of the glutamic acid residues of these proteins optimises their function. Vitamin K occurs as either phylloquinone (vitamin K1) which is the major dietary form or menaquinones (MKs or vitamin K2) which are mainly of bacterial origin. MK4 of the vitamin K2 series has additional, carboxylation-independent, functions including the regulation of osteoblastic specific markers such as alkaline phosphatase (BALP), and osteoprotegerin (OPG) and has inhibitory effects on osteoclast activity. Several observational studies have shown that low vitamin K status is associated with low bone mineral density (BMD) and increased fracture risk, although proof of causality is lacking. The results of several placebo-controlled clinical trials of vitamin K1 and MK4 have been conflicting with some, but not all, showing a positive effect of vitamin K1 on BMD or bone turnover. Positive fracture efficacy has been demonstrated with high-dose MK4, although most trials were on Japanese women. These intervention studies may have been hampered by the study design such as inclusion of vitamin K replete subjects or healthy non-osteoporotic women. The use of vitamin K in the prevention of bone loss and/or fractures in high-risk post-menopausal women with osteoporosis who are vitamin K deplete merits further investigations. The prevalence of low vitamin K stores is high in elderly subjects with osteoporosis. Preliminary data in Japanese women suggest that combined treatment with a bisphosphonate and vitamin K, at least vitamin K2 (MK4), appears to have an additive beneficial effect on BMD and bone resorption. There have been no such studies in a caucasian osteoporotic population. We want to test the hypothesis that treatment with bisphosphonates combined with vitamin K, in vitamin K deplete elderly women with osteoporosis, may offer additional benefit on skeletal metabolism and reduction of fracture risk.

The first part will be a cross-sectional study of post-menopausal women with osteoporosis aged between 60-80 years who are on treatment with bisphosphonate. Their vitamin K status will be determined and those patients who are found to have low vitamin K concentrations defined as <0.35 ug/ml will be invited to take part in an 18 months prospective randomised placebo controlled trial.

Eligible patients will be randomised to 3 arms (35 patients in each arm). All 3 groups will continue to receive weekly oral bisphosphonate (commonly Alendronate 70 mg weekly) and adjunctive calcium/vitamin D supplements (1.0g of calcium and 800 I.U of cholecalciferol). The control arm (Group A) will receive placebo. Group B will receive 1.0mg daily of vitamin K1 and MK4 placebo. Group C will receive vitamin K2 (MK4) 45 mg daily and vitamin K1 placebo. Patients will be seen at baseline and at 3, 6, 12 and 18 months. Changes in BMD at the lumbar spine, hip, fore-arm at 18 months and the biochemical parameters at each time point will be compared between the groups.

Inclusion Criteria:

Inclusion in the cross-sectional part of the study which involves assessment of vitamin K status

1. Informed consent to screening stage : assessment of vitamin K status
2. serum vitamin K concentration < 0.35 ug/ml

Inclusion into the randomised controlled trial

1. ambulatory post-menopausal women aged between 55-85 years 2. Post-menopausal osteoporosis ( history of previous fragility fractures or BMD evidence of osteoporosis or osteopenia with at least one clinical risk factors such as low BMI, positive family history of osteoporosis) 3. Treatment with a bisphosphonate and calcium/vitamin D supplements for at least 12 months 4. Informed written consent 5. e GFR >30 ml/min 6. normocalcaemia

• Exclusion Criteria:

  1. Age <55 years, or > 85 years
  2. Male gender
  3. severe renal impairment (CKD stage 4 and 5)
  4. poor mobility (inability to walk 100 yards unaided)
  5. malabsorption (extensive bowel surgery, short bowel)
  6. generalised carcinomatosis
  7. glucocorticoid therapy
  8. inflammatory disorders (e.g. active rheumatoid arthritis, inflammatory bowel disease requiring oral glucocorticoids),
  9. endocrine diseases (e.g. primary hyperparathyroidism, hyperthyroidism).
  10. chronic liver disease
  11. current treatment with teriparatide, strontium ranelate
  12. Participation in a trial with an investigational product within the previous 3 months
  13. Serum vitamin K > 0.35 µg/ml
  14. patients on anti-coagulants such as warfarin