Vitamin K deficiency in adults leading to clinical bleeding is almost unheard of, except as a consequence of hepato-intestinal disorders which disturb the absorption or utilisation of the vitamin. Use of warfarin or other anti-coagulant drugs, i.e. vitamin K antagonists, in the management of thromboembolic disease, reduces circulating concentrations of vitamin K-dependent clotting factors. The anticoagulants inhibit the biosynthesis of prothrombin and other vitamin K-dependent factors in the liver and others factors in extrahepatic tissues, leading to clotting factor deficiencies in the body (Olson, 1999). Studies investigating the effect of anticoagulants on bone health have shown mixed results but the participants were severely ill with chronic vascular disease which would have affected physical activity, a very important factor in bone health and may be the reason why vitamin K status did not appear to be important (Weber, 2001).
Newborn infants are a special case in vitamin K nutrition for several reasons: lipids are not easily transferred across the placenta, the neonatal liver is immature with respect to prothrombin synthesis, breast milk is low in vitamin K and the infant gut is sterile at birth. As a result of this unique combination of factors infants can develop a condition known as haemorrhagic disease of the newborn (HDN). The term HDN was first used in 1894 by Townsend who described it as a self-limiting bleeding disorder of newborns of unknown origin (Shearer, 2000). Until the late 1960s HDN was thought to be a problem of the first week of life; however, two other forms have now been identified: late HDN, a more serious condition, that occurs between weeks 3 and 6 of life and early HDN occurring in the first 24 hours, probably caused by antagonistic drugs taken during pregnancy. Late HDN is rare, about 4-70 cases per 100000 births but world-wide is a significant cause of infant morbidity and mortality. Late HDN has a high incidence of intercranial haemorrhage resulting in death or severe and permanent brain damage in 50% of cases. One risk factor for both the late and classical forms of HDN is exclusive breast-feeding, which may be aggravated by the low concentrations of vitamin K in breast milk 1-10 mg/L (cf. formula milk ~50 mg/L). A second risk factor might be the 'precarious' hepatic stores of vitamin K found in newborns possibly due to poor placental transfer (Shearer, 2000). As a result of this phenomenon the COMA panel recommended all babies should be given prophylactic vitamin K at birth (Whitehead, 1991). Recent interest in late HDN has come from a reported increase in the syndrome since the early 1980s. In many developed countries the increased incidence followed a decline in vitamin K prophylaxis since its introduction in the 1940s and an increase in exclusive breastfeeding. Prophylaxis was often given as an intramuscular dose at birth but since 1992 concerns about an epidemiological association between the intramuscular route and later childhood cancer has seen a shift towards the oral route of administration (Shearer, 2000). Work published by Autret-Leca and Jonville-Bera (2001) suggest oral administration of a single dose of vitamin K protects against classical and early vitamin K deficiency bleeding but that intramuscular propha-lylaxis is needed for late HDN. Although the risk of solid tumours associated with vitamin K administration is unlikely there remains a low potential risk of lym-phoblastic leukaemia in childhood. Autret-Leca and Jonville-Bera suggest formula-fed neonates without risk of haemorrhage should receive 2mg oral dose of vitamin K followed by a second 2mg dose between days 2 and 7. Infants exclusively or nearly exclusively breast-fed should receive weekly oral administration of 2mg (25 mg/day) vitamin K, after the first two doses of the vitamin, until completion of breast-feeding. High-risk neonates (premature, neonatal disease, hepatic disease, maternal drugs inhibiting vitamin K activity) should receive the first dose intramuscularly or by a slow intravenous route. Subsequent doses should be administrated according to the clotting factor profile specific to each infant.
Despite gaps in knowledge it appears that sub-optimal vitamin K status may play a role in osteoporosis. Patients with osteoporotic fractures tend to have very low plasma levels of vitamin K. Much of this evidence is circumstantial based on associations of measures of vitamin K status with fractures or biomarkers of bone metabolism and some of the evidence is conflicting (Binkley and Suttie, 1995; Shearer, 1995; Vermeer et al, 1995). French workers have shown an age-dependent impairment of g-carboxylation of osteocalcin and strong associations of ucOC with hip fracture risk (Szulc et al, 1993) and bone mineral density (BMD) (Szulc et al, 1994). The work also showed a potential influence of vitamin D on g-carboxylation, which may indicate interactions between vitamins D and K, an area requiring further investigation. It has also been shown that 'matrix Gla protein' which is an important regulator of bone strength and growth is determined by a vitamin A response gene (Gudas et al, 1991). To what extent vitamins A and K interact at the gene level is not known.
A number of adult intervention studies have been carried out, particularly on postmenopausal women, which have shown vitamin K to be effective in reducing ucOC in serum. Takahashi et al (2001) used 89 osteoporotic patients with vertebral fractures, 24 patients with hip fractures, 43 pre- and 48 post-menopausal
Japanese women. They gave either a daily dose of 45 mg vitamin K2 alone or vitamin K2 plus1 mg 1-a hydroxyvitamin D3 or vitamin D alone. After four weeks of treatment with vitamin K alone or vitamin K plus D, ucOC was significantly decreased, but was not changed in those who received vitamin D alone. There was a disproportion of ucOC/intact OC in postmenopausal women and those with hip or vertebral fractures, vitamin K and vitamins K plus D markedly decreased the ratio of ucOC/intact OC to approximately 80%, but vitamin D did not decrease the ratio. This work confirmed previous studies (Plantalech et al, 1990; Douglas et al, 1995; Schaafsma et al, 2000).
The first intervention study to look at the influence of vitamin K on bone strength was published by Akjba et al (1991) who recruited 17 dialysis patients losing bone mass due to renal insufficiency. They supplemented the patients with 45 mg vitamin K2 for 1 year and measured bone mass at different points on the skeleton. They found loss of bone was reduced in the vitamin K group. The findings were confirmed in subsequent studies by Orimo et al (1992) who carried out a placebo-controlled trial studying 546 patients with osteoporosis to whom they gave either 45mg vitamin K2 or 1 mg 1-a-hydroxy vitamin D3 for 48 weeks. Arm BMD increased by 2.1% in the vitamin K group but decreased by 2.4% in the vitamin D group (P < 0.001); no difference in vertebral BMD was found. A more recent study reported that 45mg vitamin K2 or placebo given for 2 years to 241 osteoporotic women increased BMD and significantly reduced occurrence of new fractures (14 in the vitamin K vs. 35 in the placebo group) (Shiraki et al, 2000). Finally, preliminary data from a placebo-controlled, randomised, clinical trial on 244 post-menopausal women aged 60-85 years, after a 2 year intervention period, showed a significant increase in the BMD of the distal radius in a group receiving 200 mg vitamin Kj plus 10 mg vitamin D and 1 g calcium (Bolton-Smith et al, 2001).
Was this article helpful?