Loading...
 
Toggle Health Problems and D

Improved health due to vitamin D fortification in Finland

Vitamin D Insufficiency in Young Finnish Men: Associations with bone stress fracture and respiratory tract infections
ACADEMIC DISSERTATION ILKKA LAAKSI June 2012

Table of contents


Highlights by VitaminDWiki

  • The importance of sunlight for human health was recognised by Hippocrates in ancient Greece:
    he believed that the southern side of the hill, receiving the most sunlight each day in the northern hemisphere, was the healthiest place to live.
  • Finland added vitamin D fortification to milk (20 IU per 100 grams) and margarine in 2003
    US, Canada is 40 IU per 100 grams ( 100 IU per cup (cup = 8 ounces = 250 grams)
  • Vitamin D levels increased in one year (where insufficiency = 16 nanograms)
    from 78% deficient to 35% insufficient (worse in winter?)
    Smokers had lower levels of vitamin D (29 vs 32 nanograms)
  • Those with stress fractures had less vitamin D
  • April 2010, the Finnish National Nutrition Board recommended doubling fortification to 40 IU per 100 grams (equal to that of US and Canada)
  • RCT added 400 IU in the winter – probability of improved health was 29%
  • As of 2011 Finland recommended 400 IU if < 2 years old and for pregnancy
    and 800 IU for those > 60 years old
  • Author recommends an additional that adults get 800 IU of vitamin D in the winter {minimum for white skin, healthy, not obese, etc.}

See also VitaminDWiki

Change of Vitamin D Fortification in Denmark


LIST OF ORIGINAL PUBLICATIONS
This thesis is based on the following articles, which are referred to in the text by Roman numerals I to IV.

  • I: Laaksi IT, Ruohola J-P, Ylikomi TJ, Auvinen A, Haataja RI, Pihlajamaki HK, Tuohimaa PJ
    Vitamin D fortification as public health policy: significant improvement in vitamin D status in young Finnish men. Eur J Clin Nutr. 2006 Aug;60(8):1035-8.
  • II: Ruohola J-P1, Laaksi IT1, Ylikomi TJ, Haataja RI, Mattila VM, Sahi T, Tuohimaa PJ, Pihlajamaki HK
    Association between serum 25OHD concentrations and bone stress fractures in Finnish young men. J Bone Miner Res. 2006 Sep;21(9):1483-8. 1 These authors contributed equally to this paper
  • III: Laaksi I, Ruohola J-P, Tuohimaa P, Auvinen A, Haataja R, Pihlajamaki H, Ylikomi T
    An association of serum vitamin D concentrations <40 nmol/l with acute respiratory tract infection in young Finnish men. Am J Clin Nutr. 2007 Sep;86(3):714-7.
  • IV: Laaksi I, Ruohola J-P, Mattila VM, Auvinen A, Ylikomi T, Pihlajamaki H
    Vitamin D supplementation for the prevention of acute respiratory tract infection: a randomized, double-blinded trial among young Finnish men. J Infect Dis. 2010 Sep 1;202(5):809-14.

ABSTRACT

Vitamin D is not an actual vitamin but a secosteroid hormone produced in the skin from 7-dehydrocholesterol after exposure to sunlight's ultraviolet B radiation. Vitamin D needs to be hydroxylated twice to reach an active form that is able to regulate gene expression through binding with vitamin D receptors (VDRs) and further to vitamin-D-responsive elements (VDREs) in vitamin-D-responsive genes. The level of the major circulating form of the hormone, serum 25OHD, is used for determination of vitamin D status. Vitamin D insufficiency can be regarded as a global issue with substantial implications for health. On account of inadequate sun exposure in wintertime, vitamin D insufficiency is commonplace among all age groups in Finland. As sunlight exposure is inadequate for vitamin D production in the skin, nutrition and supplements are the main sources of vitamin D in northern countries during winter months. Upon the recommendation of the Ministry of Social Affairs and Health, vitamin D has been added to commercial milk products and margarines since February 2003 in Finland.

In the first study, we determined the effects of national policy on vitamin D fortification in young Finnish men. The study population consisted of 196 young Finnish men (18-28 yrs) whose serum 25OHD concentrations were determined with Octeia® enzyme immunoassay by IDS in January 2003 (n = 96) or in January 2004, one year after national vitamin D fortification started.

We found a 50% increase in mean serum 25OHD3 concentrations after implementation of the vitamin D fortification of dairy products. In addition, the prevalence of vitamin D insufficiency (<40 nmol/l) had decreased by 50% (from 78% in January 2003 to 35% in January 2004). The study showed that national vitamin D fortification substantially improved the vitamin D status of young Finnish men. However, 35% remained vitamin D insufficient.

The most commonly known function of vitamin D is the effect on bone mineralisation. Bone stress fractures are one of the most frequently seen types of overuse injuries in athletes and military recruits. An association was recently shown between vitamin D and bone mineral content, with a correlation between low femoral bone density and stress fractures. In the second study, we measured the serum 25OHD concentration in a population sample of military recruits to determine whether vitamin D is a predisposing factor for bone stress fractures. In this prospective study, 800 healthy Finnish military recruits with a mean age of 19 years were followed up for development of stress fractures in homogenous circumstances. Serum 25OHD concentrations were measured with enzyme immunoassay at entry into military service, and the weight, height, body mass index (BMI), physical fitness score, and result of a 12-minute running test were measured for all subjects. In all, 756 subjects had completed the study at the end of 90-day follow-up, and subjects without a fracture constituted controls. The study found 22 recruits with a stress fracture (2.9%), the incidence being 11.6 (95% CI: 6.8-16.5) per 100 person-years. In the final multivariate analysis, the statistically significant risk factor for stress fracture in conscripts was below-median serum 25OHD level (75.8 nmol/l) OR being 3.6 (95% CI: 1.1-11.1). No statistically significant associations between BMI, age, physical fitness score, 12-minute running test or smoking and bone stress fractures were found in this study population. In conclusion, a lower serum 25OHD concentration may be a generally predisposing factor for bone stress fractures.

Vitamin D has a role in innate immunity activation; the production of antimicrobial peptides following toll-like receptor (TLR) stimulation by pathogen lipopeptides is dependent on a high enough level of 25OHD. Recent evidence suggest that differences in the ability of human populations to produce vitamin D may contribute to susceptibility to microbial infection. In the third study, we explored whether an association exists between vitamin D insufficiency and acute respiratory tract infection in young Finnish men. For this prospective study, young Finnish men (n = 800) serving at a military base in Finland were enrolled. Serum 25OHD concentrations were measured in July 2002 and the subjects were followed up for six months, and the number of days of absence from duty due to respiratory infection was calculated. The mean serum 25OHD concentration was 80 nmol/l in July 2002 (n = 756). The subjects with serum 25OHD concentrations <40 nmol/l, indicating vitamin D insufficiency, had statistically significantly more days of absence from duty due to respiratory infection. Also, a statistically significant positive association between serum 25OHD concentrations and the amount of physical exercise before induction into military service was found. In addition, smoking was statistically significantly associated with lower serum 25OHD concentrations. In conclusion, the study showed that a low vitamin D level increases the risk of acute respiratory tract infections.

There is clinical evidence of an association between vitamin D insufficiency and respiratory tract infections. There is also some evidence of prevention of infections by vitamin D supplementation. In the fourth study, we determined the effect of vitamin D supplementation on the incidence of acute respiratory tract infections in young Finnish men. For this RCT, 164 healthy conscripts were enrolled. From October to March, half of them received 10 ug of vitamin D daily and half received a placebo. Smoking was adjusted for in the study's analysis. The mean serum 25OHD concentrations were 79 nmol/l in October 2005 and 72 nmol/l in March 2006 in the vitamin D group. The corresponding concentrations in the placebo group were 74 and 51 nmol/l. There was no statistically significant difference in the number of days of absence from duty (the main outcome variable) between the vitamin D and placebo group. However, the proportion of men remaining healthy throughout the six-month study period was greater in the vitamin D group (51%) than in the placebo group (36%), p = 0.045. Further, in a Cox regression analysis with adjustment for smoking, the adjusted hazard ratio (HR) for absence from duty due to a respiratory tract infection was lower in the vitamin D group (HR 0.71; 95% CI: 0.43-1.15). The RCT showed some evidence of a preventive effect of vitamin D supplementation against respiratory tract infection. Larger randomised controlled trials are warranted to explore this preventive effect.

INTRODUCTION

Vitamin D is produced in the skin from 7-dehydrocholesterol through a photolytic reaction induced by ultraviolet B (UVB) radiation from sun exposure. It is a precursor of the hormone 1,25(OH)2D that is formed in two hydroxylation reactions, first in the liver and then in the kidneys or in target organs (Holick et al. 1980). Vitamin D regulates calcium and phosphorous homeostasis in the body the process that is essential for bone mineralisation (Holick 2003). Clinically, vitamin D deficiency is known to lead to secondary hyperparathyroidism, causing rickets in children and both osteomalacia and osteoporosis in adults (Compston 1998).

Vitamin D status is determined through measurement of the major circulating form of the hormone, serum 25OHD (Utiger 1998). The scientific consensus is that vitamin D sufficiency can be defined as serum 25OHD >75 nmol/l (Dawson-Hughes et al. 2005). Vitamin D insufficiency is a global issue that can be regarded as epidemic among adults without sufficient sunlight exposure. At high northern latitudes, diet is the most important source of vitamin D during winter, because of inadequate sunlight exposure. Vitamin D insufficiency is commonplace in all age groups in the wintertime in Finland (Laaksi 2011). Since February 2003, vitamin D fortification of liquid milk products and margarines has been implemented in Finland. Furthermore, the Finnish National Nutrition Board renewed the recommendation of vitamin D fortification in April 2010.

Vitamin D regulates gene expression by vitamin D receptors (VDRs). An active form of vitamin D binds to the VDR that dimerises with the retinoic X receptor. The complex then binds to vitamin-D-responsive elements inside vitamin D-responsive genes (Rachez et al. 2000). Vitamin D contributes to bone health by increasing both intestinal calcium absorption and renal calcium reabsorption (Utiger 1998). Vitamin D is also known to decrease parathyroid hormone (PTH) secretion and enhance the differentiation of both osteoblast and osteoclast precursors. Thus it has a major effect on bone mineralisation and peak bone mass (Valimaki et al. 2004). In a randomised controlled trial (RCT), a rise in the mean 25OHD level from 53 to 74 nmol/l through vitamin D supplementation was shown to reduce fracture risk at the hip, forearm, and spine in people >65 years of age (Trivedi et al. 2003). In another study, vitamin D inadequacy was linked to increased risk of hip and other nonvertebral fractures (LeBoff et al. 1999). The findings support the hypothesis that vitamin D inadequacy predisposes young Finnish men to bone stress fractures, a common overuse injury in athletes and in those performing their military service.

Toll-like receptors (TLRs), monitoring the host for the presence of pathogens, are stimulated by pathogens. This stimulation upregulates VDR
expression and vitamin D hydroxylase genes, leading to the production of antimicrobial peptides (Wang et al. 2004). Further, the induction of cathelicidin messenger RNA (mRNA) was shown to be lower in the presence of serum from African Americans, which contains less 25OHD than does serum from Caucasians (Liu et al. 2006). In the case of respiratory pathogens accumulated in the airways, increased activation of vitamin D and then increased levels of cathelicidin mRNA have been shown in another study. Again, respiratory epithelial cells can activate vitamin D and create a microenvironment with high levels of the active form of the vitamin D upregulating the cathelicidin antimicrobial peptide gene, an important component of innate immunity in the lungs. Local vitamin D activation may be an important component of host defence. (Hansdottir et al. 2008; Hansdottir et al. 2011) This hypothesised mechanism could also explain the differences in proneness to acute respiratory tract infections in young Finnish men during their military service.

REVIEW OF THE LITERATURE

Vitamin D

History

The importance of sunlight for human health was recognised by Hippocrates in ancient Greece: he believed that the southern side of the hill, receiving the most sunlight each day in the northern hemisphere, was the healthiest place to live.


In the early 17th century, Francis Glisson, Daniel Whistler, and Arnold de Boot discovered rickets, a bone disease in children that is characterised by a poorly mineralised and deformed skeleton. Later, in the early 19th century, Jedrzej Sniadecki noticed that children living in the city of Warsaw had high incidence of rickets whilst children in rural regions did not develop the disease. He thought that increased exposure to sunlight in the countryside prevented these children from developing rickets. Further, in the late 19th century Theodore Palm noted that children living in equatorial countries did not develop rickets. It was then found that rickets incidence has a geographical background related to sunlight exposure. (Holick 1994)

In the late 19th century, Niels Finsen exposed subjects with cutaneous tuberculosis to artificial sunlight when working to find a cure for tuberculosis. He found that producing moderate sunburn to a small area of affected skin with intense light caused the superficial skin layer to peel away but exposed normal and healthy skin underneath. This phototherapy improved the skin in almost all cases. Later it was used routinely as a treatment for pulmonary tuberculosis, one for which Finsen was awarded a Nobel Prize in 1903. (Zasloff 2006)

In 1919, Edward Mellanby reported that a restrictive diet led to rickets in dogs. Investigating the cause of rickets and seeking a cure, he found an exclusively oat-based diet to induce rickets. In addition, he found that adding cod-liver oil to the diet cured the dogs within a few months, so he concluded that a component of cod-liver oil (that oats do not contain) was essential in preventing rickets. The dogs in Mellanby's experiments were raised indoors with no exposure to sunlight. This led to a historical accident: vitamin D became classified as a vitamin and not as a hormone - i.e., a compound that is synthesised in the body and acts as a chemical messenger from one cell, or group of cells, to another - in contrast to a vitamin as a biologically active organic compound that is not synthesised in the body but obtained solely through the diet. Cod-liver oil entered systematic use in Germany in 1823 and soon after that also in Finland as a folk medicine. (Holick 1994)

In 1922, Elmer McCollum discovered the compound vitamin D. He investigated the chemical composition of cod-liver oil, which was known to prevent night blindness and fractures. He was anxious to find out whether cod-liver oil retained its properties when heated and oxygenated. This was tested through heating and oxygenation of cod-liver oil. It was found to prevent fractures still but no longer night blindness, meaning that two different active compounds are involved. The compound destroyed this way was named vitamin A, and the heat-stable component became known as vitamin D. (Mohr 2009)

In 1923, Harry Goldblatt and Katherine Soames established that when 7-dehydrocholesterol is irradiated with UV light, a form of a fat-soluble vitamin is produced. Alfred Hess and Mildred Weinstock further substantiated this to be vitamin D. They also showed that wheat, lettuce and vegetable oils displayed antirachitic activity after exposure to UV irradiation. In 1924, Harry Steenbock found that animals were able to produce vitamin D and suggested that it was not a dietary constituent but UV light that was essential in protection against rickets. Steenbock concluded that irradiation of food substances is beneficial for protection against rickets in children. (Steenbock 1924; Holick 1994)

Adolf Windaus received a Nobel Prize in 1928 for the discovery that exposure to UV was responsible for vitamin D synthesis. He showed that ultraviolet light has an isomerising effect on ergosterol, a steroid component of fungal-cell membranes. He was able to determine the chemical structure of vitamin D; vitamin D2, produced by ultraviolet irradiation of ergosterol, was characterised in 1932, and the discovery of the precursor of vitamin D3 (7-dehydrocholesterol, 7-DHC) occurred in 1937, when it was noticed that only those foods containing cholesterol could cure rickets after being irradiated with UV light. Windaus also showed that the antirachitic component of cod-liver oil was identical to vitamin D3. Soon after the structure of vitamin D was characterised and a process developed for its synthesis, vitamin D was directly added to milk, making ultraviolet-irradiation of milk unnecessary. (Mohr 2009)

In the early 20th century, rickets was a significant world health problem. In industrialised cities of northern Europe and the north-east United States, it was epidemic and >80% of young children who died (of whatever cause) had clinical manifestations of rickets (Holick 1994). As it had been shown that rickets could be cured by exposure to sunlight or ingestion of UV-irradiated food, there were two competing concepts in the prevention of rickets (McCollum et al. 1995).

Metabolism and mechanism of action

The chemical structure of vitamin D is closely related to that of classic steroid hormones (Norman 2008). There are two forms of vitamin D: the main source in humans is skin photosynthesis following exposure to sunlight producing vitamin D3 (cholecalciferol), and the other is vitamin D2 (ergocalciferol) produced by ultraviolet radiation in a variety of plant materials and yeasts (Norman 1998). Sunlight is the main source of vitamin D and usually contributes up to 90% of serum concentration of vitamin D. The remaining 10% comes from dietary intake and vitamin D supplements. In fur-bearing animals and many birds, sunlight is not able to reach the skin; vitamin D is synthesised in their fur or feathers, and these animals then eat the resulting cholecalciferol by licking the fur or rubbing the beak on the feathers. (Lehmann et al. 2010)

Solar UV radiation consists of UVA, UVB, and UVC, depending on the wavelength of the photons. Almost all UVC photons are usually absorbed completely by the ozone layer and do not reach the earth at all. In the two innermost layer of the skin's epidermis, the stratum basale and stratum spinosum, solar ultraviolet-B radiation at a wavelength of 290-315 nm, with its maximum spectral effect at 297 nm, penetrates the skin and converts 7-DHC (provitamin D3) to previtamin D3. Less than 1% of previtamin D3 is formed at UVA wavelengths. The action spectra for erythema and vitamin D3 production are similar, but maximal synthesis is achieved at suberythemogenic UVB doses, meaning a few minutes of summer sun exposure (MacLaughlin et al. 1982; Wolpowitz et al. 2006). Longer exposure does not increase one's stores of vitamin D but does increase the possibility of DNA damage. Approximately two thirds of human cutaneous 7-dehydrocholesterol can be found in the epidermal layer of the skin, the remaining third being in the dermis (Lehmann et al. 2010). The amount of previtamin D3 is less than 15% of the substrate 7-DHC available (Webb et al. 1988). In ageing, the amount of 7-DHC begins to decline; when exposed to the same amount of sunlight, a person who is 70 years of age makes only VV of the vitamin D that a 20-year-old person can make (MacLaughlin et al. 1985; Holick et al. 1989).

Previtamin D3 is thermodynamically unstable and is isomerised to form vitamin D3 by a non-enzymatic and highly temperature-dependent process. The formation of vitamin D3 is relatively rapid, reaching a maximum within 12-24 hours after UVB exposure. Half of the previtamin D3 can isomerise to vitamin D3 within 2.5 hours in the skin. Vitamin D3 moves to the extracellular space and binds to vitamin D binding protein (DBP), entering the circulation. (Lehmann et al. 2010)

Both the intensity of the UV radiation and the level of skin pigmentation regulate the rate of previtamin D3 formation but not the maximal level achieved. The intensity of ultraviolet radiation from sunlight varies with the season, latitude, and time of day. The farther from the equator, the less solar exposure to produce D3 there is in the course of a year, as the rays enter at a more oblique angle (zenith angle) and more UVB photons are absorbed by the ozone layer, because of the greater distance. Generally, at a latitude below 37°, vitamin D synthesis occurs in the skin throughout the year. On the other hand, in the early morning or late afternoon, vitamin D3 production is ineffective in the skin even in summer, because of the oblique zenith angle. (Webb et al. 1988; Holick 2004) Both clothing and use of sunscreen prevent vitamin D3 production (Matsuoka et al. 1987; Matsuoka et al. 1992). Melanin is an effective natural sunscreen that absorbs UVB photons. People with greater skin pigmentation require longer sunlight exposure for vitamin D3 production. (Jorgensen et al. 2010)

In the event of excess UVB radiation, previtamin D3 will be degraded into inactive sterols lumisterol and tachysterol, or be retransformed to 7-DHC. In addition, vitamin D3 will be converted to inactive suprasterols I and II and 5,6 transvitamin D3, with avoidance of vitamin D intoxication. (Holick et al. 1979; Holick et al. 1980; Holick et al. 1981)

Nutritional vitamin D from the diet or dietary supplements is transported from the intestine in chylomicrons via lymph veins and released into the liver (Bouillon et al. 1998). Both vitamin D synthesised in the epidermis and from the diet or dietary supplements are biologically inert and require hydroxylations by mitochodrial P450s in order to become active. Firstly, vitamin D is hydroxylated by 25-hydroxylase (CYP27A1) present in the liver to form 25-hydroxyvitamin D3 (25OHD3), the major circulating form of vitamin D. The production of 25OHD3 is not significantly regulated but is mainly dependent on substrate concentration, and approximately 40% of vitamin D3 is converted into 25OHD3 (Kato 2000). After hydroxylation, 25OHD3 enters the circulation. It has a half-life of about 15 days (Jones 2008).

The 25OHD3 is transported to the kidneys for successive hydroxylation by 1a-hydroxylase (CYP27B1) to form the most biologically active form of vitamin D, 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] and finally to target tissues for biological response (Holick et al. 1980; Norman 1998). The serum level of 1,25(OH)2D3 is usually 75-200 pmol/l, being a thousandth that of 25OHD3 (25-200 nmol/l). The serum half-life of calcitriol is 10-24 hrs (Levine et al. 1985). There is evidence that also 25OHD3 has hormonal activity (Lou et al.2004).

The expression of CYP27B1 has been documented in extrarenal tissues also: the skin, macrophages, lymph nodes (Hewison et al. 2000), normal and malignant colon tissue (Matusiak et al. 2005), and normal and malignant breast tissue (McCarthy et al. 2009), among other tissue types. This hydroxylation is tightly upregulated by parathormone (i.e., PTH), hypocalcaemia, and hypophosphataemia and downregulated by hypercalcaemia, hyperphosphataemia, and 1,25(OH)2D3 itself (Hewison et al. 2000).

Inactivation of 25OHD3 and/or 1,25(OH)2D3 is catalysed by a mitochondrial P450 (CYP24A) that 24-hydroxylates these to form 24,25(OH)2D3 or by 1a,24,25(OH)3D3 principally in the proximal tubules of the kidneys (Guryev et al. 2003).

There has been an assumption of equivalence of vitamin D2 and D3, as these forms differ only in the side chain to the sterol skeleton. However, in recent studies orally administered vitamin D3 has increased serum 25OHD up to 1.7 times more efficiently than does vitamin D2 when administered in molarly equal amounts for two weeks. In a study with single doses of vitamin D3 or vitamin D2, they led to equal increase in serum 25OHD levels, but those levels persisted or even continued to rise after three days in the vitamin D3 group while they declined and reached the baseline value in two weeks in the vitamin D2 group (Trang et al. 1998; Armas et al. 2004). Vitamin D2 has been shown to have a diminished ability to bind to DBP in the blood and a shorter shelf life as a supplement (Houghton et al. 2006). There is a trend to replace vitamin D2 with vitamin D3 in fortified dairy products or supplementation (Thacher et al. 2010).

Biological actions of vitamin D are mediated through a vitamin D receptor that belongs to the nuclear hormone receptor family. This has been found in cells involved in maintaining calcium homeostasis but also in others, such as a number of cancer cells (Christakos et al. 2011). As bound to 1,25(OH)2D3, the VDR forms a heterodimer with a retinoid X receptor (RXR). This complex acts as a transcription factor depending on the coregulatory proteins that are attached to the complex. These complexes will be bound to vitamin D response elements in the promoter region of target genes in which they modulate their transcription. (Kato 2000; Goltzman 2010)

In addition, as other hormones do, vitamin D mediates its effects partially through nongenomic actions by activating several signal transduction systems. In genomic action, the regulation of gene expression by vitamin D is mediated by VDRs and takes place within hours. It is likely that nongenomic response is mediated by a specific membrane-bound VDR and occurs within seconds to minutes. Increases in intracellular calcium levels, stimulation of intestinal calcium transport flux, and activation of protein kinase C are examples of the nongenomic action of vitamin D. (Lehmann et al. 2010)

Physiological actions of vitamin D

Effect on bone and calcium metabolism

The principal function of vitamin D is to maintain mineral homeostasis both by increasing calcium and phosphorus absorption from the intestine and by enhancing their renal reabsorption (Holick 2003). If normal serum calcium cannot be maintained by intestinal absorption, increased PTH secretion mobilises bone calcium and induces the synthesis of 1,25(OH)2D3. In children, vitamin D deficiency inhibits the calcification of cartilage and causes rickets. In adults, by contrast, the bone matrix (the osteoid) is not mineralised, causing osteomalacia and osteoporosis due to a reduction in bone mass. An increased risk of hip and other fractures in adults is a known consequence of vitamin D deficiency. (Compston 1998; Utiger 1998; Lips et al. 2011)

A major consequence of the loss of VDR activity is a defect in intestinal calcium and phosphate absorption, which is the primary cause of decreased bone mineralisation. For example, the phenotype of VDR-knockout (KO) mice includes rickets, osteomalacia, and increased PTH secretion, but when the mice are fed a high-calcium, high-phosphorus diet, calcium and PTH levels are normalised and osteomalacia and rickets are prevented. It is clear that vitamin D is vital for mineral homeostasis. (Christakos et al. 2011)

Transcellular transport of calcium is activated by 1,25(OH)2D3 in the duodenum, where VDR is expressed in the highest concentration and upregulates genes responsive for transportation of calcium. Most of the absorption at low and moderate levels of calcium is due to this active transcellular transport. During higher calcium intake, calcium is also absorbed in the intestine by passive diffusion paracellularly (Heaney et al. 2005). Usually calcium intake and absorption are directly related, but when calcium intake is lowered, the percentage from calcium intake that is absorbed can increase from 25% to 37% because of increased serum concentrations of PTH and 1,25(OH)2D3 (Dawson-Hughes et al. 1993; Hunt et al. 2007). A change in serum 25OHD concentration from 50 to 86 nmol/l has been shown to cause a statistically significant increase in calcium absorption in the intestine (Heaney et al. 2003).

Vitamin D has also been proved to have direct effects on bone. in vitro studies have shown that 1,25(OH)2D3 can stimulate the formation of osteoclasts by upregulating osteoclast differentiating factor (Margolis et al. 2010).

Other actions of vitamin D

Vitamin D may also have rather interesting and important non-classical actions. The active form of vitamin D, 1,25(OH)2D, has been shown to promote insulin secretion, inhibit cell proliferation, and stimulate cell differentiation (Bikle 2008). Vitamin D has apoptotic effects on prostate cancer cells in vitro. It has effects on cancer invasion and angiogenesis (Ylikomi et al. 2002). In epidemiological studies, vitamin D has been negatively associated with breast and colon cancers (Cannell et al. 2006). Also, vitamin D deficiency in infancy has been linked to diabetes (Hypponen et al. 2001; Oilinki et al. 2012), hypertension, multiple sclerosis (Correale et al. 2009; Pierrot-Deseilligny et al. 2010), and some other cancers (Holick 2004). Vitamin D insufficiency has also been associated with depression in men and with severe asthma exacerbation in childhood (Brehm et al. 2010; Lee et al. 2011).

Vitamin D metabolites have clinical utility in that 1,25(OH)2D3 and analogues are successfully used for treatment in cases of, for example, hyperproliferative skin disease and psoriasis (Guryev et al. 2003).

The role of vitamin D in immunity

Innate immunity is one part of the first-line barriers in host defence and responds rapidly to microbes that operate as an activator of the system. Antimicrobial peptides (AMPs) are evolutionarily ancient weapons in the battle against bacteria, viruses, and fungi that are available for activation even before the body has recognised an antigen (Zasloff 2002). AMPs have been found both in epithelial tissues and in phagocytic blood cells, in which they damage the lipoprotein membranes of microbes (Laube et al. 2006). In the moist airways, these proteins are secreted as a thin layer of fluid lying above the apical surface of the epithelium and below the mucous layer covering the epithelial surface. Binding of microbes to the epithelium provokes the expression of AMPs, creating a barrier that is chemically lethal to microbes. Some AMPs are secreted constitutively but others, such as LL-37 (cathelicidin), as a response to stimuli from microbial components (Zasloff 2002). Cathelicidin acts also as an attractant for macrophages and neutrophils (Laube et al. 2006).

Interaction with the immune system is one of the non-classic effects of vitamin D. Liu et al. were able to show that synthesis of vitamin D by sunlight exposure upregulates the expression of AMPs. Specifically, the stimulation of toll-like receptors, which monitor the host for the presence of microbe antigens, by pathogens upregulates the expression of two genes in monocytes - a VDR and CYP27B1, converting 25OHD3 to active 1,25(OH)2D3 - and leads to the production of AMPs. The incubation of activated monocytes with 1,25(OH)2D3 produce cathelicidin in a dose-dependent manner. (Liu et al. 2006) Cathelicidin concentrated in phagocytic vacuoles destroys M. tuberculosis in in vitro conditions, and this effect is more pronounced after macrophages are exposed to 1,25(OH)2D3. The gene-encoding cathelicidin has a vitamin-D-responsive element (VDRE) within the promoter region, and it is positively regulated by vitamin D (Wang et al. 2004). This gene and CYP27B1 are both expressed by the airway epithelium, indicating that active vitamin D can be produced locally within the lungs, leading to the production of cathelicidin (Hansdottir et al. 2011). Local vitamin D activation might be an important component of innate immunity in the lungs (Hansdottir et al. 2008).

Vitamin D regulates the activation of human T cells, controls T cell antigen receptor signalling, and enhances the recognition of microbe antigens by T lymphocytes. Human T cells have very low expression of signalling proteins (phospholipase C), correlating with low T cell antigen receptor (TCR) responsiveness. TCR triggering upregulates PLC expression and correlates with greater TCR responsiveness and T cell activation. The induction of PLC is dependent on vitamin D and expression of VDRs as a VDR bound to 1,25(OH)2D3 activates the gene-encoding PLC. (von Essen et al. 2010)

Vitamin D and respiratory tract infections

The respiratory mucosal epithelia, including those of the middle ear and pharyngotympanic tube, secrete antimicrobial peptides as part of innate immunity that inhibit the growth of otitis media pathogens: Haemophilus influenzae, Moraxella catarrhalis, and Streptococcus pneumoniae (Lee et al. 2004; Vlaminckx et al. 2005).

Higher vitamin D status might protect against influenza. The prevalence of respiratory infections showed a strong seasonal pattern inversely related to 25OHD concentrations in a cohort study with 7,000 over-45-year-old participants. Furthermore, after adjustment for adiposity, lifestyle, and socioeconomic factors, each 10 nmol/l rise in serum 25OHD concentration was associated with a 7% lower risk of infection in the study. (Berry et al. 2011)

In a secondary analysis of the Third National Health and Nutrition Examination Survey with almost 19,000 participants, serum 25OHD concentrations were found to be inversely associated with recent upper respiratory tract infection (URTI) after adjustment for season, body mass index (BMI), smoking, asthma, and chronic obstructive pulmonary disease (COPD). An independent association between lower serum 25OHD and URTI was revealed and URTI was reported by 24% of subjects with 25OHD <25 nmol/l, by 20% with 25OHD 25-75 nmol/l, and by 17% with serum 25OHD >75 nmol/l.
(Ginde et al. 2009)

There are some randomised placebo-controlled and double-blind trials of vitamin D supplementation for prevention of respiratory tract infections in adults. A lower rate of self-reported URTI or influenza was observed in the intervention arm as compared to the placebo group in a trial with 208 healthy postmenopausal African American women who were given 20 ug of vitamin D3 daily or a placebo for two years. After the increase of vitamin D3 dosage from 20 ug to 50 ug daily or a placebo for 12 months, the effect was magnified. (Aloia et al. 2007) A follow-up study with 162 healthy adults given 50 ug of vitamin D3 or a placebo daily for 12 weeks showed no benefit with respect to the incidence and the severity of URTI symptoms for the vitamin D group as opposed to the placebo group. However, there was a statistical tendency to favour the vitamin D group.(Li-Ng et al. 2009) Further, Avenell et al. failed to show a statistically significant difference between the vitamin D and placebo group in the secondary end point of self-reported infection rate in a study with 3,500 elderly subjects who were given 20 ug of vitamin D3 or a placebo for more than two years. The study was part of the Randomised Evaluation of Calcium or Vitamin D (RECORD) trial. (Avenell et al. 2007)

Stress fractures

Fractures induced by cyclic loading of normal bones with abnormal forces are described as stress fractures (Anderson et al. 1996). These fractures are common overuse injuries in military populations. At the beginning of Finnish military service, all conscripts perform eight-week basic training, with gradually increasing levels of strain. When completing this training, they are expected to maintain their fitness for battle after a 15 km march or ski journey, with a 25 kg military pack, on two consecutive days. In a study of Finnish male military recruits, 15 of the 179 recruits (8.4%) developed a stress fracture during military service (Valimaki et al. 2005). In the US Army's basic combat training (8 weeks), the reported incidence of stress fractures among male trainees has ranged from 0.9 to 5.2% (Jones et al. 2002). Still, the prevalence of these fractures can be up to 49% in military populations; it is largely dependent on the physical demands of training (Niva et al. 2009).

Stress fractures occur especially near the beginning of military service, when the bones are not adapted to increased mechanical loads. Anatomically, the stress fractures are found mainly (95%) in the lower extremities: the tibial shaft and metatarsal bones. Stress fractures of the pelvis, hip, thigh, and knee are rare but are among the high-risk injuries associated with a tendency for protracted healing and complications such as dislocation of the femoral neck leading to avascural necrosis. (Pihlajamaki et al. 2006; Mattila et al. 2007)

Stress fractures occur as a result of microdamage after repetitive strains to a normal bone (Daffner et al. 1992). Bone is dynamic tissue and is able to adapt its shape and size in response to excessive strain. Bone remodelling is cyclic removal of bone by osteoclasts (resorption) followed by the deposition of new bone by osteoblasts (formation). In bone modelling, the bones are reshaped again by the activation of osteoblasts and osteoclasts. These bone-strengthening processes may last up to six months, and up to 10% of the skeleton undergoes remodelling each year. Unloading of the skeleton will follow rapid bone loss and osteoporosis. (Kiuru et al. 2004; Brandi 2009)

Both a poor aerobic fitness level (Shaffer et al. 1999) and low muscle strength (Beck et al. 2000) at the start of a physical training programme, increased age (Gardner et al. 1988), female sex (Jones et al. 2002), smoking (Altarac et al. 2000), high or low foot-arch height (Simkin et al. 1989), valgus knees ('knock-knees') (Cowan et al. 1996), and low bone mineral density (Pouilles et al. 1989) are some risk factors for development of stress fractures among military populations.

Localised pain after a recent increase in physical training activity is a common symptom of a stress fracture. Usually, the pain eases during rest. Although the clinical findings may be exiguous, tenderness upon palpation and localised oedema might be found. One-leg standing or jumping may reveal high-risk femoral-neck stress fracture. (Kiuru et al. 2004)

For the diagnosis of a stress fracture, plain radiography should be performed initially and repeated after two weeks. In the case of urgent diagnosis for high-risk stress fractures, magnetic resonance imaging needs to be used for its higher sensitivity. (Kiuru et al. 2004; McClung et al. 2010) Treatment of stress fractures includes rest and cessation of the physical activity for 3-8 weeks, depending on the severity and the localisation of the fracture. Return to physical training should be gradual. High-risk stress fractures require surgical consultation and surgery. The prognosis with non-invasive and conservative treatment is usually excellent for non-displaced-femoral-neck stress fractures, though the use of non-weight-bearing crutches is essential. In contrast, the displacement of a femoral-neck stress fracture is treated surgically, because of the risk of serious complications such as avascular necrosis. (Kiuru et al. 2004)

Vitamin D as a factor for prevention of stress fractures

Calcium and phosphate homeostasis has a key role in bone mineralisation and peak bone mass development. Vitamin D is known to contribute to this homeostasis by increasing intestinal calcium absorption, enhancing renal calcium reabsorption, decreasing parathyroid hormone secretion, and affecting both osteoblast and osteoclast differentiation (Holick 2004).

With serum 25OHD concentration below 78-90 nmol/l, there is an increase in serum PTH, showing vitamin D insufficiency (Krall et al. 1989; Chapuy et al. 1997). The increased PTH increases bone turnover and fragility (Szulc et al. 2003). Importantly, in a recent northern European study with about 700 subjects pathologic mineralization defects of bone were seen in crista iliaca biopsies of those with serum 25OHD <75 nmol/l (Priemel et al. 2010). Low 25OHD levels are associated with reduced bone mineral content (BMC) and cortical thickness of the femoral neck in young men (Szulc et al. 2003; Valimaki et al. 2004). In addition, vitamin D supplementation increases bone mineral density (Baeksgaard et al. 1998). In the population study with >13,000 individuals including both younger and older adults BMD increased continuously with higher 25OHD levels. Interestingly, this rise was seen throughout the reference-range of 22.5-94 nmol/l contributing to the development of higher peak bone mass in young men. (Bischoff-Ferrari et al. 2004) Clinically, low vitamin D status is a determinant of increased osteoporotic fracture risk (Lips et al. 2010).

The number of prospective studies of vitamin D insufficiency and stress fractures is quite limited. In a study of over 2,500 Israeli soldiers, 25OHD levels were lower in patients with scintigraphically proved high-grade stress fractures (63 nmol/l) than in control subjects (75 nmol/l) (Givon et al. 2000). In the case of serum 25OHD <50 nmol/l, the risk of stress fracture of the tibia and fibula was double that of those with serum 25OHD >100 nmol/l in a cohort study of 1,200 female US Navy recruits. That study also found an inverse dose-response gradient between 25OHD and stress-fracture risk. (Burgi et al. 2011)

The results as to the association between low calcium intake and stress fracture development are controversial (Myburgh et al. 1990; Cline et al. 1998). In a Finnish study with 220 male conscripts, PTH levels were higher in those who developed stress fractures than among controls. There was no statistically significant difference in the vitamin D status of the groups, though (Valimaki et al. 2005).

In a randomised double-blind and placebo-controlled intervention trial of vitamin D supplementation for the prevention of stress fractures with >5,200 female US Navy recruits, the participants were given either 2,000 mg of calcium and 20 ug of vitamin D or a placebo daily during eight weeks' basic training. The intervention reduced the incidence of stress fractures by 20%, according to intention-to-treat analysis (Bouillon 2008; Lappe et al. 2008). Further, a recent study found that two polymorphisms in the VDR may independently increase the risk of stress fractures (Chatzipapas et al. 2009; McClung et al. 2010).

Vitamin D insufficiency

Serum 25OHD is the best marker for vitamin D status, since both high serum concentrations in the summer and low concentrations in the winter have been observed at northern latitudes, reflecting the amount of exposure to the sun (Chapuy et al. 1997). Vitamin D sufficiency can be detected from the increase in serum parathyroid hormone secretion; it starts rising at 25OHD cut-off levels of 78-90 nmol/l (Krall et al. 1989; Chapuy et al. 1997; Harkness et al. 2005). Additionally, it seems that the serum 25OHD threshold for PTH increase is lower in African Americans than among Caucasians (Aloia et al. 2010). Because of the relationship of skin pigmentation to vitamin D synthesis, serum 25OHD levels are highest in Caucasians, lowest in African Americans, and intermediate in Hispanics (Zadshir et al. 2005). On the other hand, PTH has an inverse relationship to skin pigmentation; African Americans have the highest and Caucasians the lowest plasma levels of PTH (Aloia et al. 2006).

Sunlight is the main source of vitamin D and produces it very effectively in skin cells: full-body exposure to UVB radiation of 15-20 minutes is able to produce up to 250 ug of vitamin D (Stamp et al. 1977). Of note is that 1 ug of vitamin D orally administered each day increases circulating serum 25OHD levels from 0.6 to 1.2 nmol/l (Heaney 2007).

In northern countries, sun exposure is inadequate for vitamin D production because of the greater distance through the atmosphere in wintertime (Norman 1998). Therefore, dietary intake and vitamin D supplements are sources of vitamin D from October to March in Finland. The foods that naturally contain substantial amounts of vitamin D are fish-liver oils, fish, mushrooms, and egg yolks. Among seafood, the lamprey, pike-perch, whitefish, Baltic herring, and salmon are excellent sources of vitamin D3, though, farmed salmon contain only 1/4 the vitamin D3 found in wild salmon. (Lu et al. 2007; Lehmann et al. 2010)

Vitamin D insufficiency can be regarded as epidemic among adults with inadequate exposure to sunlight (Holick 2003). It is common in children during wintertime in Europe. In elderly people, it lasts throughout the year, and serum 25OHD3 concentrations in institutionalised people are even lower (McKenna et al. 1998; Schleithoff et al. 2006). Also, vitamin D insufficiency has been very commonplace among young men, young girls, and healthy adults in winter in Finland (Lamberg-Allardt et al. 2001; Valimaki et al. 2004). High prevalence of vitamin D insufficiency and secondary hyperparathyroidism have been found in African immigrants to Finland (Islam et al. 2012).

The fortification of liquid milk products (0.5 ug / 100 ml) and spreads since February 2003, following the recommendation of the Finnish Ministry of Social Affairs and Health, has improved vitamin D status and increased vitamin D intake in Finland. However, approx. 20% of Finnish people still had inadequate vitamin D status in 2004 in winter, especially young girls, young adults, and middle-aged women (Lehtonen-Veromaa et al. 2008). It seems that vitamin D fortification of milk products did not resolve hypovitaminosis D in Finland (Valimaki et al. 2007). In April 2010, the Finnish National Nutrition Board recommended that vitamin D fortification be increased to 1 ug / 100 ml in liquid milk products and 20 ug / 100 g of fat spreads (but not butter) (Pietinen et al. 2010).

There is growing consensus on boundaries between stages of vitamin D inadequacy: <25 nmol/l for deficiency, 25 to 49.9 nmol/l for insufficiency, 50 to 75 nmol/l for hypovitaminosis, and >75 nmol/l for sufficiency (Dawson-Hughes et al. 2005). With these thresholds, over 90% of the more pigmented and around 75% of the Caucasian population in the USA are vitamin D insufficient. The prevalence of vitamin D insufficiency has also doubled in 10 years in these populations. (Adams et al. 2010)

Finally, American Endocrine Society recommends the screening of serum 25OHD in patients at risk for vitamin D deficiency. The society do not recommend population screening for vitamin D, though. Based on the Endocrine Society's consensus vitamin D deficiency is defined as a serum 25OHD below 50 nmol/l and vitamin D insufficiency as a serum 25OHD of 50-75 nmol/l. (Holick et al. 2011) On the other hand, the 2011 Report on Dietary Reference Intakes for Ca and Vitamin D from the Institute of Medicine in the USA concludes that scientific evidence supports a key role of vitamin D in skeletal health. However, for extraskeletal outcomes, including immune disorders, the evidence is still inconsistent. According to the report existing evidence suggests that vitamin D sufficiency is achieved at serum 25OHD levels of >50 nmol/l.

There is also evidence that very high values are not associated with greater benefit but U-shaped associations can be observed showing risks at both low and high levels of vitamin D for some outcomes (Ross et al. 2011).

AIMS OF THE STUDY

The aims of the present study were:
1. To study vitamin D status and the effect of vitamin D fortification as a national health policy on serum vitamin D concentrations in young Finnish men (I).
2. To study the association between bone stress fractures and low serum vitamin D concentrations in young Finnish men (II).
3. To study the association between vitamin D insufficiency and acute respiratory tract infections in young Finnish men (III).
4. To evaluate the effect of vitamin D supplementation on the prevention of acute respiratory tract infections in young Finnish men (IV).

MATERIAL, SUBJECTS, AND METHODS

Subjects and methods (I)

For the study, 196 Finnish men (18-28 yrs) undergoing military training in the Finnish Defence Forces as conscripts were recruited. A representative sample of subjects (n = 96) were recruited at health examinations covering all conscripts (n = 690) in January 2003. Similarly, 100 subjects were recruited, representing all of the conscripts (n = 1,158), in January 2004, nearly a year after the start of nationwide vitamin D fortification (February 2003). The volunteers were asked to participate in the examination a unit at a time, and each man was invited until the target number had been reached. Use of vitamin D supplements and staying in sunny southern countries in the last three months were used as exclusion criteria. The men all had passed their initial medical examination as healthy.

Serum 25OHD concentrations were determined with an OCTEIA® enzyme immunoassay by IDS (Immunodiagnostic Systems) in January 2003 (n = 96) and in January 2004 (n = 100), and serum 25OHD concentrations below 40 nmol/l were regarded as indicating vitamin D insufficiency. All volunteers provided written informed consent. The Ethics Committee of the Pirkanmaa Hospital District approved the study.

Subjects and methods (II, III)

For the prospective studies, 800 young Finnish men (range: 18-28 yrs; mean: 19.8 yrs) undergoing military training in the Finnish Defence Forces as recruits of the same infantry unit were randomly selected and enrolled for a study in July 2002. Because of failed samples, incomplete follow-up data, and interrupted military services, the total study population was composed of 756 subjects. They were receiving no medication, and their medical examination found them to be healthy. Those taking vitamin D supplements or having travelled in southern or sunny countries in the preceding three months were excluded from the study.

All Finnish men fall due for mandatory six-, nine-, or 12-month national service at the age of 18. Military service is voluntary for women. Annually, on average, 26,500 male and 500 female recruits undergo military training. The military conditions with respect to physical activity, nutrition, clothing, living quarters, and exposure to sunlight were homogenous for the study population. All volunteers provided written informed consent. The Ethics Committee of the Pirkanmaa Hospital District approved the study.
Background information
A structured questionnaire was used to elicit information on age, smoking, amount of physical exercise before military service (hrs/wk), physician-diagnosed lactose-intolerance, education, geographical location (northern or southern Finland), and area of residence (urban or rural). Height and weight were measured to determine body mass index (in kg/m2).

The information about aerobic fitness and muscle strength used in this study was obtained during the first weeks of military service. Aerobic fitness was measured by means of a 12-minute running test. Muscle strength was assessed with five metrics (distance of a horizontal jump and number of sit-ups, push-ups, pull-ups, and back-lifts), with an index of 0 to 3 employed for each. Information about daily smoking was assessed from questionnaires.

Because most stress fractures occur during the first eight-week basic training period common to all military recruits, the subjects were followed up to identify possible stress injuries to bone for 90 days (Jones et al. 1989). The subjects were informed about the study and instructed to seek medical attention in the event that symptoms suggestive of bone stress injuries occurred. Those who by clinical examination with careful history-taking were suspected to have sustained a bone stress injury underwent plain radiographic imaging with accepted radiological assessment (Kiuru et al. 2004; Lappe et al. 2008). The grey cortex sign, endosteal callus, periosteal callus, and fracture line were accepted as the radiographic signs marking a stress injury to bone. In addition, cases of normal findings on plain radiography with medical advice sought for the second time on account of prolonged exercise-induced pain in the pelvic area or the lower extremities were referred for MRI examination at the main military hospital. The patients with stress fractures on MRI were included in the study. The remainder of the subjects under observation constituted controls for the stress fracture cases.

Identification of respiratory infections

Medical records for all participants covering the first six months of military service were abstracted, and physician-diagnosed respiratory tract infections (i.e., sinusitis, tonsillitis, otitis, bronchitis, pneumonia, pharyngitis, and laryngitis) were recorded. As the incidence index we used the number of days of absence from duty due to respiratory tract infection.

Methods

Blood samples for determination of serum 25OHD concentration were drawn at entry into military service at the beginning of July 2002. After coagulation at room temperature for one hour, the samples were centrifuged at 2,000 g for 20 minutes at room temperature for serum separation. Then the serum samples were frozen and stored at -20 °C for later analysis. Total serum 25OHD concentrations were measured with an OCTEIA® enzyme immunoassay by IDS (Immunodiagnostic Systems). According to the manufacturer, the IDS intra- and inter-assay CVs for the assay are <8% and <10%, respectively, with a precision range of 5.3-380 nmol/l (CV <10%). In addition, the cross-reactivities for 25OHD3 and 25OHD2 are 100% and 75%, respectively. The 25OHD enzyme immunoassay correlates well with a radioimmunoassay (RIA), r2 = 0.82.

Subjects and methods (IV)

For the placebo-controlled double-blinded study, 164 young Finnish male volunteers (18-28 yrs) were recruited. They were conscripts in an infantry unit of the Finnish Defence Forces (the Pori Brigade, in south-west Finland) and were undergoing mandatory periodic military training. They represented the general conscript population of the Finnish Defence Forces. The trial was performed from October to March. Inclusion criteria for the study were lack of regular medication and passing of the initial medical examination as healthy. The exclusion criteria were the use of supplementary vitamin D, multivitamins, and cod-liver oil. In all, 164 men (of a total of 400 men entering that infantry unit) (41%) volunteered to participate and met the inclusion criteria. Then the subjects were randomly assigned to the intervention group (n = 80), receiving 10 \ig of vitamin D3 (Minisun®, Verman) daily, or the control group (n = 84), receiving a placebo (Pharmia, a capsule identical in size and form to the active preparation). Computer-generated random numbers were used for random allocation. For the subjects, the conditions related to physical activity, nutrition, clothing, accommodation, and exposure to sunlight were homogeneous during the military service. The Ethics Committee of Tampere University Hospital, Finland, approved the study. It was registered in ClinicalTrials.gov as NCT00973583. Written informed consent was obtained from all voluntary participants.

Identification of respiratory infections

The primary outcome variable was the number of days of absence from duty because of acute respiratory tract infection. Self-reported symptoms of acute respiratory tract infection (cough, runny nose, sore throat, fever, or common-cold symptoms) and hospitalisation due to acute respiratory tract infection were secondary outcome variables. The symptoms were evaluated four times during the study. Medical records for all participants covering the first six months of military service were reviewed, and any acute respiratory tract infection diagnosed (i.e., sinusitis, tonsillitis, otitis, bronchitis, pneumonia, pharyngitis, and laryngitis) was recorded. All acute respiratory tract infections diagnosed were treated in the garrison hospital. Both the subjects and the personnel treating patients were blinded to treatment allocation.

Methods

After computer-generated random allocation, blood samples were drawn from 73 subjects at the start of the study (in October 2005) and again from 108 subjects in March 2006 to determine the serum 25OHD concentrations. The blood samples were coagulated at room temperature for an hour and centrifuged at 2,000 g for 20 min at room temperature for serum separation. The samples were frozen and stored at -20 °C. For total serum 25OHD concentrations, an OCTEIA® enzyme immunoassay kit (Immunodiagnostic Systems) was used.

To determine plasma parathyroid hormone concentrations, 104 subjects were selected at random at the end of the study. Plasma parathyroid hormone concentrations were measured by electrochemiluminescence (Elecsys PTH Kit; Roche Diagnostics), and Elecsys PTH CalSet (Roche Diagnostics) was used for calibration.

Statistical analysis

In the articles: Vitamin D fortification as public health policy: significant improvement in vitamin D status in young Finnish men and An association of serum vitamin D concentrations <40 nmol/l with acute respiratory tract infection in young Finnish men (I, III)
Serum 25OHD concentration was expressed as means ± SDs. Differences in means between independent samples were tested through the use of Student's t-test or one-way analysis of variance (ANOVA). The number of days of absence from duty due to respiratory infection was expressed as medians and as lower and upper quartiles (Q1 and Q3, respectively) because the distribution was skewed. Poisson regression analysis was used to explain this variable plus 0.5, because the variable included zeroes. Over-dispersion was estimated by the deviation divided by its df. Serum 25OHD concentrations were categorised as <40 nmol/l and >40 nmol/l for the regression analysis. The limit for statistical significance was set to be equal to 0.05 for a two-sided test. We used the software SAS for Windows (version 8.2, from the SAS Institute; Cary, NC, USA) for Poisson regression analysis and SPSS for Windows software (version 11.0, from the SPSS Institute; Chicago) for other data analysis.

In the article: Association between serum 25OHD concentrations and bone stress fractures in Finnish young men (II)

At stage 1 in the statistical analysis, the skew-continuous serum 25OHD values were divided into two categories on the basis of the median. The differences in serum 25OHD between the groups were tested by Pearson x2 testing. The results were corroborated by Mann-Whitney's U-test, using the original values. Student's t-test was used to test differences in age, BMI, height, weight, muscle strength, and 12-minute running between the groups. The limit for statistical significance was set at 0.05. Data analysis was performed with SPSS for Windows (version 11.0; SPSS Institute).

At stage 2 in the statistical analysis, logistic regression was applied to study the association between stress fractures and the statistically significant explanatory variables from stage 1. ORs were calculated with a 95% CI. Variables showing statistical significance in stage 1 (dichotomised serum 25OHD concentration, muscle strength, and 12 minutes' running) were included in the same forward stepwise regression model. Data analysis was performed with SPSS for Windows (version 12.0; SPSS Institute).

In the article: Vitamin D supplementation for the prevention of acute respiratory tract infection: a randomised, double-blinded trial among young Finnish men (IV)

By recruiting only voluntary conscripts for the study (i.e., no sampling outside the target population was used), we aimed to maximise the power. The primary analysis included all randomised subjects, following the intention-to-treat principle. Mann-Whitney U-testing was used to test differences between the groups for continuous variables, and Chi-square tests were used to assess categorical data. A two-sided p value of less than 0.05 was set as the alpha criterion.

The hazard ratio (HR) was calculated by Cox's regression analysis adjusted for influenza vaccination and smoking at baseline; the first infection was used as the end point, with censoring at premature release from duty or the end of the study, after the six-month follow-up. The frequency of missing variables at the baseline varied between 2% and 6%. For smoking, it was 20%. Sixty subjects had dropped out of the study by the end point, with no specific reason given. Data analysis was performed with SPSS for Windows (version 15.0.1, from SPSS Institute; Chicago).

SUMMARY OF THE RESULTS

Improved vitamin D status after implementation of vitamin D fortification as public health policy in Finland (I)

In a before-and-after vitamin D fortification comparison study in January 2003 (n = 96) and in January 2004 (n = 100), nearly one year after national vitamin D fortification began, with a study population of 196 young Finnish men, the mean daily intake of milk and of margarine were 8 dl and 30 g, respectively. Therefore, after fortification the study subjects were ingesting, on average, 7 ug of vitamin D [(8 dl milk x 0.5 ug/dl) + (30 g margarine x 10 ug / 100 g)], daily through fortified products.

{280 IU vitamin D daily average}
Mean serum 25OHD concentrations increased by 50% after implementation of the vitamin D fortification of dairy products, and the prevalence of vitamin D insufficiency (serum 25OHD <40 nmol/l) decreased by 50%, from 78% to 35%.

Moreover, the prevalence of vitamin D deficiency (serum 25OHD <25 nmol/l) decreased from 19% to 5%.
In addition, five subjects (5%) had serum 25OHD concentration higher than 100 nmol/l (101 -111 nmol/l) after fortification.

The association of low serum 25OHD concentration with stress fractures and acute respiratory tract infections in young Finnish men (II, III)

Low serum 25OHD concentration as a predisposing factor for stress fractures in 756 military recruits
In three-month follow-up of a population of 756 recruits, 22 subjects with stress fractures were identified (2.9%). The median serum 25OHD concentration for all the subjects was 75.8 nmol/l (range: 25.2-259.0 nmol/l).
The median serum 25OHD concentration in the group with stress fractures (median: 64.3, range: 40.1-159.0) was statistically significantly lower than that in the group of recruits without stress fractures (median: 76.2, range: 25.2-259.0, n = 734). Those with serum 25OHD levels below the median had a statistically significantly greater risk of stress fractures than those above the median.

The incidence of stress fractures was 11.6 (95% CI, 6.8-16.5) per 100 person-years. A developing stress injury became symptomatic, on average, 39 days (range: 4-70 days) after the start of military training. Thirteen bone stress injuries were located in the lower right limb and 17 in the left. In the 22 patients, plain radiography revealed 30 stress fractures in all, of which 13 (43%) were in the tibia, 10 (33%) in the metatarsal bones, three (10%) in the calcaneus, and two (7%) in the tarsal navicular bone. One fracture each was found in the inferior ramus and the femur.

The recruits with stress fractures had statistically significantly worse results in the 12-minute running test (mean: 2,480 vs. 2,670 m, p = 0.007) and in muscle strength (mean: 7 vs. 9, p = 0.025) than did controls. There was no statistically significant association between daily smoking (p = 0.85), BMI (p = 0.41), age (p = 0.27), height (p = 0.15), or weight (p = 0.70) and stress fractures. Moreover, in the multivariate regression model adjusted for all statistically significant variables from stage 1, the risk of stress fractures in subjects with serum 25OHD levels below the median was 3.6 (95% CI: 1.2-11.1) times the risk for those with concentrations exceeding the median level. In the multivariate analysis, the results for muscle strength or in the 12-minute running tests did not show statistically significant correlation with stress fractures.

Low serum 25OHD predicting respiratory tract infection

Eight hundred young Finnish men serving on a military base in Finland were enrolled for a study of whether an association exists between vitamin D insufficiency and acute respiratory tract infections. Their serum 25OHD concentrations were measured at the beginning of the study, in July 2002, and they were followed up for six months. The days of absence from duty due to respiratory infection were counted.

The mean serum 25OHD concentration was 80.2 ± 29.3 nmol/l (n = 756), and 25OHD concentrations <40 nmol/l were found in 3.6% of the subjects in July 2002. Subjects with 25OHD concentrations <40 nmol/l (n = 24) had statistically significantly (P = 0.004) more days of absence from duty due to respiratory infections (median: 4, Q1 -Q3: 2-6) than did control subjects (n = 628) (2, 0-4, incidence rate ratio: 1.63, 95% CI: 1.15-2.24). The model was adjusted for smoking (n = 169).

A non-statistically significant association was seen between BMI and serum 25OHD concentration but no association between serum 25OHD concentration and lactose-intolerance. No differences were found in vitamin D status between subjects from southern and northern Finland, or between subjects living in urban and rural areas. Education level was not associated with vitamin D status.

Positive association of regular exercise with serum 25OHD concentrations

There was a statistically significant positive association between frequency of previous physical exercise and serum 25OHD concentration (p = 0.004). Subjects who exercised over five hours per week had higher serum 25OHD concentrations (85.8 ± 30.6 nmol/l) than did subjects reporting no physical exercise (71.3 ± 32.0 nmol/l). We also found a non-statistically significant association between frequency of physical exercise and number of days absent from duty due to respiratory infection (p = 0.388, n = 264).

Negative association of smoking with serum 25OHD concentrations

Serum 25OHD concentrations were statistically significantly (P <0.001) lower in subjects who smoked (72.8 ± 26.6 nmol/l, n = 192) than in control subjects (82.9 ± 29.7 nmol/l, n = 537). There was also a trend toward a statistically significant association (P = 0.065) between smoking (n = 169) and number of days of absence from duty due to respiratory infection (incidence rate ratio: 1.18, 95% CI: 0.988-1.40, n = 629).

Vitamin D supplementation's prevention of respiratory tract infections in young Finnish men (IV)

Vitamin D supplementation with 10 jug/d and its substantial effect on vitamin D status
There was no difference in serum 25OHD concentration between the intervention (78.7 ± 14.9 nmol/l, n = 29) and placebo (74.4 ± 20.8 nmol/l, n = 44) group (p = 0.35) at the beginning of the study, in October 2005. Age, height, weight, and body mass index were also comparable between the groups at baseline, though both smoking and influenza vaccination were slightly more common in the placebo group. After daily supplementation with 10 ug of vitamin D3 or a placebo for six months, the mean serum 25OHD concentration was 71.6 ± 22.9 nmol/l (n = 58) in the intervention group and 51.3 ± 15.5 nmol/l (n = 50) in the placebo group (p <0.001).

A slight preventive effect of vitamin D supplementation of 10 jug/d on the incidence of acute respiratory tract infections

The number of days absent from duty because of respiratory tract infection (the main outcome variable) did not differ between groups - the mean number of days of absence was 2.2 (SD: 3.2) in the intervention group and 3.0 (SD: 4.0) in the placebo group (p = 0.096). However, there was an effect during the first six weeks of the study, with a mean of 0.7 (SD: 2.1) days of absence in the intervention and of 1.4 (SD: 2.6) days' absence in the placebo group (p = 0.060). After the first six weeks, there tended to be no difference between the groups.

The proportion of subjects remaining healthy throughout the six-month study period was greater in the intervention group (51.3%, n = 41) than in the placebo group (35.7%, n = 30, p = 0.045). Moreover, in a Cox regression analysis with adjustment for smoking and influenza vaccination, the adjusted hazard ratio for absence from duty due to respiratory tract infection was lower in the intervention group (HR 0.71, 95% CI: 0.43-1.15), and the number needed to treat calculated from the proportion of men without any days absent from duty was 6.4 (95% CI: 3-257).

From the questionnaire, self-reported cough (65% in the intervention group vs. 57% in the placebo group, p = 0.30), runny nose (74% vs. 75%, p = 0.86), sore throat (48% vs. 45%, p = 0.77), fever (31% vs. 38%, p = 0.36), and symptoms of the common cold (56% vs. 52%, p = 0.40) did not differ between the groups.

The mean number of days of treatment in hospital was 0.31 (SD: 1.21) per subject in the intervention group and 0.90 (SD: 2.22) in the placebo group (p = 0.06). Two subjects in the intervention group reported nausea, stomach ache, and diarrhoea during the study.
In addition, one subject in the placebo group dropped out of the study because of a facial rash.


Image See VitaminDWiki = rare allergic reaction to Vitamin D


Vitamin D supplementation with 10 jg/d - no effect on plasma PTH concentrations
Plasma PTH concentrations did not differ statistically significantly between the intervention 4.3 ± 1.3 ng/l (n = 58) and placebo 4.4 ± 1.4 ng/l (n = 50) group (p =0.55).

DISCUSSION

Vitamin D insufficiency, commonplace among young Finnish men

All Finnish men are required to complete six, nine, or 12 months of military service between the ages of 18 and 29. Each year, on average, 26,500 conscripts undergo military training. As 85-90% of Finnish men serve in the military, the conscripts represent the vast majority of all young men of the same age in Finland. The study populations for these studies were composed of these healthy young men completing military service as conscripts in the Finnish Defence Forces. The population presents homogeneity with respect to age, the conditions for physical activity, nutrition, clothing, living area, and exposure to sunlight in the military environment. The homogeneity of our study setting and population offers an optimal setting for this kind of study and can be seen as a major strength of the studies.

According to our studies, the prevalence of vitamin D insufficiency, defined as serum 25OHD <40 nmol/l, in young Finnish men was only 4% in July 2002, increased to 80% in January 2003, and diminished to 35% in January 2004 on account of nationwide vitamin D fortification. Further, vitamin D deficiency (25OHD <25 nmol/l) was seen in 19% of subjects in January 2003 and in 5% in January 2004, with mean serum 25OHD levels of 34 and 50 nmol/l, respectively. The level was 80 nmol/l in July 2002. In other studies of young Finnish men, 2% had a serum 25OHD of <20 nmol/l in July 2001, 34-39% in January 2001, and 29% in January 2004 - one year after the vitamin D fortification began. The corresponding median levels of serum 25OHD were 44, 24, and 27 nmol/l, respectively (Valimaki et al. 2004; Valimaki et al. 2007).

For vitamin D determination, one must take into account that the values differ, depending on the technique used. For total serum 25OHD concentrations, we used an OCTEIA® enzyme immunoassay kit (Immunodiagnostic Systems) while in other Finnish studies described above RIA (Diasorin) or HPLC was used. In earlier evaluation of vitamin D measurement methods, HPLC was considered the gold standard of 25OHD measurement. For the IDS enzyme-linked immunoassay, there has been mixed performance, with up to 23% overestimation until 2006, that might explain the differences between our studies and Finnish studies that used other techniques (de la Hunty et al. 2010).
The present studies showed a statistically significant positive association between regular physical exercise and serum 25OHD level. The subjects exercising regularly were vitamin D sufficient in summertime (86 nmol/l). Similarly, a recent Australian study showed statistically significantly higher prevalence of vitamin D deficiency in those who were physically inactive. In that study, vitamin D deficiency was defined as serum 25OHD <50 nmol/l and found in nearly one third of Australian adults (Daly et al. 2011). All of these studies support the conclusion that regular physical exercise is related to a greater amount of time spent outdoors and is a strong determinant of improved vitamin D status.

Obesity is thought to be a risk factor for vitamin D insufficiency. This is explained by

  • reduced outdoor physical activity; by
  • sequestration of fat-soluble serum 25OHD in adipose tissue, from which it cannot be easily released; and by
  • increased need for vitamin D in the obese because of stronger bones supporting more weight (Renzaho et al. 2011).

Again, in a large nationwide Australian study, serum 25OHD levels were statistically significantly lower in obese men and women than among those of normal weight (Daly et al. 2011). Further, vitamin D levels have been shown in another study to be substantially lower in obese people than in those of normal weight after equal doses of UVB (Wortsman et al. 2000). However, the results of other studies are somewhat contradictory. In our study, there was only a non-statistically significant association between BMI and serum 25OHD concentration. The relationship between serum 25OHD and BMI remains unclear and requires further investigation, as ethnicity, gender, and age may play a crucial role here (Renzaho et al. 2011).

The present studies showed no statistically significant differences in vitamin D status between subjects from southern and northern Finland, between subjects of different education levels, or between subjects living in urban vs. rural areas. In contrast, in a recent Australian study the highest rates of vitamin D deficiency (serum 25OHD <50 nmol/l) were found among educated subjects and at a latitude above 35° S in wintertime (Daly et al. 2011). Our studies were expected to show differences between subjects from northern and southern Finland due to the geographical location of the country, extending from roughly 60° to 70° N.

The present studies showed a statistically significant negative association between smoking and serum 25OHD. Similarly, high prevalence of low 25OHD levels was found in continuous smokers with established mild or moderate COPD in a recent study (Kunisaki et al. 2011). The relation between smoking and vitamin D status is still unclear but might be explained by the effects of nicotine metabolites on hepatic CYP27A1 activity.

The results of the present studies show that, in terms of adequacy of sunlight exposure, cutaneous synthesis of vitamin D is sufficient in summer in young Finnish men and was even improved with the national policy of vitamin D fortification of dairy products. However, vitamin D insufficiency is likely in young Finnish men without vitamin D supplementation during winter. The findings are recognised as having significant public health implications globally.

Vitamin D fortification of milk products - improved vitamin D status

but no resolution of hypovitaminosis in young Finnish men

National vitamin D fortification started in February 2003 on the recommendation of the Ministry of Social Affairs and Health in Finland. It covers liquid milk products and margarines. In practice, all manufacturers have complied with the recommendation. In our study, vitamin D status was determined before and after vitamin D fortification (January 2003 and January 2004), both times during army service. In the study, nationwide fortification substantially improved the vitamin D status of young Finnish men but did not resolve hypovitaminosis in young Finnish men, as 35% remained vitamin D insufficient. Mean serum 25OHD concentrations increased by 50%, from 33.5 to 50.2 nmol/l. In another study, with 65 subjects, the vitamin D fortification increased mean serum 25OHD concentrations to a limited extent: by 20% in young Finnish men. In that study, vitamin D status was determined first in January 2001 and then in January 2004, a few years after army service. The use of milk products was different during the army service, and in the sub-population with >7.2 dl of milk daily there was actually a >40% improvement in vitamin D status (Valimaki et al. 2007). In our study, given the average daily intake of milk and margarine in the Finnish Defence Forces, the subjects ingested 7 ug more vitamin D after the fortification than before, corresponding to the target level set by the Finnish National Nutrition Board. In earlier studies, every microgram of vitamin D taken daily has increased serum 25OHD by 0.6-2.2 nmol/l. In conclusion, national vitamin D fortification increased vitamin D status as expected in our study (Byrne et al. 1995; Heaney et al. 2003; Heaney 2007).

National fortification of dairy products might include a risk of a possibility of overdose in some individuals. Hypervitaminosis D has been reported rarely and can occur only with supplementation as skin regulates the production of vitamin D by itself. The toxicity of vitamin D is due to hypercalcemia and hyperphosphatemia. Secondary to hypercalcemia hypercalciuria, kidney failure, nausea, vomiting, confusion and weakness has been seen in hypervitaminosis D patients. (Holick 2007) The toxicity may occur in serum 25OHD levels >375 nmol/l. Case reports of hypervitaminosis D include supplement use that contained much more vitamin D than label indicated. In one case, leading to hypervitaminosis D, a patient had consumed 5000 ug of vitamin D3 daily for two months because of an error in product formulation (Klontz et al. 2007). However, in another case report the usage 3750 ug of vitamin D2 daily for many years did not cause hypervitaminosis D symptoms (Jacobsen et al. 2011). Though information on symptoms was not systematically collected via questionnaires and we did not have information on vitamin D intake at individual level, no clinical signs indicating toxicity were seen in our study. Only 5% of the subjects in January 2004 and none of the subjects in January 2003 had a serum 25OHD concentration of >100 nmol/l.

The effects of the vitamin D fortification on vitamin D status in young Finnish men (1 ug / 100 ml in liquid milk products and 20 ug / 100 g of fat spreads) that began in April 2010 after the recommendation of the Finnish National Nutrition Council still needs to be clarified (Pietinen et al. 2010).

Further, daily vitamin D supplementation of

  • 10 ug for children <2 years, pregnant and breast-feeding women,
  • 7.5 ug for 2-18-year-olds, and
  • 20 ug for those over 60 years

has been recommended by the Council since January 2011 in Finland.
The long-term effects on health remain to be demonstrated.

The association of low serum 25OHD concentration with stress fractures in young Finnish men

The study demonstrates that the subjects with stress fractures had substantially lower serum 25OHD concentrations (64 nmol/l) than did controls (76 nmol/l). The study showed that in the group with a serum 25OHD level less than the median (76 nmol/l), the risk of stress fractures was 3.6 times that of 1 in the control group. In addition, those with stress fractures had substantially poorer results in the 12-minute running test and for muscle strength than controls did in the univariate analysis. No statistically significant association was found between smoking, BMI, age, height, or weight with stress fractures.
In the multivariate analysis, the results for muscle strength or 12-minute running tests were not statistically significantly associated with stress fractures. However, there was a statistically significant association between a serum 25OHD level below the median and stress fractures also in the multivariate analysis. It is known that vitamin D increases intestinal calcium absorption and its renal reabsorption, maintaining a reasonable amount of calcium in circulation, especially for bone mineralisation (Utiger 1998). Vitamin D is also known to decrease PTH secretion and enhance both osteoblast and osteoclast differentiation (Riggs 1997; Holick 2003). Vitamin D has a major enhancing effect on peak bone mass too (Vieth 1999; Valimaki et al. 2004). In a Finnish study, stress fractures were associated with higher PTH levels but not with lower 25OHD levels (Valimaki et al. 2005). Earlier, a correlation between low BMD and stress fractures was demonstrated in military recruits (Pouilles et al. 1989). Furthermore, in the Israeli Defence Forces study, serum 25OHD was substantially lower among patients with high-grade stress fractures than among the controls (Givon et al. 2000). All of these findings strongly suggest that inadequate vitamin D is a risk factor for stress fractures in military recruits.

A negative correlation between PTH and 25OHD levels has been demonstrated in earlier studies: PTH starts to increase at serum 25OHD levels below 78-100 nmol/l (Chapuy et al. 1997; McKenna et al. 1998). Further, increased calcium absorption in the intestine has been demonstrated following a rise in serum 25OHD concentration from 50 to 86 nmol/l (Heaney et al. 2003). Interestingly, in our study recruits with serum 25OHD of <76 nmol/l exhibited substantially more stress fractures than did those with higher concentrations. The results from our study suggest that the risk of stress fractures is diminished when serum 25OHD levels exceed 78-100 nmol/l throughout the year.

A recent vitamin D supplementation RCT with 2 g calcium and 20 ug vitamin D daily for eight weeks demonstrated a 20% decrease in the incidence of stress fractures (Bouillon 2008; Lappe et al. 2008). The double risk of stress fractures of the tibia and fibula was demonstrated with serum 25OHD <50 nmol/l as compared to those with >100 nmol/l in female US Navy recruits. In the study, an inverse dose-response gradient between serum 25OHD and risk of stress fractures was demonstrated also (Burgi et al. 2011). Moreover, a recent study found two polymorphisms in the VDR to increase the risk of stress fractures independently, suggesting screening for VDR polymorphisms as a tool for identifying individuals at increased risk of stress fractures during physical and/or military training (Chatzipapas et al. 2009; McClung et al. 2010). The findings from these studies support the findings in our study.

Vitamin D insufficiency as a predictor of higher risk of acute respiratory tract infection in young Finnish men

The cohort study showed a statistically significant negative association of serum 25OHD concentration with the number of days of absence from duty due to physician-diagnosed respiratory tract infection. Because living quarters are close for recruits, respiratory tract infections are commonplace and easily transmitted in garrisons, offering ideal circumstances for this kind of study. Therefore, it is likely that low vitamin D status upon initial entry into military service contributes to proneness to respiratory tract infections.

In addition to the homogeneity of the conditions during the study for all subjects, enhancing the results' validity, the study has other strengths. Firstly, the accuracy of the number of physician-diagnosed respiratory infections identified with respect to absence from duty can be considered a positive element. Secondly, a statistically significant effect on absence from duty due to infectious diseases was found despite the fact that a relatively small number of vitamin D insufficient young men was found at the beginning of the study, in July 2002. Thirdly, the effect of smoking on respiratory infections was adjusted for the study.

Persistence of vitamin D sufficiency throughout the study was not determined; vitamin D status was assessed only at the beginning of the study, in July 2002. This can be seen as a possible weakness of the study. In addition, vitamin D was statistically significantly correlated with the amount of previous physical exercise but we found only a non-statistically significant association between frequent physical exercise and days of absence from duty due to respiratory infection, not confounding the results.

Recent evidence shows that vitamin D upregulates the expression of antimicrobial peptides that proved to inhibit invasive pneumococcal disease, meningococcal disease, and group A streptococcal disease in an earlier study (Vlaminckx et al. 2005). Further, beta defensins that belong to the AMP family were shown to inhibit the growth of otitis media pathogens (Haemophilus influenzae, Moraxella catarrhalis, and Streptococcus pneumoniae) in another study (Lee et al. 2004). A polymorphism in the VDR gene was also positively associated with respiratory-syncytial-virus-related disease in recent studies, with both South African and Dutch children (Kresfelder et al. 2011). The findings are consistent with the idea from other studies that secreted antimicrobial peptides are dependent on vitamin D and as a part of innate immunity constitute the first line of defence protecting host mucosal surfaces' against microbes (Hewison 2011; Laaksi 2011). The findings support also the finding of our cohort study that low vitamin D levels pose a risk for a respiratory tract infection.

Adequate vitamin D status was proposed to protect against influenza in a cohort study with 7,000 participants aged over 45 years. In this study, the prevalence of respiratory infections showed a strong seasonal pattern inversely related to 25OHD concentrations. With adjustment for adiposity, lifestyle, and socio-economic factors, each 10 nmol/l rise in serum 25OHD concentration was associated with a 7% lower risk of infection in the study. Vitamin D status seems to have a linear relationship with respiratory infections (Berry et al. 2011). Further, in a secondary analysis of the Third National Health and Nutrition Examination Survey (1988-1994), with almost 19,000 participants aged above 12 years, serum 25OHD concentrations were inversely associated with recent URTI. The analysis took into consideration season, BMI, smoking, asthma, and COPD, with URTIs reported by 24% of subjects with serum 25OHD <25 nmol/l, by 20% with 25-75 nmol/l, and by 17% with >75 nmol/l (Ginde et al. 2009). These findings are consistent with the results of our study.

Possible preventive effect of vitamin D supplementation on acute respiratory tract infection during military service

The present RCT with 164 young Finnish men provided some evidence of a preventive effect of vitamin D supplementation against respiratory tract infection. Firstly, the proportion of the subjects who remained healthy throughout the study period was substantially greater in the intervention than in the placebo group. Secondly, the adjusted HR for absence from duty due to respiratory tract infection was lower in the intervention than in the control group. Thirdly, the number needed to treat calculated from the proportion of men without any days absent from duty was as low as 6.4. In addition, an effect was found during the first six weeks of the study, with fewer days of absence in the intervention than in the placebo group. Also, a trend in which the mean number of days of treatment in hospital was diminished by 10 ug of vitamin D supplementation daily was found in the present study.

In a 5 month RCT conducted in Antarctica during winter when ultraviolet B radiation levels are essentially zero showed that vitamin D supplementation with 50 ug/d increased serum 25OHD level to 71 nmol/l. On the other hand in the 10 ug/d subgroup an increase up to 57 nmol/l was seen. (Smith et al. 2009) In contrast, a study in submariners showed no increase in serum 25OHD with 10 ug/d of vitamin D supplementation (Duplessis et al. 2005). Importantly, a daily 10 ug of vitamin D during winter was able to keep vitamin D status almost adequate (25OHD >75 nmol/l) while levels in subjects without supplementation remained mainly insufficient (25OHD <50 nmol/l) in our study. Again, the technique used for determination of vitamin D levels has to be taken into account.

The study had some limitations. Our main outcome variable, the number of days' absence from duty due to respiratory tract infection, did not differ between the intervention and control group. Further, self-reported symptoms of respiratory tract infection (cough, runny nose, sore throat, fever, and symptoms of the common cold) did not differ between the groups. There was also a very wide 95% CI, 3-257, in the number needed to treat calculated from the proportion of men without any days absent from duty. Because of the number of subjects who withdrew from the study, its statistical power remains limited.

The completeness of the outcome data related to respiratory infections identified and the amount of days of absence from duty is a strength of the study. Interestingly, if no subjects had withdrawn from the study, a statistically significant preventive effect of 10 ug daily vitamin D supplementation on respiratory tract infections would have been shown. On the other hand, when we used a one-sided P value in the analysis, the difference in days absent from duty between the trial arms was statistically significant.

In the present study, 10 ug of vitamin D daily did not have a statistically significant effect on the study subjects' PTH concentrations. However, in another six-month Finnish RCT, elevation of PTH concentration was inhibited by 20 ug of vitamin D daily. It is probable that the dose used in our study was not enough and that an effect on PTH concentration would have been seen with a higher supplementation dose (Viljakainen et al. 2008).

In an RCT with >200 healthy post-menopausal African American women who were given 20 ug of vitamin D daily or a placebo for two years, a lower rate of self-reported upper respiratory infection or influenza was observed in the intervention arm as compared to the placebo group. After an increase in vitamin D dosage from 20 to 50 ug of vitamin D daily for one year, the effect was even greater (Aloia et al. 2007; Li-Ng et al. 2009). This finding supports the results of our study.

Another RCT failed to show any statistically significant difference between the vitamin D and placebo group in the secondary end point of self-reported infection rate. In that study, 3,500 elderly subjects were given 20 ug of vitamin D or a placebo for over two years. However, because of the poor compliance observed, the results are complicated - only half of the subjects remained compliant with use of the study medication at two years' follow-up. In addition, a relatively inadequate increase, from 38 to 63 nmol/l, in serum 25OHD levels was shown in the intervention group after vitamin D therapy (Avenell et al.
2007).

Similarly, a recent RCT with 162 healthy adults given 50 ug of vitamin D or a placebo daily for 12 weeks during the winter and spring months showed no benefit with respect to the incidence or severity of upper respiratory infection symptoms. In the study, appropriate increase in serum 25OHD levels in the intervention group from 62 to 89 nmol/l at 12 weeks was noted. However, a statistical tendency to favour the vitamin D group was seen (Li-Ng et al. 2009).

CONCLUSIONS

Based on our study vitamin D insufficiency was found in relatively low proportion of the subjects during summer. However, the study shows that vitamin D insufficiency is commonplace in Finland during winter even after vitamin D fortification, with 35% of young Finnish men being vitamin D insufficient in January 2004. Furthermore, it is likely that vitamin D levels reach their lowest in late winter and vitamin D insufficiency can then be found in substantially more than 35% of young men. It is notable that the studies were carried out before the Finnish National Nutrition Board had given new recommendation of vitamin D fortification in 2010. In order to maintain an adequate level of vitamin D, i.e. serum 25OHD >75 nmol/l, throughout the year young Finnish men will need an additional supplementation with 20 ug of vitamin D daily during winter.

In conclusion, because of vitamin D inadequacy, there seems to be a strong likelihood of considerable risk of stress fractures among young men during military service. A good target for prevention of stress fractures might be a serum 25OHD of >75 nmol/l throughout the year in young Finnish men. Interestingly our RCT with 10 ug/d of vitamin D found no effect on plasma PTH levels in the young Finnish men. Given this and earlier intervention studies' results on improving vitamin D status, an adequate serum 25OHD level (>75 nmol/l) will be achieved through 20 ug of vitamin D supplementation daily during winter in Finland (Byrne et al. 1995; Heaney et al. 2003; Heaney 2007; Burgi et al. 2011).

The cohort study performed that low vitamin D status at entry into military service and subsequent respiratory infections have a clear statistically significant relationship. In order to prevent acute respiratory tract infections in young Finnish men, at least during military service, an additional supplementation with 20 ug of vitamin D daily will be needed during winter. Clarification of the role of vitamin D in relation to acute respiratory tract infections represents a high priority for future research.

The RCT conducted for this research provided some evidence of vitamin D supplementation acting against acute respiratory infections in young Finnish men during winter in military service. However, results from other RCTs of vitamin D for the treatment and prevention of viral upper respiratory tract infections in adult populations are somewhat controversial and suggest that a larger sample, adequate vitamin D repletion, and a longer period may be beneficial in the design of future studies, to maximise immunomodulatory effects. Further, the results of the present RCT indicate that additional supplementation, to 20 ug of vitamin D daily during winter may have the preventive effect on acute respiratory tract infections in young Finnish men during military service.

CLICK HERE for the rest of the dissertation and reference articles