You are here

Article Text

Article menu
Download PDFPDF
Workplace
Original article
Changes in vitamin D and matrix metalloproteinase-9 in submariners during a submerged patrol
  1. A Baker1,
  2. C L Wood2,
  3. A M Wood1,
  4. P Timms3,
  5. A J Allsopp1
  1. 1Institute of Naval Medicine, Alverstoke, Hampshire, UK
  2. 2Department of Child Health, Newcastle Hospitals Foundation Trust, Newcastle upon Tyne, UK
  3. 3Department of Clinical Biochemistry, St Bartholomew's Hospital, London, UK
  1. Correspondence to Dr C L Wood, Department of Child Health, Newcastle Hospitals Foundation Trust, Newcastle upon Tyne NE1 4LP, UK; claire.wood{at}ncl.ac.uk

Abstract

Objectives To quantify changes in vitamin D and matrix metalloproteinase-9 (MMP9) in submariners over a single long patrol and compare the data to a group of non-deploying servicemen from their base port.

Methods A prospective time-series analysis was performed. Blood samples were taken from 49 submariners deploying on patrol and 43 shore-side controls from the base port (naval officers from base or non-deploying submariners), following a winter ashore at latitude 56° north. Samples were drawn immediately before the submarine sailed, in January, and again in the final week of patrol 85 days later. Paired pre-patrol and late samples from each individual were assayed together and changes in vitamin D and MMP9 were assessed.

Results Mean pre-patrol vitamin D concentrations were 58 and 49 nmol/L for the controls and submariners, respectively. Mean vitamin D concentrations increased in controls as expected (mean increase 12.6 nmol/L), but not in the submariners (mean decrease 1.6 nmol/L). MMP9 levels were significantly higher in submariners pre-patrol, and increased significantly during the patrol. There was a significant inverse correlation between MMP9 and vitamin D levels (r=−0.41, p=0.01).

Conclusions This is the first study to quantify vitamin D and MMP levels in submariners. Circulating vitamin D concentrations on board were insufficient to prevent a rise in MMP. This has potential for adverse health effects and requires further study.

http://dx.doi.org/10.1136/oemed-2013-101793

Statistics from Altmetric.com

    Request Permissions

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

    What this paper adds

    • Submariners are at increased risk of vitamin D deficiency due to prolonged periods submerged.

    • Vitamin D deficiency is associated with increased levels of matrix metalloproteinases (MMPs), which can destroy tissue in chronic inflammatory and malignant disease processes.

    • This is the first study to quantify changes in vitamin D and MMPs in submariners.

    • We found a significant inverse correlation between MMP and vitamin D levels.

    • Further work is needed to establish the causal relationship between increased MMP levels and morbidity in submariners.

    Introduction

    The Royal Navy operates a nuclear-powered submarine fleet, part of which is based at the Clyde Naval Base at latitude 56° north. Nuclear power generation, advanced atmosphere control, and oxygen replenishment by electrolysis of seawater have combined to allow prolonged submerged endurance; submarine patrols now routinely exceed 60 days’ duration. Depending on type, a submarine might conduct three patrols of this length a year. There are currently 4500 serving Royal Navy submariners.1 Once trained, the majority spend their entire naval career in the submarine service.

    During patrols, submariners have no sunlight exposure at all. Onboard lighting is provided by 8 W fluorescent tubes, which emit very little of the ultraviolet B (UVB) wavelength (280–315 nm) required to initiate vitamin D synthesis in the skin. Exposure to solar UVB radiation is the main source of circulating vitamin D2, with a lesser amount derived from the diet, in particular from fish, milk or fortified foods.2 Supplies of fresh food are limited on board a submarine and the Royal Navy does not currently fortify rations with vitamin D. Previous studies that focused on the role of vitamin D in calcium metabolism have demonstrated a reduction in serum 25-hydroxyvitamin D (25(OH)D) in submariners during long submerged patrols.3 ,4

    The role of vitamin D in bone health is well established but its actions on other body systems are less understood. Vitamin D receptors are now described in almost all body tissues, and vitamin D deficiency has been implicated in many disease processes including myocardial infarction,5 diabetes,6 tuberculosis,7 muscle weakness8 and some cancers.9 ,10 Vitamin D deficiency is associated with increased levels of circulating matrix metalloproteinases (MMPs) and their tissue inhibitors (TIMPs).11 MMPs are extracellular protease enzymes which degrade the matrix that support cells in tissues, both in normal physiological processes (such as embryonic development and wound healing) and in pathological processes, where they destroy tissue in chronic inflammation and malignant disease.12 MMPs are implicated, for example, in endothelial plaque disruption and myocardial infarction.13 Circulating MMPs have been shown to fall with administration of vitamin D.11 Vitamin D may therefore have a protective effect in conditions where MMPs are implicated. MMP9 is a key member of the protease family and cleaves type 4 collagen and gelatin.

    The aim of this study was to quantify changes in vitamin D and MMP9 in submariners over a single long patrol, and compare the results to data from a group of non-deploying servicemen from their base port, in order to investigate potential effects on the health and physical performance of submariners. This information is important for the maintenance of health in this unique workforce specifically selected for absence of disease, especially as improvements in vitamin D status can be readily achieved.

    Methods

    Study design

    A prospective time-series analysis study design was used. All pre-deployment data were obtained over a 4-day period in January immediately before the submarine sailed, by two of the investigators assisted by the submarine’s medical officer. Late deployment data were obtained on day 85 of the patrol, and within the next 3 days for the control subjects by the same team.

    Subjects

    Forty-nine submariners who were onboard HMS Vigilant for the duration of the patrol, and 43 controls, not deploying to sea, gave informed consent and were recruited into the study. Thirty-five of the age- and sex-matched controls were servicemen from the base and eight were submariners from the non-deploying support crew. The Ministry of Defence Research Ethics Committee approved the study.

    Clinical assessments

    Age and anthropometric data on each subject were obtained during the initial assessment:

    • Height using a portable stadiometer (Leicester Height Scale, SECA, Birmingham).

    • Body mass (electronic Mettler Toledo scales, Fisher Ltd, UK).

    • Waist circumference in a standing posture with a measuring tape placed horizontally midway between the lowest rib and the iliac crest.

    • Systolic and diastolic blood pressure by automated sphygmomanometer (Omron MX3, Matsukaka, Japan).

    Laboratory methods

    Blood for 25(OH)D, MMP9 and C-reactive protein (CRP) analyses was drawn directly into heparinised tubes. Serum 25(OH)D was used as the single best assessment of vitamin D status.14 All samples from controls were collected in the sickbay, and centrifuged within 15–30 min (Heraeus Labofuge 400R). Pre-patrol samples from the submariners were obtained while the submarine was alongside, and were returned to the sickbay for centrifugation within 90 min. All samples from the late deployment collection were centrifuged within 90 min of collection using a compact centrifuge (Heraeus Labofuge 200) at 3000 rpm for 10 min. The serum supernatant and plasma samples from the first time point were stored at −60°C in a freezer at the Institute of Naval Medicine. On completion of the second sampling, pre and late deployment samples from the same individual were batched together and arranged in ascending official number order. This ensured that paired samples from each individual were assayed together to eliminate inter-batch variability.

    All biochemical assays were performed at the Department of Clinical Chemistry at St Bartholomew’s Hospital, London. Serum 25(OH)D levels were determined by IDS OCTEIA 25(OH)D assay (ELISA- Immuno Diagnostic Systems Limited, Boldon, Tyne and Wear; normal range 48–144 nmol/L, assay sensitivity 5 nmol/L, intra- and inter-batch variability 5.3% and 4.6%, respectively). MMP9 levels were determined by the Quantikine human MMP9 Immunoassay (ELISA- R and D Systems, Minneapolis, Minnesota, USA; normal range 13.2–105 ng/mL, sensitivity <0.156 ng/mL, intra- and inter-batch variability 2.9% and 7.9%, respectively). CRP levels were measured by Th1MULITE High Sensitivity CRP immunometric assay (Diagnostic Products Corporation (DPC), Los Angeles, California, USA; upper limit of normal l mg/L, sensitivity 0.02 mg/L, intra- and inter-batch variability 5.0% and 7.5%, respectively).

    Statistical analysis

    Data were analysed using SPSS V.11.5. Where necessary, non-parametric data were normalised using logarithmic transformation. Baseline demographics were compared using paired t tests. ANOVA was then performed to determine statistical differences between the unequally sized populations at two time points, and Pearson correlation analysis was used to determine the relationship between log vitamin D and MMP9 concentrations. Statistical significance was set at p<0.05.

    Results

    All 49 submariners completed the study. Two controls were unavailable for sampling at the second time point. The eight support crew members recruited as controls were separated out into a discrete group after initial data analysis. There were no significant differences in age, height, waist, body mass index or systolic blood pressure between groups (see table 1).

    View this table:
    Table 1

    Baseline demographics and blood levels between groups

    There was no significant difference in CRP levels between groups or time points. The mean 25(OH)D of the controls significantly increased from 57.9 nmol/L pre-patrol to 70.5 nmol/L post-patrol (lnVit D mean change=+0.19, 95% CI 0.01 to 0.38; see figure 1 and table 1). Mean 25(OH)D of the submariners showed a non-significant reduction from 49.2 nmol/L pre-patrol, to 47.6 nmol/L on day 85 (lnVit D mean change=−0.04, 95% CI 0.19 to 0.11). Pre-patrol vitamin D concentrations were not significantly different between groups, but post-deployment concentrations were significantly lower in submariners (see figure 1).

    Figure 1
    Figure 1

    Boxplot of changes in vitamin D concentration during patrol.

    The mean MMP9 of the controls and support crew did not significantly change during deployment (see figure 2 and table 1), while mean MMP9 levels of submariners increased significantly during deployment from 81.0 to 152.6 ng/mL on day 85 (lnMMP9 mean change=+0.63, 95% CI 0.32 to 0.94). There was a significant difference in MMP9 concentrations between controls and support crew and controls and submariners before patrol (see table 1).

    Figure 2
    Figure 2

    Boxplot of changes in matrix metalloproteinase-9 (MMP9) concentration during patrol.

    Twelve weeks later there was no significant difference in MMP9 concentrations between controls and support crew, but a significant difference between submariners and both control groups (p<0.001) was noted. Pearson correlation analysis was used to test for a univariate linear relationship between natural log (ln) of serum 25(OH)D and lnMMP-9. Figure 3 demonstrates the inverse linear relationship between post-deployment lnMMP9 and ln25(OH)D levels (r=−0.41, p=0.01).

    Figure 3
    Figure 3

    Correlation between post-deployment matrix metalloproteinase-9 (MMP9) and vitamin D concentration.

    Eleven controls, one member of the support crew and 17 submariners took daily vitamin supplements. Each took a preparation containing 5 µg vitamin D. There was no significant difference in 25(OH)D or MMP9 concentrations between those taking supplements, and those not taking supplements in any of the groups. Although this was not surprising due to sample size, there was no clear trend for increased vitamin D concentrations or associated MMP concentrations in the supplemented group in either the controls or submariners.

    Discussion

    Our results show that a submariner's dietary intake of vitamin D is insufficient to counteract the occupational absence of sunlight over a prolonged period submerged. The majority of vitamin D is produced when skin is exposed to UVB radiation from the sun and the amount of UVB in sunlight changes with latitude, season and time of day. The significant difference in circulating 25(OH)D between submariners and controls at the second sampling point resulted entirely from an increase observed from both control groups. The increase in controls was expected as the patrol finished in April, by which time UVB radiation levels allows vitamin D synthesis. All groups had borderline vitamin D insufficiency15 at the first sampling point, in January, which supports previous research suggesting that there is insufficient UVB radiation in Scotland between October and March for adequate vitamin D synthesis.16

    The lack of significant reduction in submariners’ vitamin D concentrations during patrol contrasts with previous studies. For example, mean 25(OH)D fell from 34.3 to 19.8 nmol/L (by 42%) over a 60-day patrol in seven British submariners deploying in June l975.3 ,17 A 30% reduction in 25(OH)D concentration was observed over 63 days in a study of 30 American submariners.18 The key reason for the contrasting findings in this study compared to previous studies is in the timing of the study period. By selecting a winter deployment, the submariner's vitamin D level pre-deployment would be much lower than a pre-deployment level in June. In the previous studies, where vitamin D concentrations in submariners significantly fell during deployment, the patrol occurred during the summer months; hence the submariners would otherwise have received adequate solar wavelength for vitamin D synthesis had they not been deployed undersea.

    Food derived vitamin D2 may also have contributed to the maintenance of vitamin D concentrations in this study compared with previous studies. It is possible that the diet of submariners has improved over the 30 years since the earlier study, in terms of both improvements in foodstuffs procured, and qualitative improvements in storage facilities onboard. Data are currently being collected regarding onboard dietary intake, with a view to routine food fortification. During deployment many submariners became vitamin D deficient with concentrations between 24 and 40 nmol/L.19 Studies suggests that an optimal 25(OH)D concentration exceeds 100 nmol/L.20 The highest recorded 25(OH)D concentration was 210 nmol/L in a Puerto Rican farmer.21 Importantly, for interventional purposes, this study also indicates that a single daily 5 µg vitamin D supplement was insufficient to compensate for UVB deprivation. Our results suggest that not only should we supplement submariners with appropriate vitamin D doses when deployed but also that we should consider the supplementation of all servicemen working in Scotland during the winter months.

    This is the first study to report on MMP9 concentrations in this occupational group. As MMP9 assay is not yet a clinical tool, no upper limit delineating underlying pathology has been defined; however some values obtained in submariners in late patrol exceed the manufacturer’s quoted reference range (13.2–105 ng/mL). The significance of a given MMP9 concentration for an individual is as yet unclear, and the balance of cause and effect unknown, but raised MMP9 might facilitate progression of pathological processes in the longer term. The limited studies to date report no excess morbidity or mortality in submariners.22 However, as follow-up increases, conditions with long latencies may be uncovered. We have shown in this study an inverse association of vitamin D with MMP9 levels. The higher MMP9 levels in submariners and support crew, but not in controls pre-patrol, is worthy of further consideration. The absence of significant difference in 25(OH)D concentrations between these groups indicates that the difference in MMP9 levels is not simply due to low vitamin D. MMP9 levels were not significantly different in the controls or support crew pre- and post-patrol but continued to rise significantly in the submariners. It may be that the fall in MMP9 concentrations observed in support crew members reflects rising 25(OH)D titres from sunlight exposure suppressing previously high MMP9 concentrations in submariners who had been on patrol the previous summer. An alternative explanation for the rise in MMP in submariners is exposure to an environmental stressor aboard the submarine. However all atmospheric constituents and contaminants stayed within the maximum permitted concentration for 90 days throughout the course of the patrol.23 The significant inverse correlation between 25(OH)D and MMP9 suggests that small reductions in 25(OH)D, while not statistically significant, may allow elevation of MMP9. Our initial hypothesis was that vitamin D suppressed intrinsic production of MMP9. Our results, however, suggest that a more complex interplay is present, with MMP9 production perhaps being triggered by an environmental stressor and suppressed to a varying degree by the presence of vitamin D. There is evidence in the literature, showing that environmental stressors such as ozone or cigarette smoke affect levels of specific MMPs, to support this.24 MMP9 production may then act as a driver in some diseases, such as malignancy and myocardial infarction that are associated with low vitamin D levels. One such environmental stressor could be atmospheric contaminants which, at the time of this study (prior to June 2007), would include those from the combustion of cigarettes. The lack of any association with CRP and deployment suggests that no acute inflammatory processes resulted from confinement for the duration of the patrol. Further investigation conducted on a patrol deploying at the end of summer would be useful to determine whether higher circulating 25(OH)D concentrations result in lower pre-patrol MMP9 concentrations and reduce or prevent a rise during patrol. Because of the unexpected maintenance of mean 25(OH)D during the studied patrol, future investigations should include individual UVB dosimetry using polysulphone badges. An assay which discriminates between 25(OH)D3 and 25(OH)D2 could also be used to distinguish between endogenously produced vitamin D and that derived from the diet.

    This study shows that the average submariner is vitamin D insufficient, both pre- and post-deployment. Two potential solutions are to replace onboard lighting with full spectrum tubes, and advise administration of higher dose oral vitamin D supplements to all submariners. As we derive most of our vitamin D from the action of UVB on our skin, the most physiologically acceptable countermeasure would be the use of UV light. A case report details use of a tanning bed to treat a long-term Crohn's disease sufferer.25 Over a 4-week period, 10 min of whole body exposure three times a week raised 25(OH)D levels from 50 to 80 nmol/L. Concerns regarding adverse health effects from UVB and the potential for acute inadvertent injury, combined with the lack of space, however, preclude the use of tanning beds. The need to limit disruption to routines dictates the use of lower dose subliminal exposures. Miniature UVB lamps (Phillips type PL-S 9W/l2) were used in a feasibility study of 10 elderly nursing home residents to deliver 0.3 standard erythema dose26 (the equivalent of a 15 min face and forearms exposure in UK summer). A significant increase in 25(OH)D was achieved between 12 and 24 weeks without other intervention and without erythema or keratitis. Exposure of the whole body to one standard erythema dose of sunlight is equivalent to ingesting 250 µg of vitamin D.27 Exposure of 10% of the body surface is equivalent to 25 µg. Work patterns aboard submarines are such that crew members are either at work or sleeping. There is little communal association. The siting of lamps requires careful consideration; with a possibility being dining areas or passageways which all must traverse. There are no published reports of the use of UV lamps aboard submarines. The National Aeronautics and Space Administration (NASA) is, however considering use of UVB fluorescent tubes to mitigate bone demineralisation in astronauts.28

    Vitamin supplementation is the other possible countermeasure. The UK guidance for adults confined indoors is 10 µg/day.2 Astronauts aboard the International Space Station receive 10 µg daily. A study of veiled Arab Moslem women living in Denmark suggests that the daily requirement for sunlight-deprived individuals should be 20–25 µg/day.29 An extensive review addressing concerns about vitamin D toxicity concluded that there is no evidence of adverse effects with 25(OH)D concentrations below 140 nmol/L, to achieve which would require daily doses of 250 µg.21 Vitamin D toxicity with hypercalcaemia requires an intake greater than 1000 µg. We would certainly recommend a dose higher than the 5 µg that some of the submariners and controls in this study were taking. These findings also have implications for other workforces where working environment precludes sunlight exposure, for example night shift workers and those who work underground, in confined space or at extremes of latitude.

    Future work should establish whether there is a causal relationship between plasma MMP9 concentration and increased risk of morbidity; and if vitamin D insufficiency/deficiency results in an elevation of MMP9 concentration. If the above links are established, it is recommended that this study be repeated to investigate MMP9 concentrations during a summer or early autumn patrol when initial 25(OH) vitamin D concentrations are presumed to be higher and now that all submarines are non-smoking. This would help establish whether vitamin D repletion prevents a rise in MMP9 and confirm the inverse relationship between vitamin D status and MMP9 concentration in this present study. Intervention strategies should be examined to determine the effects of subliminal UVB lighting and oral vitamin D supplementation on vitamin D and MMP9 plasma concentrations.

    Conclusions

    Pre-patrol 25(OH)D concentrations were within the reference range for latitude and season but at the lower limit of physiological sufficiency. The mean 25(OH)D concentrations of the controls increased by late April, as expected. The mean 25(OH)D concentrations of the submariners did not fall significantly in this study, unlike those reported previously. Oral supplementation with 5 µg vitamin D had no effect on 25(OH)D concentrations and was insufficient to prevent a rise in MMP9 concentration in submariners on patrol. MMP9 concentrations were higher in submariners than controls pre-patrol; they rose further during the patrol, fell in support crew members, and did not rise in controls. MMP9 concentrations were inversely correlated with 25(OH) vitamin D concentrations, suggesting a suppressive effect of vitamin D, although it is possible that other environmental stressors may also be contributing. The long-term significance of this rise in MMP9 on the health of submariners is, as yet, unknown and this potential for adverse health effects requires further study.

    Acknowledgments

    Surg Cdr Anthony Dew and Surg Cdr Paul Turnbull for their assistance and cooperation during the conduct of this study. © British Crown Copyright 2013/Ministry of Defence. Published with the permission of the Controller of Her Majesty's Stationery Office.

    References

    1. ↵
      1. Bowden L,
      2. Jones C,
      3. Ryan S
      . Bone mineralisation in ex-preterm infants aged 8 years. Eur J Pediatr 1999;158:658–61.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      Department of Health Report on Health and Social Subjects 41. Dietary reference values for food energy and nutrients for the United Kingdom. In: HMSO ed. London: 1991;123–7.
    3. ↵
      1. Preece MA,
      2. Tomlinson S,
      3. Ribot CA,
      4. et al
      . Studies of vitamin D deficiency in man. Q J Med 1975;44:575–89.
      OpenUrlAbstract/FREE Full Text
    4. ↵
      1. Dlugos DJ,
      2. Perrotta PL,
      3. Horn WG
      . Effects of the submarine environment on renal-stone risk factors and vitamin D metabolism. Undersea Hyperb Med 1995;22:145–52.
      OpenUrlPubMedWeb of Science
    5. ↵
      1. Scragg R,
      2. Jackson R,
      3. Holdaway IM,
      4. et al
      . Myocardial infarction is inversely associated with plasma 25-hydroxyvitamin D3 levels: a community-based study. Int J Epidemiol 1990;19:559–63.
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Bikle DD
      . Clinical counterpoint: vitamin D: new actions, new analogs, new therapeutic potential. Endocr Rev 1992;13:765–84.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Stamp TC,
      2. Round JM
      . Seasonal changes in human plasma levels of 25-hydroxyvitamin D. Nature 1974;247:563–5.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Boland R
      . Role of vitamin D in skeletal muscle function. Endocr Rev 1986;7:434–48.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Ahonen MH,
      2. Tenkanen L,
      3. Teppo L,
      4. et al
      . Prostate cancer risk and prediagnostic serum 25-hydroxyvitamin D levels (Finland). Cancer Causes Control 2000;11:847–52.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Garland C,
      2. Shekelle RB,
      3. Barrett-Connor E,
      4. et al
      . Dietary vitamin D and calcium and risk of colorectal cancer: a 19-year prospective study in men. Lancet 1985;1:307–9.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Timms PM,
      2. Mannan N,
      3. Hitman GA,
      4. et al
      . Circulating MMP9, vitamin D and variation in the TIMP-1 response with VDR genotype: mechanisms for inflammatory damage in chronic disorders? QJM 2002;95:787–96.
      OpenUrlAbstract/FREE Full Text
    12. ↵
      1. Nagase H,
      2. Woessner JF Jr.
      . Matrix metalloproteinases. J Biol Chem 1999;274:21491–4.
      OpenUrlFREE Full Text
    13. ↵
      1. Dollery CM,
      2. McEwan JR,
      3. Henney AM
      . Matrix metalloproteinases and cardiovascular disease. Circ Res 1995;77:863–8.
      OpenUrlFREE Full Text
    14. ↵
      1. Zerwekh JE
      . Blood biomarkers of vitamin D status. Am J Clin Nutr 2008;87:1087S–91S.
      OpenUrlAbstract/FREE Full Text
    15. ↵
      1. Malabanan A,
      2. Veronikis IE,
      3. Holick MF
      . Redefining vitamin D insufficiency. Lancet 1998;351:805–6.
      OpenUrlPubMedWeb of Science
    16. ↵
      1. Sullivan SS,
      2. Rosen CJ,
      3. Halteman WA,
      4. et al
      . Adolescent girls in Maine are at risk for vitamin D insufficiency. J Am Diet Assoc 2005;105:971–4.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Davies DM,
      2. Morris JE
      . Carbon dioxide and vitamin D effects on calcium metabolism in nuclear submariners: a review. Undersea Biomed Res 1979;6(Suppl):S71–80.
      OpenUrlPubMedWeb of Science
    18. ↵
      1. Gilman SC,
      2. Biersner RJ,
      3. Bondi KR
      . Effect of a 68-day submarine patrol on serum 25-hydroxyvitamin D levels in healthy men. Int J Vitam Nutr Res 1982;52:63–7.
      OpenUrlPubMed
    19. ↵
      1. Thomas MK,
      2. Lloyd-Jones DM,
      3. Thadhani RI,
      4. et al
      . Hypovitaminosis D in medical inpatients. N Engl J Med 1998;338:777–83.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Pearce SH,
      2. Cheetham TD
      . Diagnosis and management of vitamin D deficiency. BMJ 2010;340:b5664.
      OpenUrlFREE Full Text
    21. ↵
      1. Vieth R
      . Vitamin D supplementation, 25-hydroxyvitamin D concentrations, and safety. Am J Clin Nutr 1999;69:842–56.
      OpenUrlAbstract/FREE Full Text
    22. ↵
      1. Inskip H,
      2. Snee M,
      3. Styles L
      . The mortality of Royal Naval submariners 1960–89. Occup Environ Med 1997;54:209–15.
      OpenUrlAbstract/FREE Full Text
    23. ↵
      Unpublished Military Internal Report FF 4135/23. HMS Vigilant: Atmosphere control report and log 29, 2 Aug 2002.
    24. ↵
      1. Fortino V,
      2. Maioli E,
      3. Torricelli C,
      4. et al
      . Cutaneous MMPs are differently modulated by environmental stressors in old and young mice. Toxicol Lett 2007;173:73–9.
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Koutkia P,
      2. Lu Z,
      3. Chen TC,
      4. et al
      . Treatment of vitamin D deficiency due to Crohn's disease with tanning bed ultraviolet B radiation. Gastroenterology 2001;121:1485–8.
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Chuck A,
      2. Todd J,
      3. Diffey B
      . Subliminal ultraviolet-B irradiation for the prevention of vitamin D deficiency in the elderly: a feasibility study. Photodermatol Photoimmunol Photomed 2001;17:168–71.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Holick MF
      . Sunlight ‘D'ilemma: risk of skin cancer or bone disease and muscle weakness. Lancet 2001;357:4–6.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Eaton LL
      . Ultraviolet fluorescent light for vitamin D replacement on long duration space flights. Aviat Space Environ Med 2002;73:270.
    29. ↵
      1. Glerup H,
      2. Mikkelsen K,
      3. Poulsen L,
      4. et al
      . Commonly recommended daily intake of vitamin D is not sufficient if sunlight exposure is limited. J Intern Med 2000;247:260–8.
      OpenUrlCrossRefPubMedWeb of Science

    Footnotes

    • Contributors All authors participated in interpreting the data and critically reviewing the manuscript. CLW and AB wrote the first draft. AB carried out the study assessments.

    • Funding This study was supported by the Royal Navy.

    • Competing interests None.

    • Patient consent Obtained.

    • Ethics approval The Ministry of Defence Research Ethics Committee.

    • Provenance and peer review Not commissioned; externally peer reviewed.