Correspondence
Vitamin D status and ill health Philippe Autier and colleagues1 should be congratulated for providing a thorough overview and thoughtful discussion of the numerous studies showing associations between low serum concentrations of vitamin D and various causes of ill health in the context of the disappointing lack of response to oral vitamin D supplementation. The authors suggest that this finding might be attributable to reverse causation (ie, ill health lessens behaviour leading to exposure to the sun) or to systemic inflammation lowering vitamin D concentrations. We believe that they have overlooked another parsimonious and likely explanation: that circulating vitamin D is a biomarker for sunlight exposure, and that the sun is—for most people and when enjoyed in moderation—beneficial to health. Exposure to ultraviolet light is carcinogenic to the skin, and extensive research has delineated the various mechanisms involved. Despite almost a century of work however, no reports suggest that exposure to sunlight shortens life. Conversely, two recent epidemiological studies show that although increased sun exposure was associated with increased incidence of melanoma, all-cause mortality was reduced;2 and that the presence of nonmelanoma skin cancer is associated with a 50% lower odds ratio for allcause death.3 High blood pressure is the leading cause of disability-adjusted lifeyears, accounting for 18% of all deaths worldwide. Population blood pressure correlates directly with latitude and is lower in summer than winter. We have shown that human skin contains large stores of nitrogen oxides, which are mobilised to the systemic circulation by exposure of the body to ultraviolet A light, causing arterial vasodilatation and a decrease in blood pressure.4 This effect is independent of vitamin D. We also noted that the suppressive effects of ultraviolet radiation on immune-driven
diseases such as atopic dermatitis and asthma are independent of circulating vitamin D concentration, at least in animals.5 Vitamin D might also be involved in these processes, but the sun seems beneficial to health via a mechanism of action unrelated to vitamin D. Public health advice on sun exposure, driven by dermatological concerns over skin cancer, has failed to take into account the possible benefits of sunlight. These benefits have become conflated with vitamin D, which we suggest only accounts for some of the positive aspects of sun exposure. Ultraviolet radiation has a wide range of biological effects other than carcinogenesis. In view of the epidemiological and mechanistic data showing improvements induced by sun in some of the major causes of disability-adjusted life-years, advice on healthy sun exposure needs to be reconsidered. We declare that we have no competing interests.
Martin Feelisch, Shelley Gorman, *Richard B Weller [email protected] Clinical and Experimental Sciences, Faculty of Medicine, University of Southampton, Southampton General Hospital, Southampton, UK (MF); Telethon Institute for Child Health Research, The University of Western Australia, Perth, WA, Australia (SG); and University of Edinburgh, MRC Centre for Inflammation Research, Edinburgh, UK (RBW) 1
2
3
4
5
Autier P, Boniol M, Pizot C, Mullie P. Vitamin D status and ill health: a systematic review. Lancet Diabetes Endocrinol 2014; 2: 76–89. Yang L, Lof M, Veierod MB, Sandin S, Adami HO, Weiderpass E. Ultraviolet exposure and mortality among women in Sweden. Cancer Epidemiol Biomarkers Prev 2011; 20: 683–90. Brondum-Jacobsen P, Nordestgaard BG, Nielsen SF, Benn M. Skin cancer as a marker of sun exposure associates with myocardial infarction, hip fracture and death from any cause. Int J Epidemiol 2013; 42: 1486–96. Liu D, Fernandez BO, Hamilton A, et al. UVA irradiation of human skin vasodilates arterial vasculature and lowers blood pressure independently of nitric oxide synthase. J Invest Dermatol 2014; published online Feb 20. DOI:10.1038/jid.2014.27. Gorman S, Scott NM, Tan DH et al. Acute erythemal ultraviolet radiation causes systemic immunosuppression in the absence of increased 25-hydroxyvitamin D3 levels in male mice. PLoS One 2012; 7: e46006.
www.thelancet.com/diabetes-endocrinology Vol 2 April 2014
In their Review,1 Philippe Autier and colleagues reasoned that without evidence from randomised controlled trials, we cannot assume a causal association between vitamin D and prevention of chronic disease. We look forward to the results of large trials that are underway, but we hope that the results will not lead to years of debate, as happened with the Women’s Health Initiative study.2–4 We should be mindful that individuals who are most deficient in vitamin D are the ones most likely to benefit from supplementation, and exercise caution when extrapolating to the general population (most of whom have adequate vitamin D status for their needs, at least in the summer and autumn at high latitudes). Despite concern about vitamin D deficiency in winter, no data exist to support the notion that the seasonal fall in circulating 25(OH)D to less than optimum concentrations in the winter months is directly associated with poor health outcomes. In this regard, small studies might provide a better controlled system for several analytical outcome measurements, with more frequent face-to-face contact with participants. Our randomised controlled trial5 of 300 older women tested two daily vitamin D doses (400 IU and 1000 IU) against placebo, with 87% completing the final visit, and was unique in that it was specifically designed to account for changes due to season, and participants attended follow-up twice a month for one year. The women, although outwardly healthy, started the trial with low vitamin D status (median 25[OH]D 34 nmol/L). Our study design uncovered a seasonal pattern of blood pressure that was independent of vitamin D treatment. As systolic blood pressure went down in the summer (by a mean of 6·6 mm Hg [SD 10·8]) and back up in the winter, the decrease was associated with increased vitamin D status in the placebo group. Thus, a causal association might have been assumed, but the same pattern was noted in both vitamin D treatment groups. e8
Correspondence
No changes occurred in markers of cardiovascular risk, glucose sensitivity, or inflammation (apart from a small change in apolipoprotein B100 that was of debatable clinical significance). Our study suggests that vitamin D status might not only be a marker of ill health, as concluded by Autier and colleagues,1 but also an indicator of other effects related to season or sunlight levels that might benefit long-term health. We declare that we have no competing interests.
*Helen M Macdonald, Adrian D Wood, William D Fraser, William G Simpson [email protected] School of Medicine and Dentistry University of Aberdeen, Foresterhill, Aberdeen AB25 2ZD, UK (HMM, ADW); Norwich Medical School, University of East Anglia, Norwich, UK (WDF); and Department of Clinical Biochemistry, Aberdeen Royal Infirmary, Foresterhill, UK (WGS) 1
2
3
4
5
Autier P, Boniol M, Pizot C, Mullie P. Vitamin D status and ill health: a systematic review. Lancet Diabetes Endocrinol 2014; 2: 76–89. Jackson RD, LaCroix AZ, Gass M, et al, for the Women’s Health Initiative Investigators. Calcium plus vitamin D supplementation and the risk of fractures. N Engl J Med 2006; 354: 669–83. Bolland MJ, Grey A, Gamble GD, Reid IR. Calcium and vitamin D supplements and health outcomes: a reanalysis of the Women’s Health Initiative (WHI) limited-access data set. Am J Clin Nutr 2011; 94: 1144–49. Prentice RL, Pettinger MB, Jackson RD, et al. Health risks and benefits from calcium and vitamin D supplementation: Women’s Health Initiative clinical trial and cohort study. Osteoporos Int 2013; 24: 567–80. Wood AD, Secombes KR, Thies F, et al. Vitamin D3 supplementation has no effect on conventional cardiovascular risk factors: a parallel-group, double-blind, placebo-controlled RCT. J Clin Endocrinol Metab 2012; 97: 3557–68.
In their review,1 Philippe Autier and colleagues examined the evidence base relating vitamin D status and supplementation to various health outcomes, and concluded that available data are insufficient to deduce a causal role for vitamin D in such conditions. This conclusion is concordant with our own findings, which resulted from a similar approach with a systematic review and metaanalysis, in the examination of the role of vitamin D in pregnancy in terms of health of the mother and child. Funded by UK National Institute for e9
Health Research Health Technology Assessment, and with standard UK Centre for Reviews and Dissemination procedures, we identified 76 articles for review from 16 842 citations (excluding duplications).2 We noted evidence of positive associations between numerous health outcomes and maternal 25-hydroxyvitamin D (25[OH]D) status, including birthweight (the pooled regression coefficient after adjustment for potential confounding factors was 5·63 g [95% CI 1·11–10·16] change per 10% change in maternal 25[OH]D serum concentration in a meta-analysis of three observational studies), offspring cord blood or postnatal calcium concentrations (meta-analysis of six intervention studies, which were all deemed to be at high risk of bias, with concentrations a mean of 0·05 mmol/L (95% CI 0·02, 0·05) greater in offspring of mothers supplemented versus unsupplemented with vitamin D in pregnancy, and offspring bone mass (positive relationship with maternal serum 25(OH)D in pregnancy, in observational studies deemed to be of good quality, but which did not permit meta-analysis). Overall, we noted substantial heterogeneity between the studies and conflicting evidence for most outcomes. Indeed, we noted no convincing evidence for any association between maternal vitamin D status and conditions previously linked to vitamin D deficiency, such as offspring asthma, atopy, type 1 diabetes, or blood pressure. Our conclusion echoed that of both Autier and colleagues1 and the editors of The Lancet Diabetes & Endocrinology,3 in that insufficient evidence exists on which to base any clinical recommendations, and that high quality randomised trials are needed.4 Such trials are now ongoing, for example the vitamin D and omega-3 trial (VITAL) study in the USA, and our own randomised controlled trial in pregnant women (MAVIDOS),5 which aims to optimise offspring bone mass.
Investigations such as these should help to answer the questions posed by previous observational studies. Hopefully, in the next 5–10 years, the evidence supporting or refuting a role for vitamin D supplementation in human health will become much more clearly documented. We declare that we have no competing interests.
*Nicholas C Harvey, Cyrus Cooper [email protected] MRC Lifecourse Epidemiology Unit, University of Southampton, Southampton SO16 6YD, UK (NCH, CC); NIHR Southampton Biomedical Research Centre, University of Southampton and University Hospital Southampton NHS Foundation Trust, Southampton, UK (NCH, CC); and NIHR Musculoskeletal Biomedical Research Unit, University of Oxford, Oxford, UK (CC) 1
2
3.
4 5
Autier P, Boniol M, Pizot C, Mullie P. Vitamin D status and ill health: a systematic review. Lancet Diabetes Endocrinol 2014; 2: 76–89. Harvey NC, Holroyd C, Ntani G, et al. Vitamin D supplementation in pregnancy: a systematic review. HTA Journal (in press). The Lancet Diabetes & Endocrinology. Vitamin D: chasing a myth? Lancet Diabetes Endocrinol 2014; 2: 1. Harvey NC, Cooper C. Vitamin D: some perspective please. BMJ 2012; 345: e4695. Harvey NC, Javaid K, Bishop N, et al. MAVIDOS maternal vitamin D osteoporosis study: study protocol for a randomized controlled trial. Trials 2012; 13: 13.
In their Review,1 Philippe Autier and colleagues, point out several limitations related to studies linking vitamin D status and health outcomes and suggest that 25-hydroxyvitamin D (25[OH]D) concentrations might be a biological marker of deteriorating health instead of having a causal role. However, no mention is made that the studies reviewed did not take into account serum bioavailable concentrations of 25(OH)D, which might be a more appropriate marker of vitamin D status than total serum concentration. 2 Notably, Powe et al2 showed that some polymorphisms in the vitamin D-binding protein that are prevalent in black Americans, but also present in white Americans, could explain why some individuals with low serum concentrations of total 25(OH)D can show normal values of bioavailable 25(OH)D. These findings might
www.thelancet.com/diabetes-endocrinology Vol 2 April 2014
Vitamin D status and ill health Philippe Autier and colleagues1 should be congratulated for providing a thorough overview and thoughtful discussion of the numerous studies showing associations between low serum concentrations of vitamin D and various causes of ill health in the context of the disappointing lack of response to oral vitamin D supplementation. The authors suggest that this finding might be attributable to reverse causation (ie, ill health lessens behaviour leading to exposure to the sun) or to systemic inflammation lowering vitamin D concentrations. We believe that they have overlooked another parsimonious and likely explanation: that circulating vitamin D is a biomarker for sunlight exposure, and that the sun is—for most people and when enjoyed in moderation—beneficial to health. Exposure to ultraviolet light is carcinogenic to the skin, and extensive research has delineated the various mechanisms involved. Despite almost a century of work however, no reports suggest that exposure to sunlight shortens life. Conversely, two recent epidemiological studies show that although increased sun exposure was associated with increased incidence of melanoma, all-cause mortality was reduced;2 and that the presence of nonmelanoma skin cancer is associated with a 50% lower odds ratio for allcause death.3 High blood pressure is the leading cause of disability-adjusted lifeyears, accounting for 18% of all deaths worldwide. Population blood pressure correlates directly with latitude and is lower in summer than winter. We have shown that human skin contains large stores of nitrogen oxides, which are mobilised to the systemic circulation by exposure of the body to ultraviolet A light, causing arterial vasodilatation and a decrease in blood pressure.4 This effect is independent of vitamin D. We also noted that the suppressive effects of ultraviolet radiation on immune-driven
diseases such as atopic dermatitis and asthma are independent of circulating vitamin D concentration, at least in animals.5 Vitamin D might also be involved in these processes, but the sun seems beneficial to health via a mechanism of action unrelated to vitamin D. Public health advice on sun exposure, driven by dermatological concerns over skin cancer, has failed to take into account the possible benefits of sunlight. These benefits have become conflated with vitamin D, which we suggest only accounts for some of the positive aspects of sun exposure. Ultraviolet radiation has a wide range of biological effects other than carcinogenesis. In view of the epidemiological and mechanistic data showing improvements induced by sun in some of the major causes of disability-adjusted life-years, advice on healthy sun exposure needs to be reconsidered. We declare that we have no competing interests.
Martin Feelisch, Shelley Gorman, *Richard B Weller [email protected] Clinical and Experimental Sciences, Faculty of Medicine, University of Southampton, Southampton General Hospital, Southampton, UK (MF); Telethon Institute for Child Health Research, The University of Western Australia, Perth, WA, Australia (SG); and University of Edinburgh, MRC Centre for Inflammation Research, Edinburgh, UK (RBW) 1
2
3
4
5
Autier P, Boniol M, Pizot C, Mullie P. Vitamin D status and ill health: a systematic review. Lancet Diabetes Endocrinol 2014; 2: 76–89. Yang L, Lof M, Veierod MB, Sandin S, Adami HO, Weiderpass E. Ultraviolet exposure and mortality among women in Sweden. Cancer Epidemiol Biomarkers Prev 2011; 20: 683–90. Brondum-Jacobsen P, Nordestgaard BG, Nielsen SF, Benn M. Skin cancer as a marker of sun exposure associates with myocardial infarction, hip fracture and death from any cause. Int J Epidemiol 2013; 42: 1486–96. Liu D, Fernandez BO, Hamilton A, et al. UVA irradiation of human skin vasodilates arterial vasculature and lowers blood pressure independently of nitric oxide synthase. J Invest Dermatol 2014; published online Feb 20. DOI:10.1038/jid.2014.27. Gorman S, Scott NM, Tan DH et al. Acute erythemal ultraviolet radiation causes systemic immunosuppression in the absence of increased 25-hydroxyvitamin D3 levels in male mice. PLoS One 2012; 7: e46006.
www.thelancet.com/diabetes-endocrinology Vol 2 April 2014
In their Review,1 Philippe Autier and colleagues reasoned that without evidence from randomised controlled trials, we cannot assume a causal association between vitamin D and prevention of chronic disease. We look forward to the results of large trials that are underway, but we hope that the results will not lead to years of debate, as happened with the Women’s Health Initiative study.2–4 We should be mindful that individuals who are most deficient in vitamin D are the ones most likely to benefit from supplementation, and exercise caution when extrapolating to the general population (most of whom have adequate vitamin D status for their needs, at least in the summer and autumn at high latitudes). Despite concern about vitamin D deficiency in winter, no data exist to support the notion that the seasonal fall in circulating 25(OH)D to less than optimum concentrations in the winter months is directly associated with poor health outcomes. In this regard, small studies might provide a better controlled system for several analytical outcome measurements, with more frequent face-to-face contact with participants. Our randomised controlled trial5 of 300 older women tested two daily vitamin D doses (400 IU and 1000 IU) against placebo, with 87% completing the final visit, and was unique in that it was specifically designed to account for changes due to season, and participants attended follow-up twice a month for one year. The women, although outwardly healthy, started the trial with low vitamin D status (median 25[OH]D 34 nmol/L). Our study design uncovered a seasonal pattern of blood pressure that was independent of vitamin D treatment. As systolic blood pressure went down in the summer (by a mean of 6·6 mm Hg [SD 10·8]) and back up in the winter, the decrease was associated with increased vitamin D status in the placebo group. Thus, a causal association might have been assumed, but the same pattern was noted in both vitamin D treatment groups. e8
Correspondence
No changes occurred in markers of cardiovascular risk, glucose sensitivity, or inflammation (apart from a small change in apolipoprotein B100 that was of debatable clinical significance). Our study suggests that vitamin D status might not only be a marker of ill health, as concluded by Autier and colleagues,1 but also an indicator of other effects related to season or sunlight levels that might benefit long-term health. We declare that we have no competing interests.
*Helen M Macdonald, Adrian D Wood, William D Fraser, William G Simpson [email protected] School of Medicine and Dentistry University of Aberdeen, Foresterhill, Aberdeen AB25 2ZD, UK (HMM, ADW); Norwich Medical School, University of East Anglia, Norwich, UK (WDF); and Department of Clinical Biochemistry, Aberdeen Royal Infirmary, Foresterhill, UK (WGS) 1
2
3
4
5
Autier P, Boniol M, Pizot C, Mullie P. Vitamin D status and ill health: a systematic review. Lancet Diabetes Endocrinol 2014; 2: 76–89. Jackson RD, LaCroix AZ, Gass M, et al, for the Women’s Health Initiative Investigators. Calcium plus vitamin D supplementation and the risk of fractures. N Engl J Med 2006; 354: 669–83. Bolland MJ, Grey A, Gamble GD, Reid IR. Calcium and vitamin D supplements and health outcomes: a reanalysis of the Women’s Health Initiative (WHI) limited-access data set. Am J Clin Nutr 2011; 94: 1144–49. Prentice RL, Pettinger MB, Jackson RD, et al. Health risks and benefits from calcium and vitamin D supplementation: Women’s Health Initiative clinical trial and cohort study. Osteoporos Int 2013; 24: 567–80. Wood AD, Secombes KR, Thies F, et al. Vitamin D3 supplementation has no effect on conventional cardiovascular risk factors: a parallel-group, double-blind, placebo-controlled RCT. J Clin Endocrinol Metab 2012; 97: 3557–68.
In their review,1 Philippe Autier and colleagues examined the evidence base relating vitamin D status and supplementation to various health outcomes, and concluded that available data are insufficient to deduce a causal role for vitamin D in such conditions. This conclusion is concordant with our own findings, which resulted from a similar approach with a systematic review and metaanalysis, in the examination of the role of vitamin D in pregnancy in terms of health of the mother and child. Funded by UK National Institute for e9
Health Research Health Technology Assessment, and with standard UK Centre for Reviews and Dissemination procedures, we identified 76 articles for review from 16 842 citations (excluding duplications).2 We noted evidence of positive associations between numerous health outcomes and maternal 25-hydroxyvitamin D (25[OH]D) status, including birthweight (the pooled regression coefficient after adjustment for potential confounding factors was 5·63 g [95% CI 1·11–10·16] change per 10% change in maternal 25[OH]D serum concentration in a meta-analysis of three observational studies), offspring cord blood or postnatal calcium concentrations (meta-analysis of six intervention studies, which were all deemed to be at high risk of bias, with concentrations a mean of 0·05 mmol/L (95% CI 0·02, 0·05) greater in offspring of mothers supplemented versus unsupplemented with vitamin D in pregnancy, and offspring bone mass (positive relationship with maternal serum 25(OH)D in pregnancy, in observational studies deemed to be of good quality, but which did not permit meta-analysis). Overall, we noted substantial heterogeneity between the studies and conflicting evidence for most outcomes. Indeed, we noted no convincing evidence for any association between maternal vitamin D status and conditions previously linked to vitamin D deficiency, such as offspring asthma, atopy, type 1 diabetes, or blood pressure. Our conclusion echoed that of both Autier and colleagues1 and the editors of The Lancet Diabetes & Endocrinology,3 in that insufficient evidence exists on which to base any clinical recommendations, and that high quality randomised trials are needed.4 Such trials are now ongoing, for example the vitamin D and omega-3 trial (VITAL) study in the USA, and our own randomised controlled trial in pregnant women (MAVIDOS),5 which aims to optimise offspring bone mass.
Investigations such as these should help to answer the questions posed by previous observational studies. Hopefully, in the next 5–10 years, the evidence supporting or refuting a role for vitamin D supplementation in human health will become much more clearly documented. We declare that we have no competing interests.
*Nicholas C Harvey, Cyrus Cooper [email protected] MRC Lifecourse Epidemiology Unit, University of Southampton, Southampton SO16 6YD, UK (NCH, CC); NIHR Southampton Biomedical Research Centre, University of Southampton and University Hospital Southampton NHS Foundation Trust, Southampton, UK (NCH, CC); and NIHR Musculoskeletal Biomedical Research Unit, University of Oxford, Oxford, UK (CC) 1
2
3.
4 5
Autier P, Boniol M, Pizot C, Mullie P. Vitamin D status and ill health: a systematic review. Lancet Diabetes Endocrinol 2014; 2: 76–89. Harvey NC, Holroyd C, Ntani G, et al. Vitamin D supplementation in pregnancy: a systematic review. HTA Journal (in press). The Lancet Diabetes & Endocrinology. Vitamin D: chasing a myth? Lancet Diabetes Endocrinol 2014; 2: 1. Harvey NC, Cooper C. Vitamin D: some perspective please. BMJ 2012; 345: e4695. Harvey NC, Javaid K, Bishop N, et al. MAVIDOS maternal vitamin D osteoporosis study: study protocol for a randomized controlled trial. Trials 2012; 13: 13.
In their Review,1 Philippe Autier and colleagues, point out several limitations related to studies linking vitamin D status and health outcomes and suggest that 25-hydroxyvitamin D (25[OH]D) concentrations might be a biological marker of deteriorating health instead of having a causal role. However, no mention is made that the studies reviewed did not take into account serum bioavailable concentrations of 25(OH)D, which might be a more appropriate marker of vitamin D status than total serum concentration. 2 Notably, Powe et al2 showed that some polymorphisms in the vitamin D-binding protein that are prevalent in black Americans, but also present in white Americans, could explain why some individuals with low serum concentrations of total 25(OH)D can show normal values of bioavailable 25(OH)D. These findings might
www.thelancet.com/diabetes-endocrinology Vol 2 April 2014
