Calcium supplements with or without vitamin D and risk of cardiovascular events: reanalysis of the Women’s Health Initiative limited access dataset and meta-analysis

We have already responded for the BMJ print edition to several of the
correspondents.1 Here, we respond to other issues raised.

Several correspondents have raised issues regarding the statistical
analysis. We wish to point out that the paper was formally reviewed by an
independent external senior academic statistician as part of the BMJ's
review process.

There was no significant interaction between supplement type and risk
of cardiovascular events with calcium (P>0.3), but the data for calcium
citrate supplements are much fewer.

We disagree with Burrill, who suggests that the participants in the
trials in our meta-analyses (postmenopausal women and older men) are not
those for whom calcium is commonly recommended. Calcium supplementation is
recommended by leading authorities 2,3 for improving skeletal health in
such individuals. In considering the risks and benefits of calcium
supplements in the populations studied in the trials in our meta-analysis,
it is not appropriate to invoke the absolute benefit of calcium and
vitamin D on fracture in a group of vitamin D deficient elderly women,
living in residential care, with very high fracture risk. The participants
in contributing trials in our meta-analysis had much lower fracture risk,
and consequently a much lower absolute benefit from calcium. The
appropriate analysis is outlined in our paper: in populations of community
-dwelling older men and women, the NNT with calcium or CaD for 5 years to
cause 1 MI/stroke was 178, and to prevent 1 fracture was 302. Treating
1000 people with calcium or CaD for 5 years would cause an additional 6 MI
or strokes, and prevent 3 fractures.4 Older, frailer groups are at higher
risk of both fractures and heart attacks. In a recent RCT in this
population group, calcium supplements increased the risk of all cause and
cardiovascular mortality, 5,6 suggesting that the balance of risk versus
benefit remains negative.

Vavasour suggests that the increased risk of cardiovascular events
with calcium might have been because the study cohorts in our meta-
analysis were vitamin K deficient. No measurements of vitamin K status, or
clotting or carboxylation status were carried out, but we have no reason
to think that the participants were at particular risk of diseases or
symptoms associated with vitamin K deficiency.

Moss says that participants in the trials had adequate dietary
calcium intake of approximately 800 mg/day, which may well be true.7 Her
claims that magnesium, vitamin C, vitamin K and omega three fatty acids,
and that specific ratios of individual nutrients, are required for
skeletal health are not supported by rigorous evidence. She implies that
co-supplementing calcium with magnesium would prevent adverse effects of
calcium, and that prescribing should be individualised. The onus falls on
those who make such claims to show that, when prescribed in this manner,
calcium supplements are effective and safe.

We have previously addressed most of Prince and colleagues'
comments.1 They also raise the issue of a supposed discrepancy between
numbers of events in our previous meta-analysis4 and the primary
publication.8 The apparent difference is due to the inclusion of strokes
reported from death certificates and hospital admission data in our meta-
analysis, and the primary paper reporting strokes from hospital admission
data only. This information is clearly stated in both papers.

Sabino asks whether arterial calcium scores are available. A substudy
of coronary calcium scores was carried out in WHI, but these data are not
available in the publicly available database for us to analyse further. We
have previously reported that calcium supplements did not affect
prevalence of aortic calcification over 5 years in older women, or the
coronary calcium score over 2 years in middle aged and older men,9 though
Daly reported that calcium-fortified milk increased abdominal aortic
calcification in an RCT in men.10

Champion and colleagues provide interesting data on calcium use in
the intensive care unit, but our findings have little relevance to that
situation.

Razzaque suggests that our findings might be mediated by increased
phosphate levels, but calcium supplements lower not increase serum
phosphate.11

Fisher and Davis argue that because a number of factors that
putatively affect mineral metabolism were not analysed at baseline, the
results of the ensuing RCTs are invalid. If correct, this argument would
invalidate any RCT. However, the argument is specious, because the
randomization process balances the treatment groups for all covariables,
whether measured or unmeasured, particularly so in large RCTs. They
suggest that our findings might be caused by secondary
hyperparathyroidism, but this is unlikely because calcium supplements
decrease PTH levels,12 which the authors state should lower cardiovascular
risks. We disagree with their view that observational studies should be
given greater weighting than meta-analyses of RCTs, and we have previously
highlighted13 that the underpowered trial of Lewis is included in our meta
-analyses, and that the results of the Wang meta-analysis are entirely
consistent with our much larger analysis. Finally, Fisher and Davis's
claim that there is no evidence for effectiveness of treatments for
osteoporosis without co-administration of calcium is not correct. Both
estrogen and clodronate administered without calcium reduced fracture
incidence.14-16 An RCT of calcium plus alendronate versus alendronate
alone reported no difference in effects on bone density.17 An RCT of
zoledronate in which calcium was not used18 showed similar effects on bone
density and bone turnover to trials of zoledronate in which calcium was
used.19-21 The implication that calcium supplements are required for
efficacy of anti-resorptive agents is not evidence-based. No trials have
been conducted that demonstrate that an anti-resorptive agent plus calcium
has superior efficacy to that of the anti-resorptive agent alone.

Tran and Nguyen suggest that the rationale for our study was weak,
based upon the baseline characteristics in WHI. However, the rationale was
clearly outlined in our paper, and endorsed by the editorialists: "The
decision to focus on the WHI participants who did not use calcium
supplements is reasonable given that the effects would be difficult to
discern if both study arms had access to what was intended to be a
randomised intervention."22 To recap, we previously reported that calcium
supplements, administered alone, increase the risk of MI in a meta-
analysis of RCTs.4 Those data raised the important question as to whether
co-administration of vitamin D modulated the vascular risk of calcium
supplementation. In WHI CaD, personal calcium supplement use of up to 1g
daily was permitted outside the protocol, and the majority of participants
were taking personal calcium supplements at baseline. As calcium
supplements increase the risk of MI, the frequent use of the study agent
(calcium) in the control group, could mask effects of that agent in the
trial.

We disagree with Tran and Nguyen that models for interval-censored
data should be applied to our study, in which data were not interval-
censored, or that multi-state models to account for competing risk of
death in observational data should be applied to RCTs. They also argue
that adjustment for multiple statistical tests should be undertaken in
safety assessments. This approach is contrary to the recommendations of
the European Medicines Agency, who state: "In the case of adverse effects,
p-values are of very limited value as substantial differences will raise
concern ... it can generally be stated that an adjustment for multiplicity
is counterproductive for considerations of safety."23

Tran and Nguyen state that we carried out 18 hypothesis tests and
estimate the likelihood of each result being a false positive of >0.81.
Once again, they appear to have misunderstood the study and results:
because calcium supplements increased the risk of MI by 30% in a previous
study, we hypothesised that personal use of calcium supplements may have
obscured adverse effects of CaD on cardiovascular risk in WHI CaD. We
observed significant interactions between personal calcium supplement use,
CaD, and cardiovascular events, and increased risks of cardiovascular
events (13-22%) in women not using personal calcium supplements,
supporting our hypothesis. These analyses were conducted using recommended
approaches for subgroup analyses, as outlined in the paper, and were
interpreted in a hierarchical fashion i.e. results within sub-groups were
only considered when a significant interaction had already been
demonstrated. Thus, there were 9 pre-planned tests, 3 of which produced a
significant result, far more than expected by chance. It is unclear why
Tran and Nyugen have assumed that the prior probability of an increase in
cardiovascular risk with calcium supplements is so low, when it is not
consistent with available evidence, but it is that assumption which
determines their estimates of the likelihood of a false positive. A more
realistic assumption of the prior probability (50% or chance) produces
estimates for false positive results using the methods proposed by Tran
and Nyugen which are in broad agreement with estimates we reported in the
paper.

Despite its size and duration, the WHI CaD trial had insufficient
power to detect small effect sizes in subgroups. For this reason we
concluded that the WHI results by themselves were not definitive, but when
taken together with existing data, they contribute to a very consistent
pattern of increased risk of cardiovascular events with calcium
supplements. This consistency of effect, which spans several large RCTs of
calcium supplements, with and without vitamin D, is very unlikely to be a
chance finding (false positive). By failing to consider the context of the
current work, Tran and Nyugen inappropriately dismiss clinically relevant
harmful consequences of a commonly prescribed medication. We wonder what
level of evidence Tran and Nyugen would require before accepting the
possibility of harm from an agent, since clinical trials configured to
investigate adverse effects of interventions as primary outcomes are
neither ethical nor feasible.

We conclude that a reassessment of the role of calcium supplements in
osteoporosis treatment and prevention is justified, because the marginal
anti-fracture efficacy is likely to be more than offset by cardiovascular
harm.

1. Bolland MJ, Avenell A, Grey A, Gamble GD, Reid IR. Authors'
response. BMJ 2011;342:d3544.

2. American Association of Clinical Endocrinologists. Medical
guidelines for clinical practice for the prevention and treatment of
postmenopausal osteoporosis. Endo Pract 2003;9:545-64.

3. Dawson-Hughes B. A revised clinician's guide to the prevention
and treatment of osteoporosis. J Clin Endocrinol Metab 2008;93:2463-5.

4. Bolland MJ, Avenell A, Baron JA, Grey A, Maclennan GS, Gamble GD,
et al. Effect of calcium supplements on risk of myocardial infarction and
cardiovascular events: meta-analysis. BMJ 2010;341:c3691.

5. Sambrook PN, Cameron ID, Chen JS, Cumming RG, Durvasula S,
Herrmann M, et al. Does increased sunlight exposure work as a strategy to
improve vitamin D status in the elderly: a cluster randomised controlled
trial. Osteoporos Int 2011.

6. Reid IR, Bolland MJ, Sambrook PN, Grey A. Calcium
supplementation: Balancing the cardiovascular risks. Maturitas 2011;69:289
-95.

7. IOM (Institute of Medicine). Dietary Reference Intakes for
Calcium and Vitamin D. Washington, DC: The
National Academies Press, 2011.

8. Baron JA, Beach M, Mandel JS, van Stolk RU, Haile RW, Sandler RS,
et al. Calcium supplements for the prevention of colorectal adenomas.
Calcium Polyp Prevention Study Group. N Engl J Med 1999;340:101-7.

9. Wang TK, Bolland MJ, van Pelt NC, Horne AM, Mason BH, Ames RW, et
al. Relationships between vascular calcification, calcium metabolism, bone
density, and fractures. J Bone Miner Res 2010;25:2501-9.

10. Daly R, Ebeling P, Khan B, Nowson C. Effects of calcium-vitamin
D3 fortified milk on abdominal aortic calcification in older men:
retrospective analysis of a 2-year randomised controlled trial. J Bone Min
Res 2009;25 suppl 1:S65.

11. Reginster JY, Denis D, Bartsch V, Deroisy R, Zegels B,
Franchimont P. Acute biochemical variations induced by four different
calcium salts in healthy male volunteers. Osteoporos Int 1993;3:271-5.

12. Reid IR, Ames R, Mason B, Reid HE, Bacon CJ, Bolland MJ, et al.
Randomized controlled trial of calcium supplementation in healthy,
nonosteoporotic, older men. Arch Intern Med 2008;168:2276-82.

13. Bolland MJ, Avenell A, Baron JA, Grey A, Reid IR, MacLennan GS,
et al. Calcium and cardiovascular risk- author response. BMJ
2010:http://www.bmj.com/content/341/bmj.c3691/reply#bmj_el_240991.

14. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C,
Stefanick ML, et al. Risks and benefits of estrogen plus progestin in
healthy postmenopausal women: principal results From the Women's Health
Initiative randomized controlled trial. JAMA 2002;288:321-33.

15. Anderson GL, Limacher M, Assaf AR, Bassford T, Beresford SA,
Black H, et al. Effects of conjugated equine estrogen in postmenopausal
women with hysterectomy: the Women's Health Initiative randomized
controlled trial. JAMA 2004;291:1701-12.

16. McCloskey EV, Beneton M, Charlesworth D, Kayan K, deTakats D,
Dey A, et al. Clodronate reduces the incidence of fractures in community-
dwelling elderly women unselected for osteoporosis: results of a double-
blind, placebo-controlled randomized study. J Bone Miner Res 2007;22:135-
41.

17. Bonnick S, Broy S, Kaiser F, Teutsch C, Rosenberg E, DeLucca P,
et al. Treatment with alendronate plus calcium, alendronate alone, or
calcium alone for postmenopausal low bone mineral density. Curr Med Res
Opin 2007;23:1341-9.

18. Grey A, Bolland M, Wattie D, Horne A, Gamble G, Reid IR.
Prolonged antiresorptive activity of zoledronate: a randomized, controlled
trial. J Bone Miner Res 2010;25:2251-5.

19. Black DM, Delmas PD, Eastell R, Reid IR, Boonen S, Cauley JA, et
al. Once-yearly zoledronic acid for treatment of postmenopausal
osteoporosis. N Engl J Med 2007;356:1809-22.

20. Lyles KW, Colon-Emeric CS, Magaziner JS, Adachi JD, Pieper CF,
Mautalen C, et al. Zoledronic Acid and Clinical Fractures and Mortality
after Hip Fracture. N Engl J Med 2007;357:1799-809.

21. McClung M, Miller P, Recknor C, Mesenbrink P, Bucci-Rechtweg C,
Benhamou CL. Zoledronic acid for the prevention of bone loss in
postmenopausal women with low bone mass: a randomized controlled trial.
Obstet Gynecol 2009;114:999-1007.

22. Abrahamsen B, Sahota O. Do calcium plus vitamin D supplements
increase cardiovascular risk? BMJ 2011;342:d2080.

23. Committee for Proprietary Medicinal Products. Points to consider
on multiplicity issues in clinical trials. 2002:CPMP/EWP/908/99

In a secondary analysis of the WHI data, Bolland and colleagues found
that calcium and vitamin D (CaD) supplement was associated with an
increased risk of clinical myocardial infarction (MI) among women not on
personal supplementation of calcium (P=0.05)[1]. The authors call for a re
-assessment of the role of calcium supplements in osteoporosis prevention.
The somewhat unexpected finding has generated considerable publicity and
anxiety among patients. However, having carefully considered the data, we
are concerned that the conclusion might have been compromised by the risk
of false positive finding and multiple testing from the interaction
analysis.

The rationale for investigating whether the effect of CaD on CVD
outcomes varied by personal use of calcium seems weak. There was no
significant difference between CaD and placebo group in terms of total
daily calcium intake (1148 vs. 1154 mg/day, P=0.40) and daily supplemental
calcium use (322.6 mg/day vs. 326.2mg/day, P=0.48)[2]. There was no
interaction effect between total calcium intake and treatment on CVD
outcomes (P=0.52)[3]. Moreover, the WHI study was not powered to find an
interaction effect[4]. Although an interaction analysis can be useful,
results of such an analysis should be treated as exploratory or hypothesis
generating rather than as a hypothesis confirmation.

A consequence of any interaction analysis is the problem of multiple
testing. It is well-known that the likelihood of false positive finding
increases with the number of tests. The issue of false positive finding is
recognised by the authors when they remark that "if the effect of calcium
and vitamin D was unrelated to personal calcium use and the end points
were independent, the probability of at least one false positive
interaction test is <40%". However, this assertion is strictly
incorrect, because (a) the endpoints are not independent (e.g., clinical
MI is a component of total MI, which is expected to be correlated with
stroke); and (b) the probability of false positive finding is dependent on
the prior probability of a true association, statistical power, and
observed variability of effect size[5, 6].

Because of the correlation among endpoints the simple per-endpoint
analysis as done by Bolland et al is not optimal. A multistate model[7]
should have been used to adjust for the correlation among endpoints and to
take into account the problem of competing risk of death[8]. Without
accounting for the correlation among endpoints and the competing risk, it
is very difficult to interpret the statistical significance (or lack of
significance) of the relative risk for myocardial infarction or indeed any
single endpoint.

In the presence of multiple tests of hypothesis, there is a real
possibility of false positive finding. In the analysis, the authors
conducted 18 tests of hypothesis (not including 9 interaction tests). A
simple adjustment of P values by the Hommel's method[9] showed that the
minimum P value was actually 0.17 (not 0.04). Moreover, if the prior
probability of association is 1/18 ? 0.05, and the upper 95% confidence
interval of relative risk is 1.5, using a well-known method[10] we
estimated that the finding of association between CaD and clinical MI
among women not on personal calcium supplement (with relative risk being
1.22 and 95% CI: 1.0 to 1.5) has a probability of false positive of 0.91.
In fact, the probability of false positive for all other "positive
associations" ranged between 0.81 and 0.96.

When the data were analysed per randomization (i.e., no
stratification by personal use of calcium), we found that the relative
risk for clinical MI in the CaD was 1.07 (95% CI: 0.92 to 1.23; P = 0.38).
And, when this result was subsequently included in an updated meta-
analysis, the pooled relative risk for clinical MI was 1.11 (95% CI: 0.99
to 1.25; P = 0.07). The absolute risk difference was 0.13% which is
equivalent to a number needed to treat of 769. These results clearly show
that the association between CaD and clinical MI is neither statistically
significant nor clinically important.

Moreover, there is an issue of data integrity. Indeed, there are
substantial discrepancies between the data used in the analysis[1] and
those originally reported by WHI investigators[3]. For instance, the
number of total MI cases on CaD group included in the analysis was 415,
but the actual number reported by WHI investigators was 411. Similarly,
the number of MI cases in the control group was 389 in the analysis, which
was lower than that reported (390) in the original publication. The
number of strokes was also lower in the analysis compared with the
original data. These discrepancies could profoundly affect the analysis
result. Among the statistically significant hazard ratios, the lower end
of the 95% confidence interval (CI) was close to 1, suggesting that the
association is either marginal or unstable. Indeed, just adding 2 MI cases
in the control group, the result is statistically insignificant.

False positive finding is highly prevalent in the medical
literature[11]. There are many causes of false positive finding, including
- among others - modest effect sizes, marginal P values, multiple tests of
hypothesis. There is a call for a more stringent P value criteria in
clinical research[11]. The Bolland et al's analysis has all the "risk
factors" for a false positive finding.

In summary, the association between CaD supplement and myocardial
infarction (among women not on personal supplementation of calcium) has a
very high likelihood of being a false positive finding. Moreover, a re-
analysis of the meta-analysis suggest that there is inadequate evidence
for the conclusion that CaD supplement is associated with increased risk
myocardial infarction or indeed any CVD outcome. Therefore, the call for a
reassessment of the role of calcium supplements in osteoporosis is not
justified.

References

1. Bolland MJ, Grey A, Avenell A, Gamble GD, Reid IR. Calcium
supplements with or without vitamin D and risk of cardiovascular events:
reanalysis of the Women's Health Initiative limited access dataset and
meta-analysis. BMJ 2011;342:d2040.

2. LaCroix AZ, Kotchen J, Anderson G, Brzyski R, Cauley JA, Cummings
SR, et al. Calcium plus vitamin D supplementation and mortality in
postmenopausal women: the Women's Health Initiative calcium-vitamin D
randomized controlled trial. J Gerontol A Biol Sci Med Sci 2009;64(5):559-
67.

3. Hsia J, Heiss G, Ren H, Allison M, Dolan NC, Greenland P, et al.
Calcium/vitamin D supplementation and cardiovascular events. Circulation
2007;115(7):846-54.

4. The Women's Health Initiative Study Group. Design of the Women's
Health Initiative clinical trial and observational study. Control Clin
Trials 1998;19(1):61-109.

5. Ball RD. Quantifying evidence for candidate gene polymorphisms:
Bayesian analysis combining sequence-specific and quantitative trait loci
colocation information. Genetics 2007;177(4):2399-416.

6. Wakefield J. A Bayesian measure of the probability of false
discovery in genetic epidemiology studies. Am J Hum Genet 2007;81(2):208-
27.

7. Commenges D. Inference for multi-state models from interval-
censored data. Stat Methods Med Res 2002;11(2):167-82.

8. Putter H, Fiocco M, Geskus RB. Tutorial in biostatistics:
competing risks and multi-state models. Stat Med 2007;26(11):2389-430.

9. Hommel G. A stagewise rejective multiple test procedure based on a
modified Bonferroni test. Biometrika 1988;75:383-386.

10. Wacholder S, Chanock S, Garcia-Closas M, El Ghormli L, Rothman N.
Assessing the probability that a positive report is false: an approach for
molecular epidemiology studies. J Natl Cancer Inst 2004;96(6):434-42.

11. Ioannidis JP, Tarone R, McLaughlin JK. The False-positive to
False-negative Ratio in Epidemiologic Studies. Epidemiology 2011;22(4):450
-6.

Dear Editor,

I read the recently published article related to increase risk of
cardiovascular events following calcium supplements with or without
vitamin D [1]. The authors speculated that the cumulative effects of
modest, yet prolonged, increase of serum calcium levels following
supplements is one of the likely mechanisms of cardiovascular
calcification and associated complications.

Calcium/vitamin D supplements are commonly prescribed to treat or
prevent progression of osteoporosis, particularly among older women. The
beneficial effects of such supplements, however, are minimized by the
potential risk of harmful cardiovascular events [2, 3]. In fact, it is
reported that calcium supplements can increase the risk of myocardial
infarction and stroke, as high as up to 30% [1, 4]. In contrast, there are
studies that claimed that calcium supplements had no major effect on the
incidence of atherosclerotic vascular diseases [5]. In a 7 years follow up
study, daily intake of calcium (1000 mg) and vitamin D (400 IU)
supplementation neither increase nor decrease the risk of coronary heart
disease or cerebrovascular disorders, in the healthy postmenopausal women
[6].

While, the debate of beneficial effects of calcium/vitamin D
supplementation to the osteoporotic patients is yet to be settled, I
believe that cardiovascular harmful effects following calcium/vitamin D
supplementation may be partly related to their phosphate balance. Since
both calcium and vitamin D can influence phosphate metabolism, it is
important to determine whether patients at risk of developing
cardiovascular complications, following calcium/vitamin D supplementation,
may have altered phosphate balance. Of relevance, Tonelli et al. have
shown an association between baseline fasting serum phosphate level and
increase age-, race-, and sex-adjusted risk of death, and cardiovascular
events in individuals with prior myocardial infarction; of particular
importance, most of the patients in this study had serum phosphate levels
within the normal range, along with relatively normal kidney function [7].
For instance, individuals with serum phosphate levels 3.5-3.9 mg/dl had an
adjusted hazard ratio for death of 1.27 (95% confidence interval, 1.02-
1.59) compared with those with serum phosphate levels <3.5 mg/dl [7].
Please note that normal serum phosphate level is usually range from 2.4 to
4.1 mg/dl. It is therefore, likely that in certain disease conditions,
even the upper limit of normal range of serum phosphate may exert harmful
effects to influence overall survival. Indeed, animal studies have shown
that serum phosphate levels is one of the most important in vivo
determinants of cardiovascular calcification, as lowering serum phosphate
levels can reduce or eliminate cardiovascular calcification, even in the
presence of extremely high serum calcium and 1,25-dihydroxyvitamin D
levels [8]. These results provide compelling in vivo evidence that
suggests that reducing "phosphate burden" should be a critical therapeutic
priority for minimizing the risk of vascular calcification and associated
complications. In fact, human and animal studies have convincingly
demonstrated the toxic effects of phosphate in accelerating various
pathologies, ranging from vascular calcification to tumor formation and
aging that eventually influence overall survival [9-13].

Another important issue that is mostly neglected in long-term follow
up studies is the failure to dissociate the deleterious effects of chronic
ingestion of unrestricted amounts of phosphate, in patients prescribed
with calcium/vitamin D supplementation. Particularly, the effects of
chronic unrestricted ingestion of high phosphate-containing processed
foods and fizzy drinks on cardiovascular heath require careful analysis,
as experimental studies have shown numerous harmful effects of such foods
and drinks [14, 15]; some of these tissue injuries are also attributed to
phosphate toxicity [11]. Finally, maintaining and monitoring the phosphate
balance in individuals taking calcium supplements with or without vitamin
D may be important for a healthy life and longevity, as phosphate
imbalance may outweigh the benefits of the supplements [16].

References

1. Bolland MJ, Grey A, Avenell A, Gamble GD, Reid IR. Calcium
supplements with or without vitamin D and risk of cardiovascular events:
reanalysis of the Women's Health Initiative limited access dataset and
meta-analysis. BMJ 2011;342:d2040.

2. Reid IR, Bolland MJ. Calcium supplementation and vascular disease.
Climacteric 2008;11(4):280-6.

3. Reid IR, Bolland MJ, Avenell A, Grey A. Cardiovascular effects of
calcium supplementation. Osteoporos Int 2011;22(6):1649-58.

4. Bolland MJ, Barber PA, Doughty RN, Mason B, Horne A, Ames R, et
al. Vascular events in healthy older women receiving calcium
supplementation: randomised controlled trial. BMJ 2008;336(7638):262-6.

5. Lewis JR, Calver J, Zhu K, Flicker L, Prince RL. Calcium
supplementation and the risks of atherosclerotic vascular disease in older
women: results of a 5-year RCT and a 4.5-year follow-up. J Bone Miner Res
2011;26(1):35-41.

6. Hsia J, Heiss G, Ren H, Allison M, Dolan NC, Greenland P, et al.
Calcium/vitamin D supplementation and cardiovascular events. Circulation
2007;115(7):846-54.

7. Tonelli M, Sacks F, Pfeffer M, Gao Z, Curhan G. Relation between
serum phosphate level and cardiovascular event rate in people with
coronary disease. Circulation 2005;112(17):2627-33.

8. Ohnishi M, Nakatani T, Lanske B, Razzaque MS. In vivo genetic
evidence for suppressing vascular and soft-tissue calcification through
the reduction of serum phosphate levels, even in the presence of high
serum calcium and 1,25-dihydroxyvitamin d levels. Circ Cardiovasc Genet
2009;2(6):583-90.

9. Razzaque MS. The dualistic role of vitamin D in vascular
calcifications. Kidney Int 2011;79(7):708-14.

10. Block GA, Hulbert-Shearon TE, Levin NW, Port FK. Association of
serum phosphorus and calcium x phosphate product with mortality risk in
chronic hemodialysis patients: a national study. Am J Kidney Dis
1998;31(4):607-17.

11. Ohnishi M, Razzaque MS. Dietary and genetic evidence for
phosphate toxicity accelerating mammalian aging. FASEB J 2010;24(9):3562-
71.

12. Jin H, Xu CX, Lim HT, Park SJ, Shin JY, Chung YS, et al. High
dietary inorganic phosphate increases lung tumorigenesis and alters Akt
signaling. Am J Respir Crit Care Med 2009;179(1):59-68.

13. Eddington H, Hoefield R, Sinha S, Chrysochou C, Lane B, Foley RN,
et al. Serum phosphate and mortality in patients with chronic kidney
disease. Clin J Am Soc Nephrol 2010;5(12):2251-7.

14. Milei J, Otero Losada M, Gomez Llambi H, Grana DR, Suarez D,
Azzato F, et al. Chronic cola drinking induces metabolic and cardiac
alterations in rats. World J Cardiol 2011;3(4):111-6.

15. Celec P, Palffy R, Gardlik R, Behuliak M, Hodosy J, Jani P, et
al. Renal and metabolic effects of three months of decarbonated cola
beverages in rats. Exp Biol Med (Maywood) 2010;235(11):1321-7.

16. Razzaque MS. Phosphate toxicity: new insights into an old
problem. Clin Sci (Lond) 2011;120(3):91-7.

We read with interest the controversial article by Bolland [1]. Calcium plays a complex role in cardiac physiology, and clinical data
are contradictory. Concerning long term treatment, calcium can reasonably
be used carefully to high risk osteoporosis patients or not used at all
until proof of safety. Calcium metabolism is indeed the keeper of
myocyte function. In short term infusion, calcium (chloride or gluconate)
has undoubtedly inotropic and vasopressor effects [2; 3]. Hypocalcaemia
has been associated with mortality but calcium supplementation has never
improved the outcome in critically ill patients [4].

In our medical and surgical intensive care unit, calcaemia was
monitored cautiously. In a retrospective study about 248 patients with
cardiogenic shock hospitalised (unpublished data), hypocalcaemia was
present in 56% of the patients. Hypocalcaemia was associated with length
of stay (p=0,003), extra-renal replacement, mechanical ventilation and
Simplified Acute Physiologic Score (SAPS2; p=0,027). For death there is a
strong trend (p=0,053) but no significant association. However, in a
multivariate analysis the association was far from significant making
hypocalcaemia rather an indicator of severity. The calcium blood level was
corrected in 72% of the patients in a median time of 24 [8 to 72] hours.

Hypercalcaemia was noticed in 6% of the admissions, was iatrogenic in 12%
of the patients and was never severe. Ischemic events or troponin release
were similar in any group (hypercalcemia, hypocalcemia, corrected or not).

The correction of the hypocalcaemia, the time from correction and
hypocalcaemia were not associated with any improvement in haemodynamic
parameters, mechanical ventilation or length of stay. Only serious
arrhythmias and the need for dobutamine infusion were associated with a
longer correction of the hypocalcaemia. Therefore, calcium supplementation
does not seem to improve dramatically the evolution or lead to ischemic
injury.

Calcium supplementation is indicated during acute care in case of
hypocalcaemia but the threshold is unknown and the effect on mortality or
other clinical variables is still not sure [5]. Calcium metabolism is
a complex cascade involving apoptosis and many phenomena difficult to
approach in routine practice. Until ionised calcium targets are
scientifically determined, calcium supplementation is left to the physician's
discretion during cardiac impairment.

1- Bolland MJ, Grey A, Avenell A, et al. Calcium supplements with or
without vitamin D and risk of cardiovascular events: reanalysis of the
Women's Health Initiative limited access dataset and meta-analysis. BMJ
2011; DOI: doi:10.1136/bmj.d2040.

2- Steeds RP, Channer KS. Drug treatment in heart failure. BMJ
1998;316:567?8.

3- Burn JH, Gibbons WR. Part played by calcium in sympathetic
stimulation. Br Med J. 1964;1(5396):1482-3.

4- Forsythe RM,Wessel CB, Billiar TR, Angus DC, Rosengart MR.
Parenteral calcium for intensive care unit patients. Cochrane Database
Syst Rev. 2008 Oct 8;(4):CD006163.

5- Jankowski S, Vincent JL. Calcium administration for cardiovascular
support in critically ill patients: when is it indicated? J Intensive Care
Med. 1995;10(2):91-100.