Calcium, Heart Attack and Osteoporosis

.
A recent meta-analysis published in the British Medical Journal by Bolland, et. al., finds that there is a 30% increase in the risk of myocardial infarction (MI) in women taking calcium supplements for osteoporosis (“Effect of calcium supplements on risk of myocardial infarction and cardiovascular events: meta-analysis”, BMJ 2010;341:c3691). If supported by other research, this could be big news, as millions of women are doing just that. A few “bullets”:

· 15 studies were reviewed by the meta-analysis comprising comparisons of 12,000 women either taking or not taking calcium supplementation for osteoporosis. For some of these studies results were available for individual women, and some for just the group as a whole, but the results were similar.

· Women receiving calcium had 30% more MIs than those who were not. Other end points: stroke, death from cardiovascular disease, and death overall, did not show significant differences, although they did show trends toward reduction in the non-calcium groups.

· The studies reviewed in the meta-analysis were all of women taking calcium but not taking vitamin D supplementation with it.

· The studies were not done for the purpose of looking at cardiovascular mortality; the data were re-analyzed and other sources of data were used to look at the outcome events.

The authors recommend that women who are taking calcium without vitamin D for osteoporosis stop doing so unless they are also taking a drug that treats osteoporosis, such as a bisphosphonate (which have their own risks, although in women with osteoporosis these are usually outweighed by the benefits) or selective estrogen receptor modulator (SERM) like raloxifene, usually used for breast cancer treatment (links are to a few different websites, including WebMD, FDA, and BreastCancer.org; they are representative although not definitive, and there are, of course, other sites). They note that their results are similar to other studies of women taking calcium alone, although a Women’s Health Initiative (WHI) study on women taking both calcium and vitamin D did not show any effect on the incidence of coronary artery disease. The authors suggest possible reasons for the difference, including protective effects of vitamin D, the younger age of the WHI participants (mean of 62 vs. 75 for the meta-analysis), and the interesting, but slightly confusing fact, that the WHI study had a much higher percent of women who were taking calcium before the study began (“non-protocol”): 54% vs. 1.2%. If taking calcium is associated with more MIs, why would women who were taking more calcium before the study have lower rates? And yet, the authors note that “Interestingly, the only study in our analysis that reported a relative risk of less than 1.0 for myocardial infarction with calcium also had high non-protocol use of calcium supplements.”

In their accompanying editorial, “Calcium supplements in people with osteoporosis”, BMJ 2010;341:c3856), JGF Cleland, K Witte and S Steel go farther than the authors of the meta-analysis, saying clearly in their sub-head “Should not be given without concomitant treatment for osteoporosis”, even when given with vitamin D. Their justification is the lack of good evidence for improved outcomes, including pathologic fractures, with the use of calcium and vitamin D. “Calcium supplements, given alone, improve bone mineral density, but they are ineffective in reducing the risk of fractures and might even increase risk, they might increase the risk of cardiovascular events and they do not reduce mortality. They seem to be unnecessary in adults with an adequate diet. Given the uncertain benefits of calcium supplements, any level of risk is unwarranted.” With regard to vitamin D, they say “Vitamin D supplements might reduce the risk of falls, might have important clinical effects on cardiovascular function, do not increase mortality, and may mitigate the trend to excess mortality seen with calcium supplements alone. However, no conclusive data are available to show that current doses of vitamin D supplements with or without calcium supplements reduce the rates of fracture, and meta-analyses found evidence of substantial reporting bias.”

The editorialists emphasize that while calcium does increase bone density, this is a surrogate variable while the issues of fractures and mortality are the true outcomes, an issue I have addressed several times recently (Rosiglitazone and the "Holy Grail", July 16, 2010; Statins and scientific integrity, July 6, 2010 ). They say “Surrogate measures may be useful in pilot studies but become problematic when they become the goal of treatment.” They are quite rigorous in looking and risk and benefit, noting even that exercise, while perhaps a good way to increase bone strength, “also carries risk”. They cite Kanis, et. al., from 2002[1], but it should be obvious that exercise can have risk.

The last part of the editorial is, however, more concerning to me. The authors call for greater demonstration that drugs will have positive effects on important outcome variables (a good thing) but they then worry that such requirements will be so burdensome as to stifle research: “Requiring companies to show before licensing that treatments for chronic diseases such as osteoporosis, diabetes, and hypertension reduce long term disability and death could lead to a cessation of research in these areas. The cost and commercial risk would be too high.” They then call for an extension of patents on these drugs to 50 years, similar to the Berne convention for copyrights on a song. The presumption is that this would be long enough for the companies to make back their money. Obviously, however, this also means that consumers would have to pay the higher costs for patent, rather than generic-equivalent drugs, for much longer.

Amazingly these authors, despite citing no conflicts of interests (which might explain such a position if they in fact held patents or were being paid honoraria by pharmaceutical manufacturers) dispense with such concerns in a single sentence “Lower prices for innovative drugs could be negotiated.” By whom? How? What would be the effect on the consumer? All I can imagine is that because they are British, and in Britain there is a National Health Service which charges a fixed fee to patients for all drugs, that they are thinking only of cost to the NHS and have no idea how much the cost of patented drugs is to Americans. Which, as Americans know, can be phenomenally high. (Example: generic alendronate, the oldest bisphosphonate, costs roughly $40 a month for either 35mg [recommended for prevention of osteoporosis] or 70mg [recommended for treatment of osteoporosis] per week doses, while the brand name, Fosamax ® costs about twice that; for those not available generically, risendronate (Actonel ®) costs 3 times as much, and ibandronate (Boniva ®) costs about $350 a month; all prices wholesale from ePocrates and www.drugstore.com.) Taking drugs that you need for a chronic disease is very different from downloading a song!

Of course, this is another strong argument for having a national health insurance plan that covers everyone. In the meantime, while we will wait for the certain flurry of responses and comments, not taking calcium unless one is also taking a bisphosphonate or similar osteoporosis treatment drug, seems prudent; taking vitamin D, without calcium, for its other benefits, is probably still a good idea.

[1] Kanis JA, Brazier JE, Stevenson M, Calvert NW, Lloyd JM. Treatment of established osteoporosis: a systematic review and cost-utility analysis. Health Technol Assess 2002;6:1-146.
.