13 Dec, 16 | by EBM
By Dr. Geoffrey Modest
A recent guideline from the National Osteoporosis Foundation and the American Society for Preventive Cardiology, with support from an independent evidence review team from Tufts University, determined that calcium supplementation, with or without vitamin D, had no relationship to cardiac health (for the recommendations see doi:10.7326/M16-1743; for the full document see doi:10.7326/M16-116).
- “calcium with or without vitamin D intake from food or supplements has no relationship (beneficial or harmful) with the risk for cardiovascular and cerebrovascular disease, mortality, or all-cause mortality in generally healthy adults at this time”
- Calcium intake should not exceed the National Academy of Medicine recommendations of 2000-2500 mg/d
- Obtaining calcium from food is preferred to taking supplements
- This recommendation is supported by a review of the human studies and is supported mechanistically and pathologically in animal studies on high-calcium diets (no biological mechanism supports the association between calcium intake and cardiovascular disease).
- It is clear that calcium and vitamin D are necessary for adequate bone health. However several recent studies have questioned whether fractures were reduced by supplementation in older adults. There even have been reports that cardiovascular disease, including MIs and strokes, may be worse by supplementation (more below).
- This review included 4 randomized controlled trials, one nested case-control study and 26 cohort studies.
- Of note, very few studies looked at calcium intakes of greater than 1600 mg per day. One study that did do so, found that there was no increased cardiovascular disease (CVD) or mortality in those at the highest level of calcium by the combination of foods and supplements, but a somewhat increased risk in those on supplements only. However, for strokes there was a lower risk for all types of consumption.
- There are several general concerns raised by all of these dietary studies:
- They are usually dependent on dietary recall/food frequency questionnaires, often at only a few points in time, and dietary recall itself is not necessarily so accurate
- It is really impossible in observational studies to isolate specific dietary ingredients or vitamins. Although analyses try to look at likely confounders, there are undoubtedly unanticipated ones (are those who have more calcium and vitamin D in their diet more likely to be health-conscious and also have a lower morbidity/mortality related to that? Or, contrarily, do those who are the least health-conscious take supplements to boost their calcium and vitamin D because they just heard on the news that this was important and might counteract their less healthy lifestyle?). It is interesting that in the one study with the highest levels of calcium intake, there was divergence between those who had calcium by foods versus supplements in terms of CVD and mortality. Perhaps those on supplements had generally less healthy diets? Or is there a fundamental physiologic difference between getting calcium from food versus pill (which appears to be different for CVD and stroke outcomes)??
- The dietary studies themselves are often given equal weight, when some are better studies in others. For example, it is notable in the above that even the inconsistent association between calcium and/or vitamin D and CVD outcomes is more apparent in subgroup analyses than in the overall trials. And some showing lack of benefit for bone studied people with already adequate 25(OH)D levels to begin with (i.e. supplements might not do much more).
- One cited concern is that increased calcium might lead to more vascular calcification. But this was derived from studies of persons with pretty severely impaired renal function on several meds, and not the general population.
- As noted in prior blog (see http://blogs.bmj.com/ebm/2016/11/21/primary-care-with-geoffrey-modest-md-lessons-ive-learned-from-looking-at-the-medical-literature/ ), there are real concerns about the value of meta-analyses/systematic reviews. Of course, there also limitations of single trials, or the latest trial that makes it into the journals and popular press. This current study was a truly independent meta-analysis, conducted by a well-respected group, and the evaluation of the individual studies they included pretty impressively confirmed that calcium intake is safe, even up to the 2000 to 2500 mg per day range (but the studies are pretty limited on this, as noted above).
- The current meta-analysis did raise pretty serious methodologic concerns about 2 recently published reports finding adverse cardiovascular outcomes with calcium supplementation with or without vitamin D
- There is also some confusion in the medical literature about the benefits of vitamin D in terms of fall prevention (see http://blogs.bmj.com/ebm/2015/05/27/primary-care-corner-with-geoffrey-modest-md-vitamin-d-and-falls-in-the-elderly/ which critiques a recent study). The USPSTF most recent guidelines (prior to this study) still noted “that vitamin D supplementation is effective in preventing falls in community-dwelling adults aged 65 years or older who are at increased risk for falls”.
So, putting this all in perspective, I think that it is clear that bone health requires adequate vitamin d and calcium levels. Guidelines differ in their specific recommendations:
- Some suggest achieving a 25(OH) vitamin d level of 20 ng/ml and some suggest 30 ng/ml.
- The USPSTF in 2013 found insufficient evidence to recommend >400 IU vitamin D or >1000 mg of calcium (though it warns against supplementation with <= 300 IU vitamin D or <= 1000 mg calcium)
- The Endocrine Society (see doi: 10.1210/jc.2011-0385) suggested checking 25(OH) vitamin d levels in people at high risk for deficiency, with deficiency defined as <20ng/ml (50nmol/L), and in general recommends:
- For children <1yo, at least 400 IU/d; for 1-18 yo, at least 600 IU/d. but not enough reliable data to raise the 25(OH)D level to 30 ng/ml
- For those 19-50yo, at least 600 IU/d, but might need 1500-2000 IU/d to achieve consistent levels >30 ng/ml
- For those 50-70yo, at least 600 IU/d,but might need 1500-2000 IU/d to achieve consistent levels >30 ng/ml
- For those >70yo: at least 800 IU/d. but might need 1500-2000 IU/d to achieve consistent level
- For pregnant/lactating women, at least 600 IU/dbut might need 1500-2000 IU/d to achieve consistent levels >30 ng/ml
- And more aggressive therapy (e.g. 2000IU/d) for those who are vitamin D deficient
- For calcium, the NIH recommends (which is pretty similar to the Institute of Medicine):
- Age 1-3: 700mg/d
- 4-8 yo: 1000 mg/d
- 9-19 yo 1300 mg/d
- 31-70yo 1000 mg/d
- But 51-70yo women and everyone >70yo: 1000 mg/d
I personally do suggest to patients that they consume a high calcium diet, but this is often limited by cultural or other circumstances (e.g., lactose intolerance). And, it is hard for those living in the Northeast to get adequate sunlight for adequate vitamin D levels (and I still shoot for 30 ng/mL as a target). So, though I would prefer all calcium and vitamin D coming from natural sources (diet, sunlight), most of my patients are on supplements. And usually taking 1 tablet of calcium 600mg combined with vitamin D 400IU twice a day is adequate (though I do have some high risk patients, including those with low Bone Mineral Density, who need 2000 IU of vitamin D/day). And I do check 25 (OH) vitamin D levels in those I think are at high risk (BMD, history of fragility fracture, medications, very limited outside sun exposure, etc.).