calcium and vitamin d

there have been a slew of articles recently on calcium and vitamin D.  i will review several of them. but, be forewarned, there are lots of articles out there which come to pretty strikingly different conclusions

1. new recommendations from US prevent services taskforce on vit d and calcium to prevent fractures in adults (see vit d calc recs uspstf 2013 in the dropbox). basic conclusion is that 

-there is insufficient evidence to assess benefits/harms of vit d/calc supplementation in preventing fractures for premenopausal women and men

-there is insufficient evidence to assess benefits/harms of daily supplementation with greater than 400 IU vit D3 and greater than 1000 mg calcium for primary prevention of fracture in noninstitutionalized postmenopausal women

-recommends against daily supplementation with <= 400 IU of vit D3 and <=1000mg of calcium for primary prevention of fracture in noninstitutionalized postmenopausal women (note double-negative)

  -prior recommendation by uspstf to prescribe vit d supplementation for those >65yo who are at increased risk of falls

 

 

caveats: they, as usual, require high level studies to make a clear recommendation (ie, the fact that there are not great studies by their criteria does not necessarily mean that the intervention is bad, just that they think that there are insufficient data to recommend).  again, this uspstf recommendation is only for fracture prevention. as noted, they have a previous recommendation supporting vitamin D in preventing falls.  there are lots of other data out there (see the plethora of articles under titles of vit d or calcium in the dropbox which reach a variety of conclusions, including for example vit d and calcium reduce fx jcem2007.full, which was a meta-anal of 9K elderly patients finding that vit d 700-800 IU plus calcium but not alone led to an 18% hip fracture risk. other articles in dropbox address other clinical issues for which vitamin d supplementation might be beneficial).

 

2. NIH study of 388K men and women aged 50-71 from the AARP Diet and Health Study (see calcium and cad  jama 2013 in dropbox) looked at dietary and supplemental calcium and mortality from cardiovasc dz, heart dz, stroke. assessed baseline dietary (from 124-item food frequency questionnaire) and supplemental ca intake (from MVIs and individual ca supplements, including antacids) in 1995-6, followed 12 years. supplements used by 51% of men and 70% of women.  7900 and 3900 deaths in men and women, respectively. 

-review of their data showed that dietary calcium intake varied from 463-1336 mg/d (means of the first and fifth quintiles) in men and 391-1170 in women; supplemental calcium mean of 289 mg/d in men and 554 in women. in those who took supplemental calcium (who may be fundamentally different from those who did not), there was in fact no difference in dietary calcium intake between supplement users and nonusers. they did not break down the sources of the calcium by dairy vs veges (spinach, kale, soybeans, white beans also with lots of calcium)

-in men. supplemental calcium intake (>1000 mg/d) assoc with 20% inc in CVD deaths, esp with heart dz (not significant 14% increase in stroke); in men with 400-1000mg/d, significant 9% increase. no diff if up to 400mg/d calcium

-in women, no assoc bet supplemental ca and CVD deaths (nonsignfif 6% increase)

-dietary calcium not related to CVD deaths in either men or women

--other studies of note: calcium as phosphate-binder in pts wtih ESRD have increased coronary artery calcifications; other studies i've sent out include calcium and cad in men.pdf which found no harm in calc supplements in men and calcium and cad in women.pdf found decrease in CAD events in those on either dietary or supplemental calcium. the Womens Health Initiative found calcium was detrimental (calcium and cad WHI bmj 2011). the EPIC study (european prospective investigation into cancer and nutrition, see calcium cad heart 2012) found dec in MI by 31% in those with higher dietary calcium but increased MI by 86% in those on supplements.

pretty unclear why the male/female differences in the NIH study.there was a big finnish study of 10.5K women which showed a 24% inc in CHD in those on calcium supplements, with or without vit d (ref: pentti k, et al. Maturitan 2009; 63: 73-8.)  editorialist (see calcium and cad jama edit 2013 in dropbox), who cited the finnish study, comments that "calcium-rich foods, such as low-fat dairy foods, beans, and green leafy vegetables, which contain not only calcium but also a cocktail of essential minerals and vitamins" make more sense.  this seems to be a recurrent theme in the US. the rather reductionist approach of trying to isolate a micronutrient, purify it and then package it seems often not to work (eg b-carotene, which in many epidemiologic studies was effective in preventing cancer, heart disease..... but in isolation seems to increase the risk of lung cancer, for example, in smokers. similarly mixed data with vitamin E, homocysteine lowering vitamins...)

3. longitudinal cohort study in sweden of 61.5 K women, a mammography cohort followed 19 years, who had calcium intake estimated based on a food frequency questionnaire at baseline and again 10 years later as well as supplement intake (see calcium and cad mortality sweden bmj 2013 in dropbox ).  overall risk of all cause mortality in those taking >1400 mg calcium/d, comparing those taking between 600-1000 mg calcium/d, was 40% higher, cardiovascular dz inc 49% and ischemic heart dz 114%, but no change in stroke. they also noted higher death rates in those on <600mg/d, though with certain sensitivity analyses, this increase disappeared. no signif difference if calcium from food or supplements.but in those taking >1400 mg/d and also supplements, the increase was 157% higher for all-cause mortality. in sweden, 1/4 of women take supplements (much lower than in US) and the average calcium in take in the lowest quartile (<600 mg/d) was 533 and in the highest (>1400 mg/d) was 2137 mg -- so baseline calcium intake in sweden is significantly higher than here. the above associations were not influenced by vitamin D intake.

so, what makes sense? one could pretty easily conclude whatever one wants. and remember that these are not intervention studies based on randomized controlled trials, but are observational cohort studies. personally, i think the following is probably reasonable:

1. increasing calcium from dietary sources, both dairy and non-dairy probably makes sense whenever possible.

2. taking calcium supplements should be reserved for those who really cannot increase dietary calcium (lactose intolerant, live in food desert such as much of dorchester, can't or unwilling to change diet, etc) and probably should be limited to 1000 mg/d. total calcium intake should probably be less than 1400 mg/d

3. vitamin d data for many of the purported benefits (cancer reduction, improved immunologic function, etc etc -- see the many papers in the dropbox) is still pending, so hard to make very strong recommendation. there are pretty consistent data on falls prevention. and vitamin d is hard to get from foods, unless it is supplemented (unlike calcium).  i think the arguments for supplementation are still pretty strong (for review: see vit d. Endocrine Guidelines 2011 in dropbox) with goal 25-OH vit d in the 20-30 range. (they recommend 20 as their floor recommendation, but there are some small trials showing secondary hyperparathyroidism in those with levels up to the high 20's.  when these recommendations came out, i spoke with mike hollick, the lead author who is at bmc, and he thought that 30 was still his target but that the 20 was a compromise solution of the panel)