vitamin d: no dec OA, dec insulin, more preeclampsia

3 recent articles on vitamin d (there are lots of them in the dropbox)

 

1. JAMA article (see vit d knee OA progression not help in dropbox) which randomized 146 people twith symptomatic knee osteoarthritis to high dose vitamin d (2000 IU/d) vs placebo for 2 years and evaluated symptoms and progression of cartilage loss. results:

    -baseline vit d increased from 23 ng/ml to 39 ng/ml with therapy

    -nonsignificant decrease in sx in vit d group vs placebo

    -nonsignificant diff in cartilage volume loss, with sl less volume loss in vit d group

    -if look at subset with lower vit d at baseline,  more improvement in both sx and cartilage with vit d than with placebo, but still nonsignificant

 

so, not so impressive results. reasons? maybe vit d is overrated. maybe better to treat people before they get symptomatic OA. maybe better results if patients with lower baseline vit d (they did find better though nonsignif results in that subgroup.  i must admit that i very rarely see baseline vit d in the low 20 range, more often in the low teen or lower range. and, the recent endocrine society guidelines --see vit d Endocrine Guidelines 2011.pdf in dropbox -- targets 20 ng/ml as the cutpoint of deficiency, so the mean in this study at baseline was above this.)

 

2. Am J Clin Nutrition article this month (see vit d and insulin sens adol am j clin nutr 2013 in dropbox) on insulin sensitivity and vit d deficiency in obese adolescents. prior very small studies have shown (though not consistently) that vit d def assoc with metabolic syndrome. in this study, (also small), 35 obese adolescents (ave age 14, BMI ave of 40, 25-OH vit d of 20 ng/ml) randomized to high dose vit d (4000 IU/d) vs placebo and followed 6 months. results:

    -increase in 25-OH vit d by 20 vs 3 ng/ml in intervention vs placebo groups

    -no diff in inflammatory markers (IL-6, TNF-a,CRP) or BMI
    -signif dec in fasting insulin levels in intervention group (-6.5 vs +1.2) -- this was more impressive than changes in the fasting glucose levels

    -HOMA-IR (a calculation reflecting insulin resistance which includes fasting insulin and fasting glucose) decreased 1.4 vs increased 0.3

    -leptin-to-adiponectin ratio (this ratio most strongly reflects the gold standard in assessing insulin resistance -- the hyperinsulinemic/euglycemic clamp. adiponectin has strong insulin-sensitizing properties and leptin resistance is common in obesity) dec 1.4 vs inc 0.1

 

comments: several studies in adults and a few in kids show lower vit d in obese people (they attribute to fat solubility and sequestering the vit d in the fat cells.  may also be that overweight people also get less sun exposure). this is small study, but pretty impressive results. surprising result that there was no change in inflammatory markers.

 

3. BMJ meta-anal of assoc of maternal vit D in pregnancy and neonatal outcomes (seevit d and prenatal outcomes bmj 2013 in dropbox). 31 studies, 22K pregnant women. somewhat variable definitions of vit d def (typically <20 or <30). findings:

    -increase in gestational diabetes in women who were vit d deficient (vit d< 30) , OR 1.49

    -increase in pre-eclampsia in women who were vit d deficient (vit d< 30) , OR 1.79

    -increase in infants who were small-for-gestational-age (vit d<15), OR 1.85

    -smaller subset of studies which looked at bacterial vaginosis (3 studies) with 4.4-fold increased risk with vit d<15

    -4 studies looked at birthweight. vit d<15 assoc with 130gm dec wt but no diff in ht or head circumference

 

comments: observational study. women with higher vit d levels might be taking care of themselves overall better than those with lower levels. the endocrine society report mentioned above suggests that pregnant women have 25-OH vit d levels >30.

 

overall comments. in spite of appearances to the contrary, i do not think that vitamin d is the be-all-and-end-all supplement that will cure the ills of the world. but i am pretty impressed with

    1. the remarkably low vit d levels in our community (10-20% of patients without detectable levels, total of probably 80-90% with levels under 20 -- and i have been testing pts regularly for the past 10-15 years),

    2. the ubiquity of vitamin d receptors throughout the body

    3. the lab studies showing vit d is assoc with immune function/large numbers or epidemiol studies showing relation with cancers, autoimmune dz, diabetes, efficacy of anti-TB meds..... (many of these studies are in the dropbox and i have sent out previously. there are a few reviews which go into detail as well).

 

by the way, the measurement of vit d is confusing. some labs use ng/ml (as i do above), but some others (esp non-US) use nmol/L. the conversion is 1 ng/ml = 2.5 nmol/L (so 20 ng/ml=50 nmol/L)