Noncalcemic adverse effects and withdrawals in randomized controlled trials of long-term vitamin D2 or D3 supplementation: a systematic review and meta-analysis

Context: Recent randomized controlled trials (RCTs) provide evidence for a possible beneficial impact of vitaminD supplementation on health outcomes beyond bone health, but there are few reviews of noncalcemic adverse effects from long-term supplementation. Objective: The aims of this systematic review of vitaminD supplementation in RCTs were as follows: to determine whether all adverse effects, when combined, are reported equally between treatment arms; to identify the most common noncalcemic adverse effects reported; and to ascertainwhether withdrawal rates, as amarker of clinical adverse effects, differ between treatment arms. Data Sources: The MEDLINE Ovid, Embase, and Cochrane Library databases were searched systematically up to May 2016. Study Selection: Randomized controlled trials thatmet the following criteria were selected: administered vitamin D-2 or D-3 supplements for a minimum supplementation or follow-up period of 24 weeks, had a placebo/control group, and were conducted among adults (>= 18 y). Data Extraction: Two researchers independently screened studies for eligibility, extracted data, and carried out quality assessment of selected studies. A total of 128 studies with 52 297 participants were identified. A random-effects model was used to calculate risk ratios in ameta-analysis. Results: Long-termvitaminD2 or D3 supplementation, compared with placebo, did not increase all adverse effects, when combined, as reported in 62 studies with 19 389 participants (relative risk [RR] = 0.97; 95% CI, 0.92-1.02). Vitamin D also did not increase the risk of the most common noncalcemic adverse effects: gastrointestinal symptoms were reported in 27 studies with 9189 participants (RR = 1.01; 95% CI, 0.87-1.17), and dermatological symptoms were reported in 8 studies with 1695 participants (RR = 1.33; 95% CI, 0.82-2.15). VitaminD did not increase withdrawals from 123 studies with 41 861 participants (RR = 1.03; 95% CI, 0.96-1.09). However, participants given vitamin D were more likely to report withdrawals than those given placebo in studies in which calcium was given in both arms (RR = 1.16; 95% CI, 1.02-1.33) when compared with participants in studies in which calcium was not given in either arm (RR = 1.00; 95% CI, 0.95-1.06; P for interaction = 0.009). Conclusions: Overall, these findings suggest that vitaminD, by itself, does not increase the risk of noncalcemic adverse effects.