The authors have declared that no competing interests exist.
Conceived and designed the experiments: BDR JLM PL. Performed the experiments: BDR JLM PL. Analyzed the data: BDR JLM PL. Wrote the paper: BDR JLM PL.
There is growing evidence that vitamin D plays a role in the pathogenesis of asthma but it is unclear whether supplementation during childhood may improve asthma outcomes.
The objective of this systematic review and meta-analysis was to evaluate the efficacy and safety of vitamin D supplementation as a treatment or adjunct treatment for asthma.
We searched MEDLINE, Embase, CENTRAL, and CINAHL through July 2014.
We included RCTs that evaluated vitamin D supplementation in children versus active control or placebo for asthma.
One reviewer extracted data and one reviewer verified data accuracy. We qualitatively summarized the main results of efficacy and safety and meta-analyzed data on comparable outcomes across studies. We used GRADE for strength of evidence.
Main planned outcomes measures were ED visits and hospitalizations. As secondary outcomes, we examined measures of asthma control, including frequency of asthma exacerbations, asthma symptom scores, measures of lung function, β2-agonist use and daily steroid use, adverse events and 25-hydroxyvitamin D levels.
Eight RCTs (one parallel, one crossover design) comprising 573 children aged 3 to 18 years were included. One study (moderate-quality, n = 100) reported significantly less ED visits for children treated with vitamin D. No other studies examined the primary outcome (ED visits and hospitalizations). There was a reduced risk of asthma exacerbations in children receiving vitamin D (low-quality; RR 0.41, 95% CI 0.27 to 0.63, 3 studies, n = 378). There was no significant effect for asthma symptom scores and lung function. The serum 25(OH)D level was higher in the vitamin D group at the end of the intervention (low-quality; MD 19.66 nmol/L, 95% CI 5.96 nmol/L to 33.37 nmol/L, 5 studies, n = 167).
We identified a high degree of clinical diversity (interventions and outcomes) and methodological heterogeneity (sample size and risk of bias) in included trials.
Randomized controlled trials provide some low-quality evidence to support vitamin D supplementation for the reduction of asthma exacerbations. Evidence on the benefits of vitamin D supplementation for other asthma-related outcomes in children is either limited or inconclusive. We recommend that future trials focus on patient-relevant outcomes that are comparable across studies, including standardized definitions of asthma exacerbations.
There is growing evidence that suggests a relationship between vitamin D and asthma. Based on results from epidemiologic and animal studies, vitamin D may play a role in the pathogenesis of asthma via its effects on the innate and adaptive immune system. [
Formal guidelines on the vitamin D requirements for the healthy population have largely focused on evidence for skeletal manifestations of vitamin D deficiency. [
We sought to systematically search and critically assess the evidence from clinical trials on the efficacy and safety of vitamin D supplementation as a treatment or an adjunct to other asthma treatments to improve asthma outcomes, including ED visits and hospital admissions, measures of asthma control, and 25-hydroxyvitamin D levels in children.
We identified references by searching MEDLINE via the Ovid interface, Embase, CINAHL and the Cochrane Central Register of Controlled Trials (CENTRAL) since inception through July 2014 using regular search terms related to vitamin D and asthma and Medical Subject Headings (MeSH) search terms in MEDLINE. Regular search terms used were: vitamin D, cholecalciferol, calcitriol, ergocalciferol, 25-hydroxyvitamin D2, child, infant, preschool, adolescent, and asthma. MeSH search terms used were: “Vitamin D OR 25-hydroxyvitamin D2 OR Cholecalciferol OR Ergocalciferol” AND “Child OR Infant OR Preschool OR Adolescent” AND “Asthma”. We manually searched the reference lists of published reviews and meta-analyses for additional studies. Full search strategies are listed in the Supporting Information (see
We considered randomized controlled trials (RCTs) with parallel or cross-over design that evaluated the effect of vitamin D supplementation alone or as an adjunct to other forms of asthma treatment versus active control or placebo in children aged 0 to 18 years with asthma. Studies met the following criteria: (1) participants: children with diagnosed asthma (doctor’s diagnosis and/or objective criteria); (2) interventions: all preparations of vitamin D as oral supplement, at any dose and for any duration, including preparations containing supplemental vitamin D; (3) comparisons: compared to placebo, no supplementation or standard care; (4) outcomes: at least one of the primary or secondary outcomes examined in the current review. The primary outcome was ED visits and/or hospital admissions for asthma exacerbations and secondary outcomes included frequency of asthma exacerbations, asthma severity or symptom scores, lung function (forced expiratory volume in one second [FEV1], peak expiratory flow [PEF]), β2-agonist use, daily steroid use, days lost from work/school, nocturnal awakening, serum 25-hydroxyvitamin D concentration, withdrawal from trial due to adverse events, and adverse events (such as vitamin D toxicity). Eligible trials investigating the efficacy and/or safety of vitamin D as an adjunct to other forms of asthma treatment were also included. We excluded studies that looked at maternal vitamin D supplementation during pregnancy or lactation or those looking at fish oil supplementation without information on vitamin D dosage.
Two investigators (BR, PL) independently scanned the abstract of every retrieved record for eligibility. The same two reviewers investigated those eligible for inclusion further as full text.
For studies that satisfied the inclusion criteria, one reviewer (BR) independently extracted data using a standardized form developed
Quantitative data from studies that reported on comparable outcome measures were combined by conducting a meta-analysis using the change in outcomes from baseline or outcomes reported at the end of the study intervention period. For continuous outcomes, the weighted mean difference (WMD) between the intervention and control group were used as measures of effect size. Weights were assigned based on the inverse variances of the effect size estimates. If studies reported continuous outcomes as different units, we calculated the standardized mean difference (SMD) between the two groups. We combined data from the included studies using a random effects model. Random effects models are more conservative and account for variability both within and between included studies. For dichotomous outcomes the random effect model was used to estimate the pooled relative risk (RR) and 95% confidence interval (CI).
We examined statistical heterogeneity between results of different studies by the overlap of their confidence interval (95% CI). We estimated heterogeneity where applicable by the Cochrane Q (Chi-square test) and the
We identified 1086 eligible citations from the four databases searched (see
Study | Population | Intervention | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Source | RCT Design | Country (Latitude) | Age range (years) | N | Recruitment setting | Diagnosis of asthma | Type | Regimen | Duration in months (start-end season) | Co-intervention | Outcomes measured | Baseline and follow-up assessments (months) |
Baris at al 2014 [ |
Parallel | Turkey (40°N) | 5 to 15 | 50 |
Subspecialty medical clinic | Medical history, physical examination findings, postbronchodilator changes in FEV1 | D3 | 650 IU/d | 12 (w—w) | SCIT | AE, AS, ICS, LF, VD, AdvE | -1, 0, 6, 12 |
Yadav et al 2013 [ |
Parallel | India (29°N) | 3 to 14 | 100 | Subspecialty medical clinic | Physician diagnosis by GINA guidelines | D3 | 60 000 IU/mo | 6 (sp—w) | None | ED, AE, AS, ICS, LF | Monthly |
Darabi et al 2013 [ |
Parallel | Iran (32°N) | 6 to 14 | 80 | Subspecialty medical clinic | Newly diagnosed asthma | D3 | 500 IU/d | 6 (w/sp—sp/su) | Fluticasone 500 μg/d | AE, AS, LF VD | NR |
Lewis et al 2012 [ |
Parallel | United States (41°N) | 6 to 17 | 30 | University medical center | By physician |
D3 | 1 000 IU/d | 12 (w—w) | None | AS, LF, VD | 0, 6, 12 |
Majak et al 2011 [ |
Parallel | Poland (52°N) | 5 to 18 | 48 | Subspecialty medical clinic | Newly diagnosed asthma; sensitive only to house dust mites | D3 | 500 IU/d | 6 (f—su) | Budesonide 800 mg/d | AE, AS, LF VD | 0, 2, 4, 6 |
Urashima et al 2010 [ |
Multicentric Parallel | Japan (36°N) | 6 to 15 | 430 |
Hospitals and private practices | By physician | D3 | 1 200 IU/d | 4 (w- sp) | None | AE, AdvE | 0, 4 |
Majak et al 2009 [ |
Parallel | Poland (52°N) | 6 to 12 | 54 |
Subspecialty medical clinic | IgE-dependent asthma with regular symptoms; disease duration ≥2 years | D3 | 1 000 IU/week in a single dose | 12 (sp—su) | Prednisone (20mg) + SIT | AS, ICS, LF, VD | 7 months pre-study, 0, 3, 12 |
Schou et al 2003 [ |
Crossover | Denmark (55°N) | 6 to 14 | 17 | NR | Diagnosis by international guidelines‡; treated with glucocorticoid 1 year before study entry | D3 | 600 IU/d | 1 (w—w)) | Budesonide, Turbulhaler, vitamin A, B | AS, ICS, LF, β2, VD | 0, 1 |
Abbreviations: w, winter; sp, spring; su, summer; f, fall; NR, not reported; HDM, house dust mites; SIT, specific immunotherapy; ED, emergency department visits; AE, asthma exacerbations; AS, asthma severity/symptoms; ICS, inhaled corticosteroid use; β2, β2 agonist use; LF, lung function; VD, serum 25(OH)D concentration; AdvE, adverse events.
a55 participants were randomized into three groups: pharmacotherapy group, SIT group, SIT + vitamin D group. Only SIT (n = 15) and SIT + vitamin D (n = 17) were included in this review (n = 32).
b430 participants were randomized into vitamin D or placebo group. Only those who were previously diagnosed with asthma were used in the calculation of the RR presented in the meta-analysis (n = 230).
c54 participants were randomized into three groups: placebo group, steroid group, steroid + vitamin D group. Only steroid group (n = 18) and steroid + vitamin D group (n = 18) were used (n = 36).
dInformation obtained directly from authors.
We summarized the assessment of the study limitations of each included trials consistent with the GRADE guidelines [
Outcomes |
||||||||
---|---|---|---|---|---|---|---|---|
Study | ED visits for asthma | Asthma Control | Lung function | Serum 25(OH)D | ||||
Exacerbations |
Severity /symptoms |
Steroids use | β-2 agonist use | FEV1 | PEF | |||
Baris et al, [ |
- | NS | NS | NS | - | NS | NS | |
Yadav et al, [ |
- | - | - | |||||
Darabi et al, [ |
- | NS | - | - | NS | - | NS | |
Lewis et al, [ |
- | - | NS | - | - | NS | - | NS |
Majak et al, [ |
- | NS | - | - | NS | - | NS | |
Urashima et al, [ |
- | - | - | - | - | - | - | |
Majak et al, [ |
- | - | NS | NS | - | NS | - | |
Schou et al, [ |
- | - | NS | NS | NS | NS | NS |
Abbreviations: NS, non-significant effect; FEV1, forced expiratory volume in 1 second; PEF, peak expiratory flow;
*Shows effect direction between intervention and control at the end of follow-up.
†Asthma symptoms score based on different scoring systems.
‡As defined by asthma exacerbation attacks.
Outcome based on two different units of measurement: number of children experiencing asthma exacerbation and number of asthma exacerbation during treatment period.
Yadav and Mittal reported a statistically significant reduction in the mean number of emergency visits in the vitamin D versus the placebo group (P = 0.015) when comparing the groups at monthly intervals up to the end of the study at 6 months. [
Asthma exacerbations were assessed in five trials, four of which that found that vitamin D supplementation decreased asthma exacerbations, [
Trials that reported on asthma symptoms used different scoring systems. There was no effect of vitamin D supplementation on asthma symptom scores in the pooled analysis of the three trials that reported scores at baseline and follow-up using SMD (
Lung function was reported as FEV1% predicted, FEV1 (volume), or as peak expiratory flow rate (PEFR). Only one study reported a significant effect of vitamin D supplementation on lung function (
In the crossover trial by Schou et al, there was no statistical difference found between vitamin D and placebo periods in the number of puffs of
Of the four trials that reported on the change in inhaled steroid usage during the treatment period, three reported no effects of vitamin D supplementation [
Three trials reported a significant increase in mean serum 25(OH)D concentration in the vitamin D group as compared to the control group (
One trial reported no adverse events, [
We found moderate-quality evidence for ED visits following asthma exacerbations and low-quality evidence for asthma exacerbation, asthma symptom scores, lung function (FEV1%, PEFR), and serum 25(OH)D, using the GRADE guidelines (
No. studies | Design | Risk of bias | Inconsistency | Indirectness | Imprecision | Publication bias | Intervention group (n) | Control group (n) | Effect | Quality | Importance |
---|---|---|---|---|---|---|---|---|---|---|---|
1 | RCT | No serious limitations | n/a | No serious indirectness | No serious imprecision | n/a | 50 | 50 | P = 0.015 |
Moderate | Critical |
6 | RCT | Serious limitations |
No serious inconsistency | Serious indirectness |
No serious imprecision | None | 257 | 250 | 0.41 (0.27 to 0.63) |
Low | Critical |
6 | RCT | Serious limitations |
No serious inconsistency | Serious indirectness |
No serious imprecision | None | 117 | 114 | No effect |
Low | Critical |
7 | RCT | Serious limitations |
No serious inconsistency | No serious indirectness | No serious imprecision | None | 167 | 164 | 0.00 (-3.17 to 3.18) |
Low | Critical |
6 | RCT | Serious limitations |
Serious inconsistency |
No serious indirectness | No serious imprecision | None | 117 | 114 | 19.66 (5.96 to 33.37) |
Low | Important |
Abbreviations: ACT, Asthma Control Test; ATAQ, Asthma Therapy Assessment Questionnaire; ACQ, Asthma Control Questionnaire; FEV1, forced expiratory volume in 1 second; PEF, peak expiratory flow rate.
aUnclear allocation concealment, blinding of participants and outcome assessors, accounting of patients and outcome events, and other risk of bias (carryover effects in crossover trial).
bUnclear allocation concealment, blinding of participants and outcome assessors, accounting of patients and outcome events.
cDifferences in interventions and outcomes measured across studies.
dNon-significant effect across studies not included in the meta-analysis.
eWeighted difference in mean (WMD) change between intervention and control group.
fRisk ratio (RR): risk of experiencing asthma exacerbation in the intervention group as compared to the control group.
gNot included in the meta-analysis; favours intervention group.
hSignificant statistical heterogeneity observed based on random effects meta-analysis.
iWeighted mean difference (WMD) at end of intervention between intervention and control group.
Our systematic review and meta-analysis provides weak evidence to support vitamin D supplementation for the reduction of asthma exacerbations. Among the studies that found reduced exacerbations, changes in serum 25(OH)D concentrations after supplementation were either not significant or not reported, which further weakens the causal interpretation of the pooled estimate of effect reported on this outcome. The evidence for other asthma-related outcomes is either limited or inconclusive, with the majority of trials demonstrating statistically non-significant improvements in asthma symptoms, lung function, and β2-agonist use and daily steroid use.
Two systematic reviews and meta-analyses of randomized controlled trials in children with asthma published recently yielded similar results. [
Our findings are also consistent with other reviews of observational studies looking at vitamin D and asthma in children. [
Several studies have provided non-experimental evidence supporting the inverse relationship between vitamin D levels and asthma exacerbations. [
We aimed to minimize bias by applying systematic methods to the process of selecting, synthesizing, and assessing the published evidence. The diversity in patient population and intervention of included trials contribute some evidence of external validity the results presented in this review.
The current review has methodological limitations that should be taken into account when interpreting the results. At the study level, it was impossible to fully assess the risk of bias for the majority of the trials largely due to incomplete reporting of the methods for allocation concealment or blinding. At the outcome level, only a small number of trials contributed data to the meta-analyses. The estimate of effect would have been strengthened if data had been available for standardized outcomes and if comparable methods of measurement or ascertainment had been used across studies (e.g. asthma exacerbations and asthma symptoms). Further, there was clinical diversity in the patient population and interventions. While race/ethnicity likely affects vitamin D concentrations in response to supplementation, [
The small sample sizes of trials included in pooled analyses resulted in poor precision of effect estimates and prevented us from conducting sensitivity. Further, because vitamin D is thought to enhance the response to steroids and to play a role in the regulation of immune function, it would have been relevant to perform sub-group analyses on trials in which vitamin D was given as an adjunct to improve the clinical efficacy of other forms of asthma treatment. However, we could not perform such analyses and the effect estimates obtained from combining trials with vitamin D given alone or as an adjunct (to steroids and/or immunotherapy) may mask this biological interaction effect. In the absence of a sufficient number of trials that would allow direct comparisons of similar interventions on standardized and clinically important outcomes, our ability to detect a causal effect for a defined intervention was limited.
RCTs provide limited evidence for a reduction in asthma exacerbations but larger trials are necessary. The current review found moderate evidence based on one study that monthly doses of 60,000 IU of vitamin D may help in preventing ED visits, although more studies are needed to corroborate this finding. The wide range of doses given across studies reflects the lack of consensus on the optimal level of total serum 25(OH)D to elucidate benefits on non-skeletal outcomes. To best inform guidelines on vitamin D supplementation in children with asthma, future trials should investigate primarily on standardized and patient-relevant outcomes that are comparable across studies. These may include standardized definitions of asthma exacerbations, and important health utilization outcomes such as ED visits and hospitalizations for asthma. Currently, several unpublished trials on the effect of vitamin D supplementation on asthma outcomes in children are in progress. Future meta-analyses should examine separately the potential for vitamin D supplementation as an adjunct to other forms of asthma treatment.
(TIFF)
(TIFF)
(TIFF)
(TIFF)
(TIFF)
(DOCX)
(DOCX)
(DOCX)
(DOCX)
We thank the authors who provided additional information on studies included in this review.