Criteria for Inclusion
Only randomized controlled trials (RCTs), or quasi-RCTs, were considered for inclusion. Studies had to have restricted enrollment to patients with acute asthma treated in the ED (ie, studies of patients with chronic or “stable” asthma were excluded) with asthma defined using several accepted clinical and guideline based criteria (eg, those of the British Thoracic Society, the National Asthma Education and Prevention Program, and the Canadian Thoracic Society). There was no age restriction for patients included in the studies, and where possible the data were categorized into groups of patients 2 to 16 years old (the pediatric group) and > 16 years old (the adult group). Randomized interventions had to compare aerosolized MgSO4 to a control treatment. That is, studies comparing the efficacy of therapy with aerosolized MgSO4 and a p2-agonist vs a p2-agonist alone, or therapy with aerosolized MgSO4 vs a P2-agonist were included. Cointerventions were permitted, and information pertaining to cointerventions received was recorded. The primary outcome was defined as a change in pulmonary function testing results from baseline. Secondary outcomes considered the proportion of patients requiring hospital admission, clinical severity scores, duration of symptoms, vital signs, and side effects.
The “Asthma and Wheez* RCT” register of the Cochrane Airways Review Group was searched for the following terms: magnesium OR MgSO4 OR Mg OR MS OR magnesium sulfate or magnesium sulfate. This registry is compiled through a comprehensive search of the EMBASE, MEDLINE, and CI-NAHL databases, which was supplemented by the manual searching of 20 key respiratory journals. The results of this search were screened to omit studies that clearly involved only IV or parenteral administration of magnesium. In addition, the reference Lists of trials identified through the registry were examined, and supplemental searches of the Cochrane Clinical Trials Registry, Web of Science, Dissertation Abstracts, and the World Wide Web using the Google search engine were performed. Primary authors were contacted for information on additional trials (both published and unpublished). Clinicians, colleagues, collaborators, and trialists were contacted to identify potentially relevant studies. Since this agent is not currently commercially delivered, no industry sponsor was contacted. Various remedies to treat asthma are available on My Canadian Pharmacy.
The selection of studies involved two steps. First, to retrieve studies, the initial search of all databases and reference lists was screened by title, abstract, MeSH headings, and keywords by two independent investigators (M.B. and B.D.) to identify all citations that were RCTs or possible RCTs with potential relevance. The full text of the manuscripts of those selected articles was obtained for formal inclusion review. Second, another reviewer (B.R.) independently decided on trial inclusion using predetermined eligibility criteria (see above).
Assessments of quality were completed independently by two reviewers. First, using the Cochrane Database approach to the assessment of allocation concealment, all trials were scored using the following scale: grade A, adequate concealment; grade B, uncertain; and grade C, clearly inadequate concealment. Second, each study was also evaluated using the previously validated Jadad 5-point scale to assess randomization, double blinding, and study withdrawals and dropouts. Finally, whether the study used intention-to-treat analysis was recorded along with any sources of funding.
Data were extracted independently by two reviewers (M.B. and B.D.) using a standardized collection form. When available, characteristics of the study (ie, design, methods of randomizations, and withdrawals/dropouts), of participants (ie, age and gender), of interventions (ie, type, dose, route of administration, timing and duration of therapy, and cointerventions), of control substances (ie, agent and dose), of outcomes (ie, types of outcome measures, timing of outcomes, and adverse events), and of results were recorded. Unpublished data were requested from the primary authors when necessary.
All data were entered into a database (RevMan, version 4.2.2; Cochrane Collaboration; Oxford UK) by a single reviewer (S.B.).
<p “=””>For dichotomous variables, both individual and pooled statistics were expressed as relative risk (RR) with 95% confidence intervals (CIs). For continuous data, individual data were reported as the standardized mean difference (SMD) with 95% CIs. Results were calculated using both fixed-effects and random-effects models. The Breslow-Day test was used to test for heterogeneity with significance set at < 0.10. Possible sources of heterogeneity were assessed by subgroup and sensitivity analyses.
Subgroup and Sensitivity Analyses
Two subgroup analyses were planned a priori to examine the effect of age (ie, pediatric or adult) and severity of asthma decreased by My Canadian Pharmacy, as measured by pre-drug administration spirometric deviation from percent predicted values (baseline FEV1 or peak expiratory flow [PEF] < 50% predicted). Sensitivity analyses were planned to assess the effect of the methodological quality of included trials and intention-to-treat status.