Bessmertny 2002.

Methods	Design: parallel randomised controlled trial. Method of randomisation: computer‐generated random numbers. Concealment of allocation: yes. Blinding: double‐blinded, placebo‐controlled. Withdrawals/dropouts: 6 (4 unable to complete spirometry, 2 inappropriate randomisation).
Participants	Location: 1 university hospital in Brooklyn, NY. Participants: 74 patients, presenting to the emergency department with acute asthma exacerbation, PEF between 40% and 80% predicted. Exclusions: smoking history > 10 pack years, known hypersensitivity to albuterol or MgSO₄, known chronic obstructive pulmonary disease, known history of renal impairment, known history of cardiac dysrhythmias, congestive heart failure or angina, fever more than 38 °C, receipt of theophylline or anti‐cholinergic within 2 h of arrival to ED.
Interventions	Treatment: albuterol 2.5 mg/3 mL nebule followed by 384 mg isotonic MgSO₄ every 20 min × 3. Control: albuterol 2.5 mg/3 mL nebule followed by normal saline every 20 min × 3.
Outcomes	Measured FEV1 every 20 minutes for 2 h. Adverse events: no serious adverse events noted.
Notes	Funding: supported by an unrestricted educational grant from Astra Pharmaceutical Company; no Astra Pharmaceutical Company products were used in the study. Mouthpieces for the spirometer were supplied at no charge from Mallinkrodt Nellcor Puritan Bennett. Circulaire nebulizers were supplied by Westmed Inc. at a reduced rate.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	An assigned third party randomised participants by means of a computer‐generated random table (1:1 randomisation) to either the treatment or control group.
Allocation concealment (selection bias)	Low risk	An assigned third party randomised participants by means of a computer‐generated random table (1:1 randomisation) to either the treatment or control group.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐blinded, placebo‐controlled. A log of the identification number and specific treatment of each participant was kept and remained closed to the investigators until the completion of the study.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Double‐blinded, placebo‐controlled. A log of the identification number and specific treatment of each participant was kept and remained closed to the investigators until the completion of the study. Outcomes were assessed every 20 minutes for 2 h.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Dropouts: 3 in each group. Albuterol plus normal saline solution (3 unable to complete spirometry); and albuterol plus magnesium (2 inappropriate randomisation, 1 unable to perform spirometry).
Selective reporting (reporting bias)	High risk	Mean values only given for FEV1, no SDs and the text reports that there were no statistically significant differences in FEV1 between the groups. The text also states "The analysis of continuous safety variables (BP, pulse rate, respiratory rate, oxygen saturation, and serum magnesium concentrations) did not demonstrate any clinically or statistically significant differences between the 2 groups at any point during the study."