The optimal management of patients with asthma who, once established on inhaled steroid therapy, require additional treatment to control symptoms remains controversial. Shrewsbury and colleagues compared the evidence for addition of salmeterol versus a large increase in the daily steroid dosage. They selected rates of exacerbation of asthma as the principal outcome with which to compare the treatment alternatives.

Electronic databases, abstracts and publications in all languages starting in 1985 were searched for studies of patients who required additional treatment once established on a dosage of up to 400 μg of inhaled beclomethasone or budesonide, or up to 200 μg of fluticasone twice daily. Quality indicators for studies included ethical approval, evidence of good clinical practice and maintenance of treatment blinding. Two researchers abstracted data from each of the nine studies. The severity of exacerbations was assessed independently by two researchers who were not aware of the adjuvant treatment selected. Whenever possible, clinical data were obtained for individual patients. Exacerbations were classified as severe if the patient required oral steroid therapy or hospital admission, moderate if an increase in inhaled steroid medication was required, and mild if an increase in any rescue medication was needed.

The studies were all parallel group comparisons. Six of the studies randomized more than 200 patients to each of the treatment groups; three studies had between 103 and 171 patients in each treatment group, and one study had 30 patients receiving salmeterol and 30 who used increased inhaled steroids as adjuvant therapy. Two studies treated patients for 12 weeks. The other seven studies treated them for 24 to 26 weeks.

Patients who received added salmeterol had higher morning peak expiratory flow (PEF) and forced expiratory volume in one second (FEV₁) levels than patients treated by increasing the steroid dosage. The mean difference in PEF (22.4 L per minute) and FEV₁ (0.10 L) were statistically significant at three and six months. Patients treated with salmeterol also showed significant improvement in the percentage of days and nights without asthma symptoms and use of rescue treatment. Compared with an increased dosage of inhaled steroid, adjuvant salmeterol therapy reduced the total number of patients with one or more exacerbations by 2.73. The authors calculate the number needed to treat as 37. Moderate or severe exacerbations were also significantly reduced, and the estimated number needed to treat was 41.

The authors conclude that in patients using low to moderate dosages of inhaled steroids who experience an increase in symptoms, the addition of salmeterol therapy improves lung function, diminishes exacerbations and reduces the need for rescue medications better than large increases in the steroid dosage.