Adding a long-acting beta₂ agonist to the regimen of patients whose asthma is not well controlled with an inhaled corticosteroid is more effective than increasing the corticosteroid dosage. Some have wondered whether it is possible to discontinue the inhaled corticosteroid once the long-acting beta₂ agonist has helped patients control their asthma. Lemanske and associates tested this hypothesis in the Salmeterol ± Inhaled Corticosteroids (SLIC) trial.

Adult patients with asthma who were younger than 60 years were included in the study if their asthma was not well controlled at the end of a six-week run-in period in which each patient received four puffs twice daily of triamcinolone by metered-dose inhaler. In the first phase after the run-in period, all patients continued using triamcinolone, and 13 of every 15 patients also received salmeterol (two puffs twice daily by metered-dose inhaler). The remaining two of every 15 patients received placebo. After two weeks, one half of the patients who were given triamcinolone plus salmeterol were randomized to continue their current medications, and one half were randomized to reduce their triamcinolone dosage by 50 percent for eight weeks and then to discontinue triamcinolone for another eight weeks. Those initially receiving triamcinolone plus placebo had similar dosage adjustments. All of these changes were done in a blinded fashion. Albuterol was used for rescue treatment, if needed.

Time-to-treatment failure was the main outcome measure. Other spirometry measures and asthma symptom reports were also collected, and an asthma-specific quality-of-life score was generated.

The SLIC study initially included 175 patients who failed to achieve adequate asthma control with triamcinolone. In the salmeterol introduction phase, 21 patients received placebo, and 154 received salmeterol; 148 patients in the salmeterol group and 19 patients in the placebo group completed this phase. Of the 148 patients in the salmeterol group, 74 were randomly assigned to continue current medications, and 74 were assigned to have reduction-discontinuation of triamcinolone. In the triamcinolone reduction and elimination phases, 13 patients withdrew for non–study-related reasons.

During the triamcinolone reduction phase, 2.8 percent of patients in the salmeterol-triamcinolone group had treatment failures, compared with 8.3 percent of patients in the salmeterol-reduced triamcinolone group. At the end of the elimination phase, these percentages increased, with 13.7 percent of patients in the salmeterol-triamcinolone group having treatment failures compared with 46.3 percent of patients in the salmeterol-discontinued triamcinolone group. Overall, the outcomes in the patients whose triamcinolone was discontinued were significantly worse than the outcomes in those who continued triamcinolone, even if it was a reduced dosage. Specifically, the relative risk of treatment failure was 2.2 (comparing the full-dose with reduced-dose triamcinolone) but rose to 4.3 when triamcinolone was discontinued.

The authors concluded that a reduction in triamcinolone dosage may be possible in most asthmatic patients who have had salmeterol added to their regimen. However, discontinuation of the inhaled corticosteroid will lead to poorer outcomes, including a treatment failure rate nearly as high as that in patients receiving placebo.