Long-acting beta(2)-agonists are prescribed for patients with persistent asthma and are sometimes used without inhaled corticosteroids (ICSs). No evidence exists, however, to support their use as monotherapy in adults with persistent asthma. To examine the effectiveness of salmeterol xinafoate, a long-acting beta(2)-agonist, as replacement therapy in patients whose asthma is well controlled by low-dose triamcinolone acetonide, an ICS. A 28-week, randomized, blinded, placebo-controlled, parallel group trial conducted at 6 National Institutes of Health-sponsored, university-based ambulatory care centers from February 1997 to January 1999. One hundred sixty-four patients aged 12 through 65 years with persistent asthma that was well controlled during a 6-week run-in period of treatment with inhaled triamcinolone (400 microg twice per day). Patients were randomly assigned to continue triamcinolone therapy (400 microg twice per day; n = 54) or switch to salmeterol (42 microg twice per day; n = 54) or to placebo (n = 56) for 16 weeks, after which all patients received placebo for an additional 6-week run-out period. Change in morning and evening peak expiratory flow (PEF), forced expiratory volume in 1 second (FEV(1)), self-assessed asthma symptom scores, rescue albuterol use, asthma-specific quality-of-life scores, treatment failure, asthma exacerbation, bronchial reactivity, and markers of airway inflammation, compared among the 3 treatment groups. During the 16-week randomized treatment period, no significant differences between the salmeterol and triamcinolone groups were observed for conventional outcomes of clinical studies of asthma therapy-morning PEF, evening PEF, asthma symptom scores, rescue albuterol sulfate use, or quality of life. Both active treatments were superior to placebo. However, the salmeterol group had more treatment failures than the triamcinolone group (13/54 [24%] vs 3/54 [6%]; P =.004), as well as more asthma exacerbations (11/54 [20%] vs 4/54 [7%]; P =.04), greater increases in median (interquartile range) sputum eosinophils (2.4% [0.0% to 10.6%] vs -0.1% [-0.7% to 0.3%]; P<.001), eosinophil cationic protein (71 [-2 to 430] U/L vs -4 [-31 to 56] U/L; P =.005), and tryptase (3.1 [2.1 to 7.6] ng/mL vs 0.0 [0.0 to 0.7] ng/mL; P<.001). The duration of benefit when patients were switched from active treatment to placebo after 22 weeks of randomized treatment was not significantly longer in the triamcinolone group than in the salmeterol group. Patients with persistent asthma well controlled by low doses of triamcinolone cannot be switched to salmeterol monotherapy without risk of clinically significant loss of asthma control.