The objectives of this study were to compare the efficacy and safety of salmeterol xinafoate (50 and 100 microg b.i.d.) with that of placebo, when added to existing therapy, in the treatment of patients with chronic obstructive pulmonary disease (COPD). Six hundred and seventy four patients were randomized to receive either salmeterol 50 microg b.i.d., salmeterol 100 microg b.i.d., or placebo treatment for a period of 16 weeks. The results showed a significant improvement in daily symptom scores noted for patients taking either 50 microg (p=0.043) or 100 microg b.i.d. salmeterol (p=0.01) compared with placebo, with a corresponding decrease in additional daytime salbutamol requirements for both salmeterol groups. The same pattern was reflected for night-time symptoms and additional salbutamol use. During treatment, forced expiratory volume in one second (FEV1) measurements improved significantly in each salmeterol group, with up to a 7% improvement observed at the end of the study. Although no difference was observed between treatment groups for the distance walked in 6 min, patients treated with salmeterol 50 microg b.i.d. were significantly less breathless than those treated with placebo after their 6 min walk, after 8 weeks (p=0.024) and 16 weeks (p=0.004) of therapy. Adverse events were similar in all three groups except for tremor, which was significantly higher in the 100 microg b.i.d. salmeterol group (p=0.005) compared both with 50 microg b.i.d. salmeterol and placebo. Salmeterol offered further positive improvement to the effect of therapy in patients with chronic obstructive pulmonary disease when added to their existing regimens. This clinical improvement was similar both with 50 and 100 microg b.i.d. dosage, although the group receiving 50 microg b.i.d. tolerated the drug better than those receiving 100 microg b.i.d. salmeterol.