We conducted a double blind, randomised, placebo-controlled, crossover study evaluating the effects of halving inhaled steroid dosage plus salmeterol, or salmeterol and tiotropium. Eighteen life-long non-smoking severe asthmatics [mean FEV(1) 1.49 l (51%)] were run-in for 4 weeks on HFA-fluticasone propionate 1000 microg daily, and were subsequently randomised to 4 weeks of either (a) HFA-fluticasone propionate 500 microg BD/salmeterol 100 microg BD/HFA-tiotropium bromide18 microg od; or (b) fluticasone propionate 500 microg BD/salmeterol 100 microg BD matched placebo. Measurements of spirometry and body plethysmography were made. Adding salmeterol to half the dose of fluticasone led to a mean improvement (95% CI) vs. baseline in morning PEF of 41.5 (14.0-69.0)l/min [p<0.05]; and RAW of 0.98 (0.14-1.8)cm H(2)O/l/s [p<0.05]. Adding salmeterol/tiotropium produced similar improvements in PEF and RAW, but also improved FEV(1) by 0.17 (0.01-0.32)l [p<0.05]; FVC 0.24 (0.05-0.43)l [p<0.05] and reduced exhaled NO by 2.86 (0.12-5.6)ppb [p<0.05]. RV and TLC were not altered by either treatment; there were no significant changes in symptoms or quality of life compared with baseline. Addition of salmeterol/tiotropium to half the dose of fluticasone afforded small, but significant improvements in pulmonary function. These effects were not associated with commensurate changes in subjective symptoms or quality of life.