Inhaled corticosteroids have been shown to effectively reduce large circadian fluctuations in peak expiratory flow (PEF). Salmeterol xinafoate (SLM), a new long-acting beta2-agonist being used in the treatment of nocturnal airway obstruction, has proved to be very effective in this respect as well. However, it is yet unknown whether using SLM alone or in combination with fluticasone propionate (FP) constitutes the best treatment. We studied, in a randomized, double-blind, parallel manner, 46 asthmatics with increased circadian variation in PEF (> or = 15%) for 6 wk to compare FP 250 microg, SLM 50 microg, and a combination of them, all given twice a day. These three treatment protocols were equally effective in improving the generally used clinical outcome parameters, i.e., the circadian variation in PEF and FEV1 and bronchial hyperresponsiveness (BHR) to methacholine (MCh) during the day and at night. FEV1 increased more at 4:00 A.M. than at 4:00 P.M. (FEV1 at both time points > 90% predicted). BHR to MCh improved with at least 1.5 doubling concentrations, thereby reducing the significant nocturnal decline in the SLM and FP group, but not in combination. The improvement in BHR to adenosine 5'monophosphate was greater (p = 0.05) when FP was combined with SLM but not when FP or SLM were used alone. Our data support the clinical view that FP, with its anti-inflammatory capacity, has greater beneficial effects as monotherapy than does SLM. However, this was detectable only by using the "indirect" stimulus adenosine 5'monophosphate, which is more specific in assessing changes in different components of airway wall inflammation than is MCh.