OBJECTIVE Whether beta(2)-adrenoceptor gene (ADRB2) polymorphisms are associated with airway responsiveness to beta(2)-agonist medications remains controversial, partly due to factors that may confound pharmacogenetic associations, including age, cigarette smoking and airway remodelling. To overcome these problems, we performed an analysis using parameters that reflected the specific bronchodilator response to beta(2)-agonists. METHODS The increases in FEV(1) after inhalation of procaterol hydrochloride (Delta FEV(1) procaterol) or oxitropium bromide (Delta FEV(1) oxitropium), and after sequential inhalation of procaterol and oxitropium (total airway reversibility), were measured in 81 Japanese patients with moderate to severe asthma. Approximately 3 kb of the DNA sequence of the coding and 5'-flanking regions of ADRB2 were genotyped by direct sequencing and PCR-restriction fragment length polymorphism assay. RESULTS The mean age of the participants was 54 years, and 38 (47%) were smokers. Although Delta FEV(1) procaterol and Delta FEV(1) oxitropium adjusted for predicted FEV(1) were not associated with ADRB2 polymorphisms, the ratio of Delta FEV(1) procaterol to total airway reversibility was significantly associated with the ADRB2 A46G genotype (P < 0.05). Patients who were homozygous for the A46 allele (arginine at amino acid 16) were more responsive than carriers of the G46 (glycine 16) allele (P = 0.008). Multivariate linear regression analysis showed that Delta FEV(1) procaterol was correlated with the number of A46 alleles (P = 0.014), and also with total airway reversibility (P < 0.001) and smoking index in current smokers (P = 0.009). CONCLUSIONS The ADRB2 A46G polymorphism was associated with a relatively greater bronchodilator responsiveness to beta(2)-agonists even in elderly asthmatic patients and smokers.