Determinants of response to bronchodilator in patients with cough variant asthma- A randomized, single-blinded, placebo-controlled study

Background: Not all patients with cough variant asthma (CVA) show responsiveness to bronchodilators (RB) in clinic. Whether there are specific clinical and pathophysiological features can indicate RB in patients with CVA needs further investigation. Thus, we aimed to investigate the RB in patients with CVA and associated factors. Methods: Forty-two CVA patients were randomized in a 2:1 ratio to receive oral bambuterol hydrochloride (10 mg, once daily, for 3 days) or matched placebo, 36 patients (24 with bronchodilator and 12 with placebo) completed the study eventually. RB was considered when cough visual analogue scale (VAS) score decreased 30% or more after 3 days treatment. The baseline clinical and pathophysiological characteristics between pa-tients with RB and patients without RB were compared. CRS was presented with the lowest concentration of capsaicin inducing at least 5 coughing (C5). Results: The responsive rate of patients with bronchodilator was significantly higher than that with placebo (62.5% vs 16.7%, p < 0.01). Patients with RB showed a significant greater mean decline of FEV1% predicted after bronchial provocation (26.7% vs 22.4%, p < 0.05) and higher geometric mean of sputum eosinophils (1.37 vs 0.69, p < 0.05) as compared with these without RB. No significant differences in sputum neutrophil, Log C5 were found between patients with RB and patients without RB. There was a moderate correlation be-tween the decline of FEV1% pred and RB (r(s) = 0.443, p < 0.05). The regression analysis showed that nocturnal cough was a predictor of RB (OR, 7.33, 95% CI: 1.11-48.26, p = 0.038). No adverse events were reported by all of the patients after the study. Conclusion: More than one-third of patients with CVA do not respond to bronchodilator treatment, indicating that the response to bronchodilator should not be a diagnostic requirement of CVA. CVA patients with higher airway responsiveness will more likely respond to bronchodilator. Cough of CVA might be elicited by different mechanisms, which suggests that CVA could be divided into two phenotypes according to the response to bronchodilators.