Effect of Oral Procaterol on Chronic Persistent Cough Following Upper Respiratory Tract Infection: Double-Blind Randomized Placebo-Controlled Trial

Persistent cough following upper respiratory tract infection (URTI) is a common problem in the clinical practice, namely post-infectious cough. The potential mechanisms are viral-induced airway epithelial damage that leads to 1) airway hyperresponsiveness and airway narrowing, 2) increase in vascular permeability resulting in airway edema, and 3) activation of inflammatory mediators from inflammatory cells resulting in airway smooth muscle contraction. It is usually spontaneously resolved, although various therapeutic trials have been used with unpredictable results. Regarding to bronchodilators, inhaled ipratropium was effective in reducing cough symptom in a small study (N=14). We conduct a double-blind randomized placebo-controlled trial to investigate the effectiveness of oral procaterol, as a bronchodilator, in non-asthmatic adult patients suffering from persistent cough post URTI.

Eligible patients who have cough lasting longer than 3 weeks post URTI with normal spirometry will be randomized to receive either placebo or procaterol (25 microgram twice daily) for 4 weeks.

The primary outcome is cough symptom score using Leicester cough questionnaire (LCQ). The secondary outcomes are pulmonary function tests (spirometry, impulse oscillometry) and exhaled nitric oxide and quality of life (SF-36). All outcomes are measured at baseline, 2 weeks, and 4 weeks. Bronchoprovocation test with methacholine is performed at baseline and 4 weeks to determine the provocative concentration of methacholine that induces falling of FEV1 >or =20%. Adverse events will be recorded every visit.

Data analysis will be in both intention-to-treat and per-protocol fashion. A linear mixed-effect regression model will be applied to assess treatment effect on LCQ score, SF-36, and lung function. Within-subject variation will be fitted in the model as random effects whereas the treatment will be considered as a fixed effect. Time at measurement (i.e., 2- and 4-week) will also be included in the mixed model by treating it as fixed-effect. Marginal treatment effects between treatments and time will be then estimated and compared.