Montelukast for post-infectious cough in adults: a double-blind randomised placebo-controlled trial
Kay Wang, Surinder S. Birring, Kathryn Taylor, Norman Fry, Alastair D. Hay, Michael Moore, Jing Jin, Rafael Perera, Andrew Farmer, Paul Little, Timothy Harrison, David Mant, Anthony Harnden
Background Post-infectious cough is common in primary care, but has no proven effective treatments. Cysteinyl leukotrienes are involved in the pathogenesis of post-infectious cough and whooping cough (pertussis). We determined the effectiveness of montelukast, a cysteinyl leukotriene receptor antagonist, in the treatment of post-infectious cough. Methods Non-smoking adults aged 16 to 49 years with post-infectious cough of two to eight weeks’ duration were recruited from 25 general practices in England. Patients were tested for pertussis (oral fluid anti-pertussis toxin IgG) and randomly assigned to montelukast 10 mg daily or image-matched placebo for two weeks. Patients chose whether to continue study medication for another two weeks. The randomisation sequence was computer-generated and stratified by general practice. Patients, healthcare professionals and researchers were blinded to treatment allocation. Effectiveness was assessed using the Leicester Cough Questionnaire to measure changes in cough-specific quality of life between baseline and two follow-up stages (two weeks and four weeks). The primary analysis was by intention to treat with imputation by last observation carried forward. Recruitment closed on 21st September 2012 and follow-up has been completed. Findings We randomly assigned 276 patients to montelukast (n=137) or placebo (n=139). Seventy patients had laboratory-confirmed pertussis (25%). Improvements in cough-specific quality of life occurred in both groups after two weeks (montelukast: mean 2•7, [95% confidence interval 2•2 to 3•3]; placebo: 3•6 [2•9 to 4•3], p=0•04) but the difference between groups was not clinically significant (mean difference -0•9 [-1•7 to -0•04]). This difference was not statistically significant in any sensitivity analyses. After two weeks, 74•1% of patients elected to continue study medication (montelukast 76•7%; placebo 71•5%). After four weeks, there were no significant between-group differences in cough-specific quality of life improvement (montelukast: 5•2 [4•5 to 5•9]; placebo: 5•9 [5•1 to 6•7]; mean difference -0•5 [-1•5 to 0•6], p=0•38) or adverse event rates (montelukast 15•3%; placebo 22•3%, p=0•14). The most common adverse events reported were increased mucus production (montelukast, n=6; placebo, n=2) and headache (montelukast, n=2; placebo, n=6). One serious adverse event was reported (placebo, n=1), which was unrelated to study medication (shortness of breath and throat tightness after severe coughing bouts). This trial is registered with EudraCT (2010-019647-19), UKCRN Portfolio (ID 8360), and ClinicalTrials.gov (NCT01279668). Interpretation Montelukast is not an effective treatment for post-infectious cough. However, the burden of post-infectious cough in primary care is high, making it an ideal setting for future anti-tussive treatment trials.