- Design
- Randomised, sham-controlled trial, 2:1 allocation
- Population
- 203 patients with COPD and chronic bronchitis
- Primary outcome
- SGRQ and CAT change at 12 months
- Effect
- SGRQ −5.9 (95% CI −11.6 to −0.2); CAT −3.4 (−5.8 to −1.0); pneumothorax 3.6% per treatment
SPRAY-CB randomised 203 people with COPD and chronic bronchitis, 2:1, to two bronchoscopies 30 to 60 days apart with mucus clearance plus metered liquid nitrogen cryospray to the airways, or to the same bronchoscopies without cryospray (sham). The idea is that freezing the abnormal surface lining allows healthier epithelium to regrow.
At 12 months, St George's Respiratory Questionnaire improved 5.9 points more with cryospray (95% CI −11.6 to −0.2), and the COPD Assessment Test 3.4 points more (−5.8 to −1.0). Exacerbations of any severity were similar (70.6% vs 67.2%); severe exacerbations were 21.3% against 26.9%. Pneumothorax occurred after 3.6% of treatments. There were six deaths with cryospray and two in controls, none judged procedure-related.
The symptom score gains are around the threshold patients notice, and the SGRQ interval runs close to zero. Severe exacerbations were not a formally tested outcome here. This is a specialised bronchoscopic procedure not available in most centres, and it is an option to watch rather than to refer for now.
- Optimise inhaled therapy, airway clearance and smoking cessation in chronic bronchitis before considering any bronchoscopic option.
- Tell patients who ask that cryospray improved symptom scores but not overall exacerbations in one trial.
- Weigh the 3.6% per-treatment pneumothorax rate against the modest symptom gain.
- Consider referral only to specialist centres with access to trials.
Why it matters
Productive cough in COPD has few targeted treatments, and this is a new mechanism under test.
Don't overread it
The lower severe exacerbation rate was not shown to be statistically significant.
The statistics, in plain English
A 4-point change in SGRQ is usually taken as the smallest change patients notice, so −5.9 is just above it, but the confidence interval reaches −0.2, close to no effect. A sham control helps, because patients did not know whether they had the active treatment.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for pulmonology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free