TREETOP Study: A Novel Lung Function Technique for the Early Detection of Chronic Obstructive Pulmonary Disease

Abstract INTRODUCTION Chronic obstructive pulmonary disease (COPD) causes major morbidity and mortality worldwide. Diagnosis relies on demonstrating airflow limitation on spirometry, which is insensitive to early changes in the small airways (where disease typically begins). Patient outcomes remain poor because irreversible pathology usually occurs before diagnosis. Computed cardiopulmonography (CCP) is a non-invasive technique that measures inhomogeneity of gas-exchange across the lung. It comprises a novel molecular flow sensor which accurately measures respired gases and a computational approach. CCP parameters, particularly the ventilation inhomogeneity index σlnCL, a sensitive marker of small-airway change, are abnormal in mild COPD and change with disease severity. Here, we report initial data from the TREETOP study, which uses CCP to study people at risk of COPD – smokers with no respiratory diagnosis and normal spirometry – alongside non-smoking controls, to investigate whether CCP can detect changes in lung function earlier than is possible using conventional tests. METHODS Current tobacco smokers of ≥10 pack-years and non-smoking controls, aged 25-60 years, with no respiratory disease, were identified through advertisement and primary care health record searches. During study visits participants reported symptoms using the COPD Assessment Test (CAT). Lung function tests were performed, including: CCP test (7 minutes breathing air followed by 5 minutes breathing pure oxygen), forced spirometry, and impulse oscillometry. Statistical analysis used t-tests. RESULTS 53 current smokers of ≥10 pack-years and 24 never-smokers were recruited. Of the smokers, 12 (23%) had abnormal spirometry (FEV1<80% predicted and/or FEV1/FVC<0.7); 1 non-smoker had abnormal spirometry. Figure 1 shows data for the 41 smokers with normal spirometry (mean age 47±9years) vs controls (46±12years). Despite normal spirometry, there were significantly more symptoms in the smokers (Fig1A p<0.0001) with 42% having a CAT score >10. FEV1 values showed significant overlap between groups (Fig1B) and no significant difference in the group means. σlnCL was significantly different between groups (Fig1C, p<0.0001) with two thirds of smokers exhibiting abnormal values (z-score>1.645). FEF25-75 and R5-20, considered conventional small-airway measures, were not significantly different between groups. Of the 13 participants with abnormal spirometry, σlnCL was abnormal in 12, while R5-20 was abnormal in 6. CONCLUSION CCP's inhomogeneity index σlnCL can detect early abnormalities in lung function in people at risk of developing clinical COPD even when spirometry is normal, when other conventional small-airway tests cannot. As such, CCP presents a realistic opportunity to bring COPD disease detection – and ultimately therapy – “upstream” of irreversible airway modelling.