Clinical Validation of a Novel Framework for Dyspnoea Investigation with CPET in COPD

Background Exertional dyspnoea remains a chief complaint of patients assessed by pulmonologists, yet current cardiopulmonary exercise testing (CPET) displays and interpretations are not geared to properly expose its key determinant: ventilatory demand-capacity imbalance. Methods We analyzed cycle ergometer CPET data from 3 cohorts of smokers and ex-smokers at risk for, or at different stages of, COPD (N=1161; 471 females; FEV 1 z-score from −5.53 to 3.06). Considering all test data points, a novel AI-based framework (Dynamic Assessment of Dyspnoea and Ventilation on Exercise, DyVe-X) quantified the severity of dyspnoea while simultaneously assessing the gradual erosion of mechanical-ventilatory reserves in subjects with varying pre-test likelihood of abnormality. Results Categories of progressively worse mechanical-ventilatory abnormalities, as defined by DyVe-X, were closely associated with increasing dyspnoea burden and decreasing exercise capacity across cohorts ( p <0.001). ∼50% of patients with preserved peak breathing reserve (BR peak )- the current criterion for ventilatory limitation – showed reduced submaximal ventilatory and/or inspiratory reserves on DyVe-X; these patients were as dyspnoeic and functionally impaired as those with low BR peak and abnormal DyVe-X. DyVe-X, but not BR peak , predicted high dyspnoea and impaired exercise tolerance across the full range of resting functional abnormalities; the odds ratio of combined “constrained” and “excessive” breathing by DyVe-X, adjusted by BR peak , for high dyspnoea burden and poor exercise tolerance were 26.5 [15.8–44.3) and 26.0 [16.6–40.7], respectively ( p <0.001). Conclusions These novel paradigms for the display and interpretation of CPET in dyspnoeic patients are poised to empower clinicians to uncover the root causes of exertional breathlessness with potentially relevant therapeutic implications.