Effect of portable non-invasive ventilation on exercise tolerance in COPD: One size does not fit all

In a cross-over RCT, portable NIV (pNIV) reduced dynamic hyperinflation (DH) compared to pursed lip breathing (PLB) during recovery from intermittent exercise in COPD, but not consistently in all subjects. In this post-hoc analysis, DH response was defined as a reduction >= 4.5 % of predicted resting inspiratory capacity with pNIV compared to PLB. At exercise iso-time (where work completed was consistent between pNIV and PLB), 8/24 patients were DH non-responders (DH: 240 +/- 40 mL, p = 0.001 greater using pNIV). 16/24 were DH responders (DH: 220 +/- 50 mL, p = 0.001 lower using pNIV). Compared to DH responders, DH non-responders exhibited greater resting DH (RV/TLC: 65 +/- 4% versus 56 +/- 2%; p = 0.028) and did not improve exercise tolerance (pNIV: 30.9 +/- 3.4 versus PLB: 29.9 +/- 3.3 min; p = 0.603). DH responders increased exercise tolerance (pNIV: 34.9 +/- 2.4 versus PLB: 27.1 +/- 2.3 min; p = 0.001). Resting RV/TLC% was negatively associated with the magnitude of DH when using pNIV compared to PLB (r = -0.42; p = 0.043). Patients with profound DH were less likely to improve exercise tolerance with pNIV. Further studies using auto-adjusted ventilators are warranted.