Dynamic Intrinsic PEEP and Respiratory Muscle Redistribution During High-Flow Nasal Cannula in Acute Exacerbation of COPD: A Physiological Proof-of-Concept Study

Abstract Background Dynamic intrinsic positive end-expiratory pressure (PEEPi) is a key determinant of inspiratory loading during exacerbation of chronic obstructive pulmonary disease (ECOPD), imposing a threshold load that must be overcome before inspiratory flow can be generated. However, the extent to which dynamic PEEPi is associated with respiratory muscle redistribution, inspiratory effort, and ventilatory efficiency remains poorly characterized. Methods In this prospective physiological proof-of-concept study, patients with ECOPD underwent a comprehensive assessment of respiratory mechanics, inspiratory effort, respiratory muscle activity, breathing pattern, dyspnea, and gas exchange at baseline and after 2 hours of HFNC delivered under standardized conditions. HFNC was delivered at 40 L/min, with FiO₂ titrated to target an SpO₂ of 92-94%. Physiological measurements included dynamic PEEPi, Pes, Pdi, Pdi sniff, Pnose, diaphragm and accessory muscle ultrasound indices, Vte, respiratory rate (RR) and Borg dyspnea score. Results Seven patients with ECOPD underwent complete physiological assessment. Higher dynamic PEEPi was associated with a pattern of respiratory muscle redistribution, characterized by lower diaphragm thickening fraction (ρ = −0.68) and greater intercostal and sternocleidomastoid thickening fractions (ρ = 0.59 and 0.81, respectively). This was paralleled by greater inspiratory effort, reflected by larger Pes, Pdi, and tidal Pnose swings (ρ = 0.57, 0.71, and 0.88, respectively). Finally, higher dynamic PEEPi was associated with lower ventilatory efficiency, reflected by lower Vte and Vte/PBW (ρ = −0.71 and −0.64, respectively), together with higher RR and Borg dyspnea score (ρ = 0.75 and 0.95, respectively). Across the T0–T1 assessment, physiological trajectories were heterogeneous, suggesting that baseline dynamic PEEPi may modulate the short-term physiological response to HFNC. Conclusions Dynamic PEEPi may identify an unfavourable neuro-mechanical phenotype in ECOPD, characterized by inspiratory threshold loading, respiratory muscle redistribution, increased inspiratory effort, and impaired ventilatory efficiency. These findings support a physiology-guided approach to respiratory support and highlight the need for individualized monitoring during HFNC.