Respiratory Mechanics and Diaphragmatic Dysfunction in COPD Patients Who Failed Non-Invasive Mechanical Ventilation

Abstract

Background Although non-invasive mechanical ventilation (NIV) is the gold standard treatment for patients with acute exacerbation of COPD (AECOPD) developing respiratory acidosis, failure rates still range from 5% to 40%. Recent studies have shown that the onset of severe diaphragmatic dysfunction (DD) during AECOPD increases risk of NIV failure and mortality in this subset of patients. Although the imbalance between the load and the contractile capacity of inspiratory muscles seems the main cause of AECOPD-induced hypercapnic respiratory failure, data regarding the influence of mechanical derangement on DD in this acute phase are lacking. With this study, we investigate the impact of respiratory mechanics on diaphragm function in AECOPD patients experiencing NIV failure. Methods Twelve AECOPD patients with respiratory acidosis admitted to the Respiratory ICU of the University Hospital of Modena from 2017 to 2018 undergoing mechanical ventilation (MV) due to NIV failure were enrolled. Static respiratory mechanics and end-expiratory lung volume (EELV) were measured after 30 mins of volume control mode MV. Subsequently, transdiaphragmatic pressure (Pdi) was calculated by means of a sniff maneuver (Pdisniff) after 30 mins of spontaneous breathing trial. Linear regression analysis and Pearson’s correlation coefficient served to assess associations. Results Average Pdisniff was 23.3 cmH2O (standard deviation 29 cmH2O) with 3 patients presenting bilateral diaphragm palsy. Pdisniff was directly correlated with static lung elastance (r=0.69, p=0.001) while inverse correlation was found with dynamic intrinsic PEEP (r=−0.73, p=0.007). No significant correlation was found with static intrinsic PEEP (r=−0.55, p=0.06), EELV (r=−0.4, p=0.3), airway resistance (r=−0.2, p=0.54), chest wall, and total elastance (r=−0-01, p=0.96 and r=0.3, p=0.36, respectively). Significant linear inverse correlation was found between Pdisniff and the ratio between Pdi assessed at tidal volume and Pdi sniff (r=−0.82, p=0.02). Conclusion The causes of extreme DD in AECOPD patients who experienced NIV failure might be predominantly mechanical, driven by a severe dynamic hyperinflation that overlaps on an elastic lung substrate favoring volume overload.