Proportional assist versus pressure support ventilation: effects on breathing pattern and respiratory work of patients with chronic obstructive pulmonary disease

Objective:To investigate the breathing pattern and the inspiratory work of breathing (WOBI) in patients with chronic obstructive pulmonary disease (CO PD) assisted with proportional assist ventilation (PAV) and conventional pr essure support ventilation (PSV). Design: Prospective controlled study. Setting: Intensive care unit of a university hospital. Patients: Thirteen COPD patients being weaned from mechanical ventilation. Interventions: All patients were breathing PSV and two different levels of PAV. Measurements and main results: During PAV (EVITA 2 prototype, Drager, Germa ny), the resistance of the endotracheal tube (R-et) was completely compensa ted while the patients' resistive and elastic loads were compensated for by approximately 80% and 50% (PAV(80) and PAV(50)), respectively. PSV was adj usted to match the same mean inspiratory pressure (Pinsp(mean)) as during P AV(80). Airway pressure, esophageal pressure and gas flow were measured ove r a period of 5 min during each mode. Neuromuscular drive (P-0.1) was deter mined by inspiratory occlusions. Mean tidal volume (V-T) was not significan tly different between the modes. However, the coefficient of variation of V -T was 10 +/- 4.%, 20 +/- 13% and 15 +/- 8% during PSV, PAV(80) and PAV(50) , respectively. Respiratory rate (RR) and minute ventilation (V-E) were sig nificantly lower during PAV(80) as compared with both other modes, but the differences did not exceed 10%. PAV(80) and PSV had comparable effects on W OBI and P-0.1, whereas WOBI and P-0.1 increased during PAV(50) compared wit h both other modes. Conclusion: Mean values of breathing pattern did not differ by a large amou nt between the investigated modes. However, the higher variability of V-T d uring PAV indicates an increased ability of the patients to control V-T in response to alterations in respiratory demand. A reduction in assist during PAV(50) resulted in an increase in WOE and indices of patient effort.