Very-Low Driving Pressure Ventilation in Patients with Acute Respiratory Distress Syndrome on Extracorporeal Membrane Oxygenation

Abstract

Rationale: The ideal ventilator settings in patients with severe acute respiratory distress syndrome (ARDS) while on Extracorporeal Membrane Oxygenation (ECMO) are unknown. ELSO recommends modest driving pressures and PEEP (i.e., “lung rest”) during ECMO to minimize ventilator induced lung injury (VILI) and improve mortality. However, even lower driving pressures (DP) may improve inflammatory and lung injury biomarker profiles, and possibly improve mortality. The focus of this study was to evaluate the feasibility and impact of very-low DP in patients with ARDS on ECMO. Methods: Patients on ECMO due to ARDS from coronavirus disease 2019 had their DP decreased from 10 to 1-5 cmH2O for two hours, as tolerated. The other ventilator settings were unchanged and were consistent with ELSO guidelines;positive end-expiratory pressure (10-15 cmH2O) and a respiratory rate 10 breaths per minute. During the low DP protocol, the transpulmonary pressure was monitored with esophageal manometry. Plasma (for IL6 and sRAGE) and urine were collected before and after the protocol. Results: Thirty patients were enrolled and ultimately 21 patients underwent the low DP protocol. Seventeen were male, aged 51 ± 9 years, median BMI of 31.7 kg/m2 (IQR 27.4-35.2), with a median SOFA score of 10 (IQR 8-12) at ICU admission. There were no complications during the low DP protocol (1-5 cm H2O). The protocol was performed a median of 16 (IQR 13.5-20) hours after ECMO initiation. Average ECMO days were 21 (IQR 12.5-41) with a survival to hospital discharge of 43% (9/21). Baseline average tidal volume (TV) was 3.4±2.6 ml/kg of predicted body weight (PBW), during the low DP protocol the TV decreased to 2.7±3.2 ml/kg of PDW. Two groups were noted-those who had spontaneous breaths vs. those who did not (see Figure 1). Ten patients during the low DP protocol continued to have unchanged TV, 5.2±2.8 to 5.4±2.7 ml/kg of PBW due to spontaneous breathing found on esophageal manometry. Eleven patients had no spontaneous breaths and TV decreased from 1.76±0.57 to 0.2±0.38 ml/kg of PBW during the low DP protocol. The spontaneous vs. nonspontaneous breathing groups has a statistically different TV before, during, and after the low DP protocol (p<0.001). Conclusions: Very-low DP is feasible in patients on ECMO support which may decrease VILI. However, in the absence of neuromuscular blockade patients might spontaneously breathe with the same transpulmonary pressure. Thus, whether low DP offers any advantage in spontaneous breathing vs. non-spontaneous breathing requires further study. (Table Presented).