Background: Extracorporeal membrane oxygenation (ECMO) has expanding indications for cardiopulmonary resuscitation including severe acute respiratory distress syndrome (ARDS). Despite the adjunct of ECMO for patients with severe ARDS, they often have prolonged mechanical ventilation and are subject to many of its inherent complications. Here, we describe patients who were cannulated for venovenous (VV) ECMO and were taken off positive pressure ventilation. Methods: This is a primary analysis of patients admitted at a tertiary medical center between the dates of August 2014 to January 2020 who were cannulated to ECMO for refractory respiratory failure. We included all patients ≥18 years old. Patients who were extubated or had a tracheostomy and taken off positive pressure while on ECMO were classified as “off positive pressure ventilation (PPV)” and were compared to patients who remained “on PPV” while on ECMO. Primary outcome was survival to hospital discharge. Secondary outcomes were ventilator free days at 30 days and 60 days after ECMO cannulation, time from cannulation to date of first out-of-bed (OOB), and hospital length of stay (LOS). Patient characteristics were derived from routine clinical information in the electronic health record (EHR). Categorical characteristics were compared using chi-square test or Fisher exact test. Continuous characteristics were compared using independent samples t-test or Wilcoxon–Mann–Whitney test. p-values were reported from all analysis. Results: Sixty-five patients were included in this retrospective analysis. Forty-eight were managed on ECMO with PPV and 17 patients were removed from PPV. Patients removed from PPV had significantly higher lung injury scores prior to cannulation (2.5 ± 0.6 vs. 1.04 ± 0.3; p = 0.031) and non-significantly longer duration of ventilation prior to ECMO (6.1 days ± 2.1 vs. 5.0 days ± 01.1; p = 0.634). One hundred percent (100%) of patients removed from PPV survived to hospital discharge compared to 45% who received PPV throughout their duration of ECMO management (p < 0.001). The mean ventilator free days at day 60 was 15 with PPV and 36 without PPV (p = 0.003). The average duration from cannulation to mobilization (i.e., out-of-bed) was 18 days with PPV and 7 days without PPV (p = 0.015). Conclusions: Patients taken off PPV while on ECMO had a very high likelihood of survival to discharge and were mobilized in half as many days. While this likely reflects patient selection, the benefit of early mobilization is well documented and the approach of extubating while on ECMO warrants further investigation.
The transition from control modes to spontaneous modes is ubiquitous for mechanically ventilated patients yet there is little data describing the changes and patterns that occur to breathing during this transition for patients on ECMO. We identified high fidelity data among a diverse cohort of 419 mechanically ventilated patients on ECMO. We examined every ventilator change, describing the differences in >30,000 sets of original ventilator observations, focused around the time of transition from control modes to spontaneous modes. We performed multivariate regression with mixed effects, clustered by patient, to examine changes in ventilator characteristics within patients, including a subset among patients with low compliance (<30 milliliters (mL)/centimeters water (cmH2O)). We found that during the transition to spontaneous modes among patients with low compliance, patients exhibited greater tidal volumes (471 mL (364,585) vs. 425 mL (320,527); p < 0.0001), higher respiratory rate (23 breaths per minute (bpm) (18,28) vs. 18 bpm (14,23); p = 0.003), greater mechanical power (elastic component) (0.08 mL/(cmH2O × minute) (0.05,0.12) vs. 0.05 mL/(cmH2O × minute) (0.02,0.09); p < 0.0001) (range 0 to 1.4), and lower positive end expiratory pressure (PEEP) (6 cmH2O (5,8) vs. 10 cmH2O (8,11); p < 0.0001). For patients on control modes, the combination of increased tidal volume and increased respiratory rate was temporally associated with significantly low partial pressure of arterial oxygen (PaO2)/fraction of inspired oxygen (FiO2) ratio (p < 0.0001). These changes in ventilator parameters warrant prospective study, as they may be associated with worsened lung injury.
About Me
Anesthesiologist and Intensivist at University of Utah
Anesthesiologist and Intensivist at University of Utah