Prone Positioning as a Rescue Therapy in Patients Undergoing Extracorporeal Membrane Oxygenation

Objectives: The aim of the study was to determine the benefit of prone positioning (PP) in patients with acute respiratory failure (ARF) on extracorporeal membrane oxygenation (ECMO) after cardiac surgery.

Methods: A review of our institutional database from April 2017 to August 2018 revealed six patients on ECMO (va-ECMO = 3, vv-ECMO = 3) who underwent PP for the treatment of ARF. Data collection included the Horowitz index (HI) and peak inspiratory pressure (PIP) at the time of PP, 6 hours after PP, 12 hours after PP, at the end of PP and 6 hours after return to supine position (SP). All blood gas analyses were taken from the right radial artery for the most reliable analysis of pulmonary oxygenation capacity. Results are presented as medians (25th-75th quartiles) or as absolute numbers and percentages.

Results: The median age was 67 years (53–70) and the median EuroSCORE II was 17.3% (7.7–43.1). Three patients had a history of COPD and two were diagnosed with pulmonary hypertension. Two patients underwent emergency surgery. The median duration of PP was 15 hours (12–20) and the median number of days after surgery where PP was implemented was 6 (4-13). We observed an increase in HI at the end of PP as compared to before PP (126 [101–187] vs. 210 [129–275], p = 0.018). This was followed by an insignificant decline in HI 6 hours after return to SP (200 (127–223), p = 0.091). In sum, the overall increase in HI compared to baseline values and values at 6 hours after return to SP indicated a significant respiratory improvement in terms of HI. However, there was no significant change in FiO2 on ECMO before and at the end of PP (0.85 [0.55–1.0] vs. 0.80 [0.40–1.0], p = 0.102). We found a nearly significant decrease in ECMO support in terms of flow before and after PP (3.1 lpm (2.8–5.6) vs. 2.5 lpm (2.0–4.4), p = 0.068). The median duration of ECMO support was 246 hours (174–376) and a total of four patients could be weaned from ECMO support. The median ICU stay was 22 days (15–32). There were no adverse events reported during the positioning of the patients. A total of four patients underwent dilatational tracheostomy and five patients suffered from nosocomial pneumonia. Thirty-day mortality was 33.3%.

Conclusions: Prone positioning can safely and effectively be used as a rescue therapy for ARF in patients on ECMO support following cardiac surgery, improving short and medium term respiratory condition and may, thus, facilitate weaning from ECMO support.