Résumé
Objectives: In patients with severe respiratory failure, invasive ventilation may deteriorate the pneumomediastinum and hypoxia. This study aimed to compare the mortality and the complications of the patients with coronavirus disease 2019 (COVID-19) related severe ARDS treated with invasive ventilation or veno-venous ECMO (VV-ECMO) to avoid intubation. We hypothesized that VV-ECMO support without prior intubation is a feasible alternative strategy to invasive ventilation. Method(s): This retrospective study evaluated patients with COVID-19 related severe respiratory failure and radiological evidence of pneumomediastinum. The primary outcome was intensive care unit (ICU) survival at 90 days. Result(s): Out of 347 patients with COVID-19 disease treated in our unit, 22 patients developed spontaneous pneumomediastinum associated with deterioration of respiratory function. In 13 patients (59%), invasive ventilation was chosen as initial respiratory support;in 9 patients (41%), VV-ECMO was chosen as initial respiratory support. The median age of the patients in the invasive ventilation group was 62 years (IQR: 49-69) compared to 53 years (IQR: 46-62) in ECMO group (P=0.31). No statistically significant difference in SAPS II score between the groups was observed (39.7 (IQR: 33.2-45.3) vs. 28.9 (IQR:28.4-34.6), P=0.06). No elevated fluid balance within the first 4 days was observed in the ECMO group compared to the invasive ventilation group (162 mL (IQR: -366-2000) vs. 3905 mL (IQR: 2068-6192), P=0.07). VV-ECMO as the initial strategy for supporting patients with severe respiratory failure and pneumomediastinum, was associated with lower 90 days mortality (HR: 0.33 95%-CI: 0.11-0.97, P= 0.04) compared to patients treated with invasive ventilation (Figure). Conclusion(s): VV-ECMO can be an alternative strategy to invasive ventilation for treating patients with severe respiratory failure and spontaneous pneumomediastinum. (Figure Presented).
Résumé
Introduction: Invasive ventilation initiation after a prolonged period of non-invasive ventilation (NIV) trial can be associated with poor outcome in coronavirus disease 2019 (COVID-19) ARDS patients. This study aimed to document our center's experience with COVID-19 ARDS patients treated with veno-venous ECMO (VVECMO) after a prolonged NIV trial period to avoid intubation. We speculated that VV-ECMO support is not associated with a worse outcome than invasive ventilation in these patients. Methods: We retrospectively reviewed 6 patients with COVID-19 ARDS who presented severe hypoxemia and pneumomediastinum after NIV (ECMO group). Twenty patients with COVID - 19 and age less than 70 years old were treated in the first wave of the national outbreak and underwent NIV trials for more than 24 h before intubation (Control group). The primary outcome was intensive care unit (ICU) survival and secondary ECMO or mechanical ventilation weaning at 28 days. Results: The age of the patients in the ECMO group was 59 years (IQR: 46 - 65) and SAPS II score 47 (IQR: 46 - 52), compared to 60 years (IQR: 51 - 66) (p = 0.71) and 48 (IQR: 45 - 54) (p = 0.63) in the control group. NIV duration before ECMO or invasive ventilation initiation was 5 days (IQR: 2 - 8) and 3 days (IQR: 1 - 5), respectively (p = 0.13). Drainage multistage femoral cannula 25 F and internal jugular infusion cannula 21 F were placed percutaneously. After cannulation, the patients received light sedation that permitted communication, active physiotherapy and oral feeding. None of the patients in the ECMO group died within 28 days after ECMO initiation (Fig. 1, Panel A) or received invasive ventilation. VV-ECMO was not associated with longer mechanical support than invasive ventilation (HR: 1.26 95%CI: 0.24 - 6.55, p = 0.77) (Fig. 1, Panel B). Conclusions: VV-ECMO can be a not inferior strategy to invasive ventilation for treating patients with COVID-19 ARDS and severe hypoxemia not responding to long trials of NIV. (Table Presented).