Effect of Respiratory Support Type and Total Duration on Weaning From Venovenous Extracorporeal Membrane Oxygenation in COVID-19 Patients

Background: We evaluated the impact of noninvasive respiratory support (NRS) and invasive mechanical ventilation duration before venovenous extracorporeal membrane oxygenation (VV-ECMO) on weaning from venovenous ECMO and survival. Methods: In a retrospective single-center study, we studied subjects with COVID-19 ARDS treated with VV-ECMO. The subjects were divided and analyzed according to the cut-off of NRS, invasive ventilation, and total duration of respiratory support. Results: We identified a cut-off of NRS duration of 4 days, invasive ventilation duration of 5 days, and total respiratory support duration of 8 days. Weaning from VV-ECMO was observed in 63% (15/24) of subjects with NRS duration <= 4 days and in 16% (4/25) of subjects with NRS > 4 days (P = .001), in 50% (17/34) of subjects with invasive ventilation duration <= 5 days, in 13% (2/15) of subjects with invasive ventilation duration > 5 days (P = .02), in 68% (13/19) of subjects with total support duration < 8 days, and in 20% (6/30) of subjects with total support duration > 8 days (P = .001). The survival probability at 200 days demonstrated a statistically significant difference in NRS and total support duration comparison (P = .001 and P = .004, respectively). We did not find a statistically significant survival difference according to invasive ventilation duration (P = .13). Conclusions: In our population, the increase in NRS and total support days before ECMO could hamper weaning from VV-ECMO support. However, due to the pandemic, the small sample size, and the lack of precise data on ventilation settings, caution should be exercised in universalizing these results.