Dear Editor,

In the earliest phases of the pandemic, the use of extracorporeal life support in patients with severe coronavirus disease 2019 (COVID-19) was associated with disastrous outcomes. However, later series have shown better results, with hospital mortality ranging from 30 to 60% [1,2,3]. These series included patients receiving support during the first wave of the pandemic. Whilst a trend towards lower mortality in the overall COVID-19 population has been observed over time [4], preliminary data from the EuroECMO registry of the EuroELSO organization suggest that outcomes after extracorporeal membrane oxygenation (ECMO) in the second wave of the pandemic have worsened [5]. The present subanalysis examines this situation.

The ECMOVIBER (use of ECMO during the coVid-19 pandemic in the IBERian peninsula) retrospective observational cohort study included data from 24 ECMO centers, 22 in Spain and 2 in Portugal. We established a cut-off date of June 30, 2020 to define the first and second waves. For more information on the study, including the statistical methodology, see the online material. A total of 319 patients received extracorporeal respiratory support due to acute respiratory distress syndrome (ARDS): 151 (47.3%) during the first wave and 168 (52.6%) during the second. Hospital mortality was significantly higher during the second wave (60.1% vs. 41.1%, p = 0.001; Figure E3, online material). Patients supported during the second wave were older, had more comorbidities and were less likely to be treated at a high-volume center (Table 1). Time between admission to the intensive care unit (ICU) and ECMO start was longer, but not time since intubation. At ECMO indication, the PaO2/FiO2 was significantly higher and levels of COVID-19-associated inflammatory biomarkers were lower. Coinfection, together with new onset pneumonia during ECMO support, was more frequent in patients during the second wave (microbiological profile in table E3, online material).

Table 1 Patient pre-ECMO characteristics, ECMO management, complications and outcomes according to the wave of the pandemic in which the support was initiated
Full size table

Although the conclusions derived from an observational study should be treated with caution, these results could be interpreted as follows. The data suggest a certain relaxation of ECMO indication criteria during the second wave, due perhaps to the less demanding context and with the wider acceptance of the use of ECMO in COVID-19 in view of the positive results of early first analyses of large international databases (https://www.euroelso.net/covid-19/covid-19-survey/ and https://www.elso.org/Registry/FullCOVID19RegistryDashboard.aspx). Thus, in the second wave, low-volume centers treated more patients and this tendency for dispersion of ECMO cases may also have negatively affected the results [3]. Another possible influence on the survival difference is the change in the COVID-19 care protocol during the pandemic: for example, the use of corticosteroids and the criteria for intubation. Patients supported during the second wave suffered more coinfections, both at initiation and during extracorporeal support, and this multi-cause lung insult may have had a significant impact on the evolution of cases. Data suggest that in these patients intubation was delayed, and this is known to have potential deleterious effects in ventilated patients.

Our results confirm a higher mortality rate in COVID-19 patients supported with ECMO during the second wave than during the first. Here, we propose possible explanations for this phenomenon, which we feel should be considered in decisions regarding the technique's indication in future patients.