May 24, 2018
Extracorporeal Membrane Oxygenation for Severe Acute Respiratory Distress Syndrome.
What was the research question?
Is early initiation of Extracorporeal Membrane Oxygenation (ECMO) in patients with severe ARDS, associated with reduction of 60-day mortality compared to standard care?
How did they do it?
A prospective, multicenter, randomized, controlled trial, with a 1:1 ratio assignment 64 centers predominantly in France
Only research staff members were blinded but not participants or clinical staff.
Calculated sample size of 331 patients for 80% power to detect 20% difference in mortality and an alpha level of 5%.
Included ARDS patients who were mechanically ventilated for less than 7 days with PaO2:FIO2 ratio <50 mmHg for >3 hours; or PaO2:FIO2 <80 mmHg for >6 hours; or arterial blood pH <7.25 with a partial pressure of arterial carbon dioxide (PaCO2) >60 mmHg for >6 hours.
Patients were randomized to receive Venovenous ECMO and low-level mechanical ventilation (124 patients) or protective lung strategy and adjuvant therapies (neuromuscular blockade, proning, recruitment maneuvers, and bicarbonate) as needed (125 patients).
Cross over for SaO2 <80% for over 6 hours.
Primary outcome was 60-day mortality and secondary end point was treatment failure (defined as a crossover to ECMO or death in patients in the control group and as death in patients in the ECMO group).
The data safety monitoring board recommended stopping the trial at the 4th interim analysis (249 out of331) for futility.
What did they find?
60-day mortality (primary outcome) was not statistically different in ECMO compared to control (35% vs. 46% p=0.09).
Treatment failure was lower in the ECMO group compared to control group (35% vs. 58%, p<0.001).
A relatively high cross over to ECMO occurred in 28% of control patients with higher mortality (57% compared to 40% in the remaining patients in the control group).
More patients in the ECMO group with severe thrombocytopenia (27% vs. 16%; ARR 11 %; 95% CI, 0 to 21), bleeding events (46% vs. 28%; ARR 18%; 95% CI, 6 to 30) and less ischemic stroke (no patients vs. 5%; ARR -5%; 95% CI, −10 to −2).
Unblinded and Underpowered to detect difference as it was stopped early.
Presumed mortality in the control group of 60% was higher than the actual mortality and a difference of 20% reduction in ARR was overestimated. This affected the sample size and the power of the study. High risk for false negative results.
A high cross over rate reduced the effectiveness of ECMO treatment (those patients were started ECMO at later time and they are much sicker patients).
A high possibility of clinician equipoise as the cross over decision was at the discretion of unblinded treating clinicians.
Possible selection bias, as the majority of ARDS patients were male and had pneumonia and septic shock.
What does it mean?
Early initiation of ECMO did not improve 60-day mortality in severe ARDS patients compared to a strategy of conventional mechanical ventilation and ECMO as a rescue therapy.
However, the study does not provide conclusive support for the benefit of ECMO in severe ARDS as secondary outcome of treatment benefit is higher in the ECMO group.