Ouweneel 2017.

Study characteristics
Methods	Study design: RCT Study grouping: parallel group Total duration of study: 2 years 4 months Number of study centres and locations: 2 centres, Academic Medical Center, Amsterdam, the Netherlands and Haukeland University Hospital, Bergen, Norway Study setting: tertiary referral centre Date of the study: 24 May 2012 to 15 September 2015
Participants	Baseline characteristics MCS Age (mean): 58 (SD 9) years Gender: 18 men; 6 women Lung function: N/A Smoking history: N/A Lactate prior to initiation (mean): 7.5 (SD 3.2) mmol/L Cause of cariogenic shock: STEMI IABP Age (mean): 58 (SD 11) years Gender: 20 men; 4 women Lung function: N/A Smoking history: N/A Lactate prior to initiation (mean): 8.9 (SD 6.6) mmol/L Cause of cariogenic shock: STEMI Inclusion criteria: AMI with ST segment elevation complicated by severe CS in the setting of immediate PCI (severe CS defined as systolic blood pressure < 90 mmHg for > 30 minutes or need for inotrope or vasopressors to maintain systolic blood pressure > 90 mmHg). To select a patient population in even worse condition, patients only qualified if they were mechanically ventilated before randomisation. Exclusion criteria: severe aorto‐iliac arterial disease with impeding placement of IABP or MCS; known severe cardiac aortic valvular disease; serious concomitant disease with a life expectancy < 1 year; participation in this study or any other trial within the previous 30 days; coronary artery bypass grafting within preceding week. Pretreatment: baseline characteristics of groups were well balanced. Number of participants randomised to intervention: 48 Number of participants lost to follow‐up: 0 Number of participants analysed: 48
Interventions	MCS vs IABP Intervention characteristics MCS Type of MCS: percutaneous Impella CP (maximum output 3.7 L/minute) Duration from diagnosis to intervention: N/A Duration of treatment: N/A Concomitant medications: 100% received inotropes/catecholamines Concomitant procedures: all participants underwent primary percutaneous intervention IABP Duration from diagnosis to intervention: N/A Duration of treatment: N/A Concomitant medications: 100% received inotropes/catecholamines Concomitant procedures: all participants underwent primary percutaneous intervention
Outcomes	Survival 30‐day survival: MCS 54%; IABP 50% 1‐year survival: not available Survival measured to: transplant: unsupported cardiac function Not described Quality of life Not described Major adverse cardiovascular events: all clinical outcomes at 6 months Cerebrovascular accident: 1 MCS; 1 IABP Major vascular complication: 1 MCS; 0 IABP Myocardial reinfarction: 1 MCS; 2 IABP Dialysis‐dependent MCS: 8/24 (33%); IABP: 7/24 (29%) Duration (median: MCS: 17 (25th to 75th percentile 5–29) days; IABP: 7 (25th to 75th percentile 2–9) days Length of hospital stay Median: MCS: 16 (25th to 75th percentile 3–26) days; IABP: 10 (25th to 75th percentile 6–24) days Length of ICU stay Median: MCS: 7 (25th to 75th percentile 3–16) days; IABP: 7 (25th to 75th percentile 4–10) days Major adverse event Major bleeding: MCS: 8 (33%); IABP 2 (8%) Haemolysis requiring extraction of the device: MCS: 2 (8%); IABP: 0 (0%) Cause of death Primary cause of death in both groups was brain damage (MCS: 41% of deceased; IABP: 50% of deceased). Death due to refractory CS occurred in 29% of deceased participants (MCS: 4/12; IABP: 3/12) Additional outcome notes Of the 24 MCS participants, 1 subsequently received Impella 5.0; 1 was already received IAVP support before randomisation (inserted before start of the primary PCI) and was subsequently randomised after PCI to MCS (protocol violation); 1 did not receive MCS as the participant showed signs of recovery after randomisation prior to device therapy. Of the 24 IABP participants, 1 subsequently received MCS and was transferred to another hospital for treatment with extracorporeal life support oxygenation; 2 participants received an alternative device, the Impella 5.0 after the IABP and 1 of those received subsequent extracorporeal life support and an LVAD at another hospital.
Identification	Sponsorship source: Academic Medical Center University of Amsterdam Country: Netherlands Setting: hospital Comments: Authors name: Dr José PS Henriques Institution: Department of Intensive Care Medicine, Academic Medical Center University of Amsterdam Email: j.p.henriques@amc.uva.nl Address: Academic Medical Center, University of Amsterdam, AMC Heart Center, Meiberg‐dreef 9, 1105 AZ Amsterdam, the Netherlands Year: 2017
Notes	Randomisation and placement of MCS or IABP took place after revascularisation except for 8 participants in whom MCS or IABP was initiated prior to revascularisation (MCS: 5; IABP: 3).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Allocation selection using a randomised 1:1 internet application.
Allocation concealment (selection bias)	Low risk	Allocation sequence adequately concealed using a randomised 1:1 Internet application.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	No description of methods of blinding for participants and personnel.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	An independent clinical event committee adjudicated the events. Imaging parameters were assessed by independent local core laboratories that were blinded to the other trial data and randomisation outcome.
Incomplete outcome data (attrition bias) All outcomes	Low risk	All outcome data were complete.
Selective reporting (reporting bias)	Low risk	All outcome data were reported.
Other bias	Low risk	Inherent risk of bias was minimised as much as possible.