STEMI-DTU

In patients with anterior STEMI without cardiogenic shock, mechanical LV unloading with Impella CP followed by a 30-minute delay before PCI did not reduce infarct size compared with immediate PCI alone and was associated with substantially more bleeding and vascular complications.

Study design

  • Prospective, multicenter, randomized controlled trial
  • 55 hospitals in the United States, Canada, United Kingdom, Germany, Italy, and Switzerland
  • 1:1 randomization
  • N = 527
  • Follow-up: 12 months
  • Primary endpoint assessed by cardiac MRI 3-5 days after PCI

Population

  • Anterior STEMI without cardiogenic shock
  • No prior MI
  • Mean age 61 years
  • 79% male
  • LAD culprit lesion in >95% of patients
  • Excluded cardiogenic shock, unwitnessed cardiac arrest, prolonged CPR, prior HF, severe aortic valve disease, prior CABG, and fibrinolysis for the index event

Interventions

  • LV unloading + delayed PCI: 262
  • Immediate PCI alone: 265
  • LV unloading performed with Impella CP via femoral access
  • Intervention arm required 30 minutes of LV unloading before PCI
  • Median mechanical support duration: 10.4 hours
  • Total ischemic time was longer in the unloading arm

Primary outcome

  • Infarct size normalized to LV mass by cardiac MRI at 3-5 days after PCI
  • 30.8% with LV unloading + delayed PCI vs 31.9% with immediate PCI alone
  • No significant reduction in infarct size
  • Primary endpoint not met
LV unloading + delayed PCI vs PCI alone
Primary endpoint: infarct size normalized to LV mass
40% 30% 20% 10% 0%
30.8%
LV unloading + delayed PCI
31.9%
PCI alone
30.8% vs 31.9% | Not significant
Primary endpoint: infarct size normalized to LV mass on cardiac MRI 3-5 days after PCI.

Secondary outcomes

  • No significant difference in key secondary efficacy outcomes
  • Composite secondary efficacy endpoint included death within 1 year, cardiogenic shock, HF, heart transplant, and extent of myocardial injury
  • No significant difference in HF-related events or other major clinical outcomes
  • Results were consistent in intention-to-treat and per-protocol analyses

Safety

  • Major bleeding or vascular complications were substantially higher with LV unloading
  • BARC 3-5 bleeding or vascular complications: 34% vs 6%
  • Treatment-related major bleeding: 30.4%
  • Most bleeding was access-site related
  • Major vascular complications occurred in 4.2%
  • Device-related bleeding or vascular complications exceeded the prespecified performance goal

Interpretation

  • STEMI-DTU does not support routine Impella CP unloading before PCI in anterior STEMI without shock
  • The strategy delayed reperfusion but did not meaningfully reduce infarct size
  • Bleeding and vascular complications were the major downside
  • Findings should not be extrapolated to cardiogenic shock, where Impella is used for hemodynamic support
  • The trial reinforces immediate PCI as standard care for anterior STEMI without shock
Kapur, N. K., Mangner, N., Aghili, N., et al. (2026). Left ventricular unloading in anterior ST-segment elevation myocardial infarction without shock: The STEMI Door-to-Unload randomized controlled trial. Journal of the American College of Cardiology. https://doi.org/10.1016/j.jacc.2026.03.071