Should Impella™ be used for MIs without shock?
Hey Chip, where's the door to unload this pallet of pumps?
Microaxial flow pumps (Impella™ being the only available product) were shown in the Danger-Shock trial (NEJM 2024) to reduce mortality in patients with cardiogenic shock due to ST-elevation myocardial infarction. The survival benefit came with the cost of significant risks for limb ischemia, bleeding, and acute kidney injury.
Almost as soon as Danger-Shock was published, pressure grew for the expansion of Impella beyond its evidence-based use case for STEMI-related cardiogenic shock.
Cardiologists commonly face gray-zone situations in which patients deteriorate during or after PCI, without frank shock. Impella was FDA-approved for mechanical circulatory support during procedures way back in 2012. After Danger-Shock, interventional cardiologists used Impellas much more frequently during high-risk procedures (and maybe some not-so-high-risk ones).
Reimbursements in the tens of thousands of dollars per Impella (to the hospital) incentivized the expansion. Cardiologists don’t receive big per-device payments for placing Impellas, but they gain massive leverage in salary negotiations with their centers based on the revenue they bring in.
Abiomed, acquired by J&J’s Medtech division in 2024 for $16B, surfed this wave well. Under the slogan “Impella Protected PCI,” the device was marketed as a safety hedge to cardiologists traumatized by patients dying on their cath lab tables, and sales grew dramatically. The PROTECT I, II, and III trials were cited to support the practice (but did so weakly, if at all).
However, to really secure the expansion into this market, i.e., patient population, higher-quality randomized trial data would be needed to establish the benefit of Impella in MI without shock.
In a refreshing (or depressing, or both) abandonment of any pretense, J&J MedTech’s own chief medical science officer served as a lead investigator of a randomized trial (‘Door to Unload’) that, if successful, could expand Impella’s FDA indication (or just clinician willingness) to MIs without shock.
Contemporaneously, the U.K. funded its own smaller trial (‘Chip’) testing Impellas for MI without shock—presumably as a due diligence counterweight against the massive expenditure that would be required of the NHS if pressure rose to include Impellas as standard care in this much larger patient population.
Gray Zone Sidebar: The Rationale for Impellas in MI Without Shock … “Unloading the LV”
Microaxial flow pumps straddle the aortic valve, whisking blood from the left ventricle to the aorta via a tiny rotor spinning at tens of thousands of times per minute.
In MIs with shock, the rationale for left ventricular assist pumps is straightforward. A patient is cold and wet due to low cardiac output —> pump more blood forward.
In MIs without shock, a new physiologic rationale is proposed: “unloading” the left ventricle to prevent the reperfusion injury that often occurs after stents are placed.
The left ventricle in the hyperacute post-MI phase is often distended and strained. Restoring blood flow to the ischemically stressed LV with percutaneous coronary intervention (stents) can paradoxically injure viable myocardium and worsen MI. Reperfusion injury post-PCI results from oxidative damage, inflammation, calcium overload, a sudden increase in contractility, and other postulated mechanisms.
Reperfusion injury risk has been inherent to PCI since its inception. “Unloading” the LV with an Impella for 30 minutes before performing PCI, it was hoped, might mitigate reperfusion injury and improve outcomes.
The ‘Door to Unload’ and ‘Chip’ Trials
In the “Door to Unload” trial, 527 patients with acute anterior STEMI were randomized to Impellas for 30 minutes before PCI vs. immediate PCI alone.
(Yes, you read that correctly: in patients with STEMI, PCI was deferred to place an Impella and run it for 30 minutes before revascularizing. Patients remained ischemic for about 47 minutes on average.)
In the Chip trial, Impellas were placed immediately before PCI in the intervention arm. Three hundred patients with STEMI and severe LV dysfunction (<35%) were randomized.
Neither trial showed a benefit in the Impella arm over usual care:
In Chip, there was a concerning trend toward harm (hazard ratio for death 1.54 in the Impella arm).
In DTU, Impellas did not improve infarct size (the primary outcome) or secondary endpoints (win ratio 1.04). All-cause mortality and cardiovascular mortality at 30 days did trend toward a benefit from Impellas (P=0.08).
A Takeaway
The Chip trial argues strongly against routine use of Impellas for patients with STEMI without shock.
Door-to-Unload’s trend toward a mortality benefit left the door cracked open for use cases for Impella in this population, which aligns with interventionalists’ strong anecdotal experience of Impella’s benefits in selected patients.
Identifying who will benefit, though, remains as difficult as ever.
An editorialist noted:
[FDA] approval and the commercial framing of these devices under the “Protected PCI” designation has reinforced a perception of procedural safety for the operator…[but] the criteria for the selection of appropriate patients remain poorly defined despite years of use.
These [trial] findings force the interventional cardiologist to occupy an uncomfortable position between two unsatisfying options — the potential burden of perceived undertreatment when a patient’s condition deteriorates or recognition that the act of doing more may itself constitute harm. Any honest appraisal must begin with that tension.
This dilemma pervades medicine, but lands hardest on proceduralists. Even if financial considerations were eliminated, this tension would remain.
References
Perera D, Ryan M, Ezad SM, et al. Left Ventricular Unloading in High-Risk Percutaneous Coronary Intervention. New England Journal of Medicine. Published online March 29, 2026. doi:https://doi.org/10.1056/nejmoa2515704
Kapur NK, Mangner N, Aghili N, et al. Left Ventricular Unloading in Anterior STEMI without Shock: The STEMI Door to Unload (DTU) Randomized Controlled Trial. JACC. Published online March 2026. doi:https://doi.org/10.1016/j.jacc.2026.03.071