Routine use of the Impella CP microaxial flow pump did not improve outcomes and may have contributed to deaths among patients undergoing high-risk percutaneous coronary intervention (PCI), according to new study results.

The use of a microaxial flow pump in complex PCI procedures, called protected PCI, has become common based on a belief that it would be safer for the patient. However, doctors should approach its use with caution, suggests the study, presented during a late-breaking session at American College of Cardiology (ACC) Scientific Session 2026 and simultaneously published in The New England Journal of Medicine.

“They should stop using it routinely,” Divaka Perera, MD, an interventional cardiologist at Guy’s and St Thomas Foundation Trust and lead author on the study, told Medscape Medical News. “Protected PCI has grown in the last 20 years…and come to be seen as a good way of improving patient outcomes. But while the term suggests that there’s a lot of evidence behind it, actually, there’s no randomized evidence at all.”

No Benefit for Protected PCI

In the CHIP-BCIS3 trial, Perera’s group assessed the use of Impella CP (Johnson & Johnson MedTech).

Researchers randomized 300 patients to elective use of the Impella pump or standard care without a pump during a planned complex PCI procedure. To be eligible for the procedure, patients must have had extensive coronary disease, severe left ventricular systolic dysfunction, and be considered a complex PCI case by several criteria.

Perera’s team analyzed outcomes after a median 22 months of follow-up by comparing each patient in the pump group to every patient in the standard-care group for a hierarchical composite outcome including time to death, time to first disabling stroke, time to first spontaneous myocardial infarction, number of cardiovascular hospitalizations, and occurrence of periprocedural myocardial injury, in that order. Using a simulation method called the win ratio for each matched pair, the investigational treatment was deemed a “winner” or a “loser.” In the event of a tie at any tier in the hierarchy, the model attempted to resolve the tie at the subsequent tier. The win ratio was defined as the total wins in the Impella group divided by the total wins in the standard-care group.

Out of more than 22,000 comparisons, 36.6% of pairings favored the microaxial flow pump and 43.0% favored standard care (win ratio, 0.85; 95% CI, 0.63-1.15).

Researchers also looked at death from any cause and death from cardiovascular causes. In the Impella group, 47 of 148 patients died from any cause and 36 died from cardiovascular causes. Among 152 patients who received standard care, 33 died from any cause and 20 died from cardiovascular causes.

Perera noted that the hazard ratio for death from any cause just missed statistical significance at 1.54 (P = .054). However, for cardiovascular death, it was 1.91 with a P value of.018.

Researchers also found that 16 out of 148 patients in the Impella group had major bleeding before discharge from the hospital compared with 11 out of 151 patients in the standard care group (risk ratio, 1.48; 95% CI, 0.71 to 3.09). Minor vascular complications occurred in 23 of the Impella patients compared with 15 of the standard care patients. Two Impella patients and 1 standard care patient had major vascular complications.

“Remember, the term was ‘protected PCI,’ and we’re now saying, actually, you’re more likely to do harm than good,” Perera said. “That is an incredibly disruptive result, and the guideline committees now have the data with which to issue some clear guidance.”

Further Studies Ongoing

Some differences in outcomes, Perera said, may be attributable to operators’ approach to revascularization in each group. One touted benefit of the Impella device is the ability to complete revascularization during the initial procedure, whereas patients receiving standard PCI without a pump typically undergo revascularization in stages. However, the study showed patients in the Impella group were more likely to experience myocardial injury during the procedure.

“The [staged] strategy looks like it’s a safer way to go,” Perera said.

He and his team are continuing to follow the study patients and plan to present results from mechanistic substudies at upcoming medical meetings.

The Impella’s manufacturer, Johnson & Johnson MedTech, is hoping for further clarity from the ongoing PROTECT IV clinical trial sponsored by its subsidiary Abiomed. “We remain committed to studying this vulnerable population using targeted therapies, shared decision-making, and support strategies that improve patient outcomes,” a Johnson & Johnson MedTech spokesperson said. In the PROTECT IV trial, the Impella CP is compared with an intra-aortic balloon pump in 1252 patients with left ventricular systolic dysfunction who qualify for complex PCI for a composite endpoint of all-cause death, stroke, myocardial infarction, unplanned revascularization, durable left ventricular assist device implant or heart transplant, or other hospitalization.

High-Risk Patients, Lengthy Procedures

The CHIP-BCIS3 results have far-reaching implications for treating patients in the high-risk category for PCI, who are typically older and have multiple comorbidities.

“The good news is we’re living longer, but sometimes [patients] show up in their seventies or eighties with lots of coronary disease, and they may not be great surgical candidates for revascularization,” said Manesh Patel, MD, president-elect of the American Heart Association and chief of the Division of Cardiology at Duke University School of Medicine, Durham, North Carolina. “We’re left thinking about the best ways to improve their quality of life.”

Patients with low cardiac output and extensive blockages often require lengthy PCI procedures. A solution such as a microaxial flow pump offers peace of mind for physicians worried about getting the patient through the procedure, said Patel, who was not involved in the study.

Roxana Mehran, MD, a professor of medicine and director of Interventional Cardiovascular Research and Clinical Trials at the Icahn School of Medicine at Mount Sinai, New York City, noted that the patients in the study were stable, and that the researchers used “really meticulous” vascular access evaluation.

In this patient population, “We need to wait for PROTECT IV, and we should stop calling these percutaneous coronary interventions ‘protected’ because we have never seen protection yet” said Mehran, who is also president of the American College of Cardiology.

In an editorial published alongside the study in NEJM, interventional cardiologists at University of Michigan, Ann Arbor, Michigan, Brahmajee Nallamothu, MD, and Brett Wanamaker, MD, wrote that the investigators provided “sufficient data to encourage a more selective approach to the use of mechanical circulatory support in high-risk PCI, particularly in the absence of clear hemodynamic instability.”

Wanamaker said some patients appear to benefit from the use of the Impella device during their PCI procedures, but that it’s difficult to distinguish those cases clinically.

“Microaxial flow pumps are not a panacea, and we need to be thoughtful and selective about which patients might benefit,” Wanamaker told Medscape Medical News. “We still have a difficult time predicting who might decompensate during a high-risk coronary intervention. Ultimately, it may require a more nuanced approach incorporating the patient’s baseline hemodynamics.”

Rapid Adoption and Approval

The CHIP-BCIS3 results raise questions among some cardiologists about whether the Impella’s approval and widespread adoption was premature. The FDA granted premarket approval in 2015 for high-risk PCI and expanded its indication in 2018. The agency based its decision to approve the Impella in high-risk PCI on the PROTECT I and PROTECT II trials.

In the single-arm PROTECT I trial, researchers evaluated the Impella 2.5 device during PCI in patients with poor left ventricular function. That trial concluded the device was safe but didn’t compare it with any other treatment. The PROTECT II trial compared the Impella 2.5 to an aortic balloon pump during PCI in 452 patients but not to standard PCI without a support device. The trial was terminated for futility.

“There was no arm in which there was no mechanical support,’ Perera said, noting that the FDA relied on substudy and subgroup analyses. “It wasn’t based on randomized evidence that the FDA approved this.”

“The larger lesson of CHIP-BCIS3 is one of humility,” John Mandrola, a cardiac electrophysiologist at Baptist Health in Louisville, Kentucky, wrote in a commentary for Medscape Medical News. “Everything about the approval and rapid growth of this device is shocking. The FDA approved the highly invasive device with no outcome data and then expanded indications with very little trial data.”

Perera reported having no relevant financial relationships. Wanamaker reported receiving consulting fees from Shockwave Medical, Boston Scientific, and Asahi Intecc USA and attended events sponsored by Abiomed. Patel reported receiving grants or consulting fees from Bayer Healthcare, HeartFlow, Jansen, Johnson & Johnson, Novartis, and PCORI. Mehran reported receiving grants from Abbott Laboratories and Alleviant; having stock in ControlRad, Elixir Medical, and STEL; and receiving consulting fees from Cordis.

Catherine Shaffer is a freelance science and medical writer with a background in molecular biology and pharmaceutical research. Her work has appeared extensively in scientific trade and mainstream publications and on public radio.

Melissa Foster contributed to this report.