Experts Hash Out Recommendations for Impella CP After DanGer Shock

BOSTON, MA—While the medical community awaits updated guidelines on the use of mechanical circulatory support (MCS) following the DanGer Shock trial, experts here at THT 2025 debated how strong any future recommendations should be and, by extension, what the effects might be across a broad swath of hospitals.

Presented last year at the American College of Cardiology (ACC) 2024 Scientific Session and published simultaneously in the New England Journal of Medicine, the trial showed that routine use of the Impella CP microaxial flow pump (Abiomed) lowered 6-month mortality in patients with STEMI complicated by cardiogenic shock compared with standard care. There was a 12.7% absolute difference, with a number needed to treat (NNT) of 8.

During a debate featuring two members of the Cardiogenic Shock Working Group, Manreet Kanwar, MD (University of Chicago, IL), and Navin Kapur, MD (Tufts University School of Medicine, Boston, MA), and a panel discussion that followed, physicians wrestled with whether Impella CP should receive the strongest Class I recommendation (ie, should be performed) or a Class IIa recommendation (ie, it is reasonable to perform).

Kanwar kicked things off, arguing that the DanGer Shock results warrant the weaker recommendation in future guidelines. While the trial was a “major win for the field of cardiogenic shock, the first positive trial,” Kanwar said, she was concerned it “could lead to some excessive extrapolation.”

DanGer Shock took more than a decade to complete, Kanwar pointed out, with the first patient enrolled in Denmark in January 2013. The trial was expanded to Germany in 2019 and then the United Kingdom in 2021 to get it done. When broken down by country, the trial results were significantly positive among the Danish cohort, but not among the patients later enrolled in Germany and the UK. It’s unclear what explains this difference, Kanwar said, but noted that more patients in the Impella CP group versus the control group received a durable LV assist device.

“So was it really a survival [benefit] of Impella CP or better, more-aggressive resource and patient management?” she asked.

Safety should be considered, too, she suggested, citing higher rates of the composite safety endpoint, bleeding, limb ischemia, kidney failure, stroke, and sepsis in the Impella CP arm of the trial. Survival bias could be at play, she acknowledged, but said these events will lead to longer lengths of stay in the hospital.

There are other issues around generalizability that need to be taken into account as well, Kanwar said. Those include the underrepresentation of women in DanGer Shock and indications that the device did not carry the same benefit in that subgroup, as well as data showing that only about a third of STEMI patients with shock today would meet inclusion criteria for the trial. Moreover, other analyses have shown that in the control arm of the trial, 30% of patients were stabilized and discharged out of the ICU within 72 hours—nearly double the proportion in the Impella CP arm—and this group had favorable outcomes.

“You have to ask yourself, is Impella CP generalizable to all STEMI shock patients or not?” Kanwar said.

“The bottom line is as much as we are excited that we have a positive trial, the study itself barely meets the significance,” Kanwar said, noting that even though the NNT was 8, the number needed to harm was 6. And the fragility score, indicative of what it would take to bring the trial from significance to nonsignificance, was 3. “If you were to really tease out the actual allocation of which device they got and follow these patients through without any crossover, the significance tends to go away,” she said.

Addressing the strength of any future recommendation, Kanwar said, “Never say never [and] never say always in medicine. And yet we are saying that every patient with STEMI shock should get an Impella [if it’s given a Class I recommendation] . . . or you will be judged for that. I ask you, is this data enough to lead to these kind of recommendations?”

A Class I Recommendation Is Warranted

Kapur, who was recently hired to be the global chief medical and scientific officer at Johnson & Johnson MedTech, which acquired Abiomed a few years ago, then made his case that a stronger Class I recommendation is justified on the basis of DanGer Shock.

Referring to his new position with industry, that’s an indicator “that I actually completely believe where we’re going with the transvalvular pump space to the point that I decided I couldn’t stay in academics for another minute. I had to go in and help the team get this done,” Kapur told THT attendees.

Kapur underscored the “compelling” survival benefit observed with Impella CP in DanGer Shock, saying the absolute difference through 6 months is “something we simply can’t ignore.” He added that the patients included in the trial represented a SCAI SHOCK C and D population. “We’re avoiding the extreme cases and we’re avoiding . . . the patients who aren’t sick enough,” he said.

When it comes to weighing the risks and the benefits, Kapur said “the preponderance of evidence would weigh towards, ‘I will survive,’” adding that the NNT to derive a survival benefit was just 5 in patients younger than 77 years old. On the risk side, there were “some very loose definitions” of events like acute kidney injury, limb ischemia, and bleeding in DanGer Shock, which was investigator-initiated, while the risk of moderate or severe bleeding in the trial was comparable to that seen in prior shock trials. “With or without MCS, cardiogenic shock is a dangerous situation to be in,” he commented.

Kapur indicated, too, that some of the safety findings could be influenced by the protocols and algorithms used at various centers.

In arguing for a Class I recommendation, he pointed out that the SHOCK trial published in 1999 missed its primary endpoint (30-day mortality) but still supported the highest level of recommendation for early revascularization in patients with cardiogenic shock based on a significant difference in a secondary endpoint (180-day mortality), with an NNT of 8. CULPRIT-SHOCK, with an NNT of 12, also resulted in a Class I recommendation.

And now, DanGer Shock comes in as the only MCS trial with a positive primary endpoint, Kapur said. “The devil’s in the details in terms of how you interpret this trial in your clinical practice when that STEMI patient comes in with anterior MI, hypotensive, and critically ill.”

He added that the trial would support a level of evidence of A or B-R considering high-quality evidence from an RCT, a meta-analysis of RCTs, and several supportive registry studies.

Kapur reflected on how management of acute MI complicated by cardiogenic shock has evolved, saying it started with early revascularization based on the SHOCK results. “Now, we’re at the first phase, after 25 years, to achieve the next step in improving survival, which is early LV unloading,” he said.

Impella Out in the Community

During a panel discussion following Manwar’s and Kapur’s talks, several physicians said a consideration around how strongly use of Impella CP should be recommended is how well most centers will be able to handle patients after it’s deployed.

Joyce Wald, DO (Hospital of the University of Pennsylvania, Philadelphia), medical director of the shock team at her center, said, “in the community, we don’t have a lot of tools for these really sick patients, so making it a [Class I recommendation] might encourage the community more to reach for this device sooner, which I think should be what we should be driving towards.”

Importantly, Wald said, “I think we should be putting in protocols for post CP placement and when to escalate. So I think the outcomes will be even better if we do that.”

Kapur agreed. “What needs to be really implemented is, as you said, a significant educational program around cardiogenic shock management,” he suggested. “What do you do on the other side of that [revascularization] once you’ve got the patient unloaded?”

Indeed, said Ajay Kirtane, MD (NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY), “the aftermath . . . is not small.”

He added: “I think all of us would agree that with the right protocol, the right device, the right patient, you’re going to get great outcomes, and that should be with a high degree of recommendation. But the way this is practiced in the community needs a major educational lift to be able to regularize the good care that we’d like to provide our patients.”

That could be a challenge, according to David Cohen, MD (St. Francis Hospital & Heart Center, Roslyn, NY, and Cardiovascular Research Foundation, New York, NY). “The next step here, regardless of whatever recommendation it gets in the guidelines, is working out these transfer protocols,” said Cohen, “because there is no way in heck that these small community hospitals without 24/7 intensivists and all of those bells and whistles are going to be able to manage these patients correctly and get the results that we want to achieve based on what we saw in DanGer Shock.”

Kapur likened some of these concerns to pushback that occurred in the early days of bringing STEMI care into community hospitals and pointed toward the efforts of the Cardiovascular Patient Outcomes Research Team (CPORT) as showing that it was possible. “CPORT is a great example of education in the community,” he said. “Getting these patients to the right center doesn’t necessarily mean they have to be in a center that has all the bells and whistles. They need to be supported, especially at the time of revasc for shock.”

About Impella CP, Cohen said, “I think we can implement it in the community, but . . . I don’t see any way to deliver the kind of care that was delivered in DanGer Shock, or in the National Cardiogenic Shock Initiative centers, in these smaller community hospitals. I’ve worked in them. There is nobody there at night. And so you just have to be really careful about figuring out what we’re going to do with those patients afterwards.” He advocated for the establishment and testing of transfer protocols to address this issue.

Alexandra Lansky, MD (Yale School of Medicine, New Haven, CT), indicated that this moment represents a starting point when it comes to implementation of Impella CP in broader practice, likening it to when door-to-balloon times were introduced for PCI and most patients were not getting treated within the recommended window. “So I think this is going to push the needle. It’s going to set standards in terms of treatment,” she said.