NEW ORLEANS – Microaxial flow pump circulatory support with the Impella device during percutaneous coronary intervention (PCI) did not help outcomes in complex patients who were not in cardiogenic shock, two randomized trials found. American College of Cardiology (ACC) annual meeting was shown.
In an alternative and largely stable but high-risk population in the CHIP-BCIS3 trial, the Bayesian win ratio favored PCI without planned circulatory support (0.85, 95% CI 0.63-1.15), compared with 43.0% win in a pairwise hierarchical comparison in that group, compared with 36.6% in patients randomized to pump support, reported Divaka Perera, MD, of King’s College London.
At the top of the hierarchy, the victory over all-cause mortality was 7 percentage points higher in the protected PCI group (HR 1.54, 95% CI 0.99–2.41). P=0.054), induced by a significantly higher risk of death from cardiovascular (CV) causes (HR 1.91, 95% CI 1.11–3.30, P=0.018).
In the STEMI Door to Unload (DTU) trial, delayed reperfusion with Impella for 30 minutes before PCI for ST-segment elevation myocardial infarction had no effect on infarct size in an intention-to-treat analysis (30.8% vs 31.9% among controls). P=0.50) or per protocol analysis (31.8% vs. 33.7%, P=0.28), reported Greg W. Stone, MD, of the Icahn School of Medicine at Mount Sinai in New York City.
Neither had any effect on clinical outcomes in secondary endpoints, with a OR of 1.04 (95% CI 0.84–1.28). P=0.73) compared with urgent coronary angiography plus PCI without temporary circulatory support device, although all-cause and CV mortality at 30 days were in favor of the intervention in this trial (both P=0.08).
The CHIP findings were published simultaneously New England Journal of Medicine (NEJM); Posted in STEMI DTU Journal of the American College of Cardiology.
chip implications
“Over the past decade, we have increasingly adopted mechanical circulatory support for high-risk PCI, particularly microaxial flow pumps, largely driven by anatomic reasoning and observational experience,” said ACC study discussant Roxana Mehran, MD, of the Mount Sinai School of Medicine in New York City. With prior inconclusive randomized evidence, “existing guidelines – American and European – remain cautious with class 2b recommendations, reflecting an area where clinical enthusiasm has clearly outstripped high-quality evidence.”
Benny Levenson, MD, PhD, of the Freie Universität Berlin and ACC press conference discussant for CHIP, said that given the lack of benefit from extracorporeal life support in the treatment of myocardial infarction-related cardiogenic shock in the ECLS-SHOCK randomized trial, the results were not really a surprise.
With these “historical” data from CHIP on elective unloading, “routine use of this strategy cannot be recommended. But perhaps the most important insight is not that unloading does not work, but rather that timing and patient selection are important,” Mehran concluded.
Panelist Ajay Kirtney, MD, of Columbia University Irving Medical Center/NewYork-Presbyterian Hospital in New York City, agreed: “There are cases where you can’t make the case without a hemodynamic support device, but the decision to put that device in requires careful consideration of the risks, as (the CHIP trial) has shown, and the potential benefits that are currently somewhat unproven. So I think we should ‘protect. Gotta move away from ‘PCI’ and bill it as PCI with hemodynamic support.”
“The use of more subtle assessment of direct hemodynamic measures – such as elevated left-sided filling pressures and reduced cardiac output – may provide better selection criteria but further evaluation is needed,” said Brahmaji Nallamothu, MD, MPH, and Brett Wanamaker, MD, of the University of Michigan in Ann Arbor. NEJM editorial.
Mehran cautioned against overly broad interpretation, noting the clear benefit of microaxial flow pump support in the Danger Shock trial in active myocardial infarction-related shock.
STEMI DTU implications
Efforts to reduce reperfusion injury have repeatedly failed, but this new failure in STEMI DTU “doesn’t close the door,” said study discussant Wayne Batchelor, MD, MHS, MBA, of the Inova Heart and Vascular Institute in Fairfax, Virginia.
“There are some other questions that need to be asked in terms of what if we are able to miniaturize these devices, eliminate or reduce vascular complications and bleeding, would that indicate efficacy – if it was deployed at the right time and the delay was the right amount – would it be successful?” He said.
“However, this tells us that it is business as usual for the cath lab right now,” Batchelor said. “Until further research is conducted, we have no justification for deploying Impulse in non-cardiogenic shock ST elevation MI.”
The key secondary safety endpoint of major device-related bleeding and vascular complications came in at 30.8%, which was above the pre-specified performance target of 26.5% that was established in collaboration with the FDA.
Stone noted that there was a lot of bleeding and hematomas, which he said had a very weak association with long-term mortality in the Impella group, suggesting that complications were much more influential in the control group.
In comparison, Pereira said that in CHIP, bleeding and vascular complications were numerically, but not statistically, more common in the microaxial flow pump group, “but the rates were actually quite low and lower than in the registries, and I don’t think it accounted for the difference in cardiovascular mortality.” He said that the really high mortality rate requires more analysis and deeper investigation.
The CHIP protocol included several measures to ensure low risk, including pre-procedure CT planning in more than two-thirds of patients on both arms, ultrasound guided micropuncture used in all patients, and percutaneous access closure with a dedicated device in the majority of these patients.
audit Report
CHIP included 300 patients from 29 UK centers who had complex coronary disease and a left ventricular ejection fraction of 35% or less (median 27%). Participants were randomized to elective unloading or standard care (mechanical circulatory support only for emergency bailout) with microaxial flow pumps in complex PCI employed during the procedure. Pereira noted that the mean BCIS-threat score was 12 on a 12-point scale, which included very high calcification, left main, and chronic total occlusion.
The STEMI DTU included 527 patients aged 18 to 85 years with no prior MI who presented with acute anterior STEMI to 55 hospitals in the US, Canada, and Europe within 1–6 hours of symptom onset. Participants were randomized to open-label treatment with Impella insertion followed by a 30-minute delay before immediate PCI versus PCI alone. The Impella device was left in for 4–24 hours at the physician’s discretion, with an average of 10.4 additional hours. This strategy resulted in an average total ischemic time of approximately 47 minutes longer than the usual care group.
Impella’s longer-than-expected length of stay with anticoagulation (mean 14 hours) and suboptimal closure techniques may have contributed to bleeding complications, Stone said. One limitation was that the STEMI DTU was not powered for clinical outcomes. Additionally, the higher than expected systolic blood pressure of approximately 143 mm Hg may have impacted efficacy, which future studies may address by combining Impella with blood pressure lowering and other adjuvant therapies to optimize loading conditions before PCI.
Mehran noted Security IV Results from a trial of Impella-assisted PCI in high-risk, complex coronary artery disease patients with reduced left ventricular function are expected within the next year.
“While (CHIP) provides clarity for contemporary stabilization of high-risk PCI, moving us away from routine use, PROTECT IV will help us determine whether a different trial design, larger population, or longer follow-up could (show) a subgroup who may benefit from it,” she said.
