乌鸦传媒

This article is a collaboration between 乌鸦传媒 and:

PHILADELPHIA -- The Impella mechanical support device, frequently used in patients in cardiogenic shock who are undergoing percutaneous coronary interventions, was more expensive than intra-aortic balloon pumps and was also associated with greater mortality in a study reported here.

In a propensity score matching study, deaths occurred in 34.1% of the patients who were treated with the intra-aortic balloon pump and 45% of those treated using the Impella device, reported Sanket Dhruva, MD, of the University of California San Francisco.

In a presentation at the annual scientific sessions of the (AHA), Dhruva said the study also found major bleeding rates of 16% among patients being treated with the intra-aortic balloon pumps and 31.3% of the patients treated with Impella.

In addition, he said, the pattern persisted when the researchers considered if the mechanical support device was implanted before or during percutaneous cardiac interventions or after the intervention.

The study identified patients with acute myocardial infarction complicated by cardiogenic shock who were treated from October 2015 through 2017. From a population of approximately two million people, the researchers created a propensity score matching of about 1,680 pairs comparing outcomes of patients who were treated with Impella and intra-aortic balloon pumps.

Patients were about age 64, 71% were men, and 22% had a previous myocardial infarction history. About 61% had left main or proximal left anterior descending blockages.

"Impella was associated with significantly higher rates of in-hospital death and major bleeding compared to intra-aortic balloon pumps," Dhruva reported. "These data provide important insights into the performance of mechanical support devices in routine clinical practice and outcomes in randomized clinical trials may differ."

In a related study, Amit Amin, MD, of Washington University School of Medicine in St. Louis, noted that the use of mechanical support devices began increasing in 2008 after the approval of Impella and had continued an upward swing almost in parallel with use of the new device, while use of the balloon pump has remained relatively flat or has declined in the past decade. Among patients undergoing percutaneous interventions, more than 30% now are Impella devices, he reported.

In addition, the cost of treating patients with mechanical support devices has increased from about $47,000 per person to $51,000 per person in the "Impella era," Amin said.

However, he added, use of the devices varies from hospital to hospital, with a six-fold variation in the likelihood that certain hospitals will use the device.

And while there are large variations in outcomes among patients who received Impella, the analysis indicated that among patients receiving Impella devices there was a 1.71-fold increased risk of death, a 2.62-fold increased risk of bleeding episodes, a 1.53-fold increased risk of acute kidney injury, and a 1.47-fold increased risk of stroke. In addition, use of Impella was associated with an increased length of hospital stay, increased time in the intensive care unit, and increased costs, Amin said.

"Prospective randomized studies are needed to clarify causative relationships between the use of mechanical support devices and outcomes in carefully defined populations," commented Ranya Sweis, MD, of Northwestern University School of Medicine in Chicago, who was not involved with the studies. "This kind of a large signal of concern should give us the impetus to start this kind of a randomized study."

"It would be unfair to characterize the Impella device as being more costly and less effective than other treatments," she told 乌鸦传媒. "It would be easy to walk away from this presentation and say 'No Impella ever.' But I don't think that would help our patients. I think we should say, 'Okay, there is a signal. Don't use this more than you think you should, but don't withhold treatment that might be lifesaving based just on registry data that is not prospective.'"

Sweis was the named discussant to put both studies in perspective but never had the opportunity to talk at the plenary session because earlier speakers went over their allotted times, squeezing her talk off the platform. The audience was referred to her slide deck available on the meeting app.

Interviewed by phone on her way to the airport, Sweis said: "These studies push us to try to get a clinical trial done that can answer the question of whether this device is useful. If you are a doctor who is looking at a patient in shock, with multiple vessel disease and other grave conditions, if you don't have a mechanical support device, that patient may not get a procedure."

"The patients who don't get support may not get treated, so does that mean more people will die? The people who are dying, won't even get a chance," Sweis continued. "The theory we have been operating on all along is that Impella gives very sick people a chance for survival. I don't think this study is a death knell for Impella, but we need a study to show that what was reported is not the case."

Seth Bilazarian, MD, medical director at Abiomed, in an e-mail response, criticized the analysis as "fundamentally flawed because it is based on poor quality, retrospective, payer coding data that lumps all indications together and is impossible to properly propensity match. In this data set, a patient walking in for an elective procedure looks the same as a patient in cardiac arrest with cardiogenic shock."

He also pointed to "more robust, previously published, FDA-audited, prospective real-world studies and randomized controlled trials, which demonstrate Impella improves survival in cardiogenic shock; reduces MACCE (death, stroke, MI and repeat revascularization) in high risk percutaneous coronary interventions, when compared to intra-aortic balloon pump; and enables more complete revascularization and improves ejection fraction and patient quality of life at 90 days."

Sweis said she was concerned about the variation of outcomes at hospitals. "That tells me there is room to improve what we are doing at the community level," she noted. She said the the study could not determine if the higher mortality with Impella was due to its use in sicker individuals. "That is something we cannot capture with this data even with the propensity scoring used in the studies. There are factors that are hard to capture," she said.