Physicians weigh the efficacy, risk of the Impella pump

Use of the Impella heart pump during percutaneous coronary intervention (PCI) rose rapidly between the 2000s and 2010s, according to a Circulation analysis—but so did its variability across hospitals and associated adverse outcomes.

The Impella, manufactured by medical device company Abiomed and FDA-approved since 2008, is one of two options for mechanical circulatory support (MCS) during PCI, first author Amit P. Amin, MD, MSc, of Washington University School of Medicine, and colleagues said in their paper. The other is an intra-aortic balloon pump (IABP), a long, skinny balloon that controls blood flow through the aorta.

Impella devices are catheterized, miniaturized ventricular assist pumps designed to provide up to 5 liters of cardiac output and reduce left ventricular preload in PCI patients who need MCS. They’ve made headlines for the past year after a post-approval study (PAS) of the Impella RP right ventricular assist device resulted in unexpectedly low survival rates, throwing the efficacy of the tech into question.

The concern was that the FDA had approved the Impella RP in 2017 based on an overwhelmingly positive pre-market approval study (PMA) of 60 patients who underwent PCI with Abiomed’s device. In that study, 73.3% of patients treated with the Impella RP survived at least 30 days after the device was removed; in the PAS released earlier this year, just 28.6% of patients met the primary survival endpoint.

A months-long investigation by the FDA revealed that, with a couple of caveats, the Impella RP system was safe when used appropriately in the right patients. Abiomed also funneled $100 million of its own funding into clinical research for the Impella device, and a recent Impella RP PAS found that when physicians followed FDA-approved protocol, patients achieved 72% survival with the device.

Amin et al. sized up the situation with their analysis of trends and variations in Impella use, clinical outcomes and costs across U.S. hospitals. Drawing from the Premier Healthcare Database, the team analyzed 48,306 patients undergoing PCI with MCS—either the Impella or an IABP—at 432 hospitals between 2004 and 2016.

The researchers initially reported that 9.9% of PCI patients treated with MCS received an Impella pump, but that number increased over time. By 2016, Impella pumps accounted for 31.9% of MCS in patients undergoing PCI.

“The overall rise in MCS use after 2008 closely paralleled a concomitant rise in patients who received Impella while the proportion of patients who received IABP remained stable or declined slightly during the post-Impella era,” the authors wrote, noting they didn’t observe increasing Impella use in patients who were more critically ill.

Amin and colleagues identified wide—greater than fivefold—variation in Impella use across hospitals, and among Impella patients there was a more than 2.5-fold variation in outcomes of bleeding. Outcomes also differed for death, acute kidney injury (AKI) and stroke, all of which were subject to 1.5-fold variation in Impella recipients. After adjusting for the propensity score, Impella use was associated with a 24% increased risk of death, 10% increased risk of bleeding and 34% increased risk of stroke.

The authors said that hospitals with higher Impella use also had higher rates of adverse outcomes and greater costs. Adverse outcomes and costs were higher in the Impella era (2008-2016) than in the pre-Impella era (2004-2007).

“Impella use is rapidly increasing among PCI patients treated with MCS, with marked variability in its use and associated outcomes,” Amin et al. wrote. “Although unmeasured confounding cannot be ruled out, when analyzed by time periods, or at the hospital level or patient level, Impella use was associated with higher rates of adverse events and costs. More data are needed to define the appropriate role of MCS in patients undergoing PCI.”

Amin and his colleagues have been criticized by some for neglecting to present study results that are fully representative of the Impella. Seth Bilazarian, Abiomed's medical director, told Cardiovascular Business that Amin et al.'s Impella dataset included patients that were on average sicker and carried higher procedural risks than patients treated with IABPs. Impella patients also reportedly had twice the rate of multi-vessel disease, chronic renal failure and NSTEMI, as well as a 50% higher rate of heart failure and diabetes.

"This observational, non-FDA audited analysis is fundamentally flawed because it is based on poor-quality, retrospective payor coding data that lumps all indications together and is impossible to properly propensity match," Bilazarian said. "In this dataset, a patient walking in for an elective procedure looks the same as a patient in cardiac arrest with cariogenic shock."