ACC.17: Cerebral embolic protection devices may not improve freedom from stroke, infarction

A randomized study found that using two FDA-approved cerebral embolic protection devices in patients undergoing surgical aortic valve replacement (SAVR) was not associated with an improvement in freedom from clinical or radiographic infarction or clinical stroke.

The devices also were not associated with an improvement in the volume of central nervous system (CNS) infarct as assessed by MRI or neurocognitive outcomes at 90 days.

The researchers assessed the CardioGard embolic protection cannula and the Embol-X intra-aortic filtration device.

Lead researcher Michael Mack, MD, of the Baylor Health Care System in Dallas, presented results of the CONSORT study in a late-breaking clinical trial session at the ACC scientific session on March 19 in Washington, D.C. The National Institutes of Health (NIH) sponsored the study.

After an interim analysis by the data and safety monitoring board, the study was stopped at 77 percent of planned enrollment. During a press conference, Mack said that the trial was not stopped because of safety issues but rather because of “anticipated futility to reach the primary endpoint.”

“We, of course, were very disappointed, not only by the outcome, but by the fact that it was stopped,” Mack said. “We appealed to the sponsor, NIH, to let us complete enrollment of the trial because we felt like a lot of the important secondary outcomes may give us a signal, but we were not able to continue enrollment.”

In this study, the researchers randomized 383 patients at 17 sites to the Embol-X group, the CardioGard group or the control group. Mack said that baseline cognitive impairment existed in one-quarter to one-third of neurologically normal patients undergoing SAVR and that a majority of the patients had evidence of radiographic infarct by MRI.

The primary endpoint occurred in 32.7 percent of patients in the CardioGard group, 27.1 percent of patients in the Embol-X group and 34.8 percent of patients in the control group. The primary endpoint was freedom from clinical or radiographic CNS infarction at 7 days post-procedure.

The researchers found the devices were associated with capturing embolic debris in that 99 percent of patients in the Embol-X group and 76 percent of patients in the CardioGard group. The devices were also associated with a reduction in delirium immediately after SAVR, including a significant reduction in in-hospital delirium in the CardioGard group and a trend toward reduced delirium in the Embol-X group.

Mack said there was increase in acute injury and adverse events in the Embol-X group compared with the other groups.

“The hypothetical reason may be that actually placement of the device causes dislodgment of debris in the aorta and embolization to the kidney,” he said. “Obviously, that’s speculation only.”

The study had a few limitations, according to Mack, including that this was the first trial evaluating these devices in study sites. He also mentioned that they diagnosed MRI infarcts with 1.5 Tesla and 3 Tesla scanners, which could have created heterogeneity. In addition, he noted that the study was underpowered for clinical stroke and other endpoints. Further, one-third of the strokes occurred after day three, and the 90-day follow-up did not adequately assess long-term neurocognitive outcomes.

Mack said he and the other researchers have had discussions with NIH about re-consenting these patients and following up with them for five years

“We have for the first time an incredible wealth of radiographic evidence of brain injury. It’s going to be interesting to see out at five years whether this does lead to longer-term dementia.”

Mack said that they have also had discussions regarding screening these patients for genetic markers to determine if they are more susceptible to brain injury.

“Despite the fact that debris was captured–which was the mechanism by which these devices are intended to work–there was no difference in the primary endpoint,” Mack said in a news release. “However, the vast majority of infarcts were only detectable by post-operative MRI, with fewer than 10 percent of patients exhibiting clinical symptoms. Moreover, trends with respect to the volume of brain infarction suggest that embolic protection device recipients were less likely to have larger infarcts. This may be important, as the risk of symptomatic stroke increases with infarct volume.”