Study Finds Clues to Acute Neurologic Events With Short-Term Impella Cardiac Support

For patients on short-term cardiac support with an Impella device, inadequate anticoagulation, duration of cardiac support and a low platelet count all may play roles in causing an ischemic stroke or intracranial hemorrhage. So finds a newly published single-institution series from Cleveland Clinic (ASAIO J. 2020;66[8]:e105-e109) analyzing outcomes among 79 patients with acute cardiogenic shock who underwent implantation of Impella, a percutaneously placed temporary left ventricular assist device (LVAD).

Ischemic stroke and intracranial hemorrhage are common causes of death and disability in patients who receive a long-term implantable LVAD, with many of these events occurring perioperatively. But the incidence of, and risk factors for, acute neurologic events in patients on short-term cardiac support with a percutaneous temporary LVAD have not been well characterized.

“Causes of these devastating acute neurologic events with short-term Impella support have not been thoroughly explored,” says stroke neurologist Ken Uchino, MD, the study’s senior and corresponding author and Head of Research and Education in Cleveland Clinic’s Cerebrovascular Center. “Our case series revealed plausible risk factors and sets the stage for further investigation.”

Study findings at a glance

Dr. Uchino and colleagues reviewed prospectively collected data on 79 consecutive patients implanted with an Impella device (Impella 2.5, 5.0, CP or RP) at Cleveland Clinic between October 2010 and July 2018. The median duration of Impella support was 8 days (range, 1-33).

Six patients (7.5%) were found to have had an acute neurologic event, as follows:

• Three ischemic strokes
• Two intracerebral hemorrhages
• One subdural hematoma

Of these six patients, four died during hospitalization, and the other two survived to heart transplantation (one underwent transplant the day before the acute neurologic event). Analysis of the six patients revealed the following:

• Their acute neurologic events occurred on day 1, 2, 3, 4, 6 and 8 after Impella implant.
• The hemorrhagic events were attributed to anticoagulant use and thrombocytopenia.
• Two of the ischemic strokes were attributed to inadequate anticoagulation, with the remaining stroke attributed to pump thrombosis.
• Age, sex, pre-implant New York Heart Association functional class and prevalence of various comorbidities did not correlate with an acute neurologic event, although smoking and a history of prior ischemic stroke were more common among the six patients who had an event than in the overall cohort.

What needs to be studied next

The incidence of ischemic or hemorrhagic stroke in this series (7.5%) was higher than in other published studies, which found a less than 2% incidence during the implantation period. The researchers note that while the reasons for this difference cannot be answered by this study, their findings focus attention on several key questions:

• How to balance risk around anticoagulation? While Impella devices require a heparin-containing continuous purge solution to prevent pump thrombosis and device failure, placing a thrombogenic pump with the possibility of anticoagulation interruption increases the risk of thromboembolism. Among the patients in this series with a neurologic event, one had embolic stroke that was likely due to pump thrombosis, and all ischemic strokes had embolic patterns consistent with thromboembolism seen with other cardiac devices. The three intracranial hemorrhages — two of which were fatal — occurred after recent heparin exposure.
• How does duration of Impella use affect risk? Other studies have suggested that risk of systemic hemorrhages increases with Impella use beyond four days, but most acute neurologic events in this series occurred within that window.
• Can the optimal platelet count be defined? Other studies have reported low platelet counts as a risk factor for poor outcomes among patients on Impella support. In this series, five patients with an acute neurologic event had thrombocytopenia, with a median platelet count among all six patients of 87 k/UL (range, 54-229). The single patient with a platelet count within normal range had a fatal intracerebral hemorrhage.

“This small retrospective study cannot determine the exact cause of neurologic events in Impella-treated cardiogenic shock patients, a population with a high risk of stroke and many confounding factors,” concludes Dr. Uchino. “But it does highlight a number of issues that will be helpful to explore in larger randomized controlled trials and prospective studies.”