Author + information
- aDepartment of Medicine and Duke Clinical Research Institute, Duke University School of Medicine, Durham, North Carolina
- bDepartment of Medicine, Division of Cardiovascular Medicine, Brigham and Women’s Hospital, Boston, Massachusetts
- ↵∗Address for correspondence:
Dr. Adam DeVore, Duke Clinical Research Institute, 2400 Pratt Street, NP-8064, Durham, North Carolina 27705.
Left ventricular assist devices (LVADs) are a lifesaving therapy for select patients with advanced systolic heart failure, but these devices remain saddled with an unacceptably high burden of adverse events from bleeding, infection, pump malfunction, and neurological events. Among the complications of durable mechanical support, stroke is the most dreaded and is an important barrier to more widespread adoption of this therapy. Rates of ischemic and hemorrhagic stroke are unacceptably high despite innovation in LVAD pump design (Table 1). Stroke is associated with increased mortality and can significantly affect quality of life, functional ability, caregiver burden, health care costs, and transplant eligibility. Moreover, stroke risk does not abate with time, and fear of this complication persists in patients and their caregivers despite the dramatic improvements in functional capacity and heart failure symptoms following LVAD.
In this issue of JACC: Heart Failure, Acharya et al. (1) provide a detailed INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support) analysis to identify pre-operative risk factors for stroke and quantify stroke-related morbidity and mortality while receiving LVAD therapy. Investigators included 7,112 patients who received a durable, continuous-flow LVAD between May 2012 and March 2015 and excluded patients receiving total artificial hearts or biventricular assist devices. Investigational devices such as the HeartMate 3 (St. Jude Medical, St. Paul, Minnesota) are not registered in INTERMACS. The study provides important insight into the burden of stroke in a real-world setting. First, the stroke incidence was high with contemporary LVADs, with 11% having at least 1 stroke, or 0.12 strokes per patient year of follow-up. Second, stroke affected both old and young LVAD recipients at similar rates. Third, stroke rates did not differ by device implantation strategy, and the rate of transplantation was cut in half among patients suffering a stroke compared with those without a stroke (15.6% vs. 33.3%, respectively). Fourth, strokes occurred both early and late after LVAD implantation, with 16% of strokes occurring within 2 weeks of implantation. Fifth, mortality was high among LVAD patients after a stroke, 1-month mortality was 36%, and 1-year mortality was 56%.
A primary aim of the study was to identify pre-operative risk factors to optimize patients before surgery and inform LVAD decision-making. Unfortunately, most of the risk factors for stroke that were identified were not modifiable and thus not relevant for preimplant optimization: female sex, increased systolic blood pressure, history of heparin-induced thrombocytopenia, pre-operative support with an intra-aortic balloon pump, and primary cause of cardiomyopathy. Controlling blood pressure after LVAD implantation has been proven to lower incidence of stroke (2,3), but it is unclear whether pre-optimization blood pressure control would further reduce this risk. Even so, these data will better inform shared decision-making for patients and their caregivers by providing a global estimate of stroke risk.
The pathophysiology of LVAD-related stroke is complex and involves patient factors (cerebrovascular disease), medical management (antithrombotic target, time in therapeutic range, blood pressure control), pump design (flow profile, sintering, shear stress), and the pump–patient interface (rotor speed, pulsatility, stasis in aortic root or carotid bulb). Although pre-implantation factors were the focus of this report, these data suggest that both post-implantation infection and gastrointestinal bleeding are associated with increased stroke risk. This bolsters evidence that a hypercoagulable state or an alteration in anticoagulation intensity can influence stroke risk. Despite the patterns of risk that emerge in the number of registered patients, these data provide limited patient-level insight into the proximate causes of stroke. The INTERMACS charter also prohibits public analysis of stroke according to pump manufacturer or implant site. This precludes an exploration of event rates or risk factors that may vary according to axial versus centrifugal flow (a surrogate for pump manufacturer), to say nothing of optimal surgical technique and medical management. Data from landmark trials suggest that pump design does play a role in the incidence of stroke (4), whereas intensive blood pressure management has proven to mitigate stroke risk in the ENDURANCE supplemental trial (3). However, stroke risk on LVAD support has not improved over time (Table 1).
Progress in many areas of LVAD care has been slow but steady and punctuated by occasional leaps forward. Consider another LVAD complication that plagued the LVAD community: pump thrombosis. A report of data from 2004 to 2013 identified an abrupt rise in the rate of pump thrombosis at 3 months after implantation, from 2.2% in March 2011 to 8.4% in January 2013 (5). The LVAD community responded by considering multiple contributing causes including lower anticoagulation intensity, reduced pump speed, and surgical implantation technique. Adoption of best practices mitigated thrombosis risk, but alteration in pump design provided the true breakthrough. The HeartMate 3, which has a fully magnetically levitated rotor, wide blood flow passages, and an intrinsic pulse designed to avert stasis within the pump, had no evidence of pump thrombosis after 6 months of support in MOMENTUM 3 (The Multicenter Study of MagLev Technology in Patients Undergoing Mechanical Circulatory Support Therapy with HeartMate 3) (6). Despite the resounding success in reducing pump thrombosis, regrettably there were no differences in the incidence of stroke, another critical component of hemocompatibility. Further insight into stroke risk will be forthcoming from the long-term support phase of MOMENTUM 3, which will provide granular data for the delicate balance between bleeding and thrombosis.
These data from Acharya et al. (1) provide a sobering estimate of contemporary stroke risk on LVADs and will inform pre-implantation conversations with patients. When it comes to stroke while receiving mechanical support, we do not yet know if renewed hope for eliminating this scourge will come from further engineering advances or from new insights into post-implantation patient management. As a field, we must redouble our efforts to optimize pump design and establish best practices for care delivery, so that future progress in mechanical support does not stall due to stroke.
For supplemental references, please see the online version of this paper.
↵∗ Editorials published in JACC: Heart Failure reflect the views of the authors and do not necessarily represent the views of JACC: Heart Failure or the American College of Cardiology.
Dr. DeVore has received research support from American Heart Association, Amgen, and Novartis. Dr. Stewart has reported that he has no relationships relevant to the contents of this paper to disclose.
- 2017 American College of Cardiology Foundation
- Acharya D.,
- Loyaga-Rendon R.,
- Morgan C.J.,
- et al.
- Teuteberg J.J.,
- Slaughter M.S.,
- Rogers J.G.,
- et al.
- ↵Milano CA. The treatment of patients with advanced heart failure ineligible for cardiac transplantation with the HeartWare ventricular assist device: results of the ENDURANCE supplemental trial. Data presented at: International Society for Heart and Lung Transplantation; April 5, 2017; San Diego, CA.
- Mehra M.R.,
- Naka Y.,
- Uriel N.,
- et al.