Some patients with advanced heart failure may benefit from the support of a CF-LVAD (continuous flow left ventricular assist device) to maintain cardiac function. However, despite improvements in the management of complications associated with the use of CF-LVADs, stroke remains a significant risk after device implantation. To better predict stroke risk, researchers successfully used machine learning algorithms.
In recent years, axial, and more recently centrifugal, continuous-flow left ventricular assist devices (CF-LVADs), have been increasingly used in patients with advanced heart failure who are waiting for a transplant, for example. These devices improve patient outcomes and cause fewer complications. However, there is one nasty complication for which the risk increases after implantation of a CF-LVAD: the risk of stroke.
“Early detection and treatment of stroke after a CF-LVAD is critical to improving outcomes. Our goal in this study was to find a way to predict the risk of future strokes that may occur after implantation of CF-LVADs,” says Dr. Nandan Mondal, assistant professor in the Michael E. DeBakey Department of Surgery at Baylor College of Medicine.
Lower level OxPhos mitochondrial proteins
Previous studies have shown that mitochondria, cellular structures that provide cells with the energy they need, play a vital role in stroke. Reduced levels of OxPhos mitochondrial proteins in white blood cells have been associated with the severity of the disease.
“In the current study, published in ASAIO Journal, we investigated whether levels of OxPhos in white blood cells can help predict stroke risk after CF-LVAD implantation,” said Jacob P. Scioscia, author of the study and medical student at Baylor conducting research in the Mondal laboratory.
Machine learning students
Scioscia heard about the project through Baylor's SOAR database. The SOAR - Student Opportunities for Advancement in Research - is dedicated to training and supporting BCM medical students in pursuing research opportunities.
“Dr. Mondal was looking for students with a background in machine learning. I have knowledge in this field and the project was related to cardiothoracic surgery, an area I am interested in. I thought it would be a good opportunity to gain experience in basic science research and meet experts in the field,” Scioscia said.
Predictive factors
For the study, the team studied 50 CF-LVAD patients; 25 had had a stroke before receiving the device and 25 had not. Blood samples were taken from all patients, both before and after CF-LVAD implantation. In those blood samples, the various OxPhos proteins in white blood cells were measured. This generated large amounts of data. Then that data was analyzed using machine learning. That led to the discovery of six predictive factors for the risk of stroke after implantation. It turned out that a decrease in OxPhos proteins could be linked to the occurrence of another stroke after implantation.
“We found that both before and after CF-LVAD implantation, the group with a previous stroke had lower levels of OxPhos proteins than the group without a previous stroke. In the group with a previous stroke, OxPhos levels were lower after CF-LVAD implantation than before,” Mondal said.
The study shows for the first time the existence of mitochondrial dysfunction in white blood cells of patients with congestive heart failure who had had a stroke, even before CF-LVAD implantation. “The findings suggest that changes in OxPhos proteins could serve as indicators, predicting new strokes after CF-LVAD,” Scioscia says.
Follow-up research
By combining experimental results with machine learning, the probability of stroke progression by OxPhos proteins could be demonstrated with a high degree of reliability. This knowledge could help improve care for these patients.
“Many factors may contribute to the risk of stroke after CF-LVAD implantation. This study is one step toward the discovery of stroke indicators. We are currently studying more. Therefore, we are continuing this study with a larger group of patients for a longer follow-up time to determine if our results are broadly applicable. This approach would give physicians and surgeons a means to anticipate complications and implement interventions prior to CF-LVAD implantation,” said Dr. Nando Mondal.
To improve care after stroke, rapid intervention by shortening the time to diagnosis, Philips and Medtronic recently entered into a strategic partnership. They slotn into the newly formed WSO Advocacy Coalition of the World Stroke Organization. This WSO-led coalition brings together diverse parties, including healthcare professionals, patient groups and policy makers, to develop strategies to address the global impact of stroke.
