Electrophysiology labs are becoming increasingly central to how patients with cardiac arrhythmias are being treated. For atrial fibrillation (AF), the most common kind of arrhythmia, the standard-of-care ablation procedure involves ablating in the upper left chamber of the heart where each of the four pulmonary veins connect.
But this procedure doesn't work for AF patients, with a failure rate as high as 40% to 50%. That is particularly the case for patients with persistent arrhythmia, which lasts longer than a week, and those who have had one or more prior procedures and develop scar tissue that can limit the effectiveness of any subsequent ablation. For those patients, Johns Hopkins University researchers have developed a personal, digital heart model that is based on contrast-enhanced magnetic resonance imaging scans.
"Atrial fibrillation is very prevalent all over the world, about 1% to 2% of the population has it. And as the population ages, it's becoming a huge burden and expense. So, in the last decade, there has been a lot of investment into new methodologies to treat this problem," Natalia Trayanova, the Murray B. Sachs Endowed Chair Professor of Biomedical Engineering and Medicine and the director of the Alliance for Cardiovascular Diagnostic and Treatment Innovation (ADVANCE) at Johns Hopkins University, told BioWorld MedTech. "In the clinic, the standard-of-care for such an arrhythmia is to burn a circle around each of these four pulmonary veins. These pulmonary veins sometimes behave in the wrong way, and, if you prevent the electrical signals coming from them, there is research showing that you will prevent the arrhythmia.
"But that doesn't work well for many persistent patients, when intermittent arrhythmia degrades into persistent arrhythmia, and they have it all the time," she continued. "The worst part is that many patients come back for repeat procedures over and over again. [E]very time the procedure is repeated, you have more and more damage to the tissue, and it ultimately does not function well." In addition, a new arrhythmia can be created. "So, it can become a vicious circle," she explained.
The technology, known as Optimal Target Identification via Modelling of Arrhythmogenesis (OPTIMA), can be used to test how the heart of an individual persistent AF patient can be expected to respond to a series of specific procedures. Tests assessing various ablation point combinations, which are typically confined to seven different places in the heart, can identify which procedure will produce the optimal results.
"I'm very optimistic that this personalized simulation-driven approach will prove to be the missing link needed to markedly improve catheter ablation outcomes in patients with more advanced forms of atrial fibrillation. This new approach may transform [the] current approach to catheter ablation of atrial fibrillation," said Hugh Calkins, a professor of medicine at Johns Hopkins Medicine and an author on the study.
The data on the best treatment for that patient is imported directly into the ablation tool to guide the subsequent procedure. A proof-of-concept study in 10 patients with persistent AF and atrial fibrosis saw only one patient return within the 300-day follow-up window for a subsequent treatment of a simpler arrhythmia. These results were published in the Aug. 19, 2019, issue of Nature Biomedical Engineering.
Electrophysiology labs are designed to do this sort of testing directly on the heart itself, but Trayanova expects that OPTIMA will enable much more vigorous and accurate testing than can be conducted on the heart currently.
"The computational prediction of ablation targets avoids lengthy electrical mapping and could improve the accuracy and efficacy of targeted AF ablation in patients while eliminating the need for repeat procedures," concludes the paper.
The U.S. FDA already has approved the next trial on the technology, which is slated to start in the fall. It will be a 160-patient, randomized trial in persistent AF and atrial fibrosis patients. If they subsequently have positive results in-hand, Trayanova expects that the technology will be ready for a licensing agreement or a spinoff to facilitate commercialization.
Trayanova didn't have the best experience with an earlier startup, Cardiosolv Ablation Technologies, where she remains as chief scientific officer. The technology was based on similar science but used in a different indication, ventricular tachycardia. Trayanova expects that this time around it will help to have much more academic data in hand prior to pursuing commercialization in a corporate setting.
"It's a paradigm change because you have a method of determining what should be the care of these patients where there is no standard-of-care – if we demonstrate better outcomes," said Trayanova. "If it works, we have a new way of treating these patients with the worst form of this disease. To me, that's a dream come true."
Advancing the virtual heart
This would be the first project for Trayanova's ADVANCE group, which launched in December. One of its primary aims is to use the underlying personalized computational modeling in the Virtual Heart platform across a number of different cardiac indications, in addition to integrating more sophisticated imaging data and supplementary patient data from electronic medical records using artificial intelligence.
The Virtual Heart platform is being developed for use in both cardiac treatment and diagnostics. Her group already is working to expand the model beyond the upper chambers. Additional early indications already in progress include prediction of the risk of sudden cardiac death, which often exhibits no signs or symptoms in advance.
The group has published a paper on risk stratification in patients after myocardial infarction, in addition to conducting research on risk prediction in myocarditis and looking to better understand the pathology underlying the deadly congenital disorder infant methemoglobinemia, also known as blue baby syndrome.
Hypertrophic myopathy and sarcoidosis are also areas of research. To better understand these, the team is integrating other imaging modalities – such as PET scans – to identify inflammation and T1 mapping for distributed fibrosis. One of these projects also uses machine learning to incorporate data from the electronic medical record, such as obesity and other clinical and lifestyle data linked to inflammation.
"We have been expanding into different diseases and incorporating different imaging modalities, as well as clinical and lifestyle data, in diseases that are more complex and involve inflammatory factors, some of these are not just heart diseases," Trayanova concluded.