Applying Engineering to Heart Disease Therapeutics

Faculty Spotlight
November 13, 2017 By Wendy Wolfson

Faculty Profile: Arash Kheradvar, M.D., Ph.D., FAHA
Professor, School of Medicine, Departments of Biomedical, Mechanical and Space Engineering

In KLAB, Dr. Arash Kheradvar’s biomedical engineering lab at the School of Engineering, a bevy of graduate students and post-docs develop technologies to combat heart disease, ranging from designing artificial heart valves to the imaging and modeling of congenital heart disease and heart failure in adults. KLAB is part of the Edwards Lifesciences Center for Advanced Cardiovascular Technology at UC Irvine. “We are trying to advance technologies that are eventually going to help patient diagnostics and treatment.” Kheradvar said.

Kheradvar, a doctor who engineers cardiovascular devices emphasizing cardiac mechanics and imaging, has so far received over 15 different U.S. and international patents on heart valves, as well as for other medical devices and methods.

“When I was a medical student, I was always thinking of what I could do to help patients in a broader setting than a particular hospital.” Kheradvar said. “The standard of care is similar in most places. I was always intrigued by the research, and the advanced technologies you could work on to elevate the standard of care and affect a larger group of patients.”

Implanting a valve.

FoldaValve, his first startup, is commercializing a transcatheter aortic valve, now in preclinical testing. “That valve is protecting the leaflets from being crunched by the stent during catheterization and transcatheter implant,” Kheradvar said. “It provides options for repositioning and retrieval, in case the procedure doesn’t go smoothly and the interventionalist decides to remove the valve.”

The valve is intended for implantation in older patients who have their aortic valve clogged by calcium deposits, a condition known as aortic valve stenosis. Transcatheter valves commonly need to be properly positioned when implanted, but calcium deposition in the native aortic valve can interfere with fit, causing paravalvular leak within the valve. “The FoldaValve delivery system is patented and provides adjustable repositioning to eliminate the leak,” Kheradvar said. As class III medical devices, artificial heart valves usually require multiple rounds of testing to determine how well they work and if they are safe. So far the devices have been shown to be safe, durable, and work in sheep. Kheradvar is now raising funding to conduct first-in-human trials.

Current bioprosthetic valves implanted in younger patients calcify faster than in older patients, and need replacement every 10-15 years—a major surgery for patients—but the alternative option, mechanical heart valves, require lifelong use of blood thinning drugs which can cause hemorrhages if the dosage is even slightly wrong.

Kheradvar’s lab is also developing patient-specific hybrid tissue-engineered heart valves using cells and tissues from patients. “We grow those tissues over a scaffold and develop a hybrid tissue-engineered heart valve that is specific to the patient,” Kheradvar said. “This potentially lowers the chance of valve failure due to calcification, and body response to foreign material. It takes up to 6 weeks to grow a valve. It would supposedly work like a native organ, as we are using the patient’s own cells.” Kheradvar’s lab grows the valve’s cell coating until the hybrid valve’s leaflets are about 500 microns thick. To ensure uniformity of coverage, researchers place the valve scaffold in a cast and inject the cells and collagen mixture.

So far, other laboratories’ prototypes of tissue engineered heart valves using resorbable and absorbable scaffolds failed under high blood pressure after the underlying matrices dissolved. But Kheradvar is pursuing a different approach by designing valves using a permanent scaffold. These valves are fully biocompatible, meaning they would not cause adverse immune reactions in the body because the scaffolds are covered by and integrated with the patient’s own cells.

According to Kheradvar, while a number of companies are working on developing better transcatheter mitral and tricuspid valve technologies, getting this type of valve right remains an unsolved problem. One major challenge is the proper anchoring of the valves to the heart. Kheradvar’s lab is now developing various prototypes of transcatheter mitral and tricuspid valves, which are currently in animal testing.

Kheradvar is also developing a 3D echocardiographic particle image velocimetry imaging technique to enable cardiologists to see and measure blood pathlines and velocity in the heart in real time. “It is cheap, non-invasive and can be performed in the physician’s office using a standard ultrasound system,” Kheradvar said. “It enables physicians to get more information. It is a continuous measurement method, and is particularly useful for pediatric patients. Currently they have to go through MRI to get similar data which requires anesthesia and major sedation for pediatric patients, but this echocardiographic technique can be performed in the doctor’s office or examining room.”

Kheradvar’s lab has also been working in the MRI 4D flow space for measuring energy dissipation, fluid dynamics, and other parameters in pediatric heart disease patients. He collaborates with pediatric cardiologists in the University Hospital Schleswig-Holstein in Kiel and in 2017, Germany’s Alexander Von Humboldt Foundation awarded Kheradvar a prestigious Humboldt Research Fellowship for Experienced Researchers.

Kheradvar has also been collaborating with Dr. Hamid Jafarkhani, chancellor’s professor of electrical engineering and computer science at UCI, to develop methods to better characterize the unique anatomies of patients with congenital heart defects. Currently, different physicians can independently calculate parameters such as ejection fraction in the same patient, and get different results. Or doctors can segment the same set of images differently even within the same day. “We call it inter- and intra-operator variability.” Kheradvar said. “So we are trying to use artificial intelligence, and in particular, deep learning, to eliminate this and come up with an automatically-derived ground truth, which is unique regardless of who works on it and when.” Their first publication1 in 2016 received over 55 citations within 18 months, an indicator of the importance of this method for clinical applications.

After receiving his M.D. from Tehran University of Medical Sciences in Iran, Kheradvar completed his Ph.D. and post doc in bioengineering from the California Institute of Technology, focusing on how fluid dynamics change during heart failure. “During the course of my Ph.D., I got very excited about the heart valves and how they work.” Kheradvar said.

At Caltech, Kheradvar was also fascinated into his thesis advisor Professor Morteza “Mory” Gharib’s avocation—studying Leonardo da Vinci’s sketches on heart valve fluid dynamics. As his first project, Kheradvar worked with Gharib to build a replica of da Vinci’s heart valve, which in 2006 became part of the exhibit, “Leonardo Da Vinci: Experience, Experiment, and Design”, at the Victoria and Albert Museum in London. “We developed a model of a heart valve based on Leonardo’s sketches,” Kheradvar said. “We showed that you can see the right vortices if you follow Leonardo’s sketches.” According to Kheradvar, Da Vinci’s sketches of the blood flow vortices in the heart were highly accurate.

1. Avendi MR, Kheradvar A, Jafarkhani H. A combined deep-learning and deformable-model approach to fully automatic segmentation of the left ventricle in cardiac MRI. Medical Image Analysis. 2016;30:108-119.