Using Computer Models to Predict How Tissue Engineered Vascular Grafts Will Work

Using Computer Models to Predict How Tissue Engineered Vascular Grafts Will Work 150 150 Kevin Mayhood

Model and experimental data strongly suggest testing TEVGs until scaffold has biodegraded

Identical tissue engineered vascular grafts (TEVGs) being tested in small-diameter veins and arteries of a mouse model performed well for 12 weeks. At 14 weeks, all TEVGs in the veins continued performing well, but all in the arteries suddenly failed.

During their effort to understand why, Christopher Breuer, MD, director of the Center for Regenerative Medicine in the Abigail Wexner Research Institute at Nationwide Children’s Hospital, and his colleagues searched the published literature and found that no one was testing TEVGs beyond 12 weeks. At 12 weeks, Dr. Breuer’s team determined, the scaffold had lost its tensile strength but remained intact.

Their analysis showed that venous scaffolds were infiltrated by host cells and over time increasingly resembled the native vein tissue. The density, organization and morphology of the arterial grafts never approached that of the native arterial tissue. The test and follow-up investigation, conducted several years ago, took six months.

Fast forward to today. A computer model developed by Dr. Breuer’s collaborator, Jay Humphrey, PhD, chair of Biomedical Engineering at Yale University, was fed data from the experiment and then predicted the exact outcomes — in just a few hours.

“The early study showed—and the computer modeling confirmed— that it’s important to study the product you’re testing until the scaffolding has gone away, otherwise you face serious risks,” Dr. Breuer says.

The modeling takes into account medical, chemical and biomechanical factors, native tissue, host responses and more to describe a complex biological process, down to a scale of one cell interacting with one polymer over time, the investigators say.

The researchers report the earlier study and the findings with computer modeling in Acta Biomaterialia.

“Ultimately, computational modeling will enable us to make grafts for patients better and faster, using fewer animal models for experimentation,” says Cameron Best, PhD, senior research associate in Dr. Breuer’s laboratory and lead study author.

The Breuer Lab makes TEVGs for congenital heart disease. Because TEVGs grow with the patient, they can eliminate the need for operations to replace standard grafts that patients outgrow, Dr. Best says. “The issue is you have to tune the grafts for each application.”

As the earlier experiment and model showed, one style of graft worked well in veins, which had high blood flow, but dilated and failed in higher-pressure arteries. The researchers have been using modeling to consider scaffolds of different materials, degradation rates, fiber diameters and more, to help them choose which designs may best work in an artery. They then test the most promising designs.

“By combining experimental data with modeling, you end up with less experimenting and a more precise design,” Dr. Breuer says.

Drs. Best and Breuer and their colleagues are also using the modeling to more fully understand the behavior of grafts in patients.

“The current model predicts the variability we see from person to person,” Dr. Breuer says. “The next step is to use each patient’s personal characteristics to personalize TEVGs.”

The researchers are focusing on one type of TEVG but suggest their computational model can be used for  valve conduits, heart valves, cartilage, bone — anything that’s tissue-engineered.


Best CA Szafron JM, Rocco KA, Zbinden J, Dean EW, Maxfield MA, Kurobe H, Tara S, Bagi PS, Udelsman BV, Khosravi R, UYi T, Shinoka T, Humphrey J, Breuer CK. Differential outcomes of venous and arterial tissue engineered vascular grafts highlight the importance of coupling long-term implantation studies with computational modelingActa Biomaterialia. 2019 Aug;94:183-194.


About the author