Modeling blood flow during CPR

Eunok Jung conducts a computer experiment in which the flow of a liquid changes direction when the frequency of pulses is changed. The black and white fluid markers represent the position at the initial and final time respectively. (Photo by Jim Richmond, enhanced by Jane Parrott) Click here for more photos.

Frank, 42, fell to the floor at home, a victim of cardiac arrest. His brother Jim immediately put his ear to Frank’s chest. Frank was not breathing; his heart had stopped beating. Jim called 911. Because he had been trained in cardiopulmonary resuscitation (CPR), he began chest compressions and mouth-to-mouth airflow in the hope of restoring his brother’s heartbeat and breathing. Tragically, Jim’s heroic efforts failed; Frank died.

CPR has been successful in restarting the hearts of people who have been electrically shocked, badly injured, or frozen, but in few instances has CPR revived victims of cardiac arrest. More than 250,000 people die from cardiac arrest each year in the United States. Yet CPR, despite its high failure rate, is used by physicians and rescue workers to preserve blood flow during cardiac arrest. If the mechanisms of blood flow in the body during CPR were better understood, it might be possible to improve CPR techniques and save more lives of victims of cardiac arrest.

At least that’s the hope of Eunok Jung, a staff member in ORNL’s Computational Mathematics Group in the Computer Science and Mathematics Division. She recently made a scientific discovery using computational simulation that is relevant to CPR. Jung conducted a computational experiment using a two-dimensional model of a rigid, doughnut-shaped tube in which one section is replaced with a flexible membrane. Earlier laboratory experiments with this fluid-filled device, which has no valves, showed that periodic squeezing of the membrane caused a flow in one direction.

Jung discovered that changing the frequency of squeezing affects not only the amount of flow but also its direction. She verified this computational finding with a physical apparatus. “I found that you can reverse the flow of fluid simply by varying the frequency of squeezing,” she says. If during CPR the heart valves remain open, then Jung’s results suggest that the rate of chest compression may partly de-termine whether CPR saves a life. Jung’s exper-iment took advantage of her Ph.D. dissertation adviser’s “immersed boundary method” for modeling the fluid dynamics of the heart. Her adviser is Dr. Charles Peskin, one of the world’s leading experts on heart modeling, who works at New York University’s Courant Institute.

Whether Jung’s finding is important to CPR may depend on which theory about the heart is correct. The cardiac compression theory says that during CPR the heart works as an active pump. The thoracic compression theory argues that during CPR the heart is a passive conduit that allows blood to flow, as a result of periodic squeezing and pressure differences between the external and internal thoracic compartments, through the cardiac valves that remain open (valveless pumping). Some imaging data suggest that the valves of the heart remain open during CPR in some instances. “These results,” Jung says, “imply that the heart is acting at least partly as a passive conduit.”

To better understand blood flow in the heart during CPR and valveless pumping in general, Jung proposes to computationally simulate the heart as a pump, as a passive conduit, and as a combination of both. She will write a number of differential equations to create three-dimensional heart models, coupled with a lumped parameter circuit model (ordinary differential equations) of the circulation that can be solved using ORNL’s supercom-puters. Her results could get at the heart of how to modify CPR techniques to save victims of cardiac arrest.