NEWS | October 2, 2019

Blood flow screening to improve detection of critical congenital heart defects in newborns

Siefkes receives an NIH grant to establish blood flow threshold

(SACRAMENTO)

Each year, 7,200 newborns in the United States are diagnosed with critical congenital heart disease (CCHD), a life-threatening heart defect. CCHD is a present-at-birth heart dysfunction that prevents the heart from pumping blood effectively or reduces the amount of oxygen in the blood.

Heather Siefkes, assistant professor of pediatric critical care at UC Davis Children's Hospital. Heather Siefkes, assistant professor of pediatric critical care at UC Davis Children's Hospital.

The standard screening tool to detect CCHD after birth and before the baby goes home is a pulse oximeter test which measures the oxygen level in the baby’s blood.

“The oxygen saturation screen using oximeters has improved CCHD detection, but unfortunately, it still misses about 900 cases annually in the U.S. alone,” said Heather Siefkes, assistant professor of pediatric critical care at UC Davis Children’s Hospital.

Siefkes, who was awarded a Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) grant last month, is working on improving CCHD detection in newborns by measuring their blood flow levels in addition to the blood oxygen levels.

“My research is looking at the potential of adding another screening measurement using the same pulse oximeter, but to detect the blood flow,” she added.

Unlike the oxygen level in blood, the blood flow measurement varies a lot. Until now, physicians do not know which blood flow threshold would indicate a possible CCHD diagnosis. Siefkes’ research will help identify this screening threshold by applying artificial intelligence techniques that accommodates for the variation in the blood flow value.

With the NICHD grant, Siefkes and her team will enroll 700 newborn babies at five hospitals, including UC Davis Children’s Hospital. Based on the blood flow screening of the babies, the researchers will develop a machine learning/artificial intelligence model to identify the perfusion (blood flow) value that can predict CCHD.

“Many years back, my mentor asked me what gave me ‘fire in the belly’?” Siefkes added. “One day during my fellowship, I received a transport call for a two-week-old baby who came to the emergency room very sick. It was clear that the baby had CCHD that went undetected by the afterbirth oximeter test. The late diagnosis meant a very bad outcome for the baby, and for me, it started a fire and a commitment to prevent this missed diagnosis from happening again.”

Siefkes is also supported by a UC Davis Clinical and Translational Science Center (CTSC) KL2 Program Scholar Award.