Jun 22, 2016

Blood Test Identifies women at Risk of Preterm Delivery as Early as 17 Weeks of Pregnancy

Faculty & Staff, Research
Professor Stephen Lye

A blood test developed by University of Toronto and international researchers has been shown to predict if a pregnant woman is at risk of delivering her baby prematurely, before the full 37 weeks of gestation.

Professor Stephen Lye

At 18 weeks, the test can predict premature delivery with 82 per cent accuracy. At 28 weeks, the accuracy rises to 86 per cent.

Premature birth remains the main cause of child-related mortality in the developed world. It occurs in five to 10 per cent of all pregnancies, but is associated with 70 per cent of all newborn deaths (excluding genetic anomalies) and up to 75 per cent of newborn disease including cerebral palsy, blindness, deafness, respiratory illness and complications of neonatal intensive care.

The study was published on June 22, 2016 in PLOS one, and was led by Dr. Stephen Lye, a professor in the departments of Obstetrics and Gynaecology and Physiology, along with researchers from Harvard University and the University of Calgary.  

 “There are treatments that can prevent preterm birth,” say Lye, who is also a Senior Scientist at the Lunenfeld-Tanenbaum Research Institute, part of the Sinai Health System. “But these treatments are only useful in a subset of women. This blood test could improve identification of women who will benefit from existing therapies. Moreover, it may also help drug studies to focus on women who are at highest risk of delivering preterm when evaluating new treatments.”

The study population is a subset of women who participated in the All Our Babies study, a community based longitudinal pregnancy cohort in Calgary, Alberta. The researchers collected paired maternal blood from pregnant women at two clinically relevant time points: approximately 17 weeks when fetal ultrasound is conducted and at approximately 27 weeks of gestation when gestational diabetes screening is performed.

The international team, consisting of clinicians, scientists and biostatisticians, used gene expression profiling and bioinformatics to develop gene sets, coupled with a patient’s clinical information such as history of preterm birth, history of abortion or anaemia, to predict whether or not a woman will deliver prematurely.