A team of researchers from the University of Warwick have developed a novel technique that can detect potential preterm birth in asymptomatic high-risk women, with up to 73% accuracy months before delivery. The research has been published in Scientific Reports.
The group of scientists used the “volatile organic compound analysis technology”, that is designed to typify the airborne chemicals. The research team has ‘trained’ the device via machine-learning techniques, so that they could identify the chemical vapour patterns from preterm birth using vaginal swabs taken during routine examinations.
Technology designed to characterise airborne chemicals has been 'trained' by researchers at @warwickmed and @WarwickEngineer to identify potential preterm births in asymptomatic high-risk women, with up to 73% accuracy.
— Warwick Newsroom (@warwicknewsroom) July 29, 2020
Their invented technique could potentially lead to a cost-effective, non-invasive, point-of-care test. This test could significantly become a part of routine care for women identified as being at risk of delivering prematurely. Moreover, it can allow the healthcare staff to better support those women during pregnancy and birth and help to reduce the risks to their baby. Preterm birth has become the leading cause of death in children under five and currently there are few accurate tools to considerably predict who is going to deliver preterm.
During the study, the protocols included initial assessment of the volatile organic compounds (VOCs) present in the vagina for a condition called bacterial vaginosis – a condition where the bacteria of the vagina have become imbalanced. Previous research and evidence have shown that bacterial vaginosis in early pregnancy is associated with an increased risk in having a preterm birth, although treating bacterial vaginosis doesn’t decrease that risk.
The technology used in the study to separate vapour molecules was done by combining two techniques that first pre-separates molecules based on their reaction with a stationary phase coating (a gas-chromatograph), which was followed by measuring their mobility in a high-electric field (an Ion Mobility Spectrometer). Then the machine learning techniques were used by the team to ‘train’ the technology to spot patterns of VOCs that were signs of bacterial vaginosis.
Following, the vaginal swabs were analysed from the expecting women, who have been attending the preterm prevention clinic as part of their routine care. These women were found to either had prior histories of preterm births or a medical condition that makes it more likely that they would deliver preterm but had shown no other indications that they would deliver preterm and were considered asymptomatic. During the 2nd and 3rd trimesters of pregnancy, the vaginal swabs were collected, and the respective outcome of all pregnancies were followed up.
The results showed that the first test had an 66% accuracy while the second, closer to the time of delivery, had an 73% accuracy. These test results implicate that 7/10 women with a positive test went on to deliver preterm and 9/10 women with a negative test delivered after 37 weeks.
Lead author of the study, Dr Lauren Lacey of Warwick Medical School and an obstetrics and gynaecology registrar at University Hospitals Coventry and Warwickshire NHS Trust said: “We’ve demonstrated that the technology has good diagnostic accuracy, and in the future, it could form part of a care pathway to determine who would deliver preterm. Although the first test taken earlier in pregnancy is diagnostically less accurate, it could allow interventions to be put in place to reduce the risk of preterm delivery; for the test towards the end of pregnancy, high risk women can have interventions put in place to optimise the outcome for baby.”