Tracking the deadly and unpredictable postpartum hemorrhage
To prevent dangerous birth complication, WashU researchers will develop wearable device to track blood loss
In the delivery room, circumstances can turn dire on a dime if the patient starts losing excessive amounts of blood. One minute they seem fine, and the next, vital signs plummet, the patient crashes, and the care team may need to scramble for a blood transfusion or perform surgery.
As is commonly the case, that alert may arrive too late, with postpartum hemorrhage being the leading cause of maternal death worldwide. The patient themself may not notice and there are few ways to easily measure the blood pouring out (or pooling in the uterus) during the delivery process. Even the most eagle-eyed doctor or nurse cannot accurately estimate blood loss just from visual indicators alone.
Modern medical sensors have their limits, noted Christine O’ Brien, assistant professor of biomedical engineering in the McKelvey School of Engineering at Washington University in St. Louis. Counting bloody rags or collecting blood in calibrated drapes may seem simplistic but remains one of the few ways to warn of excessive blood loss. A more accurate and continuous early warning system could potentially prevent 50% to 90% of those maternal deaths from hemorrhage when paired with adequate treatment.
The blood loss is “difficult to measure and if you wait on normal vital signs like heart rate or blood pressure, those can remain stable for up to 1,500 milliliters of blood loss,” she said. Postpartum hemorrhage is classified as blood loss greater than 1 liter within 24 hours of birth.
Tracking blood pressure drops doesn’t necessarily provide the right data since modern delivery rooms are constantly adding medications and fluids in the patient that might impact that measurement.
Instead, O’Brien, with a five-year $2.8 million grant from the National Heart, Lung, and Blood Institute (NHLBI) of the National Institutes of Health, will develop a wearable device that tracks blood loss through measuring cardiovascular features from a novel light-based sensor that measures changes in hemoglobin and blood flow.
This measurement produces waveforms based on how the heart is pumping and how the rest of circulatory system resists that flow. The shape of those waveforms changes based on the state of a person’s cardiovascular system. Using statistical and machine learning algorithms, they can determine which features of the waveforms are correlated with absolute blood loss.
With an affordable wearable device that fits on a wrist and tracks hemodynamics, “We’ll be able to have a much clearer signal when someone is hemorrhaging,” said O’Brien.
The device will be tested on animal models with differently pigmented skin to ensure skin color does not impact the sensor readout. In devices that run on light-based sensors, like pulse oximeters, skin color can affect the accuracy of the blood oxygen level measurements. Skin tone shouldn’t impact reading of this new light-based sensor’s waveforms, but they’ll be testing it nonetheless in both preclinical models and in a diverse mix of human test subjects to ensure the sensor works correctly under differing circumstances. That also includes testing it on subjects with varying levels of hemodilution, (hemoglobin being watered down by other fluids).
Though they are working on a way to track blood loss, the overarching aim is to prevent postpartum hemorrhage, said O’Brien. With an early warning, in many cases, there are simple, low-cost interventions, such as massaging the uterus, that could stop the bleed. Not all hospitals are fully stocked to handle an emergency blood transfusion or surgery, so it’s better to not lose the blood in the first place.
“You can find yourself in an emergency very quickly,” O’Brien said.
Click on the topics below for more stories in those areas
Back to NewsFaculty in this story
