Sepsis screening tools designed to account for the physiological changes that may occur during pregnancy perform best from 20 weeks of pregnancy through 3 days postpartum, suggests research supported by the National Institutes of Health. In early pregnancy and later in the postpartum period, tools based on general sepsis screening criteria may more accurately predict which patients are at high risk for this life-threatening complication of infection. These findings promise to help guide health care providers in selecting the best strategies to screen patients for maternal sepsis, a leading cause of pregnancy-related deaths worldwide.
The work was led by Melissa E. Bauer, D.O., of Duke University and Elliott K. Main, M.D., of Stanford University. It appears in Obstetrics & Gynecology. Funding was provided by NIH’s Eunice Kennedy Shriver National Institute of Child Health and Human Development.
Without prompt treatment, sepsis can lead to tissue damage and organ failure. However, identifying maternal sepsis early can be challenging, in part because pregnancy-related changes such as a faster heart rate and a higher white blood cell count can mask signs of the condition.
Screening tools can help health care providers identify which patients need further evaluation for sepsis. An effective screening strategy must both accurately detect patients with sepsis and have a low rate of false-positives—when the test incorrectly identifies someone without sepsis as likely to have the condition. A high false-positive rate can lead to unnecessary testing and overuse of health care resources.
Researchers set out to evaluate the performance of five widely used sepsis screening tools to identify those that are best for assessing a person’s risk for maternal sepsis at different stages of pregnancy and the early postpartum period. Those who screen positive using these tools, which are based on vital sign measurements and levels of infection-fighting white blood cells, would need further tests to confirm a diagnosis of sepsis. Two of the five tools are considered “pregnancy-adjusted,” meaning that they account for the changes in heart rate, respiratory rate, and white blood cell counts that may occur during pregnancy.
The researchers uncovered substantial variations in the performance of the five tools at different times during and after pregnancy. In one study, they assessed electronic health record data for patients who delivered between 2016 and 2021 at 59 hospitals in 12 states. For these delivery hospitalizations, the two pregnancy-adjusted tools had the lowest false-positive rates (6.9% and 9.6%) and greater than 90% sensitivity—meaning they correctly identified almost all people with sepsis.
In a second study, the investigators evaluated data from the electronic health records of pregnant patients admitted for sepsis at 67 U.S. hospitals and of patients admitted for sepsis up to 42 days postpartum at 71 U.S. hospitals. They found that, for patients admitted at less than 20 weeks of pregnancy or more than 3 days postpartum, non-pregnancy-adjusted tools were better at predicting sepsis than pregnancy-adjusted tools. However, between 20 weeks of pregnancy through 3 days postpartum, pregnancy-adjusted screening tools had the highest sensitivity and lowest false-positive rates.
Based on their findings, the authors suggest that health care providers use a pregnancy-adjusted sepsis screening tool between 20 weeks of pregnancy and 3 days postpartum. Before 20 weeks of pregnancy and after 3 days postpartum, providers should use a non-pregnancy-adjusted tool. The authors stress that these screening tools should be part of a two-step process, in which those who screen positive are then further evaluated for sepsis.
Bauer ME et al. Performance characteristics of sepsis screening tools during antepartum and postpartum admissions. Obstetrics & Gynecology DOI: 10.1097/AOG.0000000000005480 (2023)
Main EK et al. Performance characteristics of sepsis screening tools during delivery admissions. Obstetrics & Gynecology DOI: 10.1097/AOG.0000000000005477 (2023)