The Problem
The maternal mortality rate in the United States has increased from 9.9 per 100,000 births in 1999 to 26.4 per 100,000 births in 2015(1), making it one of the highest maternal mortality rates among industrialized nations. This highest risk is among non-Hispanic black women who had a mortality rate of 46 per 100,000 births in 2014(2).
More common, however, is severe maternal morbidity (SMM), which has been shown to be associated with increased risk for maternal mortality. SMM can be explained as adverse outcomes of the process of labor and delivery that result in significant short-term or long-term consequences to a woman’s health(3). The Center for Disease Control and Prevention defines SMM by 18 indicators commonly identified as peripartum complications. Severe maternal morbidity, like maternal mortality, has also significantly increased over the last 2 decades, from 49.5 per 10,000 births in 1993 to 144.0 per 10,000 births in 2014(4).
A broad number of diagnoses contribute to severe maternal morbidity and mortality. The leading causes of maternal mortality are hemorrhage, hypertensive disease, and thrombosis(5). Blood transfusion is the most common indicator of SMM and likely related to postpartum hemorrhage(3). Additionally, sepsis in the obstetric patient contributes to a large proportion of maternal mortality. In the United Kingdom, it was reported that sepsis accounted for one quarter of all maternal deaths in 2014. More recently, cardiac dysfunction has been identified as an increasingly prevalent contributor to SMM and now thought to be the leading cause of mortality (6). It is estimated that 40% of maternal deaths are preventable(7). At Duke University Hospital (DUH), ~80% of our patients are considered high risk, and therefore at increased risk for SMM. Currently, there are no systems to quickly identify women experiencing SMM. While there are a few commercial prediction systems available, none have been studied or shown to improve outcomes.
The ability to readily and reliably identify women at risk for SMM is vital to recent calls to reduce maternal morbidity and mortality. Delay in diagnosis has been identified as a factor in leading to SMM and maternal mortality. (8). In non-pregnant patients, there are a plethora of early warning systems and clinical deterioration algorithms useful in identifying patients in need of escalation of care. However, the unique physiology of pregnancy makes adoption of these algorithms in the obstetric population difficult. Often well-validated clinical deterioration systems in non-pregnant patients excluded pregnant patients in the development process and perform poorly when adapted to the obstetric population making them ineffective.
In response to recommendations from the Joint Commission and 13- state run maternal mortality review committees (including North Carolina), attempts have been made to develop maternal early warning systems to predict patients likely to require escalation of care or be admitted to the intensive care unit. One such system is Sepsis in Obstetrics Score (SOS)(9). The SOS uses the components of the Rapid Emergency Medicine Score (REM), as well as portions of the Surviving Sepsis Campaign and modifies these parameters based on expected normal values in pregnancy. SOS had an area under the curve of 0.85 (95% CI 0.76-0.95) for prediction of ICU admission for sepsis using a threshold score of 6 (9). Of note, SOS score of 6 had a positive predictive value of 15%.
In 2007, the implementation of the Maternal Early Obstetric Warning System (MEOWS) was strongly advocated in the United Kingdom(10). Singh and colleagues assessed MEOWS parameters in an obstetric population to determine performance for severe maternal morbidity. MEOWS was 89% sensitive for predicting morbidity (95% confidence interval [CI], 81%–95%) and 79% specific (95% CI, 76–82) with a positive predictive value of 39% (95% CI, 96%–99%). In the United States, the National Partnership for Maternal Safety, proposed a simplified early warning system adapted from MEOWS, the Maternal Early Warning Criteria (MEWC)(11). No scoring has been widely adopted among obstetric units in the United States in part due to low specificity in identifying at-risk patient, lack of validation in U.S. data sets, and no major U.S. obstetric society promoting adoption.
Overall, there is still work to be done to identify obstetric patients at highest risk for SMM and mortality and to effectively integrate alerts into workflows to improve patient outcomes. Freidman rightly identifies 2 key components to a useful maternal early warning system: the ability to identify patients at risk for critical illness and who benefit from timely intervention, and the ability minimize false-positive alerts so that patient care is not otherwise compromised(10). An effective system would likely be rapidly introduced into clinical care.
Our Solution
References
1. Global, regional, and national levels of maternal mortality, 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet (London, England) 2016 Oct 8;388(10053):1775-812.
2. Moaddab A, Dildy GA, Brown HL, Bateni ZH, Belfort MA, Sangi-Haghpeykar H, et al. Health Care Disparity and Pregnancy-Related Mortality in the United States, 2005-2014. Obstetrics and gynecology 2018 Apr;131(4):707-12.
3. Kilpatrick SK, Ecker JL. Severe maternal morbidity: screening and review. American journal of obstetrics and gynecology 2016 Sep;215(3):B17-22.
4. Severe Maternal Morbidity in the United States. 2017 [cited 2019 September 25]; Available from: https://www.cdc.gov/reproductivehealth/maternalinfanthealth/severematernalmorbidity.html#anchor_trends
5. Creanga AA, Berg CJ, Syverson C, Seed K, Bruce FC, Callaghan WM. Pregnancy-related mortality in the United States, 2006-2010. Obstetrics and gynecology 2015 Jan;125(1):5-12.
6. Small MJ, James AH, Kershaw T, Thames B, Gunatilake R, Brown H. Near-miss maternal mortality: cardiac dysfunction as the principal cause of obstetric intensive care unit admissions. Obstetrics and gynecology 2012 Feb;119(2 Pt 1):250-5.
7. Berg CJ, Harper MA, Atkinson SM, Bell EA, Brown HL, Hage ML, et al. Preventability of pregnancy-related deaths: results of a state-wide review. Obstetrics and gynecology 2005 Dec;106(6):1228-34.
8. Kominiarek MA, Scott S, Koch AR, Zeschke M, Cordova Y, Ravangard SF, et al. Preventing Maternal Morbidity from Obstetric Hemorrhage: Implications of a Provider Training Initiative. American journal of perinatology 2017 Jan;34(1):74-9.
9. Albright CM, Has P, Rouse DJ, Hughes BL. Internal Validation of the Sepsis in Obstetrics Score to Identify Risk of Morbidity From Sepsis in Pregnancy. Obstetrics and gynecology 2017 Oct;130(4):747-55.
10. Friedman AM. Maternal early warning systems. Obstetrics and gynecology clinics of North America 2015 Jun;42(2):289-98.
