What biomarker-guided preeclampsia care could look like

Limitations in Preeclampsia Prediction With Current Guidelines

Every obstetrician knows the routine. At an early prenatal visit, we review a patient’s history and work through the familiar risk factors for preeclampsia: prior preeclampsia, chronic hypertension, diabetes, kidney disease, autoimmune disease, nulliparity, obesity, maternal age, race, family history, and other clinical and demographic characteristics. Based on that information, we decide who should receive low-dose aspirin and who may need closer surveillance. Aspirin is an inexpensive and effective preventive intervention. But the way we decide who is at risk remains surprisingly imprecise.

Preeclampsia continues to contribute substantially to maternal morbidity, medically indicated preterm birth, and long-term cardiovascular risk.¹ Yet our prevention strategy is still largely based on population-level characteristics rather than direct measures of the biology that ultimately leads to disease. Current US recommendations use medical history and combinations of clinical and demographic risk factors to determine eligibility for aspirin prophylaxis.²

The practical problem is familiar: Most patients meet 1 or more risk criteria and never develop preeclampsia, whereas some patients who appear relatively low risk early in pregnancy subsequently develop significant disease. As precision medicine becomes routine in oncology, cardiology, and other areas of medicine, the field of obstetrics should be asking a similar question: Can we do better than a risk-factor checklist? Increasingly, the answer appears to be yes.

McElrath et al recently evaluated how well US Preventive Services Task Force (USPSTF) criteria identify patients who ultimately develop preeclampsia.³ Their findings highlight the central limitation of our current approach: Clinical risk factors cast a wide net, but they do not necessarily identify the patients who will actually develop clinically important disease.

A risk factor tells us that a patient belongs to a population in which an outcome occurs more often. It does not necessarily tell us what is happening biologically in this particular pregnancy. Consider nulliparity, obesity, or advanced maternal age. Each is associated with an increased population-level risk of preeclampsia, but none tells us whether placental dysfunction is developing in an individual patient. Conversely, a patient without obvious clinical risk factors may still develop the biologic changes that ultimately result in severe disease. This gap is where biomarkers can help.

Recent research data show that part of why our current approach may be challenging is that we are trying to predict a heterogeneous condition as though it were a single disease. Using cell-free RNA transcriptomic profiling, Elovitz et al identified biologically distinct molecular subtypes of hypertensive disorders of pregnancy.⁴ These subtypes were associated with different molecular pathways and clinical trajectories. This finding may suggest that what we call preeclampsia represents a final common clinical phenotype produced by several distinct underlying biologic pathways. That would help explain why patients with apparently similar clinical presentations can have very different courses. One patient may develop mild hypertension near term, whereas another develops early-onset disease with severe features, fetal growth restriction, or rapid clinical deterioration.

If the underlying biology differs, our ability to predict disease, and eventually prevent or treat it, will improve when we measure that biology rather than relying on a checklist of characteristics. Angiogenic biomarkers have already demonstrated that measurable biologic signals can provide useful information in patients being evaluated for suspected preeclampsia. The sFlt-1:PlGF ratio, for example, has been studied extensively as a marker of placental dysfunction and short-term disease risk.⁵ More recent data continue to demonstrate its ability to identify patients at risk of preeclampsia with severe features.⁶ The next question is whether biomarkers can move even earlier in the pathway, from helping us evaluate suspected disease to identifying biologic risk before clinical disease is apparent.

Maternal outcomes in the United States remain profoundly unequal, with Black patients experiencing substantially higher pregnancy-related mortality than White patients.³ These disparities are driven by complex and overlapping factors, including structural racism, differences in access to care, social determinants of health, and variation in how patients interact with and are treated by the health care system.

Current recommendations appropriately acknowledge these inequities. USPSTF guidance includes Black race and lower income among moderate-risk factors for preeclampsia, explicitly recognizing that these characteristics reflect social and structural conditions rather than inherent biologic susceptibility.⁷ Including these factors can help ensure that patients experiencing the consequences of structural inequity are not denied preventive therapy. But it also highlights (and perhaps perpetuates) the system’s limitations. Race is being used because we do not yet have a better way to measure the underlying exposures and biologic risk that it imperfectly represents. A biomarker-based approach offers a more equitable possibility. Imagine assessing each patient using the same objective measure of placental or maternal disease biology, rather than relying on race, income, or other demographic characteristics as proxies for risk. A patient would be classified as higher or lower risk because of what is occurring biologically in that pregnancy, not because she belongs to a population in which an adverse outcome occurs more frequently.

That does not mean biomarkers will automatically eliminate disparities. Any approach to improve the diagnosis of preeclampsia will need to address access, cost, and test performance across diverse populations. A highly predictive test that is available only to well-resourced patients or whose performance differs depending on race or ethnicity would actually widen existing disparities. Equitable implementation, therefore, must be part of biomarker development from the outset. When appropriately validated and broadly accessible, objective biomarkers would reduce at least some of the variability inherent in risk-factor ascertainment and clinical decision-making.

There is also a more immediate limitation to our current strategy of preeclampsia prevention: implementation. Our risk factor–based system requires multiple steps to work. The clinician has to identify the relevant risk factors, determine whether the patient qualifies for aspirin (either as a “recommendation” or a “consideration”), develop a plan through shared decision-making with the patient, educate the patient on the rationale for surveillance, and reinforce adherence on an ongoing basis. There are opportunities for failure at every step.

McElrath and colleagues found that real-world aspirin recommendations were not consistently aligned with USPSTF eligibility criteria, particularly among patients qualifying based on moderate-risk factors.³ The Society for Maternal-Fetal Medicine has also identified aspirin use as an important quality-improvement opportunity.⁸ Even when aspirin is recommended, taking a daily medication during pregnancy is not necessarily straightforward. Qualitative data indicate that patient concerns—including uncertainty about why aspirin is needed, worry about medication exposure during pregnancy, problems accessing medication, and difficulty maintaining daily adherence—are barriers to effective implementation.⁹

Better prediction does not solve all of these problems. But identifying a targeted, more clearly defined group of patients at increased biologic risk would allow for a consolidation of education and clinical resources around patients in whom the effort may make the most difference. That is precision prevention.

What would biomarker-guided prenatal care actually look like?

Obstetric practices already use biomarker-guided care every day. Consider gestational diabetes. We do not intensively monitor every pregnant patient because she might develop hyperglycemia. We screen, identify the group with abnormal physiology, and then place those patients into a structured care pathway that may include education, glucose monitoring, nutrition counseling, medication, and additional fetal surveillance––all in an effort to prevent poor outcomes.

A future preeclampsia pathway would look similar, and it would include much more than just low-dose aspirin. A validated biomarker obtained at an appropriate gestational age would complement clinical risk assessment and separate patients into risk groups. Patients identified as higher risk would enter a standardized pathway with reinforced aspirin counseling when appropriate, home blood pressure monitoring, patient education, more frequent assessment, laboratory surveillance when clinically indicated, and tailored maternal and fetal monitoring. Roberts et al have already described a structured care plan for patients at increased risk for preeclampsia that incorporates education, prevention, surveillance, and coordinated follow-up.¹⁰ The challenge is identifying the right population to place into that pathway. Biomarkers would potentially make targeting more precise. At the same time, patients with reassuring biologic risk assessment might avoid unnecessary testing or additional visits while still receiving routine counseling about symptoms and warning signs.

This approach is consistent with the American College of Obstetricians and Gynecologists’ recent clinical consensus on tailored prenatal care, which encourages moving away from a one-size-fits-all prenatal schedule toward care individualized to medical, pregnancy-specific, and social needs.¹¹ If prenatal care is going to become truly tailored, we need better tools to determine who needs more care, who needs different care, and when that care should begin.

For decades, much of preeclampsia care has been reactive. We identify population-level risk early in pregnancy, recommend aspirin to some patients, and then largely wait for hypertension, symptoms, laboratory abnormalities, or fetal complications to tell us that disease has developed. As Myatt argued several years ago, meaningful improvement in preeclampsia prediction will likely require moving beyond conventional clinical risk factors toward objective measures of placental dysfunction before disease becomes clinically apparent.¹² The science has advanced considerably since then.

The studies by McElrath, Elovitz, and others suggest that the future of preeclampsia care may lie in connecting molecular biology with practical clinical pathways.^3,4 Rather than asking only, “Does this patient have risk factors?” we should be asking, “Is the biology of this pregnancy moving toward disease?”

Biomarkers open the door to earlier recognition, more targeted prevention, smarter use of clinical resources, and prenatal care that responds to the biology of the patient in front of us rather than relying primarily on what happened to patients who looked similar to her in the past. That is how we begin to replace guesswork with precision.

References

1. Fink DA, Kilday D, Cao Z, et al. Trends in maternal mortality and severe maternal morbidity during delivery-related hospitalizations in the United States, 2008 to 2021. JAMA Netw Open. 2023;6(6):e2317641. doi:10.1001/jamanetworkopen.2023.17641

2. US Preventive Services Task Force; Davidson KW, Barry MJ, Mangione CM. Aspirin use to prevent preeclampsia and related morbidity and mortality: US Preventive Services Task Force recommendation statement. JAMA. 2021;326(12):1186-1191. doi:10.1001/jama.2021.14781

3. McElrath TF, Jeyabalan A, Khodursky A, et al. Utility of the US Preventive Services Task Force for preeclampsia risk assessment and aspirin prophylaxis. JAMA Netw Open. 2025;8(7):e2521792. doi:10.1001/jamanetworkopen.2025.21792

4. Elovitz MA, Gee EPS, Delaney-Busch N, et al. Molecular subtyping of hypertensive disorders of pregnancy. Nat Commun. 2025;16(1):2948. doi:10.1038/s41467-025-58157-y

5. Zeisler H, Llurba E, Chantraine F, et al. Predictive value of the sFlt-1:PlGF ratio in women with suspected preeclampsia. N Engl J Med. 2016;374(1):13-22. doi:10.1056/NEJMoa1414838

6. Berg AH, Calsavara VF, Hund M, et al. Serum soluble-fms-like tyrosine kinase 1-to-placental growth factor ratio on Elecsys immunoassay platform predicts preeclampsia with severe features in hospitalized women with hypertensive disorders of pregnancy. Am J Obstet Gynecol. 2026;234(3):834-855. doi:10.1016/j.ajog.2025.11.006

7. Hoyert DL. Maternal mortality rates in the United States, 2023. National Center for Health Statistics Health E-Stats. 2025. Accessed March 20, 2026. https://stacks.cdc.gov/view/cdc/174577

8. SMFM Patient Safety and Quality Committee. Society for Maternal-Fetal Medicine Special Statement: prophylactic low-dose aspirin for preeclampsia prevention—quality metric and opportunities for quality improvement. Am J Obstet Gynecol. 2023;229(3):B2-B9. doi:10.1016/j.ajog.2023.04.039

9. Vinogradov R, Holden E, Patel M, Grigg R, Errington L, Araújo-Soares V, et al. Barriers and facilitators of adherence to low-dose aspirin during pregnancy: a co-produced systematic review and COM-B framework synthesis of qualitative evidence. PLoS One. 2024;19(5):e0302720. doi:10.1371/journal.pone.0302720

10. Roberts JM, King TL, Barton JR, Beck S, Bernstein IM, Buck TE, et al. Care plan for individuals at risk for preeclampsia. Am J Obstet Gynecol. 2023;229(3):193-213. doi:10.1016/j.ajog.2023.04.023

11. American College of Obstetricians and Gynecologists. ACOG Clinical Consensus No. 8: tailored prenatal care delivery for pregnant individuals. Obstet Gynecol. 2025;145(5):565-577. doi:10.1097/AOG.0000000000005889

12. Myatt L. The prediction of preeclampsia: the way forward. Am J Obstet Gynecol. 2022;226(2S):S1102-S1107. doi:10.1016/j.ajog.2020.10.047

Disclosures

Reed is a member of the Clinical Advisory Board for Mirvie. This manuscript was not funded by a grant from any funding agency in the public, commercial, or not-for-profit sectors.

Acknowledgments

The author thanks Thomas McElrath, MD, PhD, and Carrie Haverty, MS, of Mirvie for reviewing the manuscript and providing editorial suggestions. All final decisions regarding manuscript content were made solely by the author.

During the preparation and revision of this work, the author used OpenEvidence and ChatGPT to assist with literature searches, organization, and language editing. The author independently reviewed the cited references, critically evaluated the evidence and interpretations presented, and takes full responsibility for the final manuscript.

Author’s Bio

Dallas Reed, MD, FACMG, FACOG, is a physician, educator, and advocate whose work bridges obstetrics and gynecology, medical genetics, reproductive health, and cancer genetics. She is an associate professor of Obstetrics/Gynecology, Medicine, and Pediatrics at Tufts University School of Medicine, and serves as Division Chief of Genetics and Director of Perinatal Genetics at Tufts Medical Center. Dr Reed earned her MD from Boston University School of Medicine, completed residency in Obstetrics and Gynecology at Bridgeport Hospital–Yale New Haven Health, and completed fellowship training in Medical Genetics at the Icahn School of Medicine at Mount Sinai. Dr Reed has focused on translating advances in genomics into scalable clinical programs that improve access, standardize care, and support precision medicine. Her clinical, educational, and research interests include prenatal and reproductive genetics, hereditary cancer, genomic sequencing in stillbirth and critically ill newborns, health equity, and culturally responsive care.