(1119) Prospective validation of USPSTF guidelines for preeclampsia prediction and clinician compliance with aspirin recommendations

Friday, January 31, 2025

4:00 PM – 6:00 PM

Submitting Author(s)

Morten Rasmussen, PhD (he/him/his)

VP of Research
Mirvie Inc.
South San Francisco, CA, United States

Presenting Author(s)

Arun Jeyabalan, MD, MS (she/her/hers)

Division Director of Maternal-Fetal Medicine
University of Pittsburgh School of Medicine
Pittsburgh, Pennsylvania, United States

Coauthor(s)

Michal A. Elovitz, MD

Hilarie L. Morgan and Mitchell L. Morgan President’s Distinguished Professor in Women’s Health
Icahn School of Medicine at Mount Sinai
New York, New York, United States
CH

Carrie Haverty, MS

Mirvie Inc.
South San Fransisco, California, United States
AM

Alison Moe, BS

Mirvie Inc.
South San Francisco, California, United States
AK

Arkady Khodursky, PhD

Mirvie Inc.
South San Francisco, California, United States
MJ

Maneesh Jain, PhD

CEO
Mirvie Inc.
South San Francisco, California, United States
ML

Manfred Lee, PhD

Mirvie Inc.
South San Francisco, California, United States
Thomas F. McElrath, MD, PhD (he/him/his)

Mirvie Brigham Women's Hospital Harvard Medical School
Boston, Massachusetts, United States
ES

Elizabeth F. Sutton, PhD

Director of Research
Woman's Hospital
Baton Rouge, Louisiana, United States
George R. Saade, MD (he/him/his)

Professor and Chair, Associate Dean for Women's Health Obstetrics and Gynecology
Macon & Joan Brock Virginia Health Sciences Eastern Virginia Medical School at Old Dominion University
Norfolk, Virginia, United States
Antonio F. Saad, MBA, MD (he/him/his)

Director of Perinatal Research Unit, MFM Fellowship Program Director
Inova Health
Falls Church, Virginia, United States
Luis D. Pacheco, MD

Professor
University of Texas Medical Branch
Galveston, Texas, United States
Joseph R. Biggio, Jr., MD, MSc (he/him/his)

System Chair, Women's Services
Ochsner Health
New Orleans, Louisiana, United States
Ebony B. Carter, MD (she/her/hers)

Associate Professor; Director, Division of Maternal Fetal Medicine
University of North Carolina
Chapel Hill, North Carolina, United States
Kara M. Rood, MD

The Ohio State University
Columbus, Ohio, United States
Antonina I. Frolova, MD, PhD (she/her/hers)

Assistant Professor of Ob&Gyn
Washington University School of Medicine
St. Louis, Missouri, United States
EP

Esther Park-Hwang

Multicare
Orlando, Florida, United States
Cynthia Gyamfi-Bannerman, MD, MS (she/her/hers)

Professor and Chair
University of California, San Diego
San Diego, California, United States
Ai-ris Y. Collier, MD (she/her/hers)

Assistant Professor
Beth Israel Deaconess Medical Center
Boston, Massachusetts, United States
William A. Grobman, MBA, MD

Professor
The Ohio State University
Columbus, Ohio, United States
Vincenzo Berghella, MD (he/him/his)

Professor, Director
Sidney Kimmel Medical College of Thomas Jefferson University
Philadelphia, Pennsylvania, United States

Objective: United States Preventive Services Task Force (TF) guidelines recommend using risk-factors for preeclampsia (PE) in order to offer low-dose aspirin (LDA) to reduce risk. However, the predictive value is unclear. We sought to prospectively evaluate the performance and utilization of TF in predicting PE and to determine if LDA was recommended based on TF in a diverse population.

Study Design: People with a singleton pregnancy enrolled in an IRB-approved prospective, observational, cohort study < 22 weeks gestation between July 2020 - June 2023. Enrollment occurred at one of 12 clinical centers or via direct-to-participant-recruitment across the United States. Data was obtained from medical records review including TF risk-factors and if LDA had been recommended (Table 1). All PE cases were adjudicated by trained MFMs.

Results: The study enrolled 5,652 diverse participants (Table 1). Using TF recommendations for prediction of PE the sensitivity is 74% and the specificity is 48%. When including one or more moderate risk-factors the sensitivity increases to 97% and the specificity drops to 12%. 2,440 (43%) received a LDA recommendation. Of those with at least one high-risk factor, 82% had LDA recommended while in those with one or more moderate risk-factors, 37.4% received LDA recommendation (Fig 1a). 3,974 (70.3%) were in the TF’s moderate risk-factors only category. For those with any risk factor, LDA was more likely to have been recommended in those who later developed PE vs those who didn’t (Fig 1b).

Conclusion: TF had poor performance as a predictive screening tool for PE in the group with moderate risk factors. There was also low compliance with LDA recommendations suggesting non-adherence to TF, practice variation, or incomplete documentation. The higher rate of PE in those recommended LDA may reflect provider identification of a higher risk subgroup, non-adherence with taking aspirin, or failure of LDA to reduce PE in certain subgroups. These data highlight the opportunity and urgent need to develop better risk assessment tools based on objective biology rather than based on clinical risk factors.