Poster Session 4
Juliana Martins, N/A (she/her/hers)
Associate Professor for Maternal Fetal Medicine
Macon & Joan Brock Virginia Health Sciences Eastern Virginia Medial School at Old Dominion University
Virginia Beach, Virginia, United States
Elizabeth Miller, MD
PGY-4 Resident
Macon & Joan Brock Virginia Health Sciences Eastern Virginia Medial School at Old Dominion University
Norfolk, Virginia, United States
Michael Bittner, N/A
Data Manager
Macon & Joan Brock Virginia Health Sciences Eastern Virginia Medial School at Old Dominion University
Norfolk, Virginia, United States
Daniel L. Rolnik, MD
Monash University
Melbourne, South Australia, Australia
Tetsuya Kawakita, MD, MS
Associate Professor
Macon & Joan Brock Virginia Health Sciences Eastern Virginia Medical School at Old Dominion University
Norfolk, VA, United States
To assess the performance of first trimester combined screening using the Fetal Medicine Foundation (FMF) model for the prediction of preterm preeclampsia in a U.S. population.
Study Design:
This was a secondary analysis of the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-be database. Diagnoses of chromosomal or structural fetal abnormality, miscarriage, or fetal death before 24 weeks of gestation, and those missing information regarding gestational age at delivery or preeclampsia were excluded. Variables in the FMF risk calculation included maternal factors, mean arterial pressure, biochemical markers, and uterine artery pulsatility index. We assessed the model performance using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve using a bootstrap method. The optimal cutoff point was determined using Liu’s method and calculated sensitivity, specificity, positive and negative predictive values (PPV, NPV), positive and negative likelihood ratio (LHR), and odds ratio (OR). Goodness of fit was evaluated by calibration plot.
Results:
The study analyzed 8,675 nulliparous women, among whom 2.18% had preterm preeclampsia. The ROC curve (Figure 1) revealed an AUC of 0.75, with a 95% confidence interval (CI) of 0.71 to 0.78 at the optimal cutoff point of 0.7%. Sensitivity of the model was 64.0% (95% CI 56.7 – 70.9), and specificity was 72.5% (95% CI 71.5 – 73.4). The PPV was 4.9% (95% CI 4.1 – 5.9), whereas NPV was 98.9% (95% CI 98.6 – 99.1). The positive LHR was 2.3 (95% CI 2.1 – 2.6), and the negative LHR was 0.5 (95% CI 0.4 – 0.6). OR for predicting preterm preeclampsia was 4.7 (95% CI 3.5 – 6.3). Overall, the FMF model underestimated the risk of preterm preeclampsia as shown by calibration plot (Figure 2).
Conclusion:
The FMF combined screening algorithm demonstrates a moderate ability to predict preterm preeclampsia in a U.S. nulliparous population, indicating a fair discriminative power. Ongoing prospective studies aim to validate this model across diverse populations, potentially leading to more tailored and effective pre-eclampsia prediction.
