Improving Preeclampsia Detection Throughout Pregnancy
Author Information
Author(s): Valeria Rolle, Petya Chaveeva, Ander Diaz-Navarro, Irene Fernández-Buhigas, Diana Cuenca-Gómez, Tanya Tilkova, Belén Santacruz, Teresa Pérez, Maria M. Gil, Simone Ferrero
Primary Institution: Hospital Universitario de Torrejón, Madrid, Spain
Hypothesis
Can repeated biomarker measurements throughout pregnancy improve the detection of preeclampsia?
Conclusion
Using repeated biomarker measurements throughout pregnancy significantly improves the prediction accuracy of preeclampsia from 50% to approximately 84%.
Supporting Evidence
- 4056 pregnancies were included in the training set and 944 in the validation set.
- The AUROC improved from 0.50 in the first trimester to 0.84 in the third trimester.
- Detection rates for preeclampsia improved significantly with repeated measurements.
- Longitudinal models allow for continuous risk assessment throughout pregnancy.
- Statistical methods considered the endogenous nature of longitudinal variables.
Takeaway
Doctors can check pregnant women multiple times to better predict if they will get a serious condition called preeclampsia, which can be harmful to both moms and babies.
Methodology
This study used a longitudinal retrospective design, analyzing data from pregnancies that included repeated biomarker measurements.
Potential Biases
Differences in maternal characteristics between training and validation datasets may introduce bias.
Limitations
The study's retrospective design may lead to non-random data loss and a relatively small number of preeclampsia cases, affecting the precision of estimates.
Participant Demographics
Participants were primarily White, with a mix of spontaneous and assisted conceptions, and varied maternal ages and BMI.
Statistical Information
P-Value
0.50
Confidence Interval
95% CI 0.73 to 0.94
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website