Maternal Characteristics for the Prediction of Preeclampsia in Nulliparous Women: The Great Obstetrical Syndromes (GOS) Study.

OBJECTIVE

Low-dose aspirin started in early pregnancy significantly reduces the risk of preeclampsia (PE) in high-risk women, especially preterm PE. This study aimed to evaluate the influence of maternal characteristics on the risk of PE in nulliparous women.

METHODS

The Great Obstetrical Syndromes (GOS) study recruited nulliparous women with singleton pregnancies at 11 to 13 weeks. The following maternal characteristics were collected: age, BMI, ethnicity, chronic diseases, smoking, and assisted reproductive technologies. Relative weight analyses were conducted, and predictive multivariate proportional hazard models were constructed. Receiver operating characteristic curve analyses with their area under the curve (AUC) were used to evaluate the value of each factor for the prediction of PE and preterm PE. The study also evaluated the SOGC guidelines for identification of women at high risk of PE.

RESULTS

Of 4739 participants, 232 (4.9%) developed PE, including 30 (0.6%) with preterm PE. In univariate analyses, only BMI was significantly associated with the risk of PE (AUC 0.60; 95% CI 0.55-0.65) and preterm PE (AUC 0.64; 95% CI 054-0.73). Adding other maternal characteristics to BMI had a non-significant and marginal impact on the discriminative ability to the models for PE (AUC 0.62; 95% CI 0.58-0.66) and preterm PE (AUC 0.65; 95% CI 0.56-0.74). At a false-positive rate of 10%, maternal characteristics could have predicted 23% of PE and 19% of preterm PE. The SOGC guidelines were not discriminant for PE (detecting 96% of PE and 93% of preterm PE with a 94% false-positive rate).

CONCLUSION

In nulliparous women, BMI is the most discriminant maternal characteristic for the prediction of PE. Maternal characteristics should not be used alone to identify nulliparous women at high risk of PE.