External validation of a model for periconceptional prediction of recurrent early-onset preeclampsia

Abstract

Objective: To validate a previously published prediction model for recurrent early-onset preeclampsia (PE). Methods: We included 229 pregnant women with a history of early-onset PE and computed their risk using the prediction model, compared the predicted risk to their pregnancy outcomes and assessed performance of the model. Results: Early-onset PE recurred in 6.6% of participants. The area under the receiver operating characteristic curve was 59% (95% CI: 45–73). The model created groups that were only moderately different in terms of their risk. Conclusions: The model’s discriminate ability was poor and predictive performance insufficient to classify women into relevant risk groups.

• Early-onset
• External validation
• Prediction
• Preeclampsia
• Recurrence.

Acknowledgements

We thank research nurses, research midwives, and administrative assistants of the Consortium for women's health and reproductivity studies, and the residents, nurses, midwives, and gynaecologists of the participating centres for their help with participant recruitment and data collection.

Ethical approval

This study has been approved by the ethical committee of the University Hospital Maastricht (Ref.no. MEC 07-2-078). A total of six academic and three non-academic hospitals participated in this study; all of them have completed their obligatory feasibility assessment procedure successfully.