Obstetrics & Gynecology · 1 h ago
Mini PIERSdynamic model predicts maternal risk in preeclampsia
A prediction model was developed and externally validated using data from 7,283 women with preeclampsia. It showed moderate discrimination across days 0 to 9, with performance varying between internal and external validation cohorts.
- Model developed using data from 6,548 women.
- External validation included 735 women with preeclampsia.
- External-validation AUC was about 0.70 on days 0–9.
- Clinical utility and generalizability require further evaluation.
Researchers developed the mini PIERSdynamic model using data from 6,548 women with preeclampsia in the PIERS studies, splitting these data into training and internal validation sets. A separate cohort of 735 women was used for external validation. The model used repeated measurements to estimate the risk of a composite outcome of maternal death or severe maternal morbidity, assigning patients to five risk strata.
Adverse outcomes occurred in 785 women (12.0%) in the development cohort and 91 (12.4%) in the external validation cohort. The daily area under the receiver-operating-characteristic curve was about 0.75 on days 0–9 in internal validation and 0.70 in external validation; peak values were 0.92 and 0.94, respectively. Observed daily event rates corresponded closely to assigned risk strata, while precision-recall performance declined as event incidence fell over time.
The findings support the model’s potential for dynamic risk prediction for up to 10 days after presentation, including in low-resource settings. The authors say further evaluation is needed to establish generalizability and clinical utility; the study does not show that using the model improves outcomes.
In practice
For women presenting with preeclampsia, mini PIERSdynamic offers a way to update estimates of adverse maternal outcome risk over as many as 10 days using repeated measurements. Its discrimination was moderate across days 0–9 in the external cohort (AUC about 0.70), although peak AUC reached 0.94. The results suggest potential for risk stratification, including in low-resource settings, but do not establish that model-guided care improves outcomes. The authors call for further assessment of generalizability and clinical utility before practice implications are established.
AI-generated clinical interpretation · Not editor-reviewed
Is this summary clinically accurate?
Help fellow clinicians: your rating sends inaccurate summaries straight to our editors.
Sign in to rate this summary →Source
Hypertension: Development and External Validation of Mini PIERSdynamic Model for Predicting the Adverse Maternal Outcome of Preeclampsia in Low-Resource Settings ↗This is an automated AI-condensed summary that has not yet been reviewed by an editor. Always consult the full item at the original source.