Minimum test tradeoff applied to uterine artery pulsatility index for antepartum prediction of preeclampsia.

OBJECTIVE

This study applies a decision analysis metric, the Minimum Test Tradeoff (MTT), to determine the minimum number of data collections of uterine artery pulsatility index (UTPI) per true positive prediction of preeclampsia to increase the expected utility of prediction.

STUDY

DESIGN

We performed a nonsystematic review of the literature for studies with prediction models for preeclampsia that included the area under the curve (AUC) of the receiver operating curve with any combination of maternal characteristics, biomarkers, and biophysical characteristics. The study needed to report the AUC both with UTPI and without UTPI.

MAIN OUTCOME MEASURES: MTT values were calculated based on the models collected from the literature search and divided into subcategories based on gestational age of the fetus.

RESULTS

The range of MTT values for each subcategory were: delivery or diagnosis of preeclampsia < 34 weeks = 1000 to no benefit, delivery or diagnosis of preeclampsia ≥ 34 weeks = 100 to no benefit, preeclampsia requiring delivery < 37 weeks = 1200 to no benefit, preeclampsia requiring delivery ≥ 37 weeks = 1000 to no benefit, preeclampsia (unspecified gestational age for diagnosis or delivery) = 100 to no benefit.

CONCLUSIONS

All subcategories resulted in large MTT values, meaning a large number of UTPI measurements per true positive prediction of preeclampsia are needed to increase the expected utility of prediction. The magnitude of the MTT indicates that UTPI is not likely a worthwhile addition to a baseline preeclampsia screening model in the general population for preeclampsia.