Short Term Prediction of Preeclampsia

Abstract

Preeclampsia (PE), a multisystem disorder in pregnancy, is a main cause of perinatal mortality and is associated with long-term maternal complications. For a long time, PE was defined as the new onset hypertension and proteinuria after 20 weeks’ gestation. It had been shown that this “gold standard definition” is not able to provide a sufficient prediction of PE-related fetal and/or maternal complications. In 2018 the International Society for the Study of Hypertension in Pregnancy recommended a broader definition of the disease. The new definition of the International Society for the Study of Hypertension in Pregnancy ruled out proteinuria as mandatory for the diagnosis of PE. This new definition increases the number of patients diagnosed as preeclamptic by nearly 21%, which is not accompanied by an increased severity of maternal outcomes. Including angiogenic biomarkers, however, has been shown to increase detection of adverse outcomes. The pathophysiology of PE is complex and not yet completely explained. Advances in prediction and diagnosis have been achieved by discovery and clinical evaluation of biomarkers, especially of placental origin. A broad spectrum of biomarkers has been tested, a few of them have been introduced into the clinical practice as of today. Especially angiogenic biomarkers that are rooted in the pathophysiology of PE have been demonstrated to be important in the prediction and diagnosis of adverse outcomes. At a cut-off value of the soluble fms-like tyrosine kinase-1 (sFlt-1)/placental growth factor (PlGF)-ratio of 85, early-onset PE <34+0 weeks of gestation can accurately be diagnosed with a sensitivity of 89% and a specificity of 97%. The Prediction of short-term outcome in pregnant women with suspected preeclampsia (PROGNOSIS) study has shown that the high negative predictive value (99.3%) of the sFlt-1/PlGF-ratio below 38 in patients with suspected PE rules out the onset of the disease within one week. PROGNOSIS Asia, evaluating the sFlt-1/PlGF-ratio cut-off of 38 in an Asian population, confirmed the excellent accuracy in prediction. Recently, the angiogenic biomarkers have been integrated in multi-marker prediction models. Digital approaches, integrating algorithm-based decision support tools paired with home monitoring devices may be the next step in enhancing predictive accuracy and thus bear the potential to reduce maternal and/or fetal morbidity and mortality and save costs for the payer in parallel. The objective of this review is to provide an overview of current methods for predicting and diagnosing PE.

Preeclampsia Predictor with Machine Learning: A Comprehensive and Bias-Free Machine Learning Pipeline

“…It shows early measurement of blood pressure can serve as proxy for risk of high blood pressure later in pregnancy. Also, information about placental analytes, which can potentially be gathered at a reasonable cost tradeoff between assessment and hospitalization, 5 allow predictions that strongly surpass the accuracy of an ACOG recommended medical history review. 33 Further validation is required for the proposed separate models for multiple time points to ensure consistency of prediction: a patient identified as high risk early in pregnancy should not be deemed low risk later without sufficient explanation.…”

“…That analytes such as PlGF, unlike blood pressure, were consistently important across the original model and the early vs late model (Figure 4), indicates their predictive power, particularly that of a rule out ability for early onset. 5 25…”

“…4 25 35 The occurrence rate of PE in this study was consistent with reported rates. 5 36 However, this meant that even with such a sizeable sample, the analysis was limited to only a couple of hundred sPE+E cases. The sub-study was also not without its limitations: analytes were only available for V1 and V2, and ultrasound data was only generally available beginning at V2.…”

“…Biochemical and biophysical markers can have an added benefit for screening for PE when used in combination with clinical characteristics taken from medical history, demographics, clinical measurements, etc. 3 5 6 7 8 Research 3 9 10 11 has suggested placental growth factor (PlGF), soluble Flt-1 (sFlt-1), pregnancy-associated plasma protein A (PAPP-A) and ultrasound measurements as clinical factors that are significant in signaling an increase in the risk of PE.…”

Prevalence and Pregnancy Outcome of Preeclampsia Using International Society for the Study of Hypertension in Pregnancy 2018 Criteria in Comparison to 2001

“…The criteria of PE have been evolving for many decades, the reason being the ability of any criteria to detect the clinical syndromes associated with it. 18 Though hypertension with proteinuria has been the commonly practised and time-tested criteria of PE, many scientific researchers have focused on other factors for diagnosis and prediction of adverse outcomes. The latest criteria recommended is one by ISSHP 2018, which has been endorsed by many societies and is currently advocated to be used in diagnosing PE.…”