Search

Objectives: To evaluate the performance of an artificial intelligence (AI) and machine learning (ML) model for first-trimester screening for pre-eclampsia in a large Asian population., Methods: This was a secondary analysis of a multicenter prospective cohort study in 10 935 participants with singleton pregnancies attending for routine pregnancy care at 11-13 +6  weeks of gestation in seven regions in Asia between December 2016 and June 2018. We applied the AI+ML model for the first-trimester prediction of preterm pre-eclampsia (<37 weeks), term pre-eclampsia (≥37 weeks), and any pre-eclampsia, which was derived and tested in a cohort of pregnant participants in the UK (Model 1). This model comprises maternal factors with measurements of mean arterial pressure, uterine artery pulsatility index, and serum placental growth factor (PlGF). The model was further retrained with adjustments for analyzers used for biochemical testing (Model 2). Discrimination was assessed by area under the receiver operating characteristic curve (AUC). The Delong test was used to compare the AUC of Model 1, Model 2, and the Fetal Medicine Foundation (FMF) competing risk model., Results: The predictive performance of Model 1 was significantly lower than that of the FMF competing risk model in the prediction of preterm pre-eclampsia (0.82, 95% confidence interval [CI] 0.77-0.87 vs. 0.86, 95% CI 0.811-0.91, P = 0.019), term pre-eclampsia (0.75, 95% CI 0.71-0.80 vs. 0.79, 95% CI 0.75-0.83, P = 0.006), and any pre-eclampsia (0.78, 95% CI 0.74-0.81 vs. 0.82, 95% CI 0.79-0.84, P < 0.001). Following the retraining of the data with adjustments for the PlGF analyzers, the performance of Model 2 for predicting preterm pre-eclampsia, term pre-eclampsia, and any pre-eclampsia was improved with the AUC values increased to 0.84 (95% CI 0.80-0.89), 0.77 (95% CI 0.73-0.81), and 0.80 (95% CI 0.76-0.83), respectively. There were no differences in AUCs between Model 2 and the FMF competing risk model in the prediction of preterm pre-eclampsia (P = 0.135) and term pre-eclampsia (P = 0.084). However, Model 2 was inferior to the FMF competing risk model in predicting any pre-eclampsia (P = 0.024)., Conclusion: This study has demonstrated that following adjustment for the biochemical marker analyzers, the predictive performance of the AI+ML prediction model for pre-eclampsia in the first trimester was comparable to that of the FMF competing risk model in an Asian population., (© 2024 The Authors. International Journal of Gynecology & Obstetrics published by John Wiley & Sons Ltd on behalf of International Federation of Gynecology and Obstetrics.)

Objectives: To (i) evaluate the applicability of the European-derived biomarker multiples of the median (MoM) formulae for risk assessment of preterm pre-eclampsia (PE) in seven Asian populations, spanning the east, southeast and south regions of the continent, (ii) perform quality-assurance (QA) assessment of the biomarker measurements and (iii) establish criteria for prospective ongoing QA assessment of biomarker measurements., Methods: This was a prospective, non-intervention, multicenter study in 4023 singleton pregnancies, at 11 to 13 + 6 weeks' gestation, in 11 recruiting centers in China, Hong Kong, India, Japan, Singapore, Taiwan and Thailand. Women were screened for preterm PE between December 2016 and June 2018 and gave written informed consent to participate in the study. Maternal and pregnancy characteristics were recorded and mean arterial pressure (MAP), mean uterine artery pulsatility index (UtA-PI) and maternal serum placental growth factor (PlGF) were measured in accordance with The Fetal Medicine Foundation (FMF) standardized measurement protocols. MAP, UtA-PI and PlGF were transformed into MoMs using the published FMF formulae, derived from a largely Caucasian population in Europe, which adjust for gestational age and covariates that affect directly the biomarker levels. Variations in biomarker MoM values and their dispersion (SD) and cumulative sum tests over time were evaluated in order to identify systematic deviations in biomarker measurements from the expected distributions., Results: In the total screened population, the median (95% CI) MoM values of MAP, UtA-PI and PlGF were 0.961 (0.956-0.965), 1.018 (0.996-1.030) and 0.891 (0.861-0.909), respectively. Women in this largely Asian cohort had approximately 4% and 11% lower MAP and PlGF MoM levels, respectively, compared with those expected from normal median formulae, based on a largely Caucasian population, whilst UtA-PI MoM values were similar. UtA-PI and PlGF MoMs were beyond the 0.4 to 2.5 MoM range (truncation limits) in 16 (0.4%) and 256 (6.4%) pregnancies, respectively. QA assessment tools indicated that women in all centers had consistently lower MAP MoM values than expected, but were within 10% of the expected value. UtA-PI MoM values were within 10% of the expected value at all sites except one. Most PlGF MoM values were systematically 10% lower than the expected value, except for those derived from a South Asian population, which were 37% higher., Conclusions: Owing to the anthropometric differences in Asian compared with Caucasian women, significant differences in biomarker MoM values for PE screening, particularly MAP and PlGF MoMs, were noted in Asian populations compared with the expected values based on European-derived formulae. If reliable and consistent patient-specific risks for preterm PE are to be reported, adjustment for additional factors or development of Asian-specific formulae for the calculation of biomarker MoMs is required. We have also demonstrated the importance and need for regular quality assessment of biomarker values. Copyright © 2019 ISUOG. Published by John Wiley & Sons Ltd., (Copyright © 2019 ISUOG. Published by John Wiley & Sons Ltd.)

Introduction: First trimester screening (FTS) is a validated screening tool that has been shown to achieve detection rates of 84%-90% for trisomies 21, 18 and 13. However, its effectiveness for different maternal ages has not been assessed. The present study aimed to assess the performance of FTS in an Asian population, and to compare its effectiveness in older (≥ 35 years) and younger (< 35 years) women. The potential use of noninvasive prenatal test (NIPT) as a contingent screening test is also examined., Methods: Data on cases of FTS performed on singleton pregnancies over a six-year period was collated from two Singapore maternal centres, National University Hospital and Singapore General Hospital. Cases that had a 1:250 risk of trisomy were considered to be screen-positive. Pregnancy outcomes were obtained from birth records or karyotype test results., Results: From 10,289 FTS cases, we obtained a sensitivity of 87.8%, a specificity of 97.6%, a false positive rate of 2.4% and a false negative rate of 0.06% for the detection of aneuploidy. The overall detection rate for trisomy 21 was 86.5%-85.7% for older women and 87.5% for younger women. The mean number of invasive tests required per case of trisomy 21 was 9.3 in younger women, 8.6 in older women and 13.5 in women with intermediate risk (1:250-1,000)., Conclusion: While the performance of FTS was similar in younger and older women, more invasive procedures were required to diagnose trisomy 21 in women with intermediate risk. It may be advantageous to offer contingent NIPT to this group of women to reduce the risk of iatrogenic fetal loss.