Prediction of twin pregnancy preeclampsia based on clinical risk factors, early pregnancy serum markers, and uterine artery pulsatility index

Objectives: To investigate whether a combination of clinical risk factors, early pregnancy serum markers, and uterine artery pulsatility index (UTPI) can be used to predict twin preeclampsia (PE). Methods: This case control study included women with twin pregnancies who had undergone obstetrics treatments and gave birth at the Huzhou Maternity and Child Health Care Hospital from October 2018 to November 2020. Patients with PE comprised study group, and patients without PE comprised control group based on selection criteria and a 1:1 ratio. Statistical analysis was performed using clinical risk factors, early pregnancy serum markers, and UTPIs, and the area under the receiver operating curve (AUC. Sensitivity, and the specificity of different combinations of these variables were calculated to predict PE in women with twin pregnancy. Results: Logistic regression analysis revealed four independent predictors for the onset of PE during twin pregnancies: first delivery (OR, 7.51; P=0.045), conception method (OR, 7.11; P=0.036), beta-HCG level (per SD OR, 2.73; P=0.026), and UTPI (OR, 0.17; P=0.043). First-delivery and IVF pregnancy methods both lead to a 7-fold increase in the PE risk during twin pregnancies. Every one sigma (standard deviation) increase in the beta-HCG level led to a 2.73-fold increase in the PE risk. Every UTPI increment by 1.0 reduces the risk of PE by 83%. The prediction efficiencies were based on an AUC of 0.837, a sensitivity of 69%, and a specificity of 92% for the clinical risk factors; an AUC of 0.800, a sensitivity of 81%, and specificity of 78% for the beta-HCG level, and an AUC of 0.814, a sensitivity of 88%, and a specificity of 65% for the UTPI. AUC was 0.928, sensitivity 85%, and a specificity 88% after applying the three types of indicators together for prediction. Conclusions: By combining early pregnancy serum markers (beta-HCG), and UTPI, the predictive value for PE during twin pregnancy is improved together with its sensitivity and specificity.

Automated summary

This study successfully predicted the occurrence of preeclampsia in twin pregnancies by combining clinical risk factors, early pregnancy serum markers, and uterine artery pulsatility index. Independent predictors identified for preeclampsia included first delivery, IVF conception, beta-HCG level, and UTPI. The combination of these factors improved the predictive value, sensitivity, and specificity for preeclampsia in twin pregnancies.