Gestational hypertension: fetal and placental characterization using MRI and ultrasound

This study aims to integrate MRI and ultrasound (US) assessments of fetal and placental growth and development in cases of gestational hypertension (GH), evaluating their potential utility for clinical management and prognostic evaluation.

A retrospective study analyzed 84 pregnant women (39 with GH, 45 controls) to compare fetal and placental parameters using MRI and US. Parameters included placental thickness (PT), relative signal intensity(rSI), relative apparent diffusion coefficient(rADC), lung-to-liver signal intensity ratio(LLSIR), and various indices of the umbilical and middle cerebral arteries. Differences between the GH and control groups were statistically examined. Binary logistic regression identified independent risk factors for GH-related adverse outcomes, leading to the creation of predictive models (MRI-only, US-only, and MRI+US). ROC curves assessed each model's predictive performance, and the DeLong test compared the area under the curve (AUC) among models.

In comparison to the control group, the GH group showed: 1) Increased PT thickness and decreased rSI and rADC values (p < 0.05). 2) Higher vascular resistance, as indicated by significantly elevated systolic-diastolic ratio (UA-S/D ratio), pulsatility index (UA-PI), and resistance index (UA-RI) of the umbilical artery (all p < 0.05), was observed. 3) Shorter biparietal diameter (BPD, p < 0.05) and reduced LLSIR (p < 0.001). 4) UA-PI, UA-S/D, PT, rADC, and rSI were identified as independent risk factors for GH through binary logistic regression. ROC analysis demonstrated that the MRI+US model (AUC = 0.89, sensitivity = 85%, specificity = 93%) significantly improved the prediction accuracy, with the combined model's AUC value being significantly different from other models (p < 0.05).

The integration of MRI and US offers a robust evaluation of placental and fetal growth and development in GH patients. Furthermore, the combined utilization of MRI and US techniques exhibits enhanced accuracy in predicting adverse pregnancy outcomes.