This retrospective study was approved by the Institutional Ethics Committee of Xiangyang No. 1 People's Hospital, Hubei University of Medicine, Hubei, China (Approval No. PHHUM/C25/H/19). Written informed consent for the use of anonymized clinical and echocardiographic data for research purposes was obtained at the time of hospitalization. The study was conducted in accordance with the ethical principles of the Declaration of Helsinki and relevant national guidelines for research involving human subjects.

A total of 240 elderly pregnant women diagnosed with pregnancy-induced hypertension syndrome (PIHS) and treated at Xiangyang No. 1 People’s Hospital between February 2019 and June 2022 were retrospectively analyzed. Participants were divided according to pregnancy outcomes into a favorable outcome group (n = 183) and an adverse outcome group (n = 57). A schematic overview of the study design, screening, and analysis procedures is illustrated in Figure 1.

Participants were enrolled based on strict eligibility conditions to ensure homogeneity of the study population and minimize confounding bias. Inclusion criteria were as follows: pregnant women aged over 35 years, with a singleton pregnancy, who had been clinically diagnosed with pregnancy-induced hypertension syndrome (PIHS) in accordance with international diagnostic criteria for hypertensive disorders of pregnancy (20). Only patients with complete clinical records and echocardiographic examination data available from the same hospitalization period were included. All diagnoses and echocardiographic assessments were confirmed by two senior obstetricians and a certified sonographer to maintain diagnostic accuracy and reproducibility.

Exclusion criteria were established to minimize confounding and ensure data reliability. Women were excluded if they had a history of psychiatric illness or cognitive impairment that could affect the accuracy of clinical data. Patients with severe uterine fibroids, ovarian cysts, or other gynecological conditions that could interfere with fetal growth or echocardiographic imaging were also excluded. Likewise, individuals with chronic infectious or systemic diseases, including HIV/AIDS, syphilis, chronic viral hepatitis, or autoimmune disorders, were excluded to reduce the influence of systemic inflammation on pregnancy outcomes. Patients with known structural heart disease, cardiomyopathy, congenital heart disease, significant valvular disease, or other pre-existing cardiac conditions affecting cardiac structure or function were excluded. Additional exclusion criteria included multiple gestations and gestational diabetes mellitus.

All postpartum women underwent echocardiographic examination within 1–3 days after delivery or pregnancy termination. A Siemens ACUSON Oxana 2 color Doppler ultrasound system (probe frequency 2–4 MHz) was used. Examinations were performed in a quiet setting with the patient in the left lateral decubitus position by an experienced sonographer who was blinded to pregnancy outcomes. Each patient underwent a single standardized echocardiographic examination.

Standard echocardiographic parameters were acquired as follows:

Structural indices: Left atrial diameter (LAD), left ventricular end-diastolic diameter (LVDd), and interventricular septal end-diastolic thickness (IVSTd) measured from the parasternal long-axis view.

Transmitral flow velocities: Early diastolic (E wave) and atrial contraction (A wave) peak velocities recorded from the apical four-chamber view; the E/A ratio was calculated.

Tissue Doppler imaging (TDI): Early diastolic mitral annular velocity (E′) measured at the septal annulus, and the E/E′ ratio derived.

Pulmonary venous flow: Pulsed-wave Doppler sampling was performed with the sample volume positioned 1–2 cm proximal to the orifice of the upper pulmonary vein (right or left) in the apical four-chamber view to obtain systolic (S wave), diastolic (D wave), and atrial reversal (Ar wave) velocities. Measurements were acquired at end-expiration, and the mean value of three consecutive cardiac cycles was used for analysis.

Myocardial performance index (Tei index): Obtained from apical four-chamber view by measuring isovolumetric contraction time (IVCT), ejection time (ET), and isovolumetric relaxation time (IVRT), calculated as Tei index = (IVRT + IVCT) / ET.

Cardiac output (CO): Derived from stroke volume (SV) × heart rate (HR), where SV = velocity–time integral (VTI) of the left ventricular outflow tract × cross-sectional area.

Ejection fraction (EF): Calculated from end-diastolic (EDV) and end-systolic (ESV) ventricular volumes using EF = (EDV − ESV) / EDV × 100%.

Each measurement was repeated across three cardiac cycles, and mean values were used for analysis.

Pregnancy outcomes were classified according to established obstetric definitions based on final delivery outcomes. Adverse pregnancy outcomes were predefined as events occurring at or beyond 20 weeks of gestation, including stillbirth (intrauterine fetal death ≥20 weeks), fetal demise, preterm birth (<37 weeks), and fetal growth restriction (estimated fetal weight <10th percentile for gestational age). Participants without any of these events were classified as having a favorable outcome. Echocardiographic assessments were performed postpartum, within 1–3 days after delivery or pregnancy termination, and the measured cardiac parameters were analyzed in relation to the documented pregnancy outcomes. Baseline demographic and clinical characteristics of the two groups are summarized in Table 1.

Demographic and obstetric data collected included maternal age, body mass index (BMI), systolic and diastolic blood pressures (SBP, DBP), gravidity and parity, years of education, and place of residence. Medical history variables included a history of pre-existing hypertension, diabetes mellitus, nephritis, and family history of hypertension (maternal or paternal). Information regarding antihypertensive treatment during pregnancy was extracted from medical records, including the use of standard obstetric antihypertensive medications administered according to institutional protocols. All clinical variables were obtained retrospectively from electronic medical records and obstetric charts and were recorded during the same hospitalization period as the echocardiographic examination.

Data were analyzed using SPSS version 23.0 (IBM Corp., Armonk, NY, USA). Continuous variables were expressed as mean ± standard deviation (SD) and compared between groups using independent-samples t-tests. Categorical variables were presented as frequencies and percentages and analyzed using the χ² test. Variables with p < 0.05 in univariate analyses were considered candidate predictors and entered into multivariable logistic regression models to identify independent risk factors for adverse pregnancy outcomes. Antihypertensive therapy during pregnancy was not included in the multivariable regression model because of its collinearity with blood pressure variables and to avoid further reduction in the events-per-variable ratio, which could increase the risk of model overfitting. To reduce the risk of overfitting, only variables with clear clinical relevance and statistical significance were retained in the final multivariable model. Baseline confounders, including history of hypertension, nephritis, and family history of hypertension, were incorporated as covariates. Model performance was evaluated using the Hosmer–Lemeshow goodness-of-fit test and Nagelkerke R². Receiver operating characteristic (ROC) curve analysis was performed to assess the predictive accuracy of echocardiographic parameters. A two-sided p-value < 0.05 was considered statistically significant.

Among the 240 elderly pregnant women with pregnancy-induced hypertension syndrome (PIHS), 183 experienced favorable pregnancy outcomes, whereas 57 had adverse outcomes. As summarized in Table 1, there were no significant differences between the two groups in terms of age, body mass index (BMI), systolic or diastolic blood pressure, history of multiple pregnancies, parity, years of education, or place of residence (all p > 0.05). However, patients in the adverse outcome group exhibited a significantly higher prevalence of pre-existing hypertension, nephritis, and a family history of hypertension (p < 0.05). These findings indicate that hereditary and chronic hypertensive conditions may predispose elderly pregnant women with PIHS to unfavorable pregnancy outcomes.

Echocardiographic evaluation of structural cardiac indices revealed no statistically significant differences between the two groups in left ventricular end-diastolic diameter (LVDd), interventricular septal thickness (IVSTd), or left atrial diameter (LAD) (p > 0.05; Table 2). This suggests that the baseline chamber dimensions and wall thickness were not primary determinants of adverse outcomes in this cohort. The absence of structural differences indicates that functional rather than morphological cardiac alterations may contribute more prominently to PIHS-related complications.

Functional echocardiographic indices differed significantly between the two groups (Table 3). The adverse outcome group demonstrated markedly lower ejection fraction (EF), stroke volume (SV), cardiac output (CO), E/A ratio, and early diastolic mitral annular velocity (E′) compared with the favorable outcome group (all p < 0.05). Conversely, E/E′ ratio and left atrial volume index (LAVI) were significantly higher in the adverse outcome group (p < 0.05). These alterations reflect diastolic dysfunction, impaired myocardial relaxation, and increased left atrial pressure load, which collectively suggest subclinical cardiac dysfunction as a contributing factor to poor maternal and fetal outcomes in elderly PIHS patients.

As shown in Table 4, analysis of pulmonary venous flow dynamics revealed that the adverse outcome group had significantly lower pulmonary vein S-wave and D-wave velocities, but higher Ar-wave velocity and Tei index values compared with the favorable outcome group (all p < 0.05). These changes indicate impaired left atrial compliance and altered ventricular filling dynamics, implying that abnormalities in pulmonary venous flow patterns and an increased myocardial performance index are potential early markers of adverse pregnancy outcomes in PIHS.

To identify independent determinants of adverse pregnancy outcomes, variables with p < 0.05 in univariate analysis were entered into a multivariate binary logistic regression model (Table 5). A history of hypertension, nephritis, and a family history of hypertension emerged as significant independent clinical risk factors (p < 0.05). Among echocardiographic parameters, higher ejection fraction (EF), cardiac output (CO), E/A ratio, and early diastolic mitral annular velocity (E′) were independently associated with a reduced risk of adverse pregnancy outcomes (OR < 1), whereas higher E/E′ ratio and left atrial volume index (LAVI) were independently associated with an increased risk of adverse outcomes (OR > 1), indicating impaired systolic and diastolic cardiac function in patients with unfavorable outcomes. The direction of these associations is consistent with the between-group comparisons shown in Tables 3, 4. The final model included only the most significant clinical and echocardiographic variables to minimize overfitting. Antihypertensive therapy during pregnancy was not entered into the model because of collinearity with blood pressure variables and concerns regarding the events-per-variable ratio. It demonstrated good calibration (Hosmer–Lemeshow p = 0.63) and strong explanatory power (Nagelkerke R² = 0.726), correctly classifying 88.3% of cases.

Receiver operating characteristic analysis was performed to evaluate the diagnostic utility of echocardiographic indices in predicting adverse pregnancy outcomes (Figures 2A–E). The area under the curve (AUC) values indicated strong predictive ability for several parameters: EF (AUC = 0.894, 95% CI 0.828–0.952), E/A ratio (AUC = 0.877, 95% CI 0.803–0.951), SV (AUC = 0.811, 95% CI 0.723–0.898), CO (AUC = 0.780, 95% CI 0.684–0.875), E′ (AUC = 0.893, 95% CI 0.826–0.959), and E/E′ ratio (AUC = 0.865, 95% CI 0.786–0.943). For pulmonary vein flow parameters, S-wave (AUC = 0.896, 95% CI 0.832–0.960), D-wave (AUC = 0.863, 95% CI 0.784–0.941), Ar-wave (AUC = 0.824, 95% CI 0.733–0.914), and Tei index (AUC = 0.791, 95% CI 0.695–0.888) also exhibited good discriminative value. Importantly, a combined diagnostic model integrating all echocardiographic parameters achieved an AUC of 0.925 (95% CI 0.872–0.989), reflecting excellent reflecting excellent discriminatory performance in distinguishing patients with adverse versus favorable pregnancy outcomes.

Representative echocardiographic images are shown in Figures 3A–E. Figure 3A shows a Doppler image of the mitral inflow velocity spectrum in a patient with an adverse pregnancy outcome, demonstrating attenuated E-wave and increased A-wave amplitude, findings suggestive of diastolic dysfunction. Figure 3B displays the corresponding image from a patient with a favorable pregnancy outcome, showing preserved diastolic filling and a normal E/A ratio. Figure 3C illustrates the echocardiographic assessment of left ventricular and atrial dimensions in a patient with an adverse outcome, revealing mild chamber dilation and increased left atrial volume. Figure 3D presents the same parameters in a patient with a favorable outcome, demonstrating normal chamber geometry and preserved systolic performance. Figure 3E provides a comparative composite image emphasizing the distinct functional patterns between favorable and adverse outcome groups. These findings collectively underscore the diagnostic relevance of functional echocardiographic indices—rather than purely structural measurements—in predicting pregnancy outcomes among elderly patients with pregnancy-induced hypertension syndrome (PIHS).