Utility of endometrial multi-vessel blood flow ultrasound parameters in predicting pregnancy outcomes

Research participants

This study enrolled women in the preconception period who visited Wenzhou People’s Hospital between June 2020 and June 2023. According to predefined inclusion and exclusion criteria, a total of 205 eligible women were included. Participants were categorized into either the pregnancy group (64 cases) or non-pregnancy group (141 cases) based on whether they achieved natural conception during follow-up. Pregnancy confirmation required ultrasound verification of intrauterine gestational sac presence, with subsequent follow-up observations continuing through 3 months of gestation.

Inclusion criteria:

1. Females aged 24–40 years;

2. Healthy women preparing for pregnancy with regular menstrual cycles (27–32 days) and normal ovulation confirmed by monitoring;

3. Normal ovarian function;

4. Presence of mature follicle development and confirmed normal ovulation;

5. Endometrial thickness≥7 mm at ovulation;

6. No use of sex hormone-related medications;

7. No history of adverse lifestyle factors (e.g., smoking, alcohol abuse);

8. Signed informed consent.

Exclusion criteria:

1. Uterine anomalies (e.g., bicornuate uterus, septate uterus);

2. Endometrial polyps, submucosal fibroids, or endometriosis;

3. History of intrauterine procedures, intrauterine adhesions, or cervical stenosis;

4. History of miscarriage, fetal demise, infertility, or related medications;

5. Active reproductive tract infections;

6. Thyroid dysfunction or severe endocrine disorders;

7. Abnormal semen parameters in the male partner or unilateral/bilateral fallopian tube obstruction.

This study received approval from the Ethics Committee of the hospital (Ethics no.: KY-202410-005q). All participants provided their signed informed consent forms prior to undergoing an ultrasound examination.

Research methods

Ultrasound examinations were conducted with Samsung Medison HERA W10 diagnostic equipment (NMPA Registration No. 20192060548, 50/60 Hz) equipped with a 5–9 MHz transvaginal probe. All the selected participants began to monitor the follicles by ultrasound on the 10th to 12th day of their menstrual cycle. They were instructed to recheck the follicles every 1–2 days according to the actual growth rate of the follicles until ovulation was confirmed to be normal. After that, the couples were instructed to have sexual intercourse. Seven to nine days post-ovulation, a transvaginal three-dimensional ultrasound examination was conducted to assess uterine and endometrial conditions. All examinations were performed by the same ultrasound physician with over five years of work experience and appropriate training, using the same equipment model and ensuring optimal image clarity.

All of the following measurements were taken three times to obtain an average value:

(1) Endometrial thickness: Endometrial thickness was measured in the sagittal plane of the uterus, approximately 1 cm from the top of the uterine cavity.

(2) Uterine artery blood flow parameters: Adjust the color gain and speed range appropriately at a point 2 cm lateral to the junction of the cervix and the uterine body until the main uterine artery blood flow signal can be clearly identified. The sampling volume is 2 mm, the sampling angle is less than 30°, and five consecutive and stable blood flow spectra are collected. The systolic/diastolic ratio (S/D), pulsatility index (PI), and resistance index (RI) of both the left and right uterine arteries were measured and averaged.

(3) Endometrial volume and power Doppler blood flow assessment (Figure 1): This evaluation must be performed using a 3D power Doppler ultrasound system (recommended with an intracavitary probe≥12 MHz). Prior to operation, complete the following standardized presets: adjust the flow velocity scale to 4–8 cm/s, pulse repetition frequency (PRF) of 0.3–0.9 kHz, Doppler frequency of 5–7 MHz, and preset a 50–100 Hz wall filter. The patient should be positioned in a modified lithotomy position (hip abduction≥120°), maintain moderate bladder filling 30 min before the examination, and be instructed to perform end-expiratory breath-holding during the data acquisition window to minimize respiratory artifacts. During 3D volume sampling, ensure the insonation angle between the ultrasound beam and blood flow direction is≤30°, employ a 120°scanning angle to cover the entire endometrial layer and a 2 mm range at the myometrial junction, select a medium frame rate (18–22 Hz), and activate automatic angle correction. Quantitative analysis is conducted using the VOCAL system with a six-plane segmentation method: set a 30 dB volume threshold to calculate endometrial volume (V), derive the flow index (FI) based on color pixel intensity integration using a 0–100 scale, determine the vascularization index (VI) through blood flow distribution density percentage, and generate the standardized vascularization-flow index (VFI). Quality control protocols require triplicate measurements averaged for final reporting, with permissible error ranges of V≤5 %, VI≤10 %, and FI≤8 %, while ensuring thermal index (TI) ≤0.4 and mechanical index (MI) ≤0.7 during single-scan durations≤90 s. All parameters must be synchronized to the mid-diastolic phase of the cardiac cycle using R-wave triggering. For special cases (e.g., intrauterine adhesions, polyps), activate high-frequency harmonic imaging (H-Flow mode) and supplement with coronal plane scanning sequences.

Figure 1:

Endometrial volume and endometrial power Doppler blood flow.

(4) Microvascular imaging of the endometrium (Figure 2): The mid-sagittal plane was selected to ensure a clear view of the endometrium, and the MV-Flow function of the device was activated. The probe was maneuvered laterally until both uterine angles were no longer visible in the same plane, and at this point, dynamic microvascular flow images of the longitudinal section of the endometrium were captured. Based on the recorded dynamic images, manual tracing was performed every 5 to 10 frames to measure the VI of the endometrium per unit area. This process was repeated 7 to 12 times to calculate an average, thus determining the endometrial VI_MV per unit area.

Figure 2:

Endometrial microvascular imaging.

Statistical analysis

SPSS 26.0 was used for statistical analyses. Quantitative data were tested for normality prior to further analysis. Data conforming to a normal distribution were expressed as the mean ± standard deviation ( x ¯ ± S ), and comparisons between groups were performed using the independent samples t-test. For data not conforming to a normal distribution, the median (interquartile range) was used, and the Mann–Whitney U test was used for intergroup comparisons. Spearman correlation analysis was employed to explore the relationships among various indicators. Receiver operating characteristic (ROC) curves were used to evaluate the predictive efficacy of each indicator for determining pregnancy results. The significance threshold was set at a p-value of<0.05 for determining statistical significance.

Comparison of age and endometrial ultrasound evaluation indicators between the pregnancy and non-pregnancy groups

Independent samples t-tests were used to compare patient age, body mass index (BMI), FI, left PI, right PI, and mean PI between the pregnancy group and non-pregnancy group. There were no statistically significant differences in age and BMI between the two groups (p>0.05). However, FI was significantly higher in the pregnancy group than that in the non-pregnancy group, while left PI, right PI, and mean PI were significantly lower, with differences between the two groups in all these indicators being statistically significant (p<0.05).

The results of the Mann–Whitney U tests for endometrial thickness, endometrial volume, VI, VFI, VI_MV, left S/D, right S/D, mean S/D, left RI, right RI, and mean RI between the pregnancy group and non-pregnancy group showed no statistically significant differences in endometrial thickness and endometrial volume between the two groups (p>0.05). However, VI, VFI, and VI_MV of women in the pregnancy group were significantly higher than those in the non-pregnancy group, while left S/D, right S/D, mean S/D, left RI, right RI, and mean RI were significantly lower than those in the non-pregnancy group, with statistically significant differences (p<0.05) (Table 1).

Correlation analysis of positive endometrial ultrasound evaluation indicators

ROC curve analysis of VI, VFI, FI, and VI_MV in predicting pregnancy

ROC curve analysis (Figure 3) showed that the area under the ROC curve (AUC) values for endometrial power Doppler ultrasound parameters in predicting pregnancy were 0.707 for VI, 0.637 for FI, 0.761 for VFI, and 0.869 for VI_MV, with all p-Values<0.05, indicating that these indicators had statistically significant pregnancy predictive capacity. The analysis revealed a pregnancy prediction threshold of 2.509 for VI, with a sensitivity of 90.6 % and a specificity of 51.8 %. For FI, the pregnancy prediction threshold was 21.579, with a sensitivity of 79.7 % and a specificity of 46.1 %. The threshold for predicting pregnancy for VFI was 0.728, with a sensitivity of 96.9 % and a specificity of 53.9 %. VI_MV had a threshold of 9.150, with a sensitivity of 92.2 % and a specificity of 68.8 % (Table 5).

Figure 3:

ROC curves of various parameters in predicting pregnancy outcomes.