Uterine Artery Doppler Disturbance Index Is Associated with Endometrial Perfusion and Clinical Pregnancy in IVF / ICSI

Objectives Uterine artery Doppler is widely used to gauge endometrial receptivity in In Vitro Fertilization (IVF)/Intracytoplasmic Sperm injection (ICSI), but a single pulsatility index (PI) measurement from bilateral uterine arteries at a single time point captures neither the full extent of inter‐side asymmetry nor the cycle‐to‐cycle variability that may better reflect a hostile peri‐implantation environment. We defined a Uterine Artery Doppler Disturbance index (UADD‐PI and UADD‐RI) as the coefficient of variation of bilateral pulsatility and resistance indices measured across 3 consecutive cycles and tested its association with 3‐D power‐Doppler endometrial perfusion and clinical pregnancy in a retrospective single‐center cohort. Methods We retrospectively reviewed the electronic medical records of 437 consecutive first IVF/ICSI cycles (173 fresh embryo transfer, 264 frozen embryo transfer) treated at a single tertiary reproductive medicine center between January 2023 and April 2026. As part of routine pre‐transfer evaluation at our center, bilateral uterine artery PI and RI were obtained on hCG‐trigger day (fresh) or progesterone day 5 (FET), and on the corresponding day of 2 preceding natural or stimulated cycles. UADD was calculated as 100 × SD/mean of the 6 measurements obtained from both uterine arteries across 3 consecutive cycles (2 sides × 3 cycles). Endometrial vascularization index (VI), flow index (FI), and vascularization‐flow index (VFI) were obtained by 3‐D power‐Doppler with VOCAL and retrieved from the picture‐archiving system. Primary outcome was clinical pregnancy; ongoing pregnancy and live birth were secondary. Multivariable logistic regression, restricted cubic splines, and 1000‐resample bootstrap internal validation were used. Results UADD‐PI was inversely correlated with VI (Spearman ρ = −0.46), FI ( ρ = −0.28), and VFI ( ρ = −0.56; all p < .001). Each 5‐percentage‐point increment in UADD‐PI was independently associated with lower clinical pregnancy (adjusted OR 0.75, 95% CI 0.64–0.89, p < .001) after controlling for age, BMI, log‐AMH, endometrial thickness, embryo grade, cycle type, and 3‐D VI. Adding UADD‐PI to a baseline model raised the AUC from 0.657 to 0.762 (Δ 0.105, p < .001); calibration was acceptable (Hosmer–Lemeshow p = .349; calibration slope 0.94; Brier 0.173 versus null Brier 0.207). The dose–response relationship was monotonic and statistically linear (non‐linearity LR χ 2 = 1.44, p = .487). Conclusions A bilateral, cycle‐aggregated UADD index captures information that a single PI measurement misses and is independently associated with endometrial perfusion and pregnancy outcome. Because the analysis is retrospective and single‐center, these findings are hypothesis‐generating; UADD is best viewed as a potential complement to—rather than a replacement for—conventional Doppler indices and requires external prospective validation before any clinical application.