Reference Ranges for Uterine Artery Pulsatility Index during the Menstrual Cycle: A Cross-Sectional Study

BACKGROUND: Cyclic endometrial neoangiogenesis contributes to changes in local vascular patterns and is amenable to non-invasive assessment with Doppler sonography. We hypothesize that the uterine artery (UtA) impedance, measured by its pulsatility index (PI), exhibits a regular pattern during the normal menstrual cycle. Therefore, the main study objective was to derive normative new day-cycle-based reference ranges for the UtA-PI during the entire cycle from days 1 to 34 according to the isolated time effect and potential confounders such as age and parity. METHODS: From January 2009 to December 2012, a cross-sectional study of 1,821 healthy women undergoing routine gynaecological ultrasound was performed. The Doppler flow of the right and left UtA-PI was studied transvaginally by colour and pulsed Doppler imaging. The mean right and left values and the presence or absence of a bilateral protodiastolic notch were recorded. Reference intervals for the PI according to the cycle day were generated by classical linear regression. RESULTS: The majority of patients (97.5%) presented unilateral or bilateral UtA notches. The crude 5th, 50th, and 95th reference percentile curves of the UtA-PI at 1–34 days of the normal menstrual cycle were derived. In all curves, a progressive significant decrease occurred during the first 13 days, followed by an increase and recovery in the UtA-PI. The adjusted 5th, 50th, and 95th reference percentile curves for the effects of age and parity were also obtained. These two conditions generated an approximately identical UtA-PI pattern during the cycle, except with small but significant reductions at the temporal extremes. CONCLUSIONS: The median, 5th, and the 95th percentiles of the UtA-PI decrease during the first third of the menstrual cycle and recover to their initial values during the last two thirds of the cycle. The rates of decrease and recovery depend significantly on age and parity.