Due to the value of UtA impedance assessments employing Doppler ultrasound, the reference ranges for the mean PI during the normal menstrual cycle (Guedes-Martins et al. 2015) and from 6 to 41 weeks in uneventful pregnancies were established and clearly revealed a general progressive gestational age-related decrement (Gómez et al. 2008; Guedes-Martins et al. 2014a). This trend has been attributed to the major changes in the placental bed that modify the properties of the UtA from a resistance vessel into a capacitance vessel (Osol and Mandala 2009).

Beyond the reference ranges for an uneventful pregnancy, a number of studies at different gestational ages have shown that UtA impedance is able to provide important predictive information about serious obstetrical disorders, such as PE. Notably, in women with gestational hypertension just before term, the incidence of elevated UtA impedance was found to be as high as 68 % and to rise to 89 % when associated with PE (Frusca et al. 2003). Other studies also observed a high incidence of abnormal UtA-PI, although at lower values, likely reflecting the different criteria for inclusion (Li et al. 2005; Meler et al. 2010).

In a group of unselected pregnant women at 22–24 weeks of gestation (Papageorghiou et al. 2001), the enhanced UtA-PI was reported to show a 69 % sensitivity for the appearance of PE with intra-uterine growth restriction (IUGR) in subsequent weeks, a value that increased to 83 % when the criteria included protodiastolic notch persistence. However, although the detection rate of PE as a result of enhanced UtA-PI was already shown to be superior to the patient’s epidemiological data detection rate (Papageorghiou et al. 2005), additional conditions had to be met. In fact, other reports indicated that the correlation between abnormal UtA impedance at 22 weeks and the establishment of PE was significant only in situations in which the fetal outcome was poor, including IUGR and preterm birth (van den Elzen et al. 1995; Aardema et al. 2004). More recently, top-decile PI values and the presence of bilateral notching were considered to have a good predictive value for an enhanced risk of stillbirth resulting from placental causes (Smith et al. 2007). As a corollary of these studies, an extensive systematic review of 74 studies including 79,547 women (Cnossen et al. 2008) with the intended purpose of evaluating the use of UtA Doppler velocimetry for the prediction of PE, concluded that UtA Doppler ultrasonography was more accurate for the prediction of PE when performed in the second trimester rather than in the first trimester. In addition, the prediction of the overall risk of PE and the risk of severe PE is significantly different in pregnant women with a low risk than in those with a high risk of developing the disease. In the group of low-risk patients, the overall risk of PE was best predicted by the presence of a second-trimester elevation of PI accompanied by UtA notching [sensitivity 23 %, specificity 99 %, positive likelihood ratio (+ve LR) 7.5, −ve LR 0.59]. Also in this low-risk group, the risk of severe PE was best predicted by either the second trimester PI (sensitivity 78 %, specificity 95 %, +ve LR 15.6, −ve LR 0.23) or bilateral notching (sensitivity 65 %, specificity 95 %, +ve LR 13.4, −ve LR 0.37). In contrast, in women at a high risk of developing PE, the overall risk of PE was best predicted by the presence of a second-trimester elevation of PI accompanied by UtA notching (sensitivity 19 %, specificity 99 %, +ve LR 21, −ve LR 0.82). Additionally, the risk of severe PE in high-risk patients was best predicted by second-trimester elevated RI (sensitivity 80 %, specificity 78 %, + ve LR 3.7, −ve LR 0.26). These findings supported the recommendation to employ PI and notching assessment in daily clinical practice (Cnossen et al. 2008).

This progressive increase in the maternal-placental blood flow during gestation is mainly due to vasodilation in part related to increased levels of 17β-estradiol, progesterone, and relaxin (Sprague et al. 2009; Vodstrcil et al. 2012). Additionally, the UtA diameter doubles in size after 20 weeks of pregnancy (Konje et al. 2001). Because blood flow within a vessel increases in proportion to the fourth power of the radius, this slight diameter increase in the UtA produces a significant blood flow capacity increase (Palmer et al. 1988; Mandala and Osol 2012) with a concomitant Doppler velocimetry increment. The downstream fall in vascular resistance leads to circumferential vessel growth, a process that nitric oxide (NO) appears to play a key role in regulating. NO is generated in the endothelium by endothelial nitric oxide synthase (eNOS) and is essential for proper endothelial function and the regulation of vascular tone. For instance, estrogen, placental growth factor (PlGF), and vascular endothelial growth factor (VEGF) augment eNOS and, consequently, NO production (Sprague et al. 2009; Mandala and Osol 2012; Grummer et al. 2009). In addition, increased soluble fms-like tyrosine kinase 1 (sFlt-1) levels inactivate and decrease circulating PlGF and VEGF concentrations and have been recognized as an important factor in PE pathogenesis (Karumanchi et al. 2005). To reverse this trend, the relative vascular insensitivity to infused angiotensin II and norepinephrine (Rosenfeld et al. 2012) also serves to increase the uteroplacental blood supply.

Abnormal UtA-PI and elevated levels of sFlt-1, reduced levels of PlGF, and an increased sFlt-1:PlGF ratio have been reported as potential predictive markers for the development of preeclampsia (Cnossen et al. 2008; Seely and Solomon 2016). However, these markers are not validated for clinical use in cases of superimposed preeclampsia.