Hemodynamic relationship between ophthalmic artery and uterine artery in pre-eclampsia: pulse wave reflection and transmission might provide the missing link.

Abstract

We read with great interest the excellent systematic review and meta-analysis on the use of ophthalmic artery (OA) Doppler in the prediction of pre-eclampsia (PE), by Kalafat et al.1. One key message is that OA Doppler has a standalone predictive value for the development of early-onset PE equivalent to that of uterine artery (UtA) Doppler evaluation. Another key message is that the findings of the review justify efforts to elucidate the hemodynamic relationship between these two seemingly unrelated maternal vessels, OA and UtA, in placental disorder with maternal circulatory dysfunction. Recently, we identified a hemodynamic model of the arterial system based on pulse wave (PW) propagation and reflection2–4 which, applied to pregnancy, may provide insight into the functional interrelationship between OA and UtA5, located opposite each other in the upper and lower body. In severe PE, simultaneous Doppler waveform changes are seen during systole in both vessels: an increased second systolic peak (P2) in the OA Doppler waveform and the appearance of a superposed systolic shoulder in the UtA Doppler waveform. According to this model, both Doppler changes may be interpreted as the result of increased PW reflection, as seen in severe PE with maternal circulatory dysfunction6–8. We would like to share two preliminary observations. Firstly, in the OA Doppler waveform presented in figure S2 of the review1, the peak identified as ‘first diastolic peak’ (marked by a yellow arrow) actually represents a systolic P2, as known from other cerebrovascular Doppler waveforms (figure 9, Masuda et al.4; figure 1, Lockhart et al.9; figure 1, Hirata et al.10; figures 1 and 2, Curtis et al.11; figures 1 and 2, Heffernan et al.12). The timing corresponds to the systolic phase, preceding the appearance of the dicrotic notch. Secondly, the appearance of a systolic shoulder (UtA-S) during the monotone systolic decline of the UtA Doppler waveform, which was classified formerly as a systolic (double) notch and identified as an ominous sign in severe PE, though scarcely referenced in the literature13,14, may actually be interpreted hemodynamically as a result of increased PW re-reflection (Figure 1)15. PWs are the result of cardiac contraction, they propagate along the systemic arterial tree and proceed to the supply arteries of more peripheral vascular territories (e.g. cerebral and uteroplacental circulation). On this transit, PW reflection occurs, thus generating a backward travelling wave16. In physiology, this concept of PW propagation and reflection is well established. Reflection will increase with vasoconstriction and vascular tone and repetitive wave reflections along the systemic arterial tree (between cardiac level and body reflection sites) may occur17,18. PW reflection from the lower body seems to originate mainly in the pelvic region (body reflection16,19) with secondary transmission to the cerebral circulation, thus generating an additional mid-systolic flow augmentation in cerebrovascular Doppler waveforms10–12,20. According to this model, the systolic P2 in the OA Doppler waveform (OA-P2) may be interpreted as a sign of PW reflection with secondary transmission to the cerebral circulation (Figure 1). This effect will be pronounced in PE due to increased vasoconstriction and vascular tone6–8,21. Moreover, in severe PE, the UtA-S is assumed to arise from re-reflection (back-and-forth propagation along the systemic arterial tree) and secondary transmission to the uteroplacental circulation (Figure 1)15. In this case, the induced flow augmentation will not reach the magnitude of a second peak due to PW attenuation during re-reflection. The utility of this model approach is substantiated by the timing of both signs of PW reflection, i.e. the systolic UtA-S coincides with the systolic OA-P25. Furthermore, a decrease of both signs of reflection, UtA-S22,23 and OA-P224, was observed in severe PE after administration of isosorbide dinitrate, a potent nitric oxide donor and nitrovasodilator, obviously due to a transient pharmacological reduction of vasoconstriction and vascular tone. In summary, hemodynamic modelling may reveal a functional relationship between OA and UtA based on increased PW reflection, with simultaneous systolic Doppler waveform changes in severe PE. UtA-S is a waveform feature that is difficult to quantify. In contrast, OA-P2 is visible in normal and pathological conditions and quantifiable throughout pregnancy (using for instance the peak ratio (PR) relative to the main systolic peak (P1) (PR = P2/P1)24, or the (cerebrovascular) flow augmentation index (FAIx = (P2 − ED)/(P1 − ED), where ED is end diastole10,12). Thus, in addition to the predictive value of OA Doppler for the development of early-onset PE, as shown in the systematic review of Kalafat et al., OA Doppler may have the potential to monitor maternal circulatory deterioration in PE.