Arterial pressure, vascular input impedance, and resistance as determinants of pulsatile blood flow in the umbilical artery

Abstract

The flow pulsatility index, the ratio of flow pulse amplitude to mean flow over the cardiac cycle, has been used to quantify pulsatility of blood flow in the umbilical artery. In experiments with fetal sheep, we showed that the flow pulsatility index in the umbilical artery is accurately estimated by the ratio of total umbilico-placental vascular resistance (mean arterial pressure divided by mean umbilical flow) divided by fundamental impedance (umbilical vascular impedance at the heart rate frequency) times the pulsatility index (pulse/mean) of the arterial pressure that drives flow through this bed. The pulsatility index of arterial pressure is primarily determined by upstream factors (e.g. heart rate) whereas fundamental impedance depends primarily on the radius and viscoelastic wall properties of the umbilical artery. An increase in resistance in the microcirculation and/or veins causes proportional changes in the flow pulsatility index because these sites have little influence on fundamental impedance. However, an increase in resistance in the highly vasoactive umbilical arteries has offsetting effects on impedance and resistance; consequently, flow pulsatility changes little even when arterial vasoconstriction markedly reduces mean flow. We conclude that when arterial pressure pulsatility is stable, a change in the flow pulsatility index provides a useful indication of a change in resistance in the microcirculation and/or veins but will not reliably detect a resistance change in the artery.

Introduction

Studies using transcutaneous Doppler ultrasound have identified highly pulsatile blood velocity waveforms in the umbilical artery of many human fetuses with intrauterine growth restriction [1], [2] and congenital anomalies [3]. Furthermore, these abnormal waveforms are highly predictive of adverse perinatal outcome [4], [5]. In the normal human fetus near term, diastolic blood velocity in the umbilical artery is approximately half that of systole. In contrast, the preterm growth restricted fetus can exhibit abnormal waveforms characterized by diastolic velocity that is very low, zero, or may even be negative between heart beats. The more pulsatile nature of these abnormal waveforms has been quantified using several indices of which the most popular in clinical practice is the pulsatility index (PI=(S-D)/M, where S is the systolic maximum, D is the diastolic minimum, and M is the mean over the cardiac cycle).

It is commonly believed that an increase in vascular resistance causes an increase in pulsatility of the arterial flow or velocity waveform in a vessel, and that if pulsatility remains unchanged then resistance too is unchanged. Indeed, one of the indices used to quantify pulsatility, the Pourcelot ratio ((S-D)/S), is sometimes referred to as ‘the resistance index’. There is considerable evidence from experiments and modeling studies to support the association between resistance and pulsatility (e.g. [6], [7]). Certainly, when resistance in the microvasculature of the fetal placenta is increased in sheep, blood flow in the umbilical artery becomes more pulsatile [6], [8], [9], [10]. However, when vasoactive agents are used to increase resistance the correlation between resistance and flow pulsatility is often absent or extremely poor [6], [11], [12] (e.g. Fig. 1). For instance, an infusion of angiotensin II into fetal sheep that increased umbilico-placental vascular resistance by a factor of six and reduced umbilical blood flow to one third of control caused no change in the pulsatility of the umbilical arterial flow waveform or in the pulsatility of the velocity waveform recorded using transcutaneous, continuous wave Doppler ultrasound [6]. In order to understand why resistance and flow pulsatility are correlated in some circumstances and not in others, it is necessary to understand the functional link between them. This is important because measuring changes in the pulsatility of umbilical arterial blood velocity waveforms is often used to infer the effect of treatments on fetal placental vascular resistance and/or blood flow in in vivo human studies and without this understanding erroneous conclusions may be drawn. As we will show, there are factors in addition to resistance that are important in determining the pulsatility of the umbilical arterial velocity waveform.