Chapter 11: DOPPLER INTERROGATION OF THE FETAL CIRCULATION

This chapter focuses on the Doppler interrogation of the fetal circulation. Other chapters in this book focus on Doppler velocimetry of the uterine arteries and fetal echocardiography. Randomized clinical trials have demonstrated that Doppler examination of the umbilical arteries can reduce perinatal mortality, and therefore this technique is no longer considered investigational. Moreover, the detection of an abnormal Doppler velocimetry of the umbilical artery allows classification of small for gestational age (SGA) fetuses into those who have "placental disease" (defined as an abnormal umbilical artery Doppler waveform) and those who have other conditions that may be responsible for their small size. A fraction of SGA fetuses are diagnosed at term and often do not have any abnormalities of the umbilical artery Doppler, but may have alterations of the cerebral circulation (anterior cerebral artery or middle cerebral artery). Therefore, practitioners need to be familiar with the sampling, interpretation, and clinical significance of these vessels. Similarly, Doppler examination of the venous system (ductus venosus, inferior vena cava (IVC), and umbilical vein) are informative of cardiac function and identify a fetus at risk for impending in utero death. Finally, examination of the peak systolic velocity of the middle cerebral artery has become important in the assessment of fetal anemia. Therefore, this chapter reviews in detail the anatomy, physiology, Doppler waveform morphology, normal ranges, and the interpretation of an abnormal waveform in clinical practice.

Umbilical artery Doppler velocimetry (UADV) has become an important clinical tool in obstetrical practice. UADV examines the impedance to blood flow in the placenta, and therefore is a test of the placental stages. The umbilical arteries are easy to sample because they are long, nonbranching, and surrounded by amniotic fluid.

Anatomy of the Umbilical Arteries

The umbilical arteries are branches of the internal iliac artery, which have a pelvic segment around the fetal bladder (Figure 11-1). Thereafter, the umbilical arteries have an upward trajectory inside the fetal abdomen as they travel to the umbilicus to become part of the umbilical cord. The vessels travel in the umbilical cord until the insertion of the placenta. An anastomosis of the umbilical arteries located approximately 3 cm from the placental insertion, acting as a pressure-equalizing system between umbilical arteries, can be demonstrated with ultrasound and is called "Hyrtl anastomosis."^1,2

The umbilical cord contains 2 arteries and 1 vein. The arteries carry deoxygenated blood from the fetus to the placenta. Such blood is enriched with oxygen in the villous tree, and then returns to the fetus through the umbilical vein. Therefore, the umbilical arteries contain blood with a lower Po₂ and pH than that in the umbilical vein.

Physiology of the Umbilical Arteries

The umbilical circulation is a low-resistance vascular bed. Blood flow to the placenta increases with advancing gestational age, and this is accomplished by a decrease in vascular resistance of the placenta. Umbilical vessels and placental vessels lack innervation, and they do not respond to vasoconstrictors.

Sampling Techniques

Doppler signals can be obtained in any segment that is visible: around the fetal bladder, in the abdominal insertion of the umbilical cord, in a free loop, or in the placental insertion of the umbilical cord. There is evidence that Doppler indices vary depending upon where the Doppler signal is obtained.^{3,4,5, and 6} In general, Doppler indices are higher in the abdominal insertion of the umbilical cord than in the placental insertion.^5,6 Although the difference is not large, a standard recommendation is that the same site be used for sampling, in particular when serial studies are being performed in a fetus at risk.

In most cases, Doppler signals are obtained from a free loop of the umbilical cord because of the ease of acquisition. The placental insertion and the abdominal insertion have the disadvantages of depending upon location. For example, it would be difficult to sample the umbilical cord at the placental insertion site in posterior placentas. Similarly, the position of the fetus may be suboptimal to sample the umbilical cord at the entry into the abdomen (when the fetus is facing down). In contrast, the umbilical arteries can be reliably identified and sampled on both sides of the bladder. This allows exclusion of a single umbilical artery (Figure 11-2) and also standardization. However, because this requires a greater level of skill, the free loop continues to be used in practice.

We obtain Doppler waveforms for analysis by placing the sample volume of a pulsed Doppler system in a manner that would include 1 umbilical artery and 1 vein (Figure 11-3A). However, when there is absence of end-diastolic velocities, we prefer to sample the umbilical artery alone to avoid missing reversal of flow.⁷ It is also important to remove the high-pass filter (wall thump filter), which eliminates low-frequency signals that are generated from the vessel wall. A well-known artifact is the generation of artifactual absent end-diastolic velocities, because the high-pass filter is on, suppressing the low end-diastolic signal (Figure 11-4A and B).

The number of waveforms to be obtained has been recommended to vary from 3 to 5.^{8,9, and 10} An important aspect is that the waveforms should be regular to obtain an accurate assessment of the vascular impedance to flow. Ideally, these waveforms should be obtained when the fetus is not breathing and not moving. Fetal breathing can cause changes in the waveform (Figure 11-5).^10,11 Similarly, an irregular fetal heart rate can also generate artifactual Doppler signals.^{11,12,13,14,15,16, and 17} It is not necessary to correct Doppler indices for the fetal heart rate when this is within normal ranges.¹⁴ The flow velocity waveform of the umbilical artery is not subject to diurnal variability,^{18,19,20, and 21} nor it is affected by maternal exercise.^22,23 A maternal supine position was reported to be associated with a significantly higher systolic-to-diastolic (S/D) ratio²⁴; however, this finding has not been confirmed.¹⁸

In the past, Doppler signals were calculated manually or using a planimeter. Today, the AutoTrace function allows calculation to be based on several waveforms. Similarly, modern equipment provides the S/D ratio, resistance index (RI), and pulsatility index (PI). These indices are angle independent. Most clinicians now use the PI to analyze waveforms. The initial studies employed the S/D ratio because it was easy to calculate (did not require integration of the entire waveform), but this issue has now been solved with modern ultrasound equipment. One advantage of using the PI is that it can be calculated when there are absent end-diastolic velocities. Under such circumstances, neither the S/D ratio nor the RI can be calculated.

Morphology of the Umbilical Artery Waveform and Changes with Gestational Age

In early pregnancy, an absent end-diastolic velocity is a normal finding (Figure 11-6A and B)^25,26; however, the presence of end-diastolic velocities is expected after 18 weeks of gestation.²⁷ From this point onward the umbilical artery waveform is characterized by forward flow during the entire cardiac cycle. Moreover, end-diastolic velocities are high, and this is consistent with low resistance to flow.^28,29

The placenta increases in size with advancing gestational age, and this is associated with an increased number of tertiary stem villi. As the placenta grows, the vascular resistance decreases. Therefore, the Doppler indices decrease as gestation progresses,^{18,20,27,30,31,32,33,34,35, and 36} and this is demonstrated in Figure 11-7 and in Table 11-1, which display the normal values for the PI of the umbilical artery.³⁷ Longitudinal reference ranges are also available for the blood velocity and PI at the intraabdominal portion, and fetal and placental ends of the umbilical artery.⁶

The Physiologic Significance of an Abnormal Umbilical Artery Waveform

Mathematical modeling of the placental circulation indicates that 50% to 60% of terminal arterial vessels must be obliterated for the PI to increase beyond its normal range. It is interesting that there is a steep increase of the PI after 80% of fractional terminal vessels are obliterated.³⁸ Therefore, an abnormal umbilical artery velocimetry reflects substantial pathology of the placental vascular territory. It is noteworthy, according to the results of mathematical modeling, that the same rate of obliteration of the vessels has a stronger effect on the PI in a smaller placenta than in a larger placenta. The effect of obliteration appears to be greater in early gestation when the placenta is smaller, rather than in the third trimester when the placenta is larger.³⁸

Patients with high Doppler indices have a significantly lower modal number of arterioles in the tertiary stem villi than that in a normal control group (7 to 8 arteries per field in the control group and 1 to 2 arteries per high-power field in the abnormal Doppler group). The pathologic lesion considered to be responsible for this is vascular sclerosis with obliteration of the small muscular arteries of the tertiary stem villi.^{39,40, and 41} Other vascular lesions associated with abnormal Doppler velocimetry of the umbilical arteries include fetal thrombotic vasculopathy, chronic villitis, and fetal vascular necrosis.^42,43 It is important to note that an abnormal umbilical artery Doppler waveform reflects placental disease and not fetal disease.

Trudinger et al have demonstrated that fetuses with abnormal umbilical artery Doppler velocimetry have evidence of higher red blood cell count and hemoglobin concentration,⁴⁴ endothelium activation,⁴⁵ platelet activation (which promotes thrombosis),⁴⁶ and platelet consumption.⁴⁷ In addition, these fetuses have an atherogenic lipoprotein profile⁴⁸ and evidence of intravascular inflammation.⁴⁹ This may be important in relating placental vascular disease, detected by umbilical artery Doppler velocimetry, to the risk for adult cardiovascular disease.⁴⁵ In other words, the fetus with abnormal umbilical artery Doppler velocimetry has similar changes to those observed in adults with atherosclerosis.

Clinical Correlations of Abnormal Umbilical Artery Doppler Velocimetry

Pregnancies with abnormal umbilical artery Doppler velocimetry are at increased risk for (1) perinatal death, (2) SGA, (3) congenital anomalies, and (4) admission to a neonatal intensive care unit. These conclusions are based on the pioneering and classic studies of Professor Trudinger in a large number of patients who had umbilical artery Doppler velocimetry because they were at risk for adverse pregnancy outcome based upon clinical indications or assessment.^7,50

Meta-analysis of randomized clinical trials in which UADV was used for the management of patients has indicated that the use of this tool results in a reduction of perinatal morbidity of approximately 38%, and therefore UADV is no longer considered an investigational tool. It is clinically indicated when the diagnosis of SGA is made, there is a history of a fetal demise, or there are obstetrical or medical complications that increase the risk of perinatal morbidity and mortality, such as preeclampsia, chronic hypertension, thrombophilic states, lupus, and other complications of pregnancy.^{51,52,53, and 54}

Small for Gestational Age

UADV is performed serially in SGA fetuses. Abnormalities of the umbilical artery waveform reflect the presence of placental rather than fetal disease. Therefore, absence or reversal of flow during the diastolic period is not by itself an indication for delivery, but rather an indication for increased surveillance to prevent in utero death.

The spectrum of abnormalities detectable in the waveforms evolves from a progressive reduction in the end-diastolic velocities (reflected by an elevated PI for gestational age), followed by the absence of the end-diastolic flow (AEDF) and, in extreme cases, the occurrence of reverse end-diastolic flow (REDF; see Figure 11-3B, and C).^{35,55,56,57,58, and 59}

There is an association between abnormalities of UADV and the fetal acid–base status (lower pH, lower Po₂, acidemia).^{60,61,62,63,64,65,66, and 67} When there is an umbilical artery PI greater than 1.5, lactate concentrations in umbilical venous blood increase sharply.⁶⁶ Moreover, SGA fetuses with abnormal UADV are more likely to experience fetal distress,^68,69 longer neonatal intensive care unit stays, and have a higher perinatal mortality.^{50,70,71, and 72} An abnormal UADV precedes the development of an abnormal fetal heart tracing by 2 to 4 weeks.^70,73

Congenital/Chromosomal Anomalies

Abnormalities in the UADV in fetuses who are not SGA should raise the suspicion of whether or not there is an undetected fetal congenital anomaly (anatomical or chromosomal).^{74,75,76,77, and 78} The spectrum of abnormalities described in association with abnormal umbilical artery Doppler indices is wide.⁷⁵ The tertiary villi of placentas from fetuses with karyotype abnormalities often display a reduction in the total vessel count, in the small muscular artery number, and in the small muscular artery to villus ratio.^77,79 This under-vascularization may represent placental immaturity as a result of arrested or delayed angiopoiesis,⁷⁹ and may predispose to the Doppler abnormalities in the umbilical artery.^{80,81, and 82}

Placental Abruption

Abnormalities of the umbilical artery Doppler velocimetry, such as a sudden increase in vascular resistance and appearance of a dicrotic notch, have been described in the presence of placental abruption.⁸³

Persistent Discordant Flow Velocity Waveforms in the 2 Umbilical Arteries

Reproducibility studies show that the waveforms obtained from the 2 umbilical arteries are usually very similar.¹⁹ These vessels are joined just before entering the placenta by a wide transverse anastomosis (Hyrtl anastomosis) and placenta pathology is usually diffuse, with consequences in the territory of distribution of both vessels. Persistently discordant waveforms recorded in the 2 arteries, with elevated vascular resistance in one vessel but not in the other one is an unusual finding, suggesting the presence of asymmetrical placental damage, such as a placental infarction⁸⁴ or placental abruption.⁸⁵

Umbilical Artery Doppler Velocimetry during Labor

Uterine contractions, artificial rupture of the membranes, and oxytocin infusion do not change the umbilical artery flow velocity waveform.^86,87 Whether administration of epidural anesthesia is associated with changes in the umbilical artery Doppler velocimetry is a subject of debate, with some investigators reporting no changes,⁸⁸ whereas others claim a reduction of the umbilical artery Doppler waveform indices.²⁴ A role for catecholamine has been proposed to explain such changes.⁸⁹

Anatomy of the Umbilical Vein

The umbilical vein carries oxygenated blood from the placenta to the fetus. Its oxygen saturation is the highest in the entire fetal circulation, because oxygen extraction by fetal tissues has not yet occurred. The umbilical vein is the longest vessel in the human fetus. Its diameter, in the free-floating portion, decreases progressively from the fetus to the placenta.⁹⁰ The vessel travels within the umbilical cord in the opposite direction of the 2 umbilical arteries and after entrance into the abdomen, courses toward the liver, and enters the organ. The vein branches into 2 vessels. The larger joins the portal vein, and together they enter the right lobe of the liver. The smaller vessel, the ductus venosus, carries oxygenated blood into the right atrium and is then shunted into the left atrium (Figure 11-8). In the newborn, the umbilical vein remains patent after birth for a few months. Closure of the umbilical vein usually occurs after closure of the umbilical arteries. After closure, it becomes the ligamentum teres, which is a fibrous cord extending from the umbilicus to the transverse fissure of the liver. There, it joins with the ligamentum venosus, which separates the left and right lobes of the liver.

Sampling Techniques

Doppler interrogation of the umbilical vein can be achieved in a free-floating loop, or in various segments of its intraabdominal portion. We usually sample the vessel in a free-floating central part of the cord, recording spectral Doppler signals from the umbilical vein and the artery (arteries) simultaneously (see Figure 11-3A).

Morphology of the Umbilical Vein Waveform and Changes with Gestational Age

Under normal circumstances, the velocity flow waveform has a nonpulsatile pattern (see Figure 11-3A). Reference ranges for the umbilical vein maximum velocity (V_max), time-averaged intensity-weighted mean velocity (V_wmean), and umbilical vein blood flow, based on longitudinal data, are available.⁶ Reference ranges for umbilical vein blood flow, based on cross-sectional data, are also available.^{91,92,93, and 94} The mean velocity in the umbilical vein is constant between 37 and 40 weeks' gestation (18.3 ± 4.0 cm/s). At this gestational age, the diameter of the vessel ranges between 5 and 7 mm.⁹¹

Definition of Umbilical Vein Pulsations

There is no agreement about the definition of umbilical vein pulsations; however, in most studies pulsations are considered to be present based on visual assessment of transient negative deflections of the venous flow velocity waveform (without a need to go below the baseline), in synchrony with the cardiac cycle (Figure 11-9).

Nakai et al⁹⁵ proposed pulsations to be diagnosed when the inflections are detected for more than 10 seconds and in more than 3 sampling sites. According to van Splunder et al,⁹⁶ a deflection in velocity can be defined as a pulsation only if it is greater than at least 10% of the time-averaged velocity. Other authors have proposed a threshold of greater than 15% in the magnitude of the reduction of the blood velocity as a reasonable cutoff.^97,98 It is possible to quantitate the magnitude of the pulsation by calculating the difference between the maximum velocity and the minimum velocity during the pulsation, expressed in centimeters per second.⁹⁹

Three types of umbilical vein pulsations have been recognized: (1) monophasic/single pulsations, in which there is 1 deflection per cardiac cycle (this represents the majority of umbilical vein pulsations); (2) biphasic/double¹⁰⁰ pulsations, in which there are 2 deflections per cardiac cycle, corresponding to systolic and diastolic peaks; and (3) triphasic pulsations,¹⁰⁰ with a positive systolic peak, a positive diastolic peak, and a reverse flow during atrial contraction. Triphasic pulsations have the worst prognosis and have been sporadically described.

Umbilical Vein Pulsations in Normal Pregnancies

Umbilical vein pulsations are physiologically present in the first trimester, but they disappear between 12 and 15 weeks of gestation.^101,102 This is considered to represent reduced conduction of the changes in intracardiac pressures due to a growing umbilical vein.¹⁰²

Transient umbilical vein pulsations can be present in normal fetuses throughout gestation.¹⁰³ Their frequency changes with gestational age (30% between 13 and 14 weeks, 5.5% at 23 to 24 weeks, and then 5% to 7.8% throughout the rest of pregnancy).⁹⁵

The criteria for a nonpathologic umbilical venous pulsation occurring in a fetus with a normal PI of the umbilical artery are the following: (1) monophasic in nature; (2) an amplitude within one-third of the maximum Doppler shift; (3) detectable only in one or two of all sampling sites; and (4) transient in nature (in other words, not observed in a subsequent examination, after an interval of at least 2 weeks).⁹⁵

Pathophysiology of Umbilical Vein Pulsations

Pulsations in the central veins are due to backward transmission of pulses, which occurs to some extent in the normal circulation.¹⁰⁴ Atrial contraction and relaxation are responsible for the pulsatile flow patterns of the ductus venosus and the IVC. However, these pulsations are not normally transmitted to the umbilical vein after the first trimester.

It is possible that pulsations in the venous system away from the heart, such as the umbilical vein, are caused by the same mechanism of transmission as those in veins which are close to the heart.¹⁰⁴ An enhanced atrial contraction, reflecting increased afterload, increased intraatrial pressure, and adrenergic drive,^105,106 may be transmitted to the IVC, the distended ductus venosus, and finally to the umbilical vein.^107,108 The ductus venosus plays a role in the transmission of pulsations to the umbilical vein. A mathematical model has demonstrated that the amplitude of the umbilical vein pulsations depends on the impedance ratio between the umbilical vein and the ductus venosus, as well as on the ratio of the mean velocity and the pulsed wave velocity in the ductus venosus.¹⁰⁹ Moreover, agenesis of the ductus venosus prevents the transmission of central venous pulsations to the umbilical vein.¹⁰⁷ This suggests that the appearance of pulsations in the umbilical vein reflects a hemodynamic change, which is transmitted from the right atrium.

The incidence of umbilical venous pulsations is higher at the umbilical ring in the abdominal wall, where the compliance is lower, than in the cord or intraabdominally.⁹⁹

In fetal lambs, umbilical vein pulsations may be obtained by cord occlusion or experimentally induced hypoxia.^106,110,111

Clinical Correlations of Umbilical Vein Pulsations

Although umbilical vein pulsations have been described after 13 weeks in normal fetuses,⁹⁵ their presence after this gestational age should raise the suspicion of cardiac failure.^{63,112,113,114, and 115} Some of the conditions in which umbilical vein pulsations have been reported include (1) fetal cardiac dysfunction,^103,116 (2) malformations and chromosomal aberrations,¹¹⁷ (3) atrioventricular anomalies,¹⁰³ (4) severe bradycardia and tachycardia,¹¹⁸ (5) nonimmune hydrops,^112,118 and (6) a hypercoiled cord.¹¹⁹

Umbilical Vein Pulsations in SGA Fetuses

Umbilical vein pulsations can be detected in advanced stages of intrauterine growth restriction, and are usually associated with severe abnormalities in the umbilical artery and ductus venosus (absent or reverse end-diastolic flow) Doppler velocimetry. Pulsations in the umbilical vein have a high positive predictive value for the identification of the fetus with acidemia,^120,121 at risk for in utero death,¹²⁰ birth asphyxia,¹²⁰ and neonatal death.¹²⁰ Double pulsations in the umbilical vein, combined with absent or reverse end-diastolic flow in the umbilical artery, reflect impaired myocardial function and advanced decompensation of cardiac function.⁹⁷

Adequate cerebral perfusion is required for intact survival. Insults that affect fetal growth, such as placental disease (defined as "absent/reverse end-diastolic velocities" of the umbilical arteries), result in changes in perfusion of the cerebral arteries. The conventional wisdom is that these changes can be demonstrated in the middle cerebral artery (MCA) and anterior cerebral artery, and that they consist in vasodilatation to maintain perfusion to the fetal brain. Consequently, Doppler interrogation of the MCA has become common in obstetrical practice. Recent observations suggest that changes in the anterior cerebral artery may occur before changes in the MCA, and that they may have functional consequences, in terms of neurological handicap.

Middle Cerebral Artery

Anatomy of the Middle Cerebral Artery

The MCA is a prominent branch of the circle of Willis (Figure 11-10). The arteries of the circle of Willis can be easily imaged with color Doppler mapping (Figures 11-11 and 11-12). The MCA has been divided into 4 segments: M1, M2, M3, and M4. The M1 segment is the most proximal to the circle of Willis, and is located before the MCA bifurcation or trifurcation (Figure 11-13). This segment has a diameter that remains relatively constant in physiological and pathological conditions, and is where most Doppler interrogation has been conducted. The MCA travels horizontally resting on the sphenoid wings on the base of the skull, and then curves upward and travels within the sylvian fissure. Its branches supply about 80% of the blood flow to the cerebral hemispheres (most of the lateral surface with exclusion of the superior part of the frontal and parietal lobes and the inferior part of the temporal lobe).¹²² Importantly, the MCA contributes to the blood supply of the internal capsule and the basal ganglia (caudate and globus pallidus).

Sampling Techniques

Doppler interrogation of the MCA is optimally performed in an axial section of the fetal brain. Imaging can be accomplished in the plane of the biparietal diameter and moving toward the base of the skull. The MCA can be visualized even without Doppler, but color flow aids in sampling (Figure 11-14). We, as others,^{123,124,125,126,127, and 128} prefer to sample the MCA in the near field and in its proximal segment (M1), approximately 2 mm after its origin. In this site, the intra- and interobserver variability is the lowest.¹²⁷ Moreover, Doppler indices are less susceptible to the effects of fetal behavioral states than those obtained by recording in the distal segment of the vessel.¹²⁶ Standardization of the sampling site is important because there are significant differences in blood flow velocity and vascular impedance between the proximal and distal segments of the MCA (lower PI in the proximal segment than in the distal third of the MCA).^{126,129,130, and 131} The far-field proximal site is the best alternative when the near-field proximal site cannot be interrogated.¹³¹

Two Doppler parameters are important: the MCA PI and the peak systolic velocity. The first is used to detect changes in vasodilatation in fetuses with placental disease and/or SGA. Peak systolic velocities are used in the diagnosis of fetal anemia.

A key consideration in obtaining reliable peak systolic velocity determinations is that the angle of insonation should be 0 degrees or as close to 0 degrees as possible. Small changes in the angle can introduce errors that can affect the estimation of the velocity and have implications for clinical management.¹³² Moreover, the reproducibility of peak systolic velocities is lower with the use of angle correction (higher interobserver variability).¹²⁷ It has been recommended that the MCA be magnified prior to sampling so that it occupies more than 50% of the image.¹²⁷

Doppler interrogation should be conducted in the absence of fetal breathing or movements, and over at least 3 to 5 uniform cardiac cycles. A pitfall in interrogating the MCA is placing an excessive amount of pressure on the fetal head with the transducer. This may result in increased impedance to flow and then artifactual absence of end-diastolic velocities (Figure 11-15).¹³³

Morphology of the Middle Cerebral Artery Waveform and Changes with Gestational Age

The MCA can be recognized using transvaginal sonography as early as the 10th week of gestation. End-diastolic velocities may be physiologically absent until the 11th week, inconstantly present between the 11th and 13th weeks, but consistently present thereafter.¹³⁴ The MCA has forward flow during the entire cardiac cycle, and the morphology of the waveform changes with gestational age.^128,135

Indices

Tables 11-2 and 11-3 and Figures 11-16 and 11-17 provide longitudinal reference ranges for the MCA peak systolic velocity and for the MCA PI as a function of gestational age.¹²⁸ Other reference ranges are available elsewhere.^{130,132,135,136,137,138,139,140, and 141} Ebbing et al¹²⁸ have also provided methods to calculate unconditional and conditional reference ranges for serial measurements of MCA Doppler indices (blood flow velocities, PI, and cerebroplacental ratio). This allows for the adjustment of the expected 50th percentile and the corresponding ranges according to the previous observation.¹²⁸ The PI of the MCA as a function of gestational age has a parabolic shape (see Figure 11-17).^{128,129, and 130,135,142,143}

Cerebroplacental Doppler Ratioh

The ratio of the PI of the MCA divided by the PI of the umbilical artery is known as the cerebroplacental Doppler ratio.¹⁴⁴ This ratio has the advantage of not being heart rate dependent (like the resistance indices) and of providing information about both placental and fetal circulations.

In normal pregnancies, the cerebral (MCA) PI is higher than the umbilical artery PI; therefore, the cerebroplacental ratio is always greater than 1. With advancing gestational age, the ratio first increases (21 weeks: mean = 1.41), reaches a peak (33 weeks: mean = 2.36), and subsequently decreases again (39 weeks: mean = 1.97) (Table 11-4).¹²⁸ Cross-sectional reference ranges for Doppler indices for this ratio are also available.^{37,145,146, and 147}

A low cerebroplacental ratio reflects a redistribution of the cardiac output to the cerebral circulation, which occurs as a fetal adaptation to placental disease ("brain-sparing effect"). The sensitivity of this ratio in the prediction of poor perinatal outcome among SGA fetuses is 90%, compared with 78% of the MCA and 83% for the umbilical artery Doppler waveform indices.¹⁴⁵ Moreover, this ratio has been claimed to be more accurate in the prediction of adverse outcome than the MCA or UA Doppler alone.^145,148

The MCA/UA PI ratio can be plotted against gestational age-specific reference ranges.^37,146,147 However, a threshold of 1.08 is generally used as an indicator of risk. Above this cutoff, cerebral placental circulation is considered normal, and below this cutoff it is abnormal and deserving of careful surveillance.^145,149 The inverse ratio (UA-PI to MCA-PI) has also been used by other investigators, and normal values are available for interpretation of this index.¹⁵⁰

Clinical Correlations of an Abnormal Middle Cerebral Artery Doppler Velocimetry

Small For Gestational Age

The increased impedance to flow in the uteroplacental vessels of fetuses with severe intrauterine growth restriction may cause hypoxemia, which induces adaptive responses in the fetal circulation to preserve adequate oxygen supply to the brain, the coronary circulation, and the adrenals.

Brain perfusion in the context of placental disease and chronic hypoxemia is maintained through adaptive hemodynamic changes, which are collectively referred to as the brain-sparing effect. Redistribution of the blood flow to the brain is accomplished by vasoconstriction in the peripheral vessels (ie, descending aorta, renal arteries), and vasodilatation of the MCA, which is detected as a decreased PI in this vessel.

Fetuses with a decreased PI in the MCA are at increased risk to have nonreassuring fetal heart rate patterns, to be admitted to the neonatal intensive care unit, and have almost a 3-fold higher risk of intrauterine death than SGA fetuses with a normal PI of the MCA.^143,151

Longitudinal observations in fetuses with brain-sparing effects show that the nadir of vasodilatation of the MCA is reached 2 weeks before the onset of antepartum late decelerations.¹⁵² In severely growth-retarded fetuses, there is an association between Doppler indices of the MCA and fetal blood Po₂.^64,153,154 The degree of vasodilatation in the MCA correlates with the degree of fetal hypoxemia; however, when the fetal Po₂ falls 2 to 4 SD below the normal for gestation, the maximum reduction of PI is reached and there is a tendency for the PI to rise.¹²³ This has been attributed either to cerebral edema¹²³ or to a decreased cardiac output in these fetuses.^155,156 Disappearance of cerebral vasodilatation with normalization of a low PI of the MCA is a sign of danger, and we have observed it before fetal death or the development of abnormal fetal heart rate tracing.^{150,156,157,158, and 159} Prolonged reverse end-diastolic velocities in the MCA is an agonal pattern.^160,161

Fetal Anemia

Anemia decreases blood viscosity and increases cardiac output. This hemodynamic phenomenon is the basis for the prenatal diagnosis of fetal anemia by measuring the peak systolic velocity in the MCA. There is an inverse correlation between fetal hemoglobin and the peak systolic velocity of the middle cerebral artery (MCA–PSV).^{132,138,162,163} The strength of the correlation is greater the more severe the anemia.¹⁶⁴ Therefore, the peak systolic velocity of the MCA has become the method of choice to diagnose, noninvasively, the presence of moderate to severe anemia.^125,132,165 The optimal (100% detection of moderate and severe anemia) threshold values for peak systolic velocity in the MCA in the prediction of fetal anemia were established to be (1) 1.29 times the median for mild anemia, (2) 1.50 times the median for moderate anemia, and (3) 1.55 times the median for severe anemia.¹²⁵ After 35 weeks of gestation, the MCA peak systolic velocity has limited value in the diagnosis of fetal anemia and the reasons for this are not clear.¹³² Differences in compliance of the vessel may contribute to reducing the diagnostic value of this parameter.¹³²

Peak systolic velocity of the MCA has also been used for the evaluation of fetal anemia due to Rh-alloimmunization and congenital parvovirus infection,^163,166 and in twin-to-twin transfusion syndrome.¹⁶⁷ The MCA–PSV is more sensitive and accurate than measurements of amniotic fluid change in optical density at a wavelength of 450 nm.¹⁶⁵

ANTERIOR CEREBRAL ARTERY

Anatomy of the Anterior Cerebral Artery

The anterior cerebral arteries (ACAs) are part of the circle of Willis (see Figure 11-10) and are connected to each other by the anterior communicating artery. The ACA supplies most of the medial surface of the cerebral cortex (frontal and parietal lobes), part of the lateral surface of the frontal and parietal lobes (approximately 1 inch), the olfactory bulb and tract, and the anterior four-fifths of the corpus callosum. In addition, perforating branches (including the recurrent artery of Heubner and medial lenticulostriate arteries) supply the anterior limb of the internal capsule and part of the basal ganglia (part of the caudate and globus pallidus). The pericallosal artery is a branch of the anterior cerebral artery.

Sampling Techniques

The appropriate plane to interrogate the ACA is an axial view of the fetal head at the level of the cerebral peduncles, the same anatomical plane as that used to acquire Doppler signals from the MCA.^168,169 An alternative is a coronal plane of the fetal head. In this plane, the pericallosal artery can be visualized traveling cranial to the ACA and with blood flow in the opposite direction.¹⁷⁰ In the axial plane, sampling after the origin of the ACA from the internal carotid yields better reproducibility than sampling in distal segments after the junction to the anterior communicating artery.¹⁷¹ An angle of insonation less than 60 degrees has been considered acceptable.¹⁶⁹ Flow velocity waveforms in the ACA can be obtained in 64% to 87.5% of cases.^91,172 The Doppler signal of the ACA can be weaker than that of the MCA; therefore, optimal flow velocities can be more difficult to obtain (Figure 11-18).¹⁶⁹

Indices

Reference ranges for the PI of the ACA are available in Figure 11-19.^169,171

Clinical Correlations of an Abnormal Anterior Cerebral Artery Doppler Velocimetry

All the 4 major cranial arteries (ACA, MCA, posterior cerebral artery [PCA], and internal carotid) undergo vasodilatation in the presence of chronic hypoxia, and therefore their PI lowers in cases of severe placental dysfunction (absent reverse end-diastolic velocity of the umbilical artery).¹⁷²

The hemodynamic changes in the cerebral circulation occur with a sequential temporal pattern, suggesting the existence of a hierarchy of different areas of the brain, with preferential and longer sparing of those areas required for long-term survival.¹⁷⁰

In a substantial proportion of fetuses, the ACA Doppler velocimetry shows significantly lower PI values before abnormalities in the MCA and the PCAs become detectable.^168,170,173 This may suggest that, in the presence of chronic hypoxia and cerebral blood flow redistribution, perfusion of the frontal lobes is preferentially preserved.¹⁶⁸

In SGA fetuses with abnormalities in UADV, abnormalities in the ACA have similar sensitivity but higher positive predictive value than abnormalities in the MCA in the prediction of adverse perinatal outcome and perinatal mortality.¹⁶⁸ Doppler interrogation of the ACA also has prognostic value in monitoring SGA fetuses with normal UADV. Recent studies suggest that the PI of the ACA is a better predictor of adverse perinatal outcome than that of the MCA (at a cutoff value set at the first tertile for gestational age for both ACA-PI and MCA-PI) with a sensitivity and specificity of 54.5% and 73.7%, respectively, versus 25% and 60% of the MCA.¹⁷⁴ In contrast to chronic hypoxia, acute hypoxia (oxytocin challenge test at term) affects both the MCA and ACA.¹⁷⁵

Fetal brain sparing has been associated with behavioral problems in infancy, and fetuses with brain circulatory redistribution to the anterior cerebral artery (high umbilical/anterior cerebral artery ratio) have a higher risk of behavioral problems and other neurologic difficulties at the age of 18 months.¹⁷⁶

The fetal circulation has 3 shunts: (1) the ductus venosus (DV), (2) the ductus arteriosus, and (3) the foramen ovale. These shunts are present during fetal life to maximize the delivery of oxygenated blood to the developing brain and to reduce pulmonary blood flow so that hyperperfusion and its consequences can be avoided. After birth, the DV is no longer necessary, and in most full-term neonates, this vessel closes functionally within minutes of birth. The mechanism of closure is thought to relate to a sudden drop in blood pressure in the umbilical vein after its collapse. Anatomical closure is generally completed 2 weeks after birth. The DV is replaced by a fibrous structure called the ligamentum venosum. Patency of the DV is responsible for a congenital porto-systemic venous shunt, which has clinical significance.

Anatomy of the Ductus Venosus

The DV is a small-diameter vessel (0.5 mm in midgestation, not exceeding 2 mm at term)¹⁷⁷ connecting the intraabdominal portion of the umbilical vein with the IVC, close to its entrance into the right atrium. The DV is localized in the liver, approximately between the right and left lobes. Figures 11-8 and 11-20 display the anatomy of the DV in relation with other vessels and structures in the liver.

Physiology of the Ductus Venosus

Approximately 20% to 50% of the umbilical venous blood, which has the highest oxygen saturation in the entire fetal circulation,¹⁷⁸ is shunted through the DV directly into the IVC. The rest of the blood within the umbilical vein goes to the portal vein.

Blood shunted through the DV into the IVC preferentially streams in the left compartment of this vessel, enters the right atrium, and crosses the foramen ovale to enter the left atrium and ventricle. This preferential streaming (via sinistra) ensures the supply of highly oxygenated blood to perfuse, through the ascending aorta, the circulation of the heart (coronary circulation) and the brain. In contrast, the blood contained in the IVC streams preferentially into the right compartment of the IVC, and enters the right atrium and ventricle, the pulmonary artery, and the ductus arteriosus (via destra) (Figure 11-21).

The preferential streaming of oxygenated blood from the DV through the IVC and foramen ovale is possible by 2 mechanisms: (1) the high kinetic energy acquired by blood during its transit within the DV (because of the umbilico-caval pressure gradient and of the smaller diameter of the DV than the umbilical vein), and (2) the anatomy of the right atrium inlet, with the foramen ovale orifice oriented towards the entrance of the IVC and the strategic insertion of the valve of the IVC, which is called the eustachian valve.

Shunting

The fraction of umbilical vein blood shunted through the DV into the IVC varies with gestational age, and it ranges between 20% and 50%.^179,180 It is close to 30% at midgestation, reaching a minimum of 18% at 18 to 20 weeks.¹⁸⁰ This indicates that most of the umbilical blood perfuses the fetal liver, before entering the IVC and the systemic circulation. However, oxygen extraction in the liver is modest (10% to 15% reduction in oxygen saturation). Experimental evidence indicates that hypoxemia causes distension of the DV.¹⁸¹ Therefore, fetuses with severe intrauterine growth restriction have significantly more shunting than those of appropriate weight for gestational age.^179,180,182

Sampling Techniques

Doppler interrogation of the DV can be achieved either in a mid-sagittal longitudinal plane of the fetal trunk (see Figure 11-8) or in an oblique transverse plane, through the upper fetal abdomen (Figure 11-22). In the mid-sagittal plane, the DV can be visualized as the continuation of the umbilical vein toward the vena cava. One of the advantages of sampling the DV in the mid-sagittal view is that color Doppler can be used to identify the vessel and that the angle of insonation is small. Alternatively, an oblique transverse view of the fetal abdomen can be obtained in the plane of the abdominal perimeter.

Visualization of the DV can be enhanced by color Doppler because the blood velocities within this vessel are high, and therefore aliasing can assist in the identification of the vessel (see Figures 11-8 and 11-22).

The sample volume should be placed at the origin of the DV from the umbilical vein. This is where color Doppler demonstrates the highest blood velocity. The angle of insonation should be as low as possible and always less than or equal to 30 degrees. High-quality Doppler interrogation needs to be conducted when the fetus is quiet and without fetal breathing movements. Because of the small diameter of the DV, we often ask mothers to hold their breath while sampling. This prevents the Doppler sample volume from falling outside the region of interest.

Morphology of the Ductus Venosus Waveform and Changes with Gestational Age

The DV waveform normally has forward flow throughout the entire cardiac cycle and a characteristic triphasic pulsatile pattern (Figure 11-23). Three waves, conventionally labeled as "S", "D," and "a" waves, are recorded during systole, early diastole, and atrial contraction (late diastole), respectively. The S wave has the highest velocity and it is followed by the D wave. The a wave has the lowest peak velocity.

Velocities in the DV are significantly higher than those recorded in the IVC. In the first trimester (11 to 12 weeks), velocities reach 20 to 30 cm/s (see Figure 11-23A). The mean peak velocity increases from 65 cm/s in the second trimester up to 75 cm/s at term (see Figures 11-23B and C).¹⁸³ This increase in ductal venous flow velocities with gestational age has been attributed to an increased cardiac compliance.

In contrast to the peak velocities, there is a decrease in the peak velocity index for veins (PVIV) and in the pulsatility index for veins (PIV) with advancing gestational age. This is consistent with a decrease in cardiac afterload, which is, in turn, the consequence of reduced placental resistance and maturation in diastolic ventricular function.

Doppler Indices

Parameters used to describe the waveform of the DV include S or peak forward velocity recorded during ventricular systole, which is the highest velocity recorded; D, which is the peak forward velocity during early diastole (passive ventricular filling); and a, which is the lowest forward velocity during atrial contraction (late diastole). Table 11-5 describes the Doppler indices used by investigators. Cross-sectional^{184,185, and 186} and longitudinal¹⁸⁷ (Figures 11-24 and 11-25, and Table 11-6) reference ranges for these indices are available.

Moreover, DeVore et al have described a DV preload index, an angle-independent measurement that allows the estimation of the right ventricular preload in the second trimester (ventricular systole — atrial systole)/ventricular systole = (S — a)/S).¹⁸⁸

Clinical Correlations of an Abnormal Ductus Venosus Doppler Velocimetry

Abnormalities in the flow velocity waveform of the DV can be detected in intrauterine growth restricted (IUGR) fetuses with abnormal umbilical artery Doppler velocimetry, as well as fetuses with structural heart disease, in which there is high pressure in the right atrium (tricuspid stenosis, Ebstein anomaly, etc).^{189,190,191, and 192}

The common denominator of these conditions is an increased volume/pressure in the right atrium, which has consequences on the venous drainage to the right heart. In such cases, the pressure oscillations in the right atrium will be transmitted to the central veins (ie, IVC) and to the DV, which will acquire a more pulsatile pattern.

The abnormalities in the DV include (1) increased PIV, (2) absence of end-diastolic flow, and (3) reverse end-diastolic flow (reverse a wave) (Figure 11-26).^{63,114,188,189,191,193,194, and 195}

The development of abnormalities in the DV Doppler velocimetry in IUGR fetuses is attributed to the progressive impaired cardiac function, which characterizes the advanced stages of this condition.^{195,196,197,198,199, and 200} Abnormalities in the DV Doppler velocimetry in IUGR fetuses are considered a major risk factor for stillbirth,¹²⁰ fetal acidemia,^{63,120,200,201, and 202} hypercapnia,²⁰⁰ perinatal mortality,^120,203 and neonatal death.¹²⁰ The performance of Doppler indices in the DV in the prediction of perinatal death is greater than that of abnormalities in the umbilical artery or evidence of brain sparing in the MCA.²⁰³ Thus, reversal of flow in the DV is a major risk factor for impending fetal death and the development of an abnormal fetal heart rate tracing, which is frequently an indication for delivery.

Anatomy of the Inferior Vena Cava

The IVC drains blood from the lower extremities and the splanchnic territories into the right atrium. The 2 iliac veins join in the pelvis to form the lower part of the IVC, which travels upward in the retroperitoneal space, on the right side of the spinal column. Just before its entrance into the right atrium, the IVC enters a funnel-like structure, which also contains the orifices of the ductus venosus and the hepatic veins. This structure is called the "venous vestibulum" (see Figure 11-20).²⁰⁴

At the entrance into the right atrium, the eustachian valve (also known as the valve of the IVC) directs oxygenated blood from the ductus venosus and the hepatic veins into the left atrium. The eustachian valve is important during fetal life because it allows oxygenated blood to be shunted from the right to the left atrium and in this way maintains adequate oxygenation of the heart (via the coronary arteries) and brain (via the brachiocephalic arteries emerging from the aortic arch). After birth, this valve becomes a rudimentary structure.

Sampling Techniques

Doppler interrogation of the IVC is optimally performed in a sagittal section of the fetus. Color Doppler will aid in visualizing the IVC traveling close to the fetal spine, parallel and to the right of the descending aorta. The standard practice is to sample the IVC close to its entrance into the right atrium (2 to 4 mm; Figure 11-27).^197,205 The distance of the sampling site from the right atrium affects the flow velocity waveform recordings as observed in Figure 11-28. However, some experts recommend sampling the IVC more distally from the venous entrance into the right atrium to avoid interference from the signals of the ductus venosus and the hepatic veins waveforms, which join the IVC in the venous vestibulum close to the right atrium.²⁰⁶ Rizzo et al reported that the highest reproducibility of the Doppler waveform of the IVC could be achieved by placing the sample gate between the entrance of the renal vein and the ductus venosus.²⁰⁷ The angle of insonation should be as close to 0 degrees as possible. Recording should be performed during fetal apnea and when there is no fetal movement. During fetal breathing, there is a gestational age-independent increase of the peak and time-averaged velocities.²⁰⁸ The success rate in obtaining good-quality flow velocity waveforms of the IVC is about 89%.^114,206

Indices

Reference ranges for the Doppler indices of the IVC are available (Figures 11-29,11-30, and 11-31).²⁰⁹

Morphology of the Inferior Vena Cava Waveform and Changes with Gestational Age

Doppler velocimetry of the IVC can be performed in early gestation.^205,210 The flow velocity waveform of the IVC has a triphasic morphology (see Figure 11-28A). Two forward flow components are recorded during atrial relaxation (S, or early systole) and atrial filling (D, or early diastole), and are followed by a third smaller reverse flow component reflecting atrial contraction (a, or late diastole). In the IVC, the presence of reverse flow during atrial contraction is a normal finding and reflects atrial function. In normal pregnancies, the percentage (expressed as a ratio of the time-velocity integral) of reverse flow during atrial contraction, relative to the combined forward flow in systole and early diastole, does not exceed 10%. Moreover, this percentage decreases with advancing gestational age, suggesting greater ventricular compliance and a reduced contribution of the atrial contraction to ventricular filling near term.^197,206,211

The time-averaged velocity increases with gestational age, reflecting the increased volume of flow in this vessel, the greater cardiac contractility, and the reduced afterload as term approaches.²⁰⁶ With advancing gestational age, the S/D ratios do not significantly change, suggesting a constant filling pattern during systole and early diastole.²¹¹ Flow velocity waveforms obtained in the IVC and superior vena cava (SVC) are similar, and so is their interpretation.²¹² However, the percentage of reverse flow during atrial contraction is greater in the IVC than in the SVC.¹⁹⁷

The preload index of the IVC decreases with gestational age (see Figure 11-31),²¹³ and is valuable in predicting fetal acidemia. Indeed, it appears superior to other Doppler indices of the IVC.²¹³

Clinical Correlations of an Abnormal Inferior Vena Cava Doppler Velocimetry

Abnormalities of the IVC velocity waveform can be detected in SGA fetuses with cardiac dysfunction that already have abnormalities in the umbilical artery Doppler velocimetry. The IVC waveform abnormalities include (1) increase in S/D ratios (both the peak velocity S/D and the time-velocity integral S/D), and (2) increase in the percentage of reverse flow and of the peak reverse velocity during atrial contraction.^197,211 These changes have been attributed to a high end-diastolic ventricular/atrial pressure that changes the characteristics of the IVC waveform.²¹¹ The percentage of reverse flow in the IVC and the preload index are good predictors of fetal acidemia and hypercapnia.^200,213

Abnormalities in the IVC velocity waveform can be detected in fetuses with a cardiac arrhythmia, including premature atrial contractions, supraventricular tachycardia, bradycardia, and heart block. The most common finding is an increase in the percentage of reverse flow during atrial systole, which is mainly attributed to overfilling of the right atrium.¹⁹⁷ The magnitude of the flow reversal depends upon the timing of the premature atrial contraction in relation to the cardiac cycle, the heart rate, and the type of conduction defect. If the premature beat occurs during ventricular systole, when the atrioventricular valves are closed, a "cannon" wave will be generated in the IVC waveform.¹⁹⁷

Anatomy of the Hepatic Veins

The hepatic veins represent the efferent venous system of the liver. There are 3 main hepatic veins: the right, the left, and the middle hepatic vein, which, in the sub-diaphragmatic area, converge with the IVC and the ductus venosus prior to opening into the right atrium. The left and middle hepatic veins run parallel to the ductus venosus, whereas the right hepatic vein runs anterior and parallel to the IVC. Knowledge of the anatomy is important because it is always possible to obtain a waveform from one of the hepatic veins.

The left lobe of the liver is supplied by oxygenated blood coming from the umbilical vein via the left portal vein. As a consequence, the middle and the left hepatic veins, which drain this territory, contain blood that is more oxygenated than that contained in the right hepatic vein.²¹⁴ The oxygenated blood from the ductus and the left hepatic veins flows directly through the foramen ovale to the left atrium.²¹⁵

Sampling Technique

The right (RHV), middle (MHV), and left (LHV) hepatic veins can be visualized in an oblique transverse plane through the upper abdomen. The sample volume should be positioned in the main stem of each vessel, as close as possible to the IVC.^184,216 The RHV can also be examined in a sagittal-coronal view of the fetal abdomen, within the right lobe of the liver (see Figures 11-20 and 11-32).¹⁸⁴

Indices

Reference values for the Doppler indices of the hepatic veins are available.^184,216,217

Morphology of the Hepatic Veins Waveform and Changes with Gestational Age

The flow velocity waveform of the hepatic veins has a 3-phasic pulsatile pattern, similar to that of the IVC. Two positive waves, an early forward flow coinciding with atrial diastole and ventricular systole (S) and a late forward flow coinciding with ventricular diastole (D), are followed by a reverse flow during atrial contraction (a) (Figure 11-33). The "a" wave reflects the right atrial pressure, as is the case for the IVC. With advancing gestational age, the a wave recorded in the right hepatic vein can display forward flow rather than reverse flow.²¹⁷

With advancing gestational age, the blood flow velocities increase, whereas the peak velocity index for the vein (PVIV) and pulsatility index for the vein (PIV) decrease. The reduction in PVIV and PIV with advancing gestation may reflect the decrease in cardiac afterload due to a decrease in placental resistance, and to maturation of diastolic ventricular function. Velocities in the hepatic veins are lower than those in the ductus venosus.¹⁸⁴ The preload index (a/S) of the hepatic veins, as described for the IVC, decreases gradually with advancing gestational age.^213,216 The preload index of each hepatic vein is higher than that of the IVC at any gestational age.²¹⁶

Clinical Correlations

In fetuses with intrauterine growth restriction, the flow velocity waveform of the hepatic veins displays similar changes to those observed in the IVC, with an increase in the percentage of reverse flow and higher peak velocities during atrial contraction.

The fetal pulmonary circulation differs from that of the adult in fundamental ways. During fetal life, approximately 20% to 25% of the combined cardiac ventricular output (50% of the right ventricular flow) is directed to the lungs and returns to the left atrium through the pulmonary veins. The right ventricle is the systemic ventricle in the fetus (in the adult, in contrast, the left ventricle is the systemic one) and the arterial pressure in the pulmonary artery is consequently the systemic pressure. The vascular resistance in the pulmonary vascular bed is high, and the lungs are not expanded and are not a site of oxygen exchange.²¹⁸ Visualization of the pulmonary veins is important and should be part of every sonographic examination. The demonstration of 1 pulmonary vein excludes total anomalous venous return, which is a neonatal emergency.

Anatomy of the Pulmonary Veins

Four pulmonary veins carry blood from the lung to the left atrium. The capillary network that surrounds the air sacs joins first in a single vessel for each lobule. These vessels join again as they approach the hilum of the lung forming a single trunk for each lobe, 3 for the right and 2 for the left lung, respectively. The vein from the middle lobe of the right lung usually joins with that of the upper lobe, so that there are 2 pulmonary veins for each lung. While traveling toward the left atrium, the right pulmonary veins pass behind the right atrium and the SVC, whereas the left pulmonary veins pass in front of the descending aorta. The right and left pulmonary veins finally enter the left atrium in its posterior wall. Individual variations are possible and the number of pulmonary veins opening in the left atrium may vary between 3 and 5 (Figure 11-34).

Sampling Techniques

Doppler interrogation of the pulmonary veins can be obtained in a cross-section of the fetal chest at the level of the 4-chamber view (see Figure 11-34). The visualization of the pulmonary veins can be optimally displayed using color Doppler and presets that detect low velocities. Although color power Doppler has been used, it has the disadvantage of not providing information about direction of flow. Therefore, it would not help to distinguish between pulmonary veins and pulmonary arteries.²¹⁹ The sample volume is commonly positioned over one pulmonary vein just before its entrance into the left atrium. However, pulmonary venous flow velocity waveforms, obtained at the entrance of the left atrium, in a point at the middle of the fetal lung and at the inner thoracic wall, are not significantly different (except for the absence of end-diastolic flow in 3% of the middle and 16% of the distal pulmonary venous waveform).²²⁰ The Doppler recording is considered intrapulmonary if the sample volume is positioned more than 5 mm from the left atrium.²²¹ The repeatability is acceptable and the coefficients of variation are less than 15% for all velocity parameters and their ratios.²²⁰ The angle of insonation should be below 10 to 20 degrees.^221,222 Interrogation is best obtained during a period of fetal apnea and over at least 5 uniform cardiac cycles. Color Doppler waveforms can be reliably obtained in about 90% of cases.²²¹ Identification of a single pulmonary vein is important because it excludes the diagnosis of total anomalous pulmonary venous return.

The Physiology of the Pulmonary Veins

Flow velocity waveforms of the pulmonary venous system are mainly determined by changes in pressure within the left atrium. There is an inverse relationship between pulmonary venous flow and left atrial pressure.^{223,224, and 225} If the left atrial pressure increases, the pulmonary venous flow decreases and vice versa. Pulmonary veins function as a reservoir of blood for the left atrium.^{223,224, and 225} Fetal breathing movements do not affect the waveforms of the pulmonary veins because the entire vessel (atrial and collecting venules) is located within the thoracic cavity and therefore is subject to changes in intrathoracic pressure.²²⁶

Morphology of the Pulmonary Veins Waveform and Changes with Gestational Age

The normal Doppler flow velocity waveform of the pulmonary veins demonstrates in most cases forward flow during the entire cardiac cycle (Figure 11-35). It is pulsatile and could be biphasic or triphasic. There are 1 or 2 forward phases during systole and 1 during diastole (S, D, and a waves). In the fetus, the end-diastolic component of the waveform (a wave) has low velocities, but they are generally forward.^218,222,227 However, some investigators have reported mild reversed flow during atrial contraction (negative a wave) in a small proportion of normal fetuses.^221,228 It is possible that Doppler sampling very near to the atrium demonstrates reversal of flow, whereas measurements within the lung parenchyma are free from any reversal of flow.²²² The pulsatile flow recorded in the pulmonary veins may be due to forward transmission of flow pulses from the pulmonary arteries²²⁹ or a consequence of the changes in left atrial pressure during the cardiac cycle.²²⁴ Through the second half of gestation there is an increase in S, D, and a peak velocities.²¹⁸ A slight decline²²⁷ or no significant changes²² during gestation have been reported for the S/D ratio. The time velocity integral also increases with advancing gestation. The rise in peak velocities is consistent with the increase in cardiac output and in pulmonary blood flow, and atrial relaxation during gestation.²²²

Indices

Reference ranges for the peak velocities in systole (S), diastole (D), and atrial contraction (a),^{218,221,227,228,230} as well as reference ranges for the venous indices (PVIV and PIV)²²² are available. Moreover, regression equations for Doppler measurements obtained in intrapulmonary branches of the pulmonary veins are also available.²²⁰

The Physiologic Significance of an Abnormal Pulmonary Vein Waveform

Fetuses with congenital heart defects may have abnormalities in the Doppler velocimetry of the pulmonary veins that reflect changes in the atrial preload. Recently, Lenz and Chaoui²³¹ provided a description of the pulmonary veins Doppler flow waveform in a large series of fetuses with cardiac defects, which were classified in 5 groups, according to the possible effect on pulmonary hemodynamic characteristics.²³¹ Among the different cardiac anomalies, the group with left atrial obstruction and restrictive foramen ovale (frequent findings in the presence of hypoplastic left heart syndrome) had a marked increase in pulsatility, with reverse flow during atrial contraction.²³¹

Doppler investigation of the pulmonary veins is of value in fetuses with hypoplastic left heart syndrome,²³² because it provides important information to identify fetuses with a restricted foramen ovale.²³² Such a condition represents a well-known prognostic factor for survival in these fetuses (both after reconstructive surgery and transplantation).^233,234 Adequate direct visualization of the atrial septum and foramen ovale may not always be possible.²³² A restricted foramen ovale is responsible for left atrial and pulmonary venous hypertension, which affects the left atrium hemodynamics and thus the morphology of the pulmonary veins waveform (increase in S velocity, decrease in D velocity, increase in S/D ratio and a velocity).^{232,235,236, and 237} The degree of restriction at the level of the atrial septum correlates with the degree of Doppler abnormalities.²³⁵ In extreme cases in which there is a completely restricted foramen ovale, the pulmonary waveform has a "to and fro" flow pattern.²³⁷

The simultaneous recordings of pulsed wave Doppler signals in branches of the pulmonary artery and vein within the inner two-thirds of the lung parenchyma has been proposed as an alternative sampling site for the assessment of arrhythmias in the fetus.^238,239 This technique allows the simultaneous recording of atrial and ventricular activity, and is relatively independent from fetal position.^238,239

Anatomy of the Aortic Isthmus

The aortic isthmus (AOI) is the vascular segment of the aorta that, in the fetus, extends between the origin of the left subclavian artery and the entry of the ductus arteriosus into the aorta (Figure 11-36). After birth, with the closure of the ductus arteriosus and its transformation into the fibrous ligamentum arteriosus, this anatomical point becomes the distal landmark of the AOI.

The AOI during fetal life has a smaller diameter than the rest of the aortic arch (the term "isthmus" means "a narrow connection") and blood flow through the isthmus is negligible until the closure of the ductus arteriosus. Reference values for diameters of the isthmus in late gestation have been obtained using a coronal view.²⁴⁰ These nomograms may be of value in the prenatal diagnosis of coarctation of the aorta.²⁴¹

Hemodynamics in the Aortic Isthmus

The cardiovascular system, both in fetal and adult life, is composed of 2 circuits. In the fetus, these circuits function in parallel, whereas in the adult, they function in series. During fetal life, the blood returning to the heart from the IVC and SVC is directed from the right atrium into the right ventricle (this is relatively deoxygenated blood). This blood exits the right ventricle through the pulmonary artery, but given the high pulmonary vascular resistance, it is directed towards the ductus arteriosus into the descending aorta and toward the "sub-diaphragmatic circulation."

In contrast to the pathway of deoxygenated blood, highly oxygenated blood coming from the placenta through the umbilical vein and the ductus venosus into the right atrium travels through the foramen ovale to the left atrium, left ventricle and aorta, the aortic arch, and the brachiocephalic arteries. This blood supplies the heart as well as the brain (Figure 11-37).²⁴²

The AOI establishes the only communication between the 2 arterial circulations perfusing in parallel the upper and lower part of the fetus: the "cephalic circulation" (left ventricle, highly oxygenated blood, brain perfusion) and the "sub-diaphragmatic circulation" (right ventricle, less oxygenated blood, splanchnic territory, lower limbs, and eventually returning to the placenta through the umbilical arteries).²⁴²

The direction and the volume of blood shunted by the AOI depend upon the gradient of pressure between the 2 circulations. This gradient is dependent upon (1) the impedance of flow of the cephalic circulation, (2) the impedance of the placental bed and the sub-diaphragmatic circulation, (3) left ventricular stroke volume, (4) right ventricular stroke volume.^242,243 During systole, the left ventricular output favors forward flow, whereas the right ventricular output favors retrograde flow. During diastole, the major determinants of isthmic flow are the vascular resistance in the "cephalic" and in the "sub-diaphragmatic" circulations (see Figure 11-37).²⁴² These hemodynamic concepts are essential to interpreting and understanding the clinical significance of Doppler signals obtained from the AOI.

Sampling Techniques

Doppler interrogation of the AOI is performed in a mid-sagittal view of the fetal trunk. Starting from an apical 4-chamber view of the heart, preferably with the spine down, the transducer is then rotated 90 degrees to obtain a sagittal view of the fetus and of the aortic arch. Sampling is recommended a few millimeters beyond the origin of the subclavian artery. The angle of insonation should be as low as possible and always less than or equal to 30 degrees.

Another way to obtain Doppler signals from the AOI is to use the standard 3-vessels view used in fetal echocardiography.^244,245 This method is easier to perform, less time-consuming, and more reproducible than obtaining signals from the longitudinal aortic arch view.²⁴⁵ Importantly, there is a high degree of reliability between PI values obtained from the longitudinal aortic arch and the 3-vessels view.²⁴⁵ Doppler interrogation should be conducted when the fetus is quiet and without fetal breathing movements. Recording should include at least 5 uniform heart cycles.

Morphology of the Aortic Isthmus Waveform and Changes with Gestational Age

The morphology of the AOI waveform changes with gestational age. Up to approximately 30 weeks, there is forward flow throughout the entire cardiac cycle. The waveform has a steep increase during systole, followed by a gradual and smooth diastolic deceleration phase. After 30 weeks, an end-systolic incisura appears. This incisura becomes deeper with advancing gestation and, toward the end of pregnancy, it results in a brief end-systolic reversal of flow. Color flow mapping is very useful in demonstrating reversal of flow in the AOI of blood that comes from the ductus arteriosus (Figure 11-38).^243,246

Doppler Indices

In contrast to other vessels, evaluation of the Doppler waveform of the isthmus is conducted with a specific index, which is called the "isthmic flow index" or IFI. The IFI provides information about the direction of diastolic, systolic, and net blood flow through the fetal AOI (Table 11-7).²⁴³ In normal pregnancies the net blood flow is always antegrade. Some reversal of flow during diastole is possible, as term approaches. However, the amount of diastolic reverse flow will never exceed the amount of forward flow during systole.

There are 3 blood flow patterns in the AOI: types I, II, and III. Type I is the most frequent pattern observed in normal fetuses, and is characterized by a net flow that is antegrade IFI > 0. Toward the end of gestation, a type II pattern may be present in normal fetuses. In these cases, the net flow is still antegrade; however, diastolic flow may be either absent or reversed. Retrograde net flow in the aortic isthmus is always pathologic and is called type III.²⁴³ Reference ranges for the IFI during gestation are provided in Table 11-8.²⁴³ The PI evaluates the impedance to flow of only 1 vascular bed, and therefore it is not useful to assess the complex hemodynamics of the AOI.^243,246

The Physiologic Significance of an Abnormal Aortic Isthmus Waveform

Conditions favoring reversal of flow in the aortic isthmus include left ventricular dysfunction, increase in placenta impedance, and reduction of the cerebral impedance to flow.²⁴² Fetuses with profound anemia may also have reversal of flow in the aortic isthmus because of extreme cerebral vasodilatation of the MCA.²⁴⁷

Clinical Correlations of an Abnormal Aortic Isthmus Doppler Velocimetry

Fetuses with severe growth restriction can develop abnormalities in the aortic isthmus Doppler waveform. The spectrum of abnormalities reflects progressive fetal hemodynamic compromise (see Table 11-7). Typically, the first change observed in the AOI waveform is reversal of flow during diastole. Eventually the magnitude of the reverse diastolic flow increases and equals or even exceeds the amount of systolic forward flow. The finding of "retrograde net flow" (reverse flow during diastole greater than forward flow during systole) is an ominous sign.

Experimental evidence suggests that "net retrograde flow" in the aortic isthmus is associated with reduced delivery of oxygen to the brain,²⁴⁸ which can lead to cerebral hypoxic damage. Indeed, in the presence of retrograde aortic isthmus net flow, the poorly oxygenated blood from the right atrium (reaching the isthmus traveling through the pulmonary artery and the ductus arteriosus) mixes with the oxygenated blood ejected by the left ventricle, which is destined to the brain curculation.²⁴⁸ The magnitude of retrograde flow, which can be estimated by the IFI, correlates to the risk of prenatal brain damage. The lower the IFI, the higher the risk.²⁴² A pattern of predominantly retrograde flow in the aortic isthmus during fetal life has been associated with a 2-fold risk of neurodevelopmental deficit between 2 and 4 years of age.²⁴⁹ An IFI of 0.7 or below is associated with the highest overall positive and negative predictive values in identifying those children who will have suboptimal neurodevelopment.²⁵⁰

SGA fetuses with an abnormal IFI have a higher risk of adverse perinatal outcome, including respiratory distress syndrome, bronchopulmonary dysplasia, grade III/IV intraventricular hemorrhage, necrotizing enterocolitis, and longer neonatal intensive care stay.²⁵¹ Moreover, the presence of retrograde net flow in the aortic isthmus is associated with a higher perinatal mortality than antegrade flow (70% versus 5%).

Doppler of the fetal circulation is performed, in part, for the identification of the fetus at risk for impending death (ie, before the decompensation occurs). Doppler abnormalities in the fetal venous system frequently are late signs of compromise and are associated with a higher risk of perinatal death.^114,252 Investigation of the aortic isthmus is of value because reversal of diastolic flow in this vessel can precede the reversal of flow in the umbilical artery.²⁵³ Similarly, 80% of fetuses with reversal of flow in the aortic isthmus may not have ductus venosus abnormalities and develop such abnormalities only after the aortic isthmus is abnormal (from 24 to 48 hours later).²⁵¹ On average, the aortic isthmus blood flow becomes abnormal 1 week earlier than that of the ductus venosus.²⁵⁴ In conclusion, Doppler interrogation of the aortic isthmus could help in identifying the fetus with impending failure of the adaptive mechanisms against hypoxia.²⁴⁹

Sampling Techniques

Interrogation of the descending aorta is best performed in a dorso-posterior or dorso-anterior mid-sagittal plane of the fetus. The transducer is oriented along the longitudinal axis of the fetal aorta so that a long segment of at least 3 cm is visualized.²⁵⁵ Doppler interrogation is possible either in a cephalad to caudal direction or in caudal to cephalad direction. By sampling from a caudal to cephalad direction, the obstruction of the beam by the ribs or limbs is avoided. Standardization of the site of sampling, a low high-pass filter, and a well-defined angle of insonation ensure maximum reproducibility.^255,256

We sample the aorta just above the diaphragm, as proposed also by other investigators.^{201,255,257,258} Velocities and resistance indices change significantly when the aorta is interrogated at the level of the ascending aorta,²⁵⁹ the aortic arch,²⁶⁰ in its intermediate thoracic part,^260,261 or in its abdominal tract (above or below the renal arteries),^{260,261,262,263,264, and 265} and this should be considered to choose the correspondingly built reference curve (Figure 11-39). Sampling should be recorded for an average of 5 to 10 uniform cardiac cycles during fetal apnea and a quiet state. Angles of insonation ranging between 30 and 70 degrees have been used in different studies.^{201,266,267, and 268} For the descending aorta, satisfactory waveforms can be obtained in 98% to 100% of cases.^114,266

Morphology of the Descending Aorta Waveform and Changes With Gestational Age

The flow velocity waveform of the descending aorta has a pulsatile forward flow throughout the whole cardiac cycle (see Figure 11-39).²⁶⁰ End-diastolic velocities are absent in the first trimester and appear at 12 to 13 weeks, suggesting lowering of the systemic vascular resistance. From this point onward, the aortic waveform begins with a steep acceleration phase leading to the peak systolic velocity. The deceleration phase, from peak systole to end diastole, is divided into a rapid and a slower component by a notch (incisura-dichrotic notch), which marks the end of systole and represents aortic and pulmonary valve closure.²⁶⁹ The systolic phase (acceleration time) is a reflection of cardiac contractility, whereas the diastolic phase reflects peripheral vascular resistance. With increasing gestation, the incisura tends to deepen and the diastolic velocities tend to increase,²⁶⁹ indicating a progressive increase in cardiac contractility and reduction of the peripheral resistances as term approaches.²⁶⁹

The resistance indices (PI, RI, and S/D ratio) of the aorta remain constant throughout gestation.^{255,261,266,270} Aortic velocities increase with gestational age^255,271 as a consequence of a reduction in placenta vascular resistances.²⁶⁹

The aortic waveform morphology is influenced by the site of sampling,^255,260,272 cardiac output and contractility of the fetal heart, blood viscosity, vascular compliance, and peripheral vascular resistance,^255,273 as well as fetal behavioral states.²⁷⁴ Interrogation of the aorta far from the fetal heart yields lower resistance indices (PI, RI) and absolute velocities.^255,260,272 This has been attributed to morphological changes in the aortic wall, and to the lower impedance to flow the more vessels have branched from the aorta.^261,275 Active sleep (fetal behavioral state 2) is associated with lower aortic resistance to flow than during quiet sleep (fetal behavioral state 1). This has been interpreted to represent greater blood supply to the skeletal musculature during active fetal movement.²⁷⁴

Indices

Reference ranges for the fetal aorta (PI, RI, V_max, mean velocity [V_mean], minimum velocity [V_min]) are available.^{37,255,263,266,267} Reference ranges for the fetal aorta PI are provided in Table 11-9.³⁷

Clinical Correlations of an Abnormal Descending Aorta Doppler Velocimetry

In SGA fetuses, the morphology of the flow velocity waveform of the descending aorta reflects the status of peripheral vasoconstriction with high downstream impedance, aiming to maintain adequate blood flow perfusion and oxygen delivery to the essential organs (brain-sparing effect).^201,269,276

The spectrum of Doppler changes include (1) a reduction of end-diastolic velocities,^269,273 (2) a progressive increase in PI,^{200,201,269,273,277} and (3) absent or reverse end-diastolic velocities.^269,277,278

Absent or reversed end-diastolic flow in the fetal aorta of SGA fetuses is widely accepted as an ominous sign of fetal compromise (Figure 11-40) and a risk factor for developing necrotizing enterocolitis,²⁷⁹ as well as pulmonary, gastrointestinal, and intraventricular hemorrhage.^278,279 This has been attributed to prolonged tissue ischemia and hypoxia. In addition, abnormalities in the hemostatic system can also be observed (eg, thrombocytopenia, coaugulopathy, and bleeding).^279,280

SGA fetuses with absent end-diastolic flow in the thoracic aorta are at increased risk of perinatal death.^277,279,281 The loss of end-diastolic velocities in the abdominal aorta may precede that in the thoracic aorta, and thus it is a more sensitive marker for fetal distress and mortality.²⁶⁵

A loss of end-diastolic velocities in the aorta may precede the development of a suspicious fetal heart rate tracing by 2 to 3 days, and abnormal findings of nearly 8 days.²⁸¹ In one study that included 100 SGA fetuses, abnormalities in the thoracic aorta were better predictors of perinatal death than those of other vessels (umbilical artery and MCA).²⁷⁷ The combined absence of end-diastolic velocities in the umbilical arteries and in the thoracic aorta has been associated with a 50% risk of perinatal death.²⁷⁷

The ratio between the mean velocity of the common carotid artery and that of the thoracic aorta (carotid V_mean/aortic V_mean), as well as the ratio between the PI of the common carotid artery and that of the thoracic aorta (carotid PI/thoracic PI) have been proposed to better reflect the diversion of flow to the fetal head and the reduction of flow to the viscera in the presence of fetal asphyxia than the results of a single vessel.²⁶⁶ Figure 11-41 displays the flow velocity waveform of the common carotid artery and of the jugular vein, which can be obtained in a para-sagittal section of the fetal neck.

Anatomy of the Renal Arteries

The renal arteries are branches of the abdominal aorta, and they originate just below the superior mesenteric artery. The arteries divide into anterior and posterior branches as they approach the hilum. Accessory renal arteries are frequent, especially in the left side. Imaging of these arteries is important in the diagnosis of bilateral or unilateral renal agenesis. Doppler interrogation has been undertaken to examine the physiology of the renal circulation as a function of gestational age and also in pathologic conditions such as SGA and oligohydramnios.

Sampling Techniques

The optimal image to sample the renal arteries is a coronal section of the fetal abdomen, which allows imaging of the aorta in a longitudinal plane as well as both kidneys. Color flow mapping is essential for reliable sampling (Figure 11-42). The pulse Doppler gate can be placed at the level of the artery's origin from the abdominal aorta²⁸²; however, other investigators recommend measuring in the renal arterial trunk away from the aorta and before any visible branches.²⁸³ Some investigators have also sampled the arteries at the level of the renal hilum.^{284,285, and 286} Recording should be performed during periods of fetal apnea, and the angle of insonation should be as low as possible and always less than 30 degrees.

Morphology of the Renal Arteries Waveform and Changes with Gestational Age

The flow velocity waveform of the renal arteries changes during gestation (Figure 11-43). End-diastolic velocities are present in approximately 55% of fetuses at 20 to 24 weeks of gestation, but they are present in approximately 80% of fetuses in the 33- to 36-week range.²⁸⁶ In normal fetuses, the peak systolic velocity and V_mean of the renal arteries increase with advancing gestational age.²⁸⁶ In contrast, the PI and RI decrease approaching term. These hemodynamic changes suggest a progressive fall in the impedance to flow, and an increase in renal perfusion.^287,288 Such findings occur at the same time as there is development of the renal glomerular and renal tubular function,²⁸⁶ which are accompanied by a gradual increase in the mean hourly urinary production occurring from 32 to 40 weeks of gestation.²⁸⁹ The PI is the Doppler index of choice because many fetuses have physiologic absence of end-diastolic velocities (Table 11-10).

IUGR is associated with a significant reduction of the peak systolic velocities²⁹⁰ and with a higher PI.^282,287 These changes correlate with the degree of fetal blood deoxygenation.²⁸⁷ It is possible that the abnormalities in the Doppler waveform of the renal arteries reflect the compensatory redistribution of blood away from the kidneys, aiming to maintain adequate blood perfusion of the brain, the coronary, and the adrenals. It is possible that such changes increase the risk for chronic renal disease later on in life in IUGR fetuses.^{291,292, and 293}

Doppler interrogation of the renal arteries has been conducted in fetuses suspected to have renal disease. A maximum peak flow systolic velocity (V_max) below −1.5 SD of the mean has been associated with poor postpartum outcome. V_max is the index of choice for this purpose because it has greater change with gestational age and has a lower measurement error than the PI and RI.^285,294

Renal hypoperfusion may be a cause of oligohydramnios. Doppler velocimetry of the renal arteries has been used in the differential diagnosis. Fetuses with oligohydramnios due to renal hypoperfusion/vasoconstriction have a higher PI,^282,295 RI,²⁹⁶ and S/D ratio²⁹⁷ in the renal arteries than those of normal controls.

Anatomy of the Adrenal Arteries

Three groups of arteries supply the adrenal: (1) superior adrenal arteries (from the phrenic artery), (2) middle adrenal arteries (from the abdominal aorta), and (3) inferior adrenal arteries (from the renal artery). Gastric, ovarian, and superior mesenteric arteries provide additional blood flow. There is a great individual variability in the number and origin of the adrenal vessels. Venous blood from the venous plexus of the medulla and from the cortex is collected into the central vein, which drains directly into the IVC on the right side, and into the renal vein on the left side.

Sampling Techniques

Doppler interrogation of the adrenal artery is performed in a transverse plane of the fetal abdomen, just above the kidneys. Color Doppler is valuable to identify the adrenal vessels. It is recommended to reduce the high-pass filter, because of the low velocities in this circulation (Figure 11-44). The middle adrenal artery arises from the aorta and courses toward the adrenal hilum. There is disagreement about the site of sampling, which can be performed either at the adrenal hilum²⁹⁸ or in proximity of the aorta, where the middle adrenal artery has its origin.²⁹⁹ The available nomogram for the PI of the adrenal artery is based on measurements obtained by sampling at the origin of the vessel from the aorta. Sampling should be performed in a period of absence of fetal breathing²⁹⁹ and movements. Recording should be performed for at least 3 cardiac cycles. There are no significant differences between the PI of the left and right middle adrenal arteries. Similarly, there are no significant differences between the PI of the inferior adrenal artery (arising from the renal artery) and middle adrenal artery (arising from the aorta).²⁹⁹ Sampling of the adrenal artery can be obtained in 80% to 90% of cases,^298,299 on average in 2 to 3 minutes.²⁹⁹

Physiology of the Adrenal Artery

Nomograms for adrenal size are available from 15 weeks to term.^300,301 In the presence of fetal hypoxia there is a significant increase in the adrenal blood flow. This has been attributed to the effects of the adrenocorticotropic hormone (ACTH), whose concentrations are increased as a consequence of activation of the pituitary-adrenal axis.³⁰²

Morphology of the Adrenal Artery Waveform and Changes with Gestational Age

The flow velocity waveform of the adrenal artery has a continuous forward flow with a high end-diastolic velocity, indicating low impedance (Figure 11-45). The PI of the normal adrenal is one of the lowest values in the arteries of a human fetus.²⁹⁹ With advancing gestation, either a significant reduction of the PI²⁹⁹ (cross-sectional study, 131 normal pregnancies, sampling at the origin of the superior adrenal artery from the aorta) or no significant changes in PI, RI, peak systolic velocity (PSV), and time-averaged maximum velocity (TAMXV) (cross-sectional study, 62 normal pregnancies, sampling at the adrenal hilum)²⁹⁸ have been reported.

Resistance (PI, RI) and velocity (PSV and TAMXV) indices are significantly higher in females of appropriate weight for gestational age. This difference between male and female fetuses is not present among fetuses with IUGR.²⁹⁸ There is a correlation between velocity indices in the adrenal artery and estimated fetal weight.²⁹⁸

Indices

Reference ranges for the PI of the adrenal artery are provided (Table 11-11).²⁹⁹

Clinical Correlations of an Abnormal Adrenal Artery Doppler Velocimetry

There are only a few studies on the Doppler velocimetry of the adrenal arteries and the results are not in agreement. Some investigators have reported that SGA neonates with an abnormal (>95th percentile for GA) adrenal artery PI have significantly higher incidence of fetal heart rate decelerations and preterm delivery when compared to SGA fetuses with normal adrenal artery PI.²⁹⁹ Other investigators, however, did not confirm these findings.²⁹⁸

Anatomy of the Superior Mesenteric Artery

The superior mesenteric artery (SMA) is a branch of the abdominal aorta and emerges below the celiac trunk. The mesenteric artery supplies the intestine from the lower part of the duodenum to about one-half of the transverse part of the colon and the pancreas.

Sampling Techniques

Doppler interrogation of the SMA can be accomplished in a para-sagittal view of the fetal abdomen, which has the advantage of allowing insonation of the vessel with a better angle. The sample gate can be placed at the origin of the vessel from the abdominal aorta³⁰³ or halfway between its origin and the abdominal wall.³⁰⁴ The angle of insonation should be as low as possible and always less than or equal to 30 degrees. Sampling should be conducted when the fetus is quiet.

Indices

The Doppler indices commonly used in the evaluation of the flow velocity waveform of the SMA include the peak systolic velocity, the time-averaged maximum velocity, and the PI. Cross-sectional^{304,305, and 306} and longitudinal³⁰³ reference ranges for these indices have been established. Tables 11-12 and 11-13 and Figures 11-46 and 11-47 display the normal values of the SMA PI and peak systolic velocity obtained by a longitudinal evaluation of 161 low-risk pregnancies.³⁰³

Morphology of the Superior Mesenteric Artery Waveform and Changes with Gestational Age

The waveform of the superior mesenteric artery is characterized by a rapid systolic velocity, followed by an initial rapid deceleration, then a more gradual velocity deceleration with a positive forward flow during diastole. The peak velocity increases, possibly because of a greater intestinal perfusion.³⁰³ The PI increases until about 30 weeks of gestation, when it reaches a plateau.³⁰³

Clinical Correlations of an Abnormal Superior Mesenteric Artery Doppler Velocimetry

Centralization of blood flow is the adaptive fetal response to hypoxemia. This consists in a preferential perfusion of vital organs (the brain, the heart, and the adrenals) at the expense of a lower perfusion of other organs (the kidneys, gut, liver, and spleen). Centralization of blood flow is mediated by a selective modulation of the resistance to flow in the different districts.

SGA infants who have experienced in utero hypoxia are more likely to develop necrotizing enterocolitis because of poor intestinal perfusion. Doppler velocimetry of the SMA in neonates who were growth restricted in the uterus shows abnormalities including a higher PI and lower systolic, mean, and end-diastolic flow velocities than controls.^307,308 Moreover, the preprandial blood flow velocity in the superior mesenteric artery is inversely related to the degree of growth retardation (r = 0.63, P < .05).³⁰⁹

A PI greater than the 97.5th percentile is associated with adverse perinatal outcomes. However, this change may occur late in the course of the disease, and is a late sign of fetal compromise with limited value in fetal surveillance.³¹⁰ Doppler indices of the SMA are not predictive of the outcome of fetuses with gastroschisis.³⁰⁶

Anatomy of the Celiac Artery

The celiac artery, also known as celiac trunk, is the first major artery originating from the abdominal aorta (Figure 11-48). It divides in 3 branches: the splenic, the hepatic, and the left gastric artery. The celiac trunk bifurcates into the splenic and the common hepatic artery, with the left gastric artery arising from the celiac trunk between the aorta and the bifurcation (86%). Variations are common. Through these branches, the celiac artery supplies blood mainly to the spleen and liver, but participates also to the vascularization of the esophagus, stomach, and pancreas. Blood returning from the area of distribution of the celiac artery is diverted to the liver via the portal venous system.

Sampling Techniques

To date there is no standardized insonation technique for this vessel. Sampling of the celiac artery can be obtained from a mid-sagittal view or axial view of the fetal abdomen. The mid-sagittal view is helpful in identifying the celiac trunk as the most proximal of the unpaired branches of the abdominal aorta. The diaphragm, the abdominal aorta, and the superior mesenteric artery are additional landmarks for the identification of this vessel. Sampling should be performed close to the origin of the vessel from the abdominal aorta, with an angle of interrogation as close to 0 degrees as possible and never greater than 30 degrees. Recording is performed during absence of fetal movements and breathing, over at least 5 uniform cardiac cycles.

Morphology of the Celiac Artery Waveform and Changes with Gestational Age

The flow velocity waveform of the celiac artery reflects the pattern of the peripheral arteries. The end-diastolic velocity in the celiac artery is always positive (between 18 and 40 weeks of gestation). The peak systolic velocity (PSV) increases throughout the second half of gestation. The curve describing the PI has an inverted U shape. This pattern is similar to that observed for the splenic artery and indicates an increasing perfusion of the splanchnic organs with advancing gestation, which is paralleled by a correspondingly relative increase in portal flow to the fetal liver.³¹¹

Indices

Longitudinal reference ranges for the celiac artery PI are provided (Figure 11-49 and Table 11-14).³¹²

Implications

The current literature lacks studies on the Doppler velocimetry of the celiac artery. One advantage of this vessel is that it is relatively easy to sample when compared to the nearby hepatic artery. The findings of a recent study, however, suggest that the hemodynamics of these 2 vessels are not closely related, and that sampling of the celiac artery cannot substitute that of the more difficult to assess hepatic artery. Further studies will determine if the Doppler velocimetry of this vessel may have clinical applications.

The spleen has hematopoietic and immunological functions and operates as a blood reservoir during fetal life.³¹³ DoppIer interrogation of the splenic artery is easy to perform and has been studied to assess fetal anemia and also hemodynamic changes in the peripheral circulation in cases of fetal growth restriction.

Anatomy of the Splenic Artery

The splenic artery is the largest branch of the celiac trunk and travels from the midline to the spleen (see Figure 11-48). It is a tortuous vessel behind the stomach. The splenic vein travels along the splenic artery, and sometimes can be imaged as well.

Sampling Techniques

Interrogation of the splenic artery is best performed in a transverse section of the fetal abdomen at the level of the stomach, with the fetal spine at 3 or 9 o'clock. Using Color Doppler, the main splenic artery can be identified as illustrated in Figure 11-50. The celiac trunk is not clearly imaged in this position because its direction is often perpendicular to the ultrasound beam. The standard recommendation is to place the sample volume in the proximal segment of the vessel close to its origin from the celiac trunk. Ideally, the sample volume should be positioned so that the angle of insonation is close to zero (up to 30 degrees is acceptable). Recording should be performed when the fetus is quiet, and it is ideal to obtain at least 5 uniform cardiac cycles.

Indices

The Doppler indices used in the evaluation of the flow velocity waveform of the splenic artery include peak systolic velocity (PSV), end-diastolic velocity (EDV), time-averaged maximum velocity (TAMXV), PI, and RI. Cross-sectional^{314,315, and 316} and longitudinal³¹² reference ranges for these indices have been described.

Morphology of the Splenic Artery Waveform and Changes with Gestational Age

The flow velocity waveform of the splenic artery reflects the pattern of peripheral arteries (Figure 11-51). Blood flow velocities in the splenic artery increase with advancing gestational age. This is compatible with a progressive increase of the fetal cardiac output with gestational age, and also with a decreased impedance to flow in the splenic vascular bed, which expands as the spleen grows. The relationship between the PI and gestational age has a parabolic inverted shape and, in this regard, is similar to that of the middle cerebral artery PI (Figure 11-52).

Clinical Significance

A few studies of the splenic artery have been reported, and they focus predominantly on the assessment of SGA fetuses and fetal anemia. SGA fetuses, particularly those with abnormalities in the umbilical artery Doppler velocimetry, have a lower mean fetal splenic artery PSV,³¹⁷ a lower PI,³¹⁵ and a lower RI³¹⁴ than normal controls. This reflects reduced impedance to flow in the splenic artery and a greater splenic perfusion, which may be necessary for the spleen to provide compensatory erythropoiesis to the fetal hypoxia.³¹⁵ The PI of the splenic artery correlates with umbilical cord blood gas and pH. However, this information can be obtained by sampling other vessels more easily accessible, such as the middle cerebral artery and the IVC.³¹⁵

Doppler velocimetry of the splenic artery has been proposed for the evaluation of fetal anemia caused by Rh-alloimmunization. Affected fetuses often have splenomegaly due to extramedullary erythropoiesis. There is a correlation between the splenic artery PSV, expressed in multiples of the median, and fetal hemoglobin concentration, both prior and after a fetal blood transfusion. A new index, the deceleration angle, has been proposed for the prediction of severe anemia in Rh-isoimmunized fetuses before the development of hydrops.³¹⁸

Anatomy

The abdominal aorta divides into the 2 common iliac arteries, which diverge and travel downward and laterally into the pelvis. At the level of the sacroiliac joints, the common iliac arteries bifurcate into 2 branches, the external and internal (hypogastric) arteries. The external iliac arteries travel downward and laterally, and enter the tight passing underneath the inguinal ligaments, becoming the femoral arteries.

The internal iliac arteries, which in the fetus are twice as large as the external iliac arteries, after surrounding the bladder, become umbilical arteries. The 2 umbilical arteries ascend along the anterior wall of the abdomen toward the umbilicus, where they enter the umbilical cord and coil around the umbilical vein.