Doppler ultrasound of the umbilical artery for fetal surveillance in singleton pregnancies

INTRODUCTION — Doppler ultrasound waveforms are derived from Doppler frequency shifts generated by sound waves reflected by red cells moving in a circulation. Analysis of a Doppler waveform yields information on the various aspects of a circulation, including the presence and direction of flow, velocity profile, volume of flow, and circulatory impedance (ie, the resistance to a pulsatile flow) [1]. In obstetrics, use of umbilical artery (UA) Doppler, in conjunction with other fetal monitoring tools and appropriate intervention, can decrease perinatal mortality in selected high-risk pregnancies [2].

Doppler investigation of the UA flow constitutes an essential component of fetal surveillance in pregnancies complicated by fetal growth restriction (FGR), hypertension, twin-twin transfusion syndrome, and twin anemia-polycythemia sequence. This topic will discuss the use of UA Doppler in singleton pregnancies complicated by FGR. The management of FGR, the safety of Doppler sonography in pregnancy, and its use in monitoring twin pregnancies are reviewed separately.

●(See "Fetal growth restriction: Evaluation".)

●(See "Overview of ultrasound examination in obstetrics and gynecology", section on 'Safety'.)

●(See "Selective fetal growth restriction in monochorionic twin pregnancies" and "Twin-twin transfusion syndrome: Screening, prevalence, pathophysiology, and diagnosis" and "Twin-twin transfusion syndrome: Management and outcome" and "Twin anemia-polycythemia sequence (TAPS)".)

UMBILICAL ARTERY DOPPLER IN UNCOMPLICATED PREGNANCIES — In normal pregnancies, the fetoplacental vascular system continues to expand with advancing gestation, which ensures sufficient placental nutrient and oxygen transfer to support fetal growth and well-being. Specifically, fetoplacental arterial impedance progressively declines and umbilical blood flow progressively rises. These normal vascular changes are reflected by a progressive decline in the umbilical artery (UA) Doppler indices. Forward flow in the UA at the end of diastole (ie, end-diastolic flow [EDF]) of the fetal cardiac cycle is noted by 14 to 16 weeks of gestation and progressively increases (image 1), thus providing an uninterrupted supply of nutrients and oxygen to the growing fetus throughout the cardiac cycle.

UMBILICAL ARTERY DOPPLER IN FETAL GROWTH RESTRICTION — In fetal growth restriction (FGR), the umbilical artery (UA) end-diastolic flow (EDF) is reduced due to underlying chronic placental insufficiency. The resulting fetal deprivation of nutrients and oxygen is accompanied by sequential fetal compensatory hemodynamic responses, which provide the basis for clinical evaluation and management of FGR [3,4]. This is often a progressive process:

●Rising impedance is initially observed in the fetoplacental circulation and is reflected in the falling EDF velocity and the rise of UA Doppler indices [3,5].

●With continuing deprivation, blood flow is redistributed to favor perfusion of the vital organs, such as the brain, heart, and adrenals, at the expense of flow to muscle, viscera, skin, and other less critical organs [6]. The cerebral redistribution is reflected in a decreased cerebroplacental ratio (CPR), which is the ratio of the middle cerebral artery (MCA) and the UA Doppler indices.

●With rising fetoplacental circulatory impedance, the EDF may become absent (AEDF) or reversed (REDF) (image 2A-C), which signify worsening fetal status. These changes in EDF occur on average one week before acute deterioration [7].

●Progressive decline of the fetal condition is further manifested by the absence or reversal of the ductus venosus atrial wave (a), diminished and eventual loss of fetal heart variability and reactivity, and the loss of fetal movement and breathing, indicating impending fetal jeopardy [6].

Although this sequence is common, it is not inevitable. In some cases, an initially high Doppler index may progressively decline with advancing gestation, which may indicate a relatively improved prognosis, but the fetus still remains at risk of adverse outcomes.

DOPPLER EXAMINATION OF THE UMBILICAL ARTERY

Procedure — The essential steps of the umbilical artery (UA) Doppler examination are described below. Comprehensive guidelines for obtaining fetal Doppler measurements are available from the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) [8].

●The patient should be counseled regarding the relevant aspects of the test, and placed in a semirecumbent position with left lateral tilt to avoid maternal inferior vena cava compression.

●The principles for safe use of ultrasound are observed with the thermal and mechanical indices set below 1, usually by default setting of the device. (See "Overview of ultrasound examination in obstetrics and gynecology", section on 'Safety'.)

●The wall filter settings for pulsed Doppler are kept at the lowest level (approximately 50 to 60 Hz) to prevent the elimination of low-velocity flow signals. (See 'Technical factors' below.)

●An initial ultrasound examination is performed to confirm that the fetus is quiescent as fetal breathing, movements, or hiccups affect the UA Doppler waveforms. Duplex Doppler ultrasound is used to identify a free-floating loop of cord for sampling; adding color Doppler may be helpful. (See 'Physiological factors' below.)

●The Doppler cursor line is aligned with the vessel axis, with the angle of insonation as close to zero as possible. The minimal angle provides the highest maximum velocity envelope. Because the ratios are virtually independent of the angle of insonation, Doppler waveforms are not corrected for angle in clinical practice. Most ultrasound systems assume a zero angle, which may be noted in the device display. (See 'Technical factors' below.)

●The Doppler sample volume, the site of Doppler interrogation, is then placed in the vascular lumen and its axial length adjusted to encompass the lumen. Switching to the spectral Doppler mode leads to the generation of Doppler waveforms (movie 1). Pulse repetition frequency scale should be adjusted so that the waveforms occupy approximately 75 percent of the Doppler panel. The sweep speed is selected to provide 6 to 10 waveforms in the window.

●The device software computes the maximum velocity envelope of the Doppler waveforms, identifies the peak systolic and end-diastolic velocities, averages usually three uniform waveforms with the highest maximum velocity envelope, and generates the indices (see 'Doppler indices' below). This process can also be accomplished manually.

Doppler indices — UA Doppler waveform analysis is based upon the following characteristics of the maximum velocity envelope (waveform 1):

●Peak systolic flow velocity (S)

●End-diastolic flow velocity (D)

●Average flow velocity over the cardiac cycle (A)

These three values are used to develop indices that measure the pulsatility of the Doppler waveform reflecting the dynamic changes in flow through the cardiac cycle. The common Doppler indices used for obstetric applications are:

●Peak systolic to end-diastolic flow ratio (S/D) [9]

●Pulsatility index (PI = S-D/A) [10]

●Resistance index (RI = S-D/S) [11]

This author has reported interobserver and intraobserver variances of 10 to 14 percent and 5 to 9 percent, respectively, for the Doppler indices measured by continuous wave Doppler [12,13]. These findings are consistent with a subsequent study that reported intraobserver coefficient of variations of 10.5, 6.8, and 13.0 percent for PI, RI and S/D, respectively, using directional guidance of pulsed duplex Doppler system [14].

Factors affecting the umbilical artery Doppler waveform — UA Doppler waveforms are influenced by procedural techniques, physiology of pregnancy, and the use of certain pharmacological agents. These are summarized below.

Technical factors

●Angle of insonation – We advise aligning the cursor along the vessel axis as much as achievable so that optimal waveforms are obtained (image 2A). The angle of insonation is the most significant technical modifier of Doppler waveforms, although the Doppler indices are virtually angle independent. The angle affects the magnitude of the Doppler shift, which will be highest when the Doppler beam path (as indicated by the cursor line) is parallel to the vessel axis and the angle is thus 0 degrees. In this setting, the Doppler shifts will measure the true velocity of blood flow.

●Sampling site – We use a free-floating loop of the cord for the Doppler measurement rather than either end of the cord. The sampling site is a significant source of variance of the Doppler indices [12]. The indices are higher at the fetal abdominal end than at the placental end of the cord [15]. The hemodynamic concept of impedance and wave reflection provides insight into this phenomenon [1]. The closer the measurement site is to the placenta, the lesser the impedance and the greater the end-diastolic flow (EDF).

However, a prospective study found that, although the PI was significantly lower in the free loop than at the UA adjacent to the fetal urinary bladder, the contribution of the measurement site to the total variability of the PI was insignificant and selection of a specific location of interrogation did not improve the variability [16].

●Wall filter setting – We utilize a wall filter setting of 50 to 60 Hz as a default. The wall filter removes high-amplitude, low-frequency Doppler signals generated by the vascular wall movements from the total Doppler flow signals. However, a higher-filter setting in UA Doppler may also remove low-frequency flow signals during the end-diastole and falsely suggest absent EDF (AEDF). Therefore, the wall filter should be set as low as achievable for a specific ultrasound device.

Physiological factors

●Gestational age – UA EDF increases progressively with advancing gestation (image 1) due to the normal exponential expansion of the fetoplacental vascular system and decline in fetoplacental flow impedance; Doppler indices decline as well [17]. Gestational age is the main contributor to the total variance of Doppler indices, ranging from 33 percent for the PI to 46 percent for the S/D [12].

●Fetal heart rate – Heart rate influences the arterial Doppler waveform and therefore affects the Doppler indices [18]. In a prospective study that quantified this effect, 15 to 18 percent of the total variance of the indices could be attributed to the heart rate, primarily mediated through the diastolic phase of the cardiac cycle [19]. However, within the normal limits of the fetal heart rate (120 to 160 beats per minute), the changes in the Doppler indices are clinically insignificant [20].

●Fetal breathing and hiccups – Doppler interrogation of the UA should be performed only in the absence of fetal breathing and hiccups. Substantial variations in the intrathoracic pressure and central hemodynamics occur during fetal breathing or hiccups [21-23], causing dynamic variability in the Doppler waveforms (image 3).

●Fetal movement – Doppler examination should be performed during fetal quiescence. Fetal movements generate high-amplitude, low-frequency Doppler signals from sources extraneous to the umbilical blood flow and interfere with a stable Doppler interrogation [1,12,16].

●Circadian rhythm – The significance of circadian rhythm in clinical practice of fetal Doppler analysis remains unproven. Although earlier investigators failed to observe any appreciable diurnal changes in the UA Doppler waveform [24], a prospective study reported significant diurnal changes in middle cerebral and umbilical flow velocities in singleton pregnancies in the third trimester [25].

●Physical activity – Most studies failed to show any changes in UA Doppler indices from mild to moderate exercises, such as bicycle ergometry, treadmill and walking, except when there were alterations in the fetal heart rate. When pregnant athletes exercised at >90 percent maternal oxygen consumption, fetal bradycardia and elevated UA PI resulted, but these changes returned to baseline shortly after the cessation of exercise [26]. In a study on yoga in pregnancy, there were no changes in UA Doppler indices after one hour of yoga practice [27].

Pharmacological factors — UA Doppler can be affected by certain pharmacological agents that may directly or indirectly affect fetal or maternal circulations.

●Antenatal corticosteroids – The most frequently encountered example is betamethasone or dexamethasone administration to pregnancies where preterm birth is anticipated: Several reports have confirmed variable (30 to 50 percent) improvements in UA Doppler. However, such changes were transient and did not impact the outcomes.

●Magnesium sulfate – Another example is the maternal magnesium sulfate administration to prevent eclampsia, which can reduce the UA PI, but its prognostic significance remains uncertain.

Interpretation of UA Doppler indices — Doppler indices are interpreted either quantitatively based on their single point threshold value or distribution, or qualitatively based on the absence or reversal of UA EDF.

Quantitative interpretation — An S/D ratio >3.0 or RI >0.6 at ≥28 weeks of gestation is the best single point threshold for identifying pregnancies at high risk of adverse outcomes [28] and is used by many clinicians. Another common approach is to use the gestational age-specific percentile distribution of the Doppler indices [29,30]. A UA Doppler index (S/D, PI, or RI) greater than 95^th percentile for gestational age is the recommended threshold for abnormal UA Doppler [29].

The choice of the best nomogram can be challenging. A systematic review of observational studies found significant heterogeneity and suboptimal methodological quality, highlighting the need for reliable reference ranges [31]. More reliable nomograms are available from prospective well-designed longitudinal studies in low-risk [14] and diverse populations [14,32], and are often incorporated in ultrasound reporting systems.

Qualitative interpretation — Absent and reversed end-diastolic flow velocities (AEDF and REDF) in the UA are signs of fetoplacental circulatory deterioration due to increased vascular impedance to flow and associated with adverse perinatal outcomes. The incidence of AEDF or REDF depends on the risk category of the pregnancy. In a systematic review of 42 studies with over 18,000 mostly low-risk or unselected populations from higher-income countries, the prevalence of AEDF or REDF varied widely, from 0.08 to 2.13 percent [33]. In high-risk pregnancies, it varies from 2 to 56 percent [34]. The wide variations can be explained by the differing standards of risk categorization, severity of the underlying obstetric complications such as growth restriction, and the technique used for Doppler imaging, especially the level of the high-pass filter setting.

In a meta-analysis of the risk of fetal death in FGR before 34 weeks of gestation, the risks of stillbirth with UA AEDF and REDF were 6.8 and 19 percent, respectively [35]. AEDF and REDF are also associated with a higher prevalence of aneuploidy and congenital anomalies (table 1) [36-38]. In a review of 1126 cases of AEDF and REDF compiled from the literature, the stillbirth and neonatal mortality rates were 17 and 28 percent, respectively [38]. Most of the deaths were related to FGR, preterm birth, fetal anomalies, and aneuploidy (especially trisomy 13, 18, and 21).

AEDF may be present intermittently. Although the perinatal outcome is better in these fetuses compared with those with persistent AEDF, the fetus still remains at risk; about 34 percent are delivered for nonreassuring fetal condition and over 50 percent have increased neonatal morbidity [39].

EVIDENCE OF EFFECTIVENESS

High-risk pregnancies — In a meta-analysis of randomized trials of fetal and umbilical Doppler ultrasound in high-risk pregnancies (16 randomized trials, over 10,000 pregnancies), the use of Doppler resulted in fewer perinatal deaths (1.2 versus 1.7 percent, risk ratio [RR] 0.71, 95% CI 0.52-0.98) and the number needed to prevent one death was 203 (95% CI 103-4352) [2]. Additional benefits included fewer inductions (RR 0.89, 95% CI 0.80-0.99) and cesarean births (RR 0.90, 95% CI 0.84-0.97). This review, however, did not specify which high-risk group would benefit most from umbilical artery (UA) Doppler surveillance. A previous meta-analysis addressed this issue and showed that the Doppler surveillance primarily benefited pregnancies complicated by fetal growth restriction (FGR) and/or hypertensive complications of pregnancy [40].

UA Doppler may be less predictive of adverse perinatal outcome in FGR pregnancies >32 completed weeks than FGR diagnosed earlier in gestation, which may be explained by evidence that late onset FGR is distinct clinically and pathophysiologically from early onset FGR [41]. The issue, however, remains controversial, especially the gestational age threshold defining the classification [42]. In late-onset FGR, fetal blood flow redistribution as reflected by middle cerebral artery (MCA) Doppler and the cerebroplacental ratio (CPR) may be superior to UA Doppler indices alone in predicting adverse perinatal outcomes [43-45]. However, there are no randomized trials supporting the effectiveness of MCA Doppler or CPR. In a prospective cohort study (TRUFFLE-2) of singleton pregnancies at risk of FGR in late-preterm gestations, cerebral flow redistribution, as reflected by the MCA Doppler <5^th percentile and high gestational age-specific UA/MCA Doppler ratio (umbilicocerebral ratio [UCR], which is the inverse of the CPR), demonstrated the highest relative risk for composite adverse outcomes [46]. Adjustment for gestational age at birth and birth weight reduced the importance of the association, but it remained statistically significant. This study highlighted the need for a randomized trial to demonstrate the effectiveness of MCA Doppler and CPR or UCR in clinical practice, and such a trial is in progress.

Low-risk and unselected pregnancies — In contrast to high-risk pregnancies, clinical trials of UA Doppler in lower-risk pregnancies have not shown any improvement in pregnancy outcome. In a meta-analysis of four trials of fetal and umbilical Doppler surveillance in over 11,000 low-risk and unselected pregnancies, perinatal death was not significantly reduced (RR 0.80, 95% CI 0.35-1.83); the only trial that assessed serious neonatal morbidity was not informative (RR 0.99, 95% CI 0.06-15.75) [47]. Given the lack of evidence, we recommend not performing UA Doppler surveillance in low-risk pregnancies.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Fetal surveillance".)

SUMMARY AND RECOMMENDATIONS

●UA Doppler indices used in obstetrics – The common Doppler umbilical artery (UA) indices used for obstetric applications are (waveform 1):

•Peak systolic to end-diastolic flow ratio (S/D)

•Pulsatility index (PI = S-D/A)

•Resistance index (RI = S-D/S)

An S/D ratio >3.0 or RI >0.6 at ≥28 weeks of gestation is the best single point threshold for identifying pregnancies at high risk of adverse outcomes. Another common approach is to use Doppler index (S/D, PI, or RI) greater than 95^th percentile for gestational age as the threshold for abnormal UA Doppler. (See 'Doppler indices' above and 'Quantitative interpretation' above.)

●Normal UA Doppler indices – In pregnancies without placental insufficiency, the fetoplacental vascular system continues to expand with advancing gestation, which results in a progressive decline in the UA Doppler indices (image 1). (See 'Umbilical artery Doppler in uncomplicated pregnancies' above.)

●UA Doppler indices in placental insufficiency – In chronic placental insufficiency, fetal deprivation of oxygen and nutrients is accompanied by sequential fetal compensatory hemodynamic responses. With rising fetoplacental circulatory impedance, UA end-diastolic flow (EDF) may become absent (AEDF) or reversed (REDF) (image 2A-C), which signifies worsening fetal status. (See 'Umbilical artery Doppler in fetal growth restriction' above.)

●Procedure – UA Doppler waveforms are influenced by procedural techniques, physiology of pregnancy, and the use of certain pharmacological agents. These factors must be recognized for accurate performance of the procedure and interpretation of results. (See 'Factors affecting the umbilical artery Doppler waveform' above.)

●Clinical use – Monitoring pregnancies at high risk of uteroplacental insufficiency, particularly those with fetal growth restriction (FGR), with UA Doppler coupled with appropriate intervention reduces perinatal mortality by approximately 30 percent. UA Doppler monitoring in lower-risk pregnancies does not improve pregnancy outcome. (See 'Evidence of effectiveness' above.)