Fetal growth restriction remains a challenging entity with significant variations in clinical practice around the world. The different etiopathogenesis of early and late fetal growth restriction with their distinct progression of fetal severity and outcomes, compounded by doctors and patient anxiety adds to the quandary involving its management. This review summarises the literature around diagnosing and monitoring early onset fetal growth restriction (early onset FGR) with special emphasis on optimal timing of delivery as guided by recent research advances.
Highlights
- •
Early and late onsets FGR have different manifestations probably as a consequence of different aetiopathologies.
- •
The middle cerebral artery (MCA) Doppler is a proxy hallmark of hypoxia.
- •
Abnormal ductus venosus Doppler waveform can reliably predict fetal acidaemia and neonatal death.
- •
Monitoring should commence from 24 weeks in women identified as being at high risk of FGR at booking.
Introduction
The diagnosis and management of fetal growth restriction (FGR) still remain an unresolved dilemma in modern day obstetrics. This is due to several reasons such as unclear diagnostic criteria, variable monitoring methods, complex confounding factors affecting different gestational ages and obstetrician and patient anxiety. These factors are further compounded by lack of robust evidence although recently progress has been made in this regard . FGR is known to be associated with significant neonatal morbidity and mortality which has large economic implications for the health service. It has been shown that improved detection of FGR could potentially save £360,000 a year in a unit with 3000 births ( www.perinatal.org.uk ). As our understanding continues to evolve, and given that there is no treatment for this condition, it is crucial to make an accurate diagnosis at a correct gestational age as this will determine the timing of delivery and the perinatal outcomes associated with it. Over the last couple of decades, it has become clear that FGR can start early in the gestation when it is termed early onset fetal growth restriction (early onset FGR); and this follows a more severe trajectory in terms of neonatal outcome as compared to late onset fetal growth restriction (late onset FGR) . Studies have shown that early onset FGR can rapidly deteriorate from umbilical artery and venous abnormalities to abnormal biophysical profile leading to interventions and significant neonatal morbidity and mortality . Furthermore, early onset FGR has shown to be associated with other pathologies such as preeclampsia (PET) and increased perinatal mortality as compared to late onset FGR which is associated with a milder form of PET . Therefore, there is a clear need to identify the gestational age cut off to be able to label FGR as early or late, given the wider implications for antenatal management and delivery. Unfortunately, studies done until now have used arbitrary cut off gestational ages to diagnose early onset FGR. However, this wide difference in the gestational age cut off might be considered acceptable in the broader medical community, given the variable guidance followed and the logistics of available diagnostic tools. Also, given the lack of robust research in early onset FGR, it is likely that the different gestational age ranges could be a result of consensus rather than scientific reasoning. However, a recent large prospective study evaluated the gestational age cut off according to perinatal mortality and adverse perinatal outcome in a cohort of 656 pregnancies with FGR . They identified 32 weeks as the diagnostic cut off for maximising differences between early onset FGR and late onset FGR, particularly in the absence of umbilical artery Doppler information.
Definition
The agreed definition of the American College of Obstetricians and Gyneoclogists (ACOG) and the Royal College of Obstetricians and Gynaecologists (RCOG) to identify a growth restricted fetus is based on either the estimated fetal weight (EFW) or abdominal circumference (AC) < 10 th centile. However it became imperative to ascertain exactly which sonographic findings are truly associated with adverse perinatal outcomes. In this regard, the PORTO study analysed 1100 pregnancies with EFW < 10 th centile and looked at various sonographic biometric and Doppler flow markers. The authors concluded that EFW < 3 rd centile along with abnormal umbilical artery Doppler velocimetry was consistently associated with adverse perinatal outcome and therefore could be taken as a definition of FGR. This definition gains further credence from the multicentre TRUFFLE study where 503 women with FGR fetuses were recruited and followed until delivery to describe the perinatal morbidity and mortality depending on the sonographic features and gestational age of diagnosis. More recently in an attempt to accurately define FGR by international consensus, a Delphi consensus methodology was used. Experts provided there opinion on various parameters to diagnose both early and late onset FGR. For early FGR, three solitary parameters of AC or EFW < 3 rd centile and absent umbilical end diastolic flow and four contributory markers of AC or EFW < 10 th centile with a PI > 95 th centile in either umbilical or uterine flow were agreed upon .
Definition
The agreed definition of the American College of Obstetricians and Gyneoclogists (ACOG) and the Royal College of Obstetricians and Gynaecologists (RCOG) to identify a growth restricted fetus is based on either the estimated fetal weight (EFW) or abdominal circumference (AC) < 10 th centile. However it became imperative to ascertain exactly which sonographic findings are truly associated with adverse perinatal outcomes. In this regard, the PORTO study analysed 1100 pregnancies with EFW < 10 th centile and looked at various sonographic biometric and Doppler flow markers. The authors concluded that EFW < 3 rd centile along with abnormal umbilical artery Doppler velocimetry was consistently associated with adverse perinatal outcome and therefore could be taken as a definition of FGR. This definition gains further credence from the multicentre TRUFFLE study where 503 women with FGR fetuses were recruited and followed until delivery to describe the perinatal morbidity and mortality depending on the sonographic features and gestational age of diagnosis. More recently in an attempt to accurately define FGR by international consensus, a Delphi consensus methodology was used. Experts provided there opinion on various parameters to diagnose both early and late onset FGR. For early FGR, three solitary parameters of AC or EFW < 3 rd centile and absent umbilical end diastolic flow and four contributory markers of AC or EFW < 10 th centile with a PI > 95 th centile in either umbilical or uterine flow were agreed upon .
Etiopathogenesis
It is imperative to understand the pathogenesis resulting in early onset FGR and late onset FGR to be able to guide management and delivery; particularly they are governed by different placental disorders. Early onset FGR stems from the reduction of the villous vascular area, typically more than 30%, occurring in the second trimester and resulting in increased resistance to the umbilical arterial flow . This usually causes fetal biometry which, in conjunction with raised umbilical artery resistance constitutes the diagnosis of early onset FGR . Early onset FGR constitutes up to 30% of all FGR and is less common than late onset FGR which constitutes up to 70% . Late onset FGR occurs in the third trimester and is more associated with impaired maturation of the villi rather than reduction in the surface area. As a result, the umbilical arterial blood flow may not necessarily be impeded as the villous immaturity does not impact on the resistance; rather, it hampers the gaseous and nutrient exchange . The fetus senses the hypoxic conditions and responds by reducing impedance in the middle cerebral artery (MCA) . Therefore, whilst umbilical artery Doppler is crucial in the diagnosis and monitoring of early onset FGR, the middle cerebral artery (MCA) Doppler is probably more useful to diagnose late onset FGR. The progression of the condition in early onset FGR can take many weeks unless pre-eclampsia supervenes, and is more variable in late onset FGR [ Figure 1 ]. In early onset FGR the sequence usually starts from abnormal Doppler velocimetry in the umbilical artery, MCA, ductus venosus (DV) followed by abnormal cardiotocograph (CTG) findings while in late onset FGR it is mainly abnormal MCA or abnormal umbilical artery Doppler velocities . However it has been reported that the time interval between absent or reversed DV ‘a’ wave velocity to fetal death could be up to a week .
Histopathological findings associated with early onset FGR and clinical correlations
The pathological basis for early onset FGR is classically related to the impaired trophoblastic invasion by the spiral arteries so it is termed as an abnormality of the villous vessels causing massive lesions of the placental structure . Although this is the basis of the association of early onset FGR with other diseases which also affect placental vessels such as preeclampsia , a rather tenuous relationship between histopathological placental lesions with FGR and PET has also been reported . Abnormal uterine artery Doppler velocimetry reflects uterine malperfusion further confirmed by placental bed biopsies showing defective spiral artery remodelling and fibrinoid necrosis . Abnormal uterine artery resistance associated with defective trophoblastic migration on placental bed biopsies is also seen in preterm PET, thus underlying the extent of similarity between these conditions . In fact convincing evidence from a recent meta-analysis has shown that uterine artery Doppler abnormalities not only reflect abnormal placental vasculature but can anticipate its progression . Studies have also reported an association between umbilical artery pulsatility index and the proportion increase in abnormal fetal stem artery thickness sometimes present in up to 90% of cases with absent EDF . It is interesting that further reports have not only confirmed this finding but also shown that changes in the fetal stem arteries progress in parallel with umbilical artery Doppler abnormalities and in most cases precede abnormal uterine artery Doppler abnormalities .
Histopathological findings associated with late onset FGR and clinical correlations
Unlike the similarity in the clinical and pathological findings of early onset FGR and preterm PET, the same association between late onset FGR and PET is less common . Late onset FGR represents a failure of the fetus to achieve its optimal growth potential, likely secondary to placental insufficiency . Studies have shown a much lower incidence of uteroplacental lesions in late onset FGR, and in the majority of cases they were deemed unremarkable . The placental lesions are generally not significant enough to translate into increased uteroplacental and fetal vascular resistance . Other villous lesions reported include fibrosis, hypovascularity and avascularity, suggestive of fetal thrombotic event but no clinical correlation has been found .
Maternal cardiovascular function and FGR
Whether placental lesions are a cause of FGR or an effect of systemic maternal cardiovascular dysfunction remains controversial. There is emerging evidence suggesting that abnormal uterine artery Doppler may reflect maternal arterial function and the contribution of placental and cardiovascular components lead to the abnormal expression of the waveform. Furthermore, FGR is associated with low maternal cardiac output and high vascular resistance . It is therefore likely that the combination of reduced placental perfusion and abnormal placental function are implicated in the pathogenesis of FGR.
Investigations
Uterine artery Doppler
Impaired placentation with aberrations of trophoblastic spiral arterial invasion is associated with FGR, pregnancy induced hypertension and PET . In the first trimester uterine artery Doppler notches are found in up to 65% of normal pregnancies but even persistent notching beyond 20 weeks has a lower positive predictive value for both PET and FGR in a high risk population as compared to women at lower risk of these conditions ( Figures 2 and 3 ). On the contrary, its negative predictive value is high at 97% in the high risk population. Predictive models developed by Fetal Medicine Foundation in which uterine artery Doppler velocimetry was combined with various biochemical markers in the first trimester have shown variable detection rates of 52.3% and 73% for a 10% false positive rate . Furthermore, a systematic review of 74 studies of PET and 61 studies of FGR has shown that Uterine artery Doppler abnormalities are better predictors of PET than FGR . A recent meta-analysis involving 55,974 women investigating the accuracy of Uterine artery Doppler in predicting FGR has shown that its sensitivity for identifying early onset FGR was only 39% while the sensitivity for predicting stillbirth was 14.5%. Therefore, risk prediction with uterine artery Doppler screening is no better than the risk prediction using other risk factors. The poor reproducibility of first trimester Uterine artery Doppler and variable results from past studies have resulted in focus on other impedance markers as predictors of FGR rather than Uterine artery Doppler alone. Combining Uterine artery Doppler flow studies with biochemical markers may improve the prediction of FGR as shown in a recent study .
Second trimester uterine artery Doppler studies have shown better predictive performance than first trimester studies , in particular for early onset PET and FGR . The concept of individualised risk prediction based on uterine artery Doppler and maternal factors was first described following uterine artery Doppler assessment in the late second trimester . Uterine artery Doppler remains a useful modality for assessing the aetiology of FGR in the second trimester.
Umbilical artery Doppler
There is clear evidence of the benefit of umbilical artery Doppler measurement in order to reduce perinatal deaths with a reduction of up to 29% in high risk pregnancies . While increased resistance to umbilical artery flow helps in the diagnosis of FGR, reversed flow can correlate with neonatal mortality . Umbilical artery Doppler is crucial in the diagnosis of early onset FGR especially as it is difficult to interpret CTG findings in fetuses under 28 weeks of gestation. Unfortunately, in common clinical practice there is still heavy reliance on CTG findings beyond 28 weeks of gestation and, by the time the tracing becomes pathological, up to 80% of fetuses are already hypoxic . There is evidence from nine randomised trials that perinatal mortality and length of hospital stay are reduced when umbilical artery Doppler monitoring is used along with CTG tracings . A European trial further noted the positive correlation between worsening changes in the umbilical artery Doppler velocimetry and perinatal outcomes . Early changes in umbilical artery Doppler velocimetry were present around 15 days prior to deterioration, whilst the most severe abnormalities i.e. reversed flow, were present around 4-5 days before delivery . The progression of the abnormal changes in the umbilical artery Doppler can be rapid in early onset FGR but absent or reversed flow after 26 weeks of gestation has been shown to have an independent impact on neurodevelopment . Children born with these abnormalities have been further shown to have impaired cognitive impairment and motor development at the age of 2 years . Umbilical artery Doppler, whilst signifying an ‘at risk’ fetus, is not good at predicting or informing timing of delivery in early onset FGR.
Middle cerebral artery Doppler
Unlike umbilical artery Doppler, MCA Doppler is a proxy for hypoxia and may be abnormal for many weeks in early onset FGR. The MCA Doppler may be useful in tracking late onset FGR independent of umbilical artery Doppler findings, as there is an association with adverse perinatal outcomes . It can also be used to predict emergency Caesarean section; a lower MCA pulsatility index (PI) and cerebro-placental ratio is associated with an increased risk of abdominal delivery by a factor of 6 as compared to MCA with a normal PI . These children have also been shown to have impaired neurodevelopment at 2 years of age . Since early onset FGR is associated with brain sparing, the interpretation of the MCA Doppler findings is unclear especially in relation to timing of delivery. Nevertheless, MCA PI less than the 5 th centile can evaluate cerebral vasodilatation and a recent study has shown that CPR <1 is associated with worse perinatal outcome . Similarly, the cerebroplacental ratio (CPR), which is the ratio of MCA PI to the umbilical artery PI, is more useful in late onset FGR, as impaired CPR is in up to 25% of late onset FGR and is associated with a worse outcome at delivery as compared to using MCA Doppler alone . Although one recent study has shown that CPR correlates with adverse perinatal outcomes, even in early onset FGR , CPR alone may not be predictive for adverse outcome nor give information regarding the optimal time of delivery, leaving uncertainty as to what factors should trigger delivery in late onset FGR .
Ductus venosus (DV) Doppler
There is strong evidence to suggest that (DV) Doppler predicts the risk of fetal death in early onset FGR and absent or reversed ‘a’ waves are associated with perinatal mortality irrespective of gestational age, with a risk up to 100% in early onset FGR . A systematic review of 18 observational studies has also shown the importance of DV Doppler velocimetry in predicting perinatal mortality . It has been shown that DV abnormalities precede changes in computerised CTG in 50% of cases whilst it predates abnormal biophysical profile (BPP) by up to 3 days . It is therefore considered a better investigation for determining the timing of delivery of fetuses especially after steroid administration. There have been 2 multicentre randomised controlled trials (RCTs) investigating timing of delivery in early onset FGR, namely GRIT and TRUFFLE . The growth restriction intervention trial (GRIT) included women before and after 30 weeks of gestation but has been criticised for timing delivery based on clinicians judgment rather than on objective signs . It did not specifically evaluate Doppler findings. The outcomes at 2 years of age in both the groups, i.e. immediate or deferred delivery following signs of fetal distress, were similar and the optimum timing of delivery was unclear. In the more recent TRUFFLE trial the average age of entry was 29 weeks of gestation and the women were randomised in one of the three groups’ reduced CTG short term variation (STV), early or late DV changes respectively. They found that, although survival without neurological impairment was similar in all three groups, the timing of delivery according to late changes in DV was associated with an improvement in developmental outcomes at 2 years of age . This trial reinforces the importance of abnormal DV changes in the management of early onset FGR.
Aortic isthmus Doppler
The oxygenated blood from the placenta enters the left atrium via the DV, left hepatic vein, inferior vena cava, right atrium and foramen ovale, and the deoxygenated blood enters the right atrium through the inferior and superior vena cava. From there, it passes into the right ventricle and is pumped out through the pulmonary artery. It passes through the ductus arteriosus to reach the aorta at the aortic isthmus. Thus, these two streams, the oxygenated and deoxygenated blood, meet substantively for the first time in the aortic isthmus. The aortic isthmus (AI) is therefore very important because it acts as a crucial interface connecting the oxygenated and deoxygenated circulations . Animal studies have shown that not only does the AI blood flow become retrograde prior to a decrease in cerebral blood flow , but in fact changes in AI flow become apparent even before changes in umbilical artery velocimetry . Human studies have also reported an association between abnormal changes in AI velocimetry and neonatal mortality and neurological morbidity in early onset FGR . Reversed AI flow could be seen as an late step in the spectrum of Doppler abnormalities however AI flow abnormalities have also been found in some cases of late onset FGR . Furthermore, abnormal AI flow precedes abnormal DV flow by a week and therefore deemed not a suitable marker to predict fetal death in early onset FGR . Abnormal AI flow in early onset FGR might be of value in predicting neurological injury but more research and training is required to incorporate into clinical practice.
Cardiotocography
Routine CTG in FGR fetuses has been reported to have a high (50%) false positive rate . Furthermore, interpretation of the fetal heart rate (FHR) poses a challenge especially in extremely small fetuses. However, significantly reduced and prolonged variability or unprovoked decelerations are considered preterminal events and therefore not useful for early identification . Bracero et al reported that computerised CTG (cCTG) could lead to 5 times fewer interventions as compared to non-computerised CTG use (9% vs. 49%). Furthermore, although the perinatal outcome was similar in both groups, patients spent less time per test in the cCTG than in the visual interpretation group. NICU stay was also slightly shorter in the cCTG group than in the visual CTG group, although this difference could be attributed to chance. There is evidence suggesting that the ability to predict fetal death is similar for the short term variability (STV) which is evaluated late changes in DV . A Cochrane database review suggested that STV correlates with acidosis and severe hypoxia, and a longitudinal study has reinforced its role as an marker of acute compromise, which coincides with abnormal DV flow velocimetry .
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
