See related article, page 428
The diagnosis and treatment of the fetus that is small for gestational age (SGA) generally includes those fetuses that have been found to have an estimated fetal weight <10th percentile by ultrasound scanning. Caring for women with these fetuses (who are also commonly referred to as intrauterine growth restricted [IUGR]) is a common problem faced by obstetric clinicians. Issues that are related to diagnosis include how to screen for fetal growth abnormalities, the accuracy of ultrasound scanning, and how to distinguish pathologic growth from a fetus or neonate who is constitutionally small. In terms of management, the use of antenatal testing, assessment with Doppler ultrasound scanning, and timing of delivery are key issues that each present challenges. I will attempt to touch on these issues.
Currently, the standard of care screening tool that is used to identify fetal growth disorders is the use of serial fundal heights. Unfortunately, this tool has a poor sensitivity and specificity; at most, 50% of SGA/IUGR fetuses are identified. This has led to the use of serial ultrasound assessment in pregnancies with even the slightest increased risk for growth abnormalities such as twins, hypertension, diabetes mellitus, lupus, advanced maternal age, smokers, illicit substance users, and those women in whom fundal height measurement is challenging (eg, obesity and women with fibroid tumors). Of course, although ultrasound scanning will increase the sensitivity for growth disorders, it also has its own inaccuracies with poor sensitivity and specificity.
The delineation between SGA and IUGR relates to semantics as well as physiology/pathophysiology. In general, the term SGA is used to refer broadly to neonates or fetuses with a birthweight or estimated fetal weight that is <10th percentile. Because the estimated fetal weight is notoriously inaccurate, we commonly use the word “suspected” as a qualifier to describe the SGA fetus, but we treat these cases just the same as if we knew the exact fetal weight. Additionally, although IUGR commonly is used interchangeably with SGA, a more nuanced use of the term is to apply it to fetuses in whom one suspects that the intrauterine environment has resulted in a fetus who is less well-grown than would be optimal. Thus, the SGA fetus at the 8th percentile whose parents are both at the 1st percentile for height is less likely to be IUGR than is the same size fetus in parents at the 99th percentile for height.
What about the fetus who has a suspected intrauterine infection such as toxoplasmosis or cytomegalovirus? Or a fetus with aneuploidy? Is that SGA or IUGR? For the infected fetus, I believe most caregivers would use the term IUGR because the fetus is smaller than would have been predicted by its baseline characteristics. For an aneuploid fetus, I have heard both terms used; although an SGA fetus is expected in many cases of aneuploidy, IUGR can also result from a dysfunctional placenta. Finally, in the setting of asymmetric (particularly head-sparing) growth, IUGR generally is suspected more than in the symmetrically small fetus. It is probably best to think of SGA as referring to a fetus or neonate whose genetic potential might lead to less growth than average; IUGR should be used to describe a fetus or neonate whose in utero growth was impinged by the intrauterine environment.
In this issue of the Journal, Hanley and Janssen deal with the issue of SGA based on ethnicity-based growth distributions. When they used a broad population-based threshold to delineate SGA, they found only mild increases in neonatal morbidity. However, when they used ethnicity-specific thresholds, they found that a smaller group of neonates was identified as growth restricted. This smaller group had a much higher risk of perinatal morbidity. This work supports the aforementioned delineation between SGA and IUGR. Because race/ethnicity provides 1 predictor of birthweight, stratification of thresholds leads to a more accurate identification of truly IUGR fetuses with higher perinatal risk. Presumably, the racial/ethnic differences are based, in turn, on population norms of height and weight, diet, and comorbidities such as hypertension and diabetes mellitus. Further, in multiparous women there is previous information regarding previous birthweight that may or may not have been pathologic but which does seem to recur. So, it may be even better to create growth curves based on even more predictive characteristics such as these. In the future, one can imagine that items such as the maternal age, height, weight, paternal height and weight, race/ethnicity, altitude, previous birthweights, fetal sex, and medical comorbidities would all be entered into the ultrasound software at the time of a sonogram to deliver a patient-specific percentile estimate.
Once fundal height screening and/or ultrasound scanning have characterized a fetus as SGA/IUGR, the next step is management. Management generally includes ongoing fetal assessment (antenatal testing with NSTs and BPPs and serial ultrasound scans for growth and Doppler interrogation). In the timing of early-term delivery document from the National Institute of Child Health and Human Development conference, the issue of delivery timing in IUGR fetuses was addressed, with uncomplicated (otherwise normal) fetuses to be delivered at 38 or 39 weeks’ gestation, those fetuses with complications such as oligohydramnios or abnormal Doppler studies to be delivered at 34-37 6/7 weeks’ gestation, and those with abnormal fetal surveillance to be delivered immediately. Within these recommendations, the fetal assessment options actually are included. Thus, discussion of such fetal assessment is necessary before the timing of delivery is considered.
At the time of the diagnosis of suspected SGA/IUGR, Doppler ultrasound scanning, particularly of the umbilical artery, should be performed. With normal blood flow, concerns about immediate stillbirth or compromise can be diminished. However, with the presence of increased systolic to diastolic ratio, absent end-diastolic blood flow, or reversed end-diastolic blood flow, delivery at an earlier gestational age is indicated. With reversed end-diastolic blood flow, the acute risk is so high that delivery at nearly any viable gestational age is recommended commonly. On the other hand, the risk of death is so high at 23, 24, or 25 weeks’ gestation that expectant management, even in such a dire situation, is also used. More recently, blood flow in the ductus venosus has also been used to help characterize the pathophysiologic condition of these fetus. In the setting of any abnormal Doppler blood flow, the key question is how will the neonate do at the current gestational age vs what is the ongoing risk over the next week and the outcome if it achieves another week of gestation. Thus, at term, there is little to be gained by expectant management, even in the early term (37-38 6/7 weeks’ gestation) period; whereas in the preterm period, there is more to be gained. In a recent study by Trudell et al, it appears that delivery of 1000 SGA fetuses at 37 weeks’ gestation as opposed to waiting until 38 weeks’ gestation would prevent 2 stillbirths and lead to 5 additional neonatal intensive care unit admissions and 3 additional cases of respiratory distress syndrome. These tradeoffs appear strongly to favor delivery at 37 weeks’ gestation but would become less so at earlier gestational ages when there would also be differences in neonatal mortality rates.
In the setting of normal Doppler assessment, ongoing fetal assessment usually is performed 1-2 times per week, whereas with abnormal Doppler assessment even more frequent evaluation is recommended. I generally would recommend either a modified or complete biophysical profile so that an amniotic fluid index is obtained. The amniotic fluid index is a crude measurement of mid- to long-term placental function; as noted in the National Institute of Child Health and Human Development document, earlier delivery with oligohydramnios in the setting of suspected IUGR is recommended. In addition to fetal testing, serial growth ultrasound scans and Doppler blood flow studies should also be performed to assess interval growth.
What is to be done with a fetus who has moved from the 8th percentile down to <3rd percentile? Although the recommendations are not clear entirely for such cases, there is very clear evidence that the subset of fetuses in the <3rd percentile range are at higher risk than the others in the entire <10th percentile range. In one recent analysis, fetuses at the 5th-9th percentile had approximately twice the risk of stillbirth as those at ≥10th percentile. Those at the 3rd-4th percentile had about twice the risk of stillbirth as those at the 5th-9th percentile. Finally, those in <3rd percentile had approximately twice the risk of stillbirth as those at the 3rd-4th percentile. Thus, if a fetus moves on its growth curve from the 8th to <3rd percentile, the risk of stillbirth increases approximately 4-fold. Again, consideration of what benefit an extra week of intrauterine maturation would provide should be weighed against this risk of stillbirth.
In the end, these recent publications all provide data that lead us towards a more evidence-based approach regarding the diagnosis and management of IUGR. However, we are still far from having the gold standard, prospective randomized trials, that will remove the impact of the confounding that is likely present in many of these current studies. Until a time when such prospective trials have been conducted and published, however, we are reliant on further observational research into stratifying the risk of suspected IUGR fetuses using patient characteristics and ultrasound parameters. While such studies produce evidence with greater uncertainty, they continue to advance our understanding of this challenging clinical issue.