Intrauterine growth restriction is diagnosed primarily through ultrasound, but pinning it down is harder than it sounds. A baby that simply measures small on a scan is not necessarily growth-restricted, and a baby whose weight looks acceptable on a chart can still be in trouble. The diagnosis depends on a combination of estimated fetal weight, abdominal circumference measurements, Doppler blood-flow studies, and sometimes blood tests drawn from the mother, all interpreted in context. Getting the distinction right matters because true IUGR carries real risks for the baby, and the surveillance and timing of delivery decisions that follow a confirmed diagnosis are very different from those for a constitutionally small but healthy fetus.
Small for Gestational Age Is Not the Same Thing
The single most common source of confusion is the overlap between “small for gestational age” (SGA) and intrauterine growth restriction. SGA is a statistical label: it means the estimated fetal weight or birth weight falls below the tenth percentile on a growth chart. Many SGA babies are perfectly healthy. They are just small, often because their parents are small. IUGR, by contrast, means the fetus is not reaching its own growth potential because something is going wrong, usually with the placenta’s ability to deliver oxygen and nutrients. A fetus can be SGA without being growth-restricted, and a fetus can be growth-restricted while still measuring above the tenth percentile if it was genetically destined to be larger.
To separate the two, clinicians look for additional red flags beyond size alone: abnormal Doppler blood-flow patterns in the umbilical artery, reduced amniotic fluid, a failure to grow between scans, or a weight estimate below the third percentile rather than just the tenth.1PubMed Central. Intrauterine Growth Restriction: Need to Improve Diagnostic Accuracy and Evidence for a Key Role of Oxidative Stress in Neonatal and Long-Term Sequelae This is why a single scan showing a small baby is never enough to make the diagnosis. Serial measurements and Doppler studies are almost always needed to determine whether you are dealing with a healthy small fetus or a genuinely compromised one.
Early Versus Late Onset
Growth restriction diagnosed before 32 weeks and growth restriction diagnosed at or after 32 weeks behave so differently that researchers now treat them as distinct conditions. Early-onset FGR tends to involve more severe placental dysfunction. It is strongly associated with preeclampsia and carries substantially higher perinatal mortality. One study evaluating the 32-week cutoff found perinatal mortality of about 7% in the early group compared with 0% in the late group, and a composite of serious adverse outcomes in roughly 13% versus 5%.2Fetal Diagnosis and Therapy. Evaluation of an Optimal Gestational Age Cut-Off for the Definition of Early- and Late-Onset Fetal Growth Restriction Early FGR is also more closely linked to preeclampsia, with the association present in about a third of early cases compared with about one in eight late cases.
The diagnostic paradox is that early-onset FGR is usually easier to identify but far more dangerous to manage, because the fetus is too premature for safe delivery. Late-onset FGR, which accounts for the majority of cases, is harder to catch on ultrasound but simpler to manage once detected because delivery is a viable option.3Journal of Fetal Medicine. Fetal Growth Restriction (FGR): How the Differences Between Early and Late FGR Impact on Clinical Management? This is one reason late FGR often goes undiagnosed until birth: the standard Doppler abnormalities that make early FGR easy to spot may not be present when growth restriction develops in the final weeks.
How Formal Diagnostic Criteria Work
An international consensus established through expert agreement set out specific criteria for diagnosing FGR, and the thresholds differ depending on when the problem appears. For early FGR (before 32 weeks), a fetus qualifies if any one of three standalone findings is present: an abdominal circumference below the third percentile, an estimated fetal weight below the third percentile, or absent blood flow at the end of the heartbeat cycle in the umbilical artery. Alternatively, the diagnosis can be made when the weight or abdominal circumference is below the tenth percentile and there is also an abnormally high resistance index in either the umbilical or uterine artery.4PubMed. Consensus definition of fetal growth restriction: a Delphi procedure
For late FGR (32 weeks and beyond), the standalone criteria are tighter: only an abdominal circumference or estimated weight below the third percentile qualifies on its own. The combination criteria require the weight or circumference to be below the tenth percentile plus at least one of several additional findings, such as a drop across two or more quartiles on the growth chart between scans or an abnormal ratio of blood flow between the brain and placenta.5PubMed. Consensus definition of fetal growth restriction: a Delphi procedure The emphasis on crossing centiles (dropping from, say, the 30th percentile to the 8th over a few weeks) is an important detail: a fetus that tracks steadily at the fifth percentile may just be small, while one that plummets from the 40th to the fifth is telling you something has changed.
Not every country uses identical criteria, though. A comparison of national guidelines from six countries found broad agreement that a weight below the tenth percentile is the starting threshold, but significant variation in whether Doppler findings are required, whether abdominal circumference is used alongside or instead of weight, and whether population-based or customized growth charts should be applied.6American Journal of Obstetrics and Gynecology. International guidelines on fetal growth restriction: a critical appraisal and comparative analysis This means the exact same fetus could be diagnosed as growth-restricted in one country and not in another, which is one of the field’s frustrating inconsistencies.
Ultrasound Biometry and Its Limits
The foundational diagnostic tool is ultrasound measurement of the fetus, particularly the estimated fetal weight (calculated from head, abdomen, and thigh-bone measurements) and the abdominal circumference on its own. The abdominal circumference is especially valued because when a fetus is nutritionally deprived, it tends to divert blood flow to the brain at the expense of the liver, and a small liver makes the belly measure small before the head does.
One study comparing different approaches to screening found that when ultrasound-derived estimated weight below the tenth percentile was used as the threshold, it was more accurate than using growth velocity (the rate of change between scans) for either abdominal circumference or weight. The discriminative power of weight alone was modest, with an area under the curve of 0.61, and combining it with growth velocities barely improved things.7PubMed. Comparing fetal biometric growth velocity versus estimated fetal weight for prediction of neonatal small for gestational age Those numbers are frank: ultrasound biometry catches many cases, but it misses some and falsely flags others. It is a screening tool, not a definitive one, which is why Doppler studies carry so much weight in confirming the diagnosis.
Even cruder screening happens before ultrasound enters the picture. Fundal height measurement, where a midwife or doctor uses a tape measure across the pregnant abdomen, catches fewer than half of small fetuses. Handheld ultrasound devices that measure the abdominal circumference in a clinic setting can do much better, identifying roughly three-quarters of SGA babies compared with about a third caught by tape measurement alone.8PubMed Central. Diagnostic accuracy of fundal height and handheld ultrasound-measured abdominal circumference to screen for fetal growth abnormalities This matters in resource-limited settings where formal ultrasound may not be routine at every visit.
What Doppler Studies Add
Doppler ultrasound measures blood flow, and it transforms the diagnosis of growth restriction from a question about size into a question about how well the placenta is working. The umbilical artery is the workhorse measurement. In a healthy pregnancy, blood flows through the umbilical cord toward the placenta with relatively low resistance. When the placenta is diseased, resistance rises, and in severe cases the forward flow of blood during the resting phase of the heartbeat (end-diastolic flow) disappears entirely or reverses direction.
Absent or reversed end-diastolic flow is among the most alarming findings in fetal medicine and is strongly linked to poor outcomes.9PubMed. Intrauterine growth restriction and absent or reverse end-diastolic blood flow in umbilical artery (Doppler class II or III) A study of 233 growth-restricted fetuses found that adverse outcomes occurred in about 17% of those with elevated but still-present flow, about 35% of those with intermittent absent flow, and 56% of those with persistently absent flow, showing a clear gradient of risk.10PubMed. Clinical significance of intermittent absent end-diastolic flow of the umbilical artery in fetal growth restriction The pulsatility index of the umbilical artery, a numerical score reflecting flow resistance, has been found to be abnormal in all confirmed IUGR cases in some studies, making it one of the most consistent diagnostic markers available.11Pakistan Journal of Health Sciences. Detection of Intrauterine Growth Retardation in 2nd And 3rd Trimester with Fetal Parameters and Umbilical Artery Doppler
The brain-sparing reflex adds another diagnostic layer. When a fetus is deprived of oxygen, it dilates blood vessels to the brain to protect it, which shows up as decreased resistance in the middle cerebral artery. Clinicians often calculate the cerebroplacental ratio, which compares blood flow to the brain against blood flow in the umbilical cord. The expectation has been that this ratio is a strong predictor of trouble in late FGR, but a systematic review found surprisingly large variation in its accuracy and was unable to confirm that it had reliable clinical value in fetuses at or beyond 32 weeks.12PubMed Central. Prognostic accuracy of cerebroplacental ratio and middle cerebral artery Doppler for adverse perinatal outcome: systematic review and meta‐analysis Newer research has explored whether alternative measurements, such as flow in the vertebral arteries, might be more specific. One study found that abnormal vertebral artery Doppler was associated with earlier delivery, lower birth weight, and higher rates of neonatal intensive care admission, suggesting it may outperform the standard middle cerebral artery measurement for identifying the sickest babies.13PubMed Central. Patterns of Brain Sparing in a Fetal Growth Restriction Cohort
Uterine Artery Doppler for Early Prediction
Rather than waiting for the fetus to show signs of compromise, clinicians can also look at the mother’s blood supply to the uterus. The uterine arteries normally undergo a dramatic drop in resistance during the first and second trimesters as the placenta burrows into the uterine wall. If this remodeling fails, resistance stays high, and the fetus is at greater risk of growth restriction and preeclampsia down the line.
Mid-pregnancy uterine artery Doppler has the strongest standalone ability to predict severe early-onset disease, with some studies reporting detection rates between 69% and 93% for severe early preeclampsia and FGR.14American Journal of Medical and Clinical Research & Reviews. Role of First- and Second-Trimester Uterine Artery Doppler in Predicting Adverse Pregnancy Outcomes: A Review of Literature First-trimester Doppler on its own is less sensitive for overall preeclampsia but performs better for the severe forms that require early delivery. There is growing interest in incorporating uterine Doppler into first-trimester screening algorithms alongside blood biomarkers, potentially flagging pregnancies at risk months before any change in fetal size becomes apparent.15PubMed. Doppler studies of placental function
The Ductus Venosus and Timing of Delivery
When early-onset FGR is confirmed, one of the hardest clinical decisions is when to deliver. Too early, and extreme prematurity poses its own severe risks. Too late, and the fetus may deteriorate further. The ductus venosus, a small vessel in the fetal liver that shunts blood toward the heart, has become a critical tool for this decision.
The TRUFFLE trial, a landmark study of early-onset FGR, randomized women to different surveillance strategies for triggering delivery. Among survivors, 95% of babies in the group that delayed delivery until ductus venosus Doppler worsened had a normal outcome, compared with 85% when delivery was triggered by electronic heart-rate monitoring alone.16American Journal of Obstetrics and Gynecology. TRUFFLE study of early fetal growth restriction A separate study found that a specific measure of ductus venosus flow velocity remained a significant predictor of neonatal death or severe illness even after adjusting for how early the baby was delivered and how small it was.17PubMed. Ductus venosus Doppler waveform pattern in fetuses with early growth restriction In practice, worsening ductus venosus flow is now one of the signals many centers use to decide it is time to deliver, even if the baby is very premature.
Blood Biomarkers From the Mother
Ultrasound and Doppler tell you what the fetus and placenta look like from the outside. Blood tests can add information about what is happening at a molecular level. The most studied biomarker pair is sFlt-1 and PlGF, two proteins involved in blood-vessel growth. A failing placenta releases too much sFlt-1 (which blocks vessel growth) and too little PlGF (which promotes it). Their ratio in the mother’s blood rises as placental disease worsens.
A systematic review and meta-analysis found that an sFlt-1/PlGF ratio above 33 had a pooled sensitivity of about 63% and specificity of about 84% for predicting FGR.18PubMed. Diagnostic capacity of sFlt-1/PlGF ratio in fetal growth restriction: A systematic review and meta-analysis That means it misses about a third of cases but is reasonably good at ruling in those it does flag. For early-onset FGR specifically, a ratio at or above 99 predicted adverse outcomes serious enough to require delivery with about 85% sensitivity and 92% specificity.19PubMed. sFlt-1/PlGF Ratio Predicts Serious Outcomes in Early-Onset Fetal Growth Restriction The rate of increase matters too: a rapidly climbing ratio day over day was significantly associated with clinical progression of the disease.20PubMed Central. Course of the sFlt-1/PlGF ratio in fetal growth restriction and correlation with biometric measurements, feto-maternal Doppler parameters and time to delivery
These biomarkers are most useful in early-onset FGR, where placental dysfunction is more severe and the protein imbalance more pronounced. They are increasingly used in clinical practice in Europe, though adoption in the United States has been slower. They do not replace ultrasound but can help clinicians decide how aggressively to monitor a pregnancy or how soon delivery may be needed.
When to Look for Infections or Genetic Problems
The traditional reflex when a baby measures small has been to order a TORCH screen, a battery of blood tests checking the mother for toxoplasmosis, rubella, cytomegalovirus (CMV), and herpes. The logic was that these infections can cross the placenta and stunt fetal growth. But the evidence suggests this is usually a waste of time. One study of 319 pregnancies with FGR found zero cases of toxoplasmosis, rubella, or herpes; only six cases (about 2%) had congenital CMV, and just two of those had growth restriction as the only abnormality.21PubMed. Significance of maternal screening for toxoplasmosis, rubella, cytomegalovirus and herpes simplex virus infection in cases of fetal growth restriction A more recent two-center study concluded bluntly that TORCH testing for isolated FGR has no diagnostic yield and should be abandoned.22PubMed. Clinical utility of maternal TORCH screening in fetal growth restriction: A retrospective two-centre study If there are other findings, such as brain calcifications or enlarged liver on ultrasound, CMV testing can be reasonable, but as a blanket screen it adds cost without helping.
Genetic testing is a different story, particularly in severe early-onset cases. When FGR appears very early with no obvious maternal or placental explanation, chromosomal abnormalities or rare genetic conditions become more likely. Chromosomal microarray analysis in one study of 40 severe early-onset cases found one clearly harmful genetic change and three variants of uncertain significance, plus cases of abnormal chromosomal segments that could point to uniparental disomy or other uncommon conditions.23PubMed Central. Severe Early-Onset Fetal Growth Restriction: The Yield of Antenatal and Postnatal Genetic Testing The yield is not enormous, but it can change management when positive. In practice, amniocentesis for genetic testing is usually considered when FGR is severe, early, and structurally abnormal on ultrasound, rather than for every small baby.
Symmetric Versus Asymmetric Growth Restriction
An older framework split growth restriction into symmetric (the baby is proportionally small all over) and asymmetric (the head is spared while the body is small). The thinking was that symmetric IUGR suggested an early insult such as infection or a genetic problem, while asymmetric IUGR pointed to late-pregnancy placental insufficiency where blood was diverted to protect the brain. This classification was widely taught for decades.
The concept is not wrong in broad strokes, but it is less clean than it was once presented. One foundational anthropometric study found that the percentage decrease in head circumference was similar in both groups, undermining the idea that the brain is truly “spared” in asymmetric cases.24American Journal of Obstetrics and Gynecology. Comparative newborn anthropometric data in symmetric versus asymmetric intrauterine growth retardation Modern pathological assessment distinguishes between reduced growth potential (what used to be called symmetric) and nutritional IUGR (formerly asymmetric), recognizing that the causes can overlap and that indicators like an elevated brain-to-liver weight ratio or thymus atrophy can be more informative than proportionality alone.25PubMed. Pathological assessment of intrauterine growth restriction The symmetric/asymmetric distinction still shows up in clinical conversations, but it is gradually being replaced by the early/late framework because the timing-based classification maps more reliably onto outcomes and management decisions.
Placental MRI as a Complementary Tool
Ultrasound sees the fetus well but gives only a rough picture of the placenta itself. MRI is being explored as a way to directly assess placental health in growth-restricted pregnancies. Placentas from IUGR pregnancies tend to appear more oval and thicker on MRI compared with the circular, flatter shape in normal pregnancies.26PubMed Central. Morphological and histopathological changes in placentas of pregnancies with intrauterine growth restriction A technique called T2* mapping, which is sensitive to oxygen levels in tissue, has shown a correlation between abnormal placental histology and low T2* values. The finding that matters: some placentas with low T2* were attached to normally sized fetuses, meaning the placenta was dysfunctional even though the baby had not yet fallen below a size threshold. Fetal size alone, the authors pointed out, is not a reliable marker of placental dysfunction.27PubMed. T2* weighted placental MRI in relation to placental histology and birth weight
Placental MRI is not yet part of routine care. It is expensive, time-consuming, and available only at academic centers. But it represents a potentially important shift: instead of inferring placental failure from fetal measurements, you can look at the placenta directly. This could eventually help catch growth restriction earlier, especially the late-onset cases that slip through standard Doppler surveillance.
Machine Learning and Circulating MicroRNAs
Two emerging technologies could reshape how FGR is diagnosed in the coming years. Machine-learning models trained on fetal cardiac measurements have shown promise in identifying late-onset FGR, exactly the subgroup that conventional methods struggle with. One such model achieved strong performance for late FGR and successfully flagged cases that had normal umbilical artery Doppler, meaning it caught pathology that traditional assessment missed entirely.28PubMed Central. Machine learning-enhanced prediction of fetal growth restriction using fetal cardiac remodeling parameters Another machine-learning approach, using features derived from fetal heart-rate patterns, reported classification accuracy above 90% for IUGR at 34 to 38 weeks.29Informatics in Medicine Unlocked. Classification of intrauterine growth restriction at 34–38 weeks gestation with machine learning models
On the molecular side, researchers are investigating microRNAs, tiny fragments of genetic material that circulate in the mother’s blood and reflect what is happening in the placenta. In pregnancies complicated by severe preterm FGR, several microRNAs regulated by low-oxygen conditions were found at levels three to six times higher than in matched controls. One microRNA, miR-373, was barely detectable in normal pregnancies but highly expressed in growth restriction, and its levels tracked with the degree of fetal oxygen deprivation estimated by Doppler.30PLOS ONE. Circulating MicroRNAs in Maternal Blood as Potential Biomarkers for Fetal Hypoxia In-Utero Other work has found dysregulated microRNAs and their gene targets as early as the first and second trimesters in pregnancies that later developed FGR, raising the possibility of a simple blood-draw screening test months before the condition becomes visible on ultrasound.31PubMed Central. Differential Expression of Maternal Plasma microRNAs and Their Respective Gene Targets Can Predict Early Fetal Growth Restriction None of these tests are ready for routine clinical use yet, but the trajectory of the research is toward earlier, less subjective, and more biologically grounded diagnosis.
Why the Diagnosis Matters Beyond Birth
Getting the FGR diagnosis right is not just about managing the pregnancy. There is substantial evidence that growth restriction leaves a lasting metabolic imprint. When a fetus is starved of nutrients, it redirects blood flow to the brain at the cost of organs like the kidneys, liver, and pancreas. Those organs develop under ischemic conditions, and the consequences can show up decades later as high cholesterol, reduced kidney function, impaired glucose tolerance, and the cluster of risk factors known as metabolic syndrome. This is sometimes called the “thrifty phenotype” hypothesis: the fetus adapts to scarcity in ways that become harmful in a world of nutritional abundance. Recognizing that a baby was genuinely growth-restricted rather than just constitutionally small could inform long-term health monitoring in childhood and beyond.