What Happens at a Growth Scan & What to Expect

A growth scan is a detailed ultrasound examination performed in the third trimester of pregnancy, typically between 28 and 36 weeks, to check whether your baby is growing as expected. During the appointment, a sonographer measures your baby’s head, abdomen, and thigh bone, uses those numbers to estimate weight, and checks the amniotic fluid and placenta. The whole process usually takes 20 to 30 minutes, though it can run longer if the baby is in an awkward position. While the scan itself feels much like the mid-pregnancy anatomy scan you may have already had, its purpose is different, and the results can lead to follow-up decisions that are worth understanding before you walk in.

Why You Have Been Referred for a Growth Scan

Not every pregnancy gets a dedicated growth scan. In many health systems, third-trimester ultrasound is reserved for pregnancies considered to be at increased risk, and current guidelines have expanded those risk categories considerably over the past decade. New care pathways rely on regular third-trimester ultrasound biometry and plotting of estimated fetal weight for pregnancies flagged as higher risk, which has substantially increased demand on ultrasound departments.

Common reasons for referral include a fundal height measurement (the tape-measure check your midwife does at routine appointments) that seems too large or too small, a history of a previous small or large baby, conditions such as pre-eclampsia or gestational diabetes, reduced fetal movements, or a multiple pregnancy. Some units now offer universal third-trimester scanning to all first-time mothers. A large cohort study found that universal scanning roughly tripled the detection of small-for-gestational-age babies compared with selective scanning based on risk factors alone, picking up about 57% versus 20%.¹ That gap has fuelled debate about whether all pregnancies, not just flagged ones, should receive a late scan.

What Happens in the Room

When you arrive, you will be asked to lie on a couch, usually slightly reclined or tilted to one side. In later pregnancy, lying flat on your back can compress a major blood vessel and cause dizziness or a drop in blood pressure, sometimes called supine hypotensive syndrome. The severity depends on the size of the uterus and the exact position of you and the baby, and it ranges from barely noticeable to feeling faint.² A wedge or rolled towel under your right hip tilts the uterus off the vessel, and most scan rooms will position you this way from the start. If you feel lightheaded at any point, tell the sonographer and they will adjust.

Gel is applied to your abdomen, and the transducer (the handheld probe) is moved across your belly, pressing gently to get clear images. You do not need a full bladder for a third-trimester scan as you might have for an early pregnancy one. The sonographer will spend most of their time freezing the image on screen, placing electronic calipers on specific landmarks, and recording measurements. You will probably see your baby on the monitor, though the image may look less recognizable than it did at your 20-week scan because the baby is larger and more tightly curled. Some practitioners will narrate what they are looking at; others work quietly and explain afterward.

The Measurements Taken

The core of a growth scan is fetal biometry: a standardized set of measurements of the baby’s body. Three measurements are taken as a minimum, often repeated across multiple images to improve reliability.

Head circumference (HC): Measured in a cross-sectional slice of the skull at the level of two brain structures called the thalami. The sonographer fits an electronic ellipse around the outer edge of the skull bone.
Abdominal circumference (AC): Measured in a cross-section of the baby’s belly at a level where the stomach bubble and a short stretch of the umbilical vein are visible. Again, an ellipse is traced around the outer skin contour.
Femur length (FL): A lengthwise view of the thigh bone, measured from one end of the shaft to the other.

These landmarks are chosen because they are reliably identifiable and because together they capture growth of the brain, soft tissue, and skeleton. The measurements follow standardized protocols set by international guidelines, including rules about caliper placement and which anatomical structures must or must not be visible in the image plane.³ Some units also measure biparietal diameter (BPD), the distance across the skull from one side to the other, though head circumference is generally considered more informative on its own.

In research settings, automated caliper-placement algorithms are being tested. One evaluation had software detect each structure and place measurements automatically on the standard three biometric variables across multiple images per pregnancy.⁴ In routine clinical practice, though, a trained sonographer still places the calipers manually.

How Your Baby’s Weight Is Estimated

The raw measurements of head, abdomen, and femur are plugged into a mathematical formula that produces an estimated fetal weight (EFW). The most widely used formula was developed by Hadlock in the 1980s and is still the standard in many units worldwide. The estimate is then plotted on a growth chart to see where the baby falls relative to other babies at the same gestational age. A baby whose estimated weight sits on the 50th centile is right in the middle of the range; one on the 10th centile is smaller than about 90% of babies at that stage.

The distinction between the formula and the chart matters more than you might expect. A recent methodological analysis of the Hadlock reference found significant internal inconsistencies: the centile lines printed on the standard chart do not perfectly match what the underlying equation actually produces. The mismatch is most pronounced at the lower end, where clinical decisions hinge. Using the chart alone, about 4% of babies at 35 to 36 weeks were classified below the 10th centile, but using the equation directly, that figure was closer to 7%.⁵ In practical terms, some babies flagged as small by one method would be missed by the other, and vice versa. Your care team will typically use whichever reference their unit has adopted, and many hospitals have moved to customized charts that adjust for your height, weight, ethnicity, and whether this is your first baby.

Checking the Amniotic Fluid

Alongside measuring the baby, the sonographer assesses the volume of amniotic fluid surrounding it. Too little fluid (oligohydramnios) or too much (polyhydramnios) can signal problems with the baby’s kidneys, with the placenta, or with other conditions. The assessment is done by ultrasound, not by any invasive procedure.

There are two main methods. One measures the single deepest vertical pocket of fluid visible on screen. The other divides the uterus into four quadrants and adds up the deepest pocket in each, giving an amniotic fluid index (AFI). Both approaches perform similarly in predicting serious problems for the baby, but the four-quadrant method diagnoses low fluid more than twice as often and leads to more labor inductions without actually improving outcomes.⁶ A multicenter randomized trial confirmed the pattern: the AFI group had roughly four and a half times the rate of oligohydramnios diagnosis compared with the single-pocket group, with no difference in neonatal intensive care admissions.⁷ Because of this, many guidelines now favor the single deepest pocket method, especially in lower-risk pregnancies. If your report mentions either measurement, the result is simply telling your team whether there is a normal cushion of fluid around the baby.

When Doppler Studies Are Added

A standard growth scan focuses on size and fluid. If the baby measures small or if there are other concerns, your team may add Doppler ultrasound, which measures the speed and direction of blood flow through specific vessels. This part of the scan uses the same probe and feels identical to you, but the sonographer switches to a mode that displays flow patterns as a waveform on screen.

The two vessels checked most often are the umbilical artery, which carries blood from the baby to the placenta, and the middle cerebral artery inside the baby’s brain. In a healthy pregnancy, blood flows easily through the umbilical cord to the placenta. When the placenta is not working well, resistance rises and the flow pattern changes. In the brain, the opposite happens: the baby redirects blood preferentially to protect the brain, a phenomenon sometimes called “brain sparing,” which shows up as lowered resistance in the middle cerebral artery.

One study comparing these two measurements in growth-restricted pregnancies found that abnormal middle cerebral artery readings had perfect specificity and positive predictive value for identifying babies born below the 10th centile, while the umbilical artery ratio offered higher sensitivity and overall diagnostic accuracy.⁸ In practice, clinicians often combine both into a ratio (the umbilicocerebral ratio), which along with the estimated fetal weight has been shown to effectively identify perinatal risk in small babies detected in the late preterm period.⁹

If growth restriction is more severe, additional vessels may be assessed, including the ductus venosus (a small vessel in the baby’s liver) and the inferior vena cava. A longitudinal study tracking the sequence of deterioration in severely growth-restricted babies found that changes in the umbilical artery typically came first, followed by redistribution to the brain, and then abnormal venous Doppler patterns later, alongside falling biophysical profile scores.¹⁰ Understanding this sequence helps clinicians judge how urgently a baby may need to be delivered.

The Placenta and Cord

During a growth scan, the sonographer will also note the position of the placenta and where the umbilical cord inserts into it. In most pregnancies, the cord attaches near the center of the placenta, giving it good access to the blood supply. A marginal insertion, where the cord meets the placenta within about a centimeter of its edge, can sometimes be associated with growth issues.¹¹ In the vast majority of cases, though, the cord insertion is unremarkable and the sonographer simply confirms its location as part of the standard checklist.

Placental location also matters if you had a low-lying placenta noted at your 20-week scan. By the third trimester, most low placentas have “moved” upward as the uterus grows, and the growth scan is an opportunity to confirm that. The cord insertion site can sometimes become harder to locate in the third trimester, particularly with a posterior placenta, though one study found this was an issue in fewer than 4% of cases.¹²

How Accurate Is the Estimated Weight

This is one of the most common questions parents have after a growth scan, and the honest answer is that fetal weight estimation is an educated guess with a known margin of error. Errors arise from equipment factors, the image plane the sonographer selects, caliper placement, and how well the baby cooperates with positioning.¹³ Maternal body habitus plays a role too: a thicker abdominal wall makes it harder to get a crisp image, which can degrade measurement precision.

Even among trained sonographers working under a rigorous quality-control protocol, measurement variability is real. In the large INTERGROWTH-21st project, which imposed strict training and image standards, the 95% limits of agreement for head circumference were about plus or minus 3 to 4%, and for abdominal circumference roughly 5 to 6%.¹⁴ Since estimated fetal weight is calculated from these measurements, the final weight estimate typically carries a margin of error of about 10 to 15% in either direction. A baby estimated at 3,000 grams could realistically weigh anywhere from roughly 2,550 to 3,450 grams at birth.

The scan is generally better at identifying trends than pinning down an exact number. A single scan that puts your baby on the 30th centile does not mean a lot in isolation, but if a follow-up scan two or three weeks later shows the baby has dropped to the 10th centile, that downward trajectory is meaningful regardless of the absolute weight.

Accuracy for Large Babies

Growth scans are sometimes ordered because a baby is suspected to be large, either from fundal height or because of gestational diabetes. The scan’s ability to spot a genuinely large baby is decent but far from perfect. A population-based study of over 26,000 births found that when the scan estimated a weight above the 90th centile, only about 41% of those babies actually turned out to be large for gestational age at birth. In the subgroup specifically scanned because a large baby was suspected, the positive predictive value was a bit better at around 50%, but the false-positive rate was 36%.¹⁵ In other words, about half the time the scan says the baby is big, it actually is. The practical consequence is that a “big baby” label from a single growth scan should be taken as a heads-up, not a certainty, and clinical decisions like early induction are weighed against the possibility that the estimate is off.

Similarly, detection of large-for-gestational-age babies through selective scanning based on clinical suspicion catches only about a quarter of them, and even universal scanning picks up just over a third.¹⁶ The scan is simply a better tool for flagging small babies than large ones.

What Happens If Something Looks Off

If the estimated weight is below the 10th centile or the growth trajectory has slowed, your care team will typically arrange closer monitoring. That might mean repeat growth scans every two to three weeks, Doppler assessments, and sometimes additional checks like cardiotocography (CTG), where sensors on your belly monitor the baby’s heart rate pattern. In more concerning cases, a biophysical profile may be done, which scores the baby on five components: muscle tone, body movements, breathing movements, amniotic fluid, and the heart rate tracing. Research on growth-restricted pregnancies has found that between 30% and 70% of those with a non-reactive heart rate tracing need the biophysical profile to clarify whether the baby is genuinely compromised, and an abnormal score in 8% to 27% of cases identifies a need for delivery not picked up by Doppler alone.¹⁷ Combining these surveillance tools seems to improve outcomes: studies using integrated monitoring report stillbirth rates of 0% to 4%, compared with 8% to 11% when monitoring is less frequent or less structured.

The decision about when to deliver a baby with suspected growth restriction balances the risks of prematurity against the risks of leaving the baby in a struggling environment. A large randomized trial comparing induction with expectant monitoring at term for growth-restricted babies found that both approaches were broadly equivalent for serious outcomes, with no increase in operative delivery rates in the induction group.¹⁸ For many parents, knowing that induction does not automatically mean a more complicated birth can be reassuring. The trial’s authors noted that it is reasonable for parents who prefer non-intervention to choose expectant management with close monitoring, while those who want to reduce the (small) risk of stillbirth may prefer induction.

Coping with Uncertainty After the Scan

Being told that your baby measures small, or large, or that anything is “not quite right” on a scan can be enormously stressful. Research into parental experiences after abnormal fetal scans has identified several layers of uncertainty. Parents often receive probabilistic information: a low centile does not mean the baby is definitely in trouble, and a Doppler reading that is “borderline” does not come with a clear-cut plan. Some information is complex and hard to interpret, and practical details about next steps, like what monitoring will involve or how decisions about early delivery would be made, are not always spelled out in the initial conversation.¹⁹

If you leave a growth scan feeling anxious, it is worth writing down your questions and contacting your midwife or obstetrician for a follow-up conversation. Ask specifically: what centile is my baby on, has the trend changed since last time, and what is the plan from here? Understanding whether you are being monitored as a precaution or because there is a genuine concern can make a significant difference to how the waiting period between scans feels.

Three-Dimensional Volume Scanning and Where the Technology Is Heading

The standard growth scan uses two-dimensional ultrasound, and for most pregnancies that is all that is needed. But researchers have been exploring whether three-dimensional imaging can improve weight estimation, particularly for babies at the extremes of size. A cross-sectional study comparing a 3D limb-volume model against the conventional 2D Hadlock formula found that the 3D approach was substantially more accurate for babies under 3,500 grams. Among those smaller babies, the 3D model produced an estimate within 5% of birth weight in about 58 to 65% of cases, compared with only 19% for the standard 2D formula.²⁰ That is a meaningful improvement, because growth-restricted babies are exactly the population where precision matters most.

Artificial intelligence is also entering the field. Automated caliper-placement tools are already being tested in clinical workflows, and machine-learning algorithms trained on large image databases are being developed to flag suboptimal image planes before the sonographer records a measurement. These tools are unlikely to replace a skilled sonographer anytime soon, but they may reduce the kind of random variability that currently contributes to the 10 to 15% error band around every weight estimate. For now, though, the growth scan you receive will almost certainly be a conventional 2D examination performed by a human operator, and the best guarantee of accuracy remains a well-trained sonographer working to standardized protocols with adequate time per patient.

Is It Safe

Diagnostic ultrasound has been used in pregnancy for over half a century, and no confirmed harmful effects at standard clinical settings have been established. That said, the machines in use today are considerably more powerful than the early models. In 1992, the FDA raised allowable output levels for ultrasound equipment substantially, a change that theoretically can increase fetal temperature by up to 2 degrees Celsius.²¹ To manage this, modern machines display the thermal index (TI) and mechanical index (MI) on screen in real time, and sonographers are trained to follow the ALARA principle: keeping exposure as low as reasonably achievable while still getting the diagnostic information needed. In practice, this means the operator keeps scan times efficient and avoids lingering in Doppler mode, which deposits more energy than standard B-mode imaging, over any one area longer than necessary. A growth scan performed at a regulated clinic following standard protocols carries negligible known risk to you or your baby.