What Is the Difference Between EDD and AUA in Pregnancy?

EDD and AUA refer to two different pieces of information on your prenatal records, and they answer slightly different questions. EDD, or Estimated Due Date, is the calendar date your provider expects you to deliver, usually calculated from the first day of your last menstrual period. AUA, or Average Ultrasound Age, is the gestational age assigned to your pregnancy based on how your baby measures on an ultrasound scan. These two figures often agree closely, but when they don’t, the gap between them triggers real clinical decisions about whether to adjust your due date, how to monitor growth, and when to consider delivery.

How Your Estimated Due Date Gets Set

The most traditional way to estimate a due date uses Naegele’s rule, a formula dating back over two centuries. It takes the first day of your last menstrual period (LMP), adds seven days, and counts forward nine calendar months. The underlying assumption is that a typical pregnancy lasts about 280 days from the start of your last period, which works out to 40 weeks.¹ The original version of Naegele’s method explicitly used the first day of menstruation as the starting point, not the last day or the end of bleeding, a detail that occasionally gets garbled.²

The problem is that Naegele’s rule assumes you have a textbook 28-day cycle and that you ovulated on day 14. Plenty of people don’t. If your cycle runs longer or shorter, or if you aren’t sure of the exact date your period started, the LMP-based EDD can be off by a week or more before anyone looks at an ultrasound. That built-in imprecision is one reason ultrasound dating has become standard practice.

How AUA Is Determined

When a sonographer performs an ultrasound, they take specific measurements of the embryo or fetus and plug them into standardized reference charts. In the first trimester, the key measurement is the crown-rump length (CRL), the distance from the top of the head to the bottom of the torso. CRL produces the smallest prediction errors of any single measurement for estimating when a baby will be born.³ International reference standards have been developed from large populations specifically to translate a CRL measurement into a gestational age in weeks and days.⁴

After the first trimester, the baby is too curled and too large for a simple head-to-rump measurement. Sonographers then switch to a set of other measurements: biparietal diameter (the width of the skull), head circumference, abdominal circumference, and femur length. Head circumference, sometimes combined with femur length, is typically used for gestational-age estimation from the mid-trimester onward when no earlier scan is available.⁵ The AUA that appears on your ultrasound report is the gestational age that these measurements produce when run through the chart’s equation.

Why the Two Often Disagree

A gap between your LMP-based EDD and the AUA on your ultrasound is common, and in many cases it simply reflects the imprecision of period-based dating. You might have ovulated a few days later than day 14, or you might be slightly off on which day your period started. A discrepancy of three or four days in the first trimester is entirely normal and usually means nothing is wrong.

Larger discrepancies, a week or more, tend to raise more questions. They might indicate irregular ovulation, a longer or shorter cycle than assumed, or occasionally bleeding in early pregnancy that was mistaken for a period. In some cases, a big gap reflects genuinely unusual fetal size rather than a dating error, and that distinction matters clinically.

Which Method Is More Accurate

For most pregnancies, first-trimester ultrasound wins. A large population study found that when gestational age was based on ultrasound data, up to 92% of deliveries fell within the 37-to-42-week window, compared with 87% when based on a reliable LMP alone. The earlier the ultrasound, the more accurate the prediction.⁶ In IVF pregnancies, where the exact date of embryo transfer is known and removes all guesswork, first-trimester ultrasound dating has been shown accurate to within roughly a day and a half of the true gestational age.⁷

That precision fades quickly as pregnancy progresses. At 14 weeks, the uncertainty in ultrasound-based gestational age estimates is about six to seven days. By 26 weeks, it expands to roughly two weeks, and in the third trimester it exceeds two weeks.⁸ This is partly why dating scans performed early carry so much more weight than those done later. First-trimester dating is most accurate because biological variability between individual fetuses has not yet come into play.⁹ In the second and third trimesters, babies of the same gestational age can differ meaningfully in size simply because of genetics, nutrition, or other factors.

When Your Provider Changes the Due Date

Obstetric guidelines are specific about when an EDD should be adjusted based on ultrasound findings and when it should not. The American College of Obstetricians and Gynecologists (ACOG) recommends that the gestational age and EDD be established as early as possible using LMP data, ultrasound, or both, and that the result be documented clearly. Subsequent changes should be reserved for rare circumstances. ACOG also considers a pregnancy without an ultrasound before 22 weeks to be “suboptimally dated.”¹⁰

In practice, ACOG and similar guidelines use specific discrepancy thresholds. If your first-trimester ultrasound’s AUA differs from your LMP-based age by more than about five to seven days, many providers will redate the pregnancy to match the ultrasound. In the second trimester, the threshold widens to about ten to fourteen days before redating is considered, reflecting the wider margin of error at that stage. By the third trimester, the discrepancy would need to be three weeks or more to justify changing dates, and most providers simply won’t do it at that point because the measurement is too imprecise.

These thresholds exist because casually changing a due date can have consequences. A Canadian clinical guideline highlights the risk that a fetus that is actually growth-restricted or abnormally large could be mistakenly “redated” to match its measurements, which would mask the very problem that needs attention.¹¹ If a growth-restricted baby measuring small at a late scan causes the provider to push the due date later, the baby might not get the surveillance or early delivery it needs.

When a Discrepancy Signals Something Real

Not every mismatch between the EDD and the AUA is a dating problem. Sometimes the baby really is bigger or smaller than expected for its age, and that gap carries clinical meaning. Research on early-pregnancy ultrasounds found that when a fetus measured seven or more days larger than expected based on the LMP date, the risk of an especially large birth weight rose substantially.¹² In other words, a discrepancy that looks like a simple dating error might actually be an early flag for macrosomia, which affects delivery planning.

The same logic runs in the opposite direction. A baby that consistently measures smaller than dates could be growing normally on its own curve, or it could be experiencing growth restriction. Distinguishing between the two is one of the harder problems in obstetrics, and it depends heavily on having an accurate anchor date. If the dating was done early and is considered reliable, a baby that falls progressively behind on later scans raises concern. If the pregnancy was never properly dated, there’s no stable baseline for comparison.

What Affects Ultrasound Accuracy Beyond Timing

Even first-trimester ultrasound is not a perfect ruler. Several factors introduce small amounts of noise into measurements. Fetal sex, maternal age, whether the mother has had previous pregnancies, and even whether the baby is in a breech position can nudge head-based age estimates by a day or two.¹³ Parental size also plays a role once measurements shift to the second trimester. Maternal body mass index correlates more strongly with abdominal circumference and estimated fetal weight, while both parents’ heights influence femur length.¹⁴

Then there is the equipment itself. The ultrasound machine, the probe, and even the specific combination of the two can shift measurements enough to change the estimated gestational age. One study found that dating differences of up to two days for first-trimester CRL and up to four days for second-trimester head width could result purely from using different machine-probe setups.¹⁵ Operator technique compounds this. Where exactly the calipers are placed, which image plane is selected, and the sonographer’s level of experience all feed into the measurement.¹⁶ None of these individual error sources is large, but they stack up, which is part of why clinicians treat ultrasound estimates as approximations rather than certainties.

The IVF Exception

If you conceived through in vitro fertilization, you have something most pregnant people don’t: a known date of embryo transfer, which pins down gestational age far more precisely than any period date or ultrasound measurement. Interestingly, even with that known date available, ultrasound-based age estimates still tend to run slightly higher than the transfer-based age. One study found a mean difference of about 0.9 days, with ultrasound consistently dating the pregnancy a bit older than the embryo transfer date would predict.¹⁷

Researchers have also tested how well standard ultrasound reference charts perform when checked against IVF dates. One study found that all six charts they examined generated dates that differed from the known IVF-based gestational age to a statistically meaningful degree, which prompted the creation of a new reference chart that more closely matched the known dates.¹⁸ This doesn’t mean existing charts are wildly wrong. The discrepancies are small, typically a day or two. But for IVF pregnancies, providers generally stick with the transfer-based date rather than switching to ultrasound dating, since the known conception date removes the uncertainty that ultrasound is designed to address.

Why Getting the Dates Right Matters for Delivery Decisions

The practical stakes of accurate pregnancy dating come into sharpest focus at the end of pregnancy, when decisions about induction and post-term management hinge on how many weeks along you are. If your due date is off by a week, a pregnancy that appears to be at 41 weeks might actually be at 40, and vice versa. That error could mean either inducing labor earlier than necessary or waiting too long when the placenta is aging.

Induction itself is not a minor event. A retrospective study found that women who were induced had a higher rate of cesarean delivery and greater use of epidural analgesia compared to those who went into labor spontaneously.¹⁹ An unnecessary induction, one triggered by a dating error that makes the pregnancy look more overdue than it is, therefore carries real costs. On the other hand, failing to intervene in a truly post-term pregnancy creates its own risks. The reliability of your EDD is the foundation on which both decisions rest.

Dating accuracy also feeds into growth monitoring throughout the second and third trimesters. Serial ultrasound scans track fetal growth velocity, which is independently associated with how the baby does at birth.²⁰ But growth velocity only means something when each measurement is plotted against an accurate gestational age. A dating error shifts every subsequent data point on the growth chart, potentially hiding a baby that is falling behind or making a normally growing baby look too small.

How Growth Velocity Fits into the Picture

Once your due date is established, later ultrasounds are not about redating. They are about growth. A single measurement tells you how big the baby is right now; serial measurements spaced weeks apart tell you how fast the baby is growing. That growth trajectory can detect problems that a one-off size estimate would miss. A baby that sits at the 20th percentile on two consecutive scans is likely just small. A baby that dropped from the 50th to the 20th percentile may be in trouble.

Recent work has shown that incorporating growth velocity into birth-weight prediction models improves their accuracy compared to using a single estimated fetal weight alone. One study found that models including velocity data produced the lowest prediction errors and had the greatest share of predictions falling within 10% of actual birth weight.²¹ For clinical decision-making, a more accurate estimate of how big the baby will be at delivery helps providers plan for complications like shoulder dystocia or decide when to deliver a growth-restricted baby.

This all loops back to the early establishment of dating. Growth monitoring is only as good as the anchor date it starts from. A first-trimester scan that pins down gestational age to within a day or two gives every subsequent measurement a reliable reference point. A pregnancy that was never scanned before the mid-second trimester starts with a fuzzier baseline, and that imprecision propagates forward through every growth assessment that follows.