How this instrument works
Estimated fetal weight (EFW) isn't measured directly — it's calculated from a handful of ultrasound measurements using a regression formula built from real pregnancies where birth weight was later compared to the prenatal measurements. This calculator uses the widely used Hadlock four-parameter formula (Hadlock et al. 1985, American Journal of Obstetrics and Gynecology), which combines biparietal diameter (BPD), head circumference (HC), abdominal circumference (AC), and femur length (FL) — the same combination of measurements a standard growth ultrasound already collects.
The percentile step is a separate calculation: the EFW is compared to Hadlock's own published mean EFW by gestational week (Hadlock et al. 1991, Radiology), using the paper's directly quoted uniform variance figure of ±12.7% (one standard deviation) throughout gestation to convert the gap between your EFW and that week's mean into a z-score, then a percentile using a logistic approximation to the normal curve — the same approximation already used by this site's IQ percentile calculator.
Three limits worth stating plainly. First, EFW formulas themselves carry real measurement uncertainty — Hadlock's own reported accuracy is on the order of a 7-8% standard deviation between estimated and actual birth weight, before this calculator's percentile step adds its own approximation on top. Second, this calculator's reference curve holds 14 representative gestational weeks (even weeks 14 through 40, taken verbatim from Hadlock 1991's published table) rather than a continuous week-by-week curve — it matches the entered gestational age to the nearest of those 14 points. Third, this is a screening tool, not a diagnostic one: a percentile in the growth-restriction or macrosomia range is a prompt for the sonographer or physician to review the scan, dating, and trend across visits — it is not, by itself, a diagnosis of any fetal growth condition.
- Enter biparietal diameter (BPD) in centimeters.
- Enter head circumference (HC) in centimeters.
- Enter abdominal circumference (AC) in centimeters.
- Enter femur length (FL) in centimeters.
- Enter gestational age at the scan in weeks (14 to 40).
- Read the estimated fetal weight, the reference mean for that week, the z-score, and the approximate percentile.
Worked example — 28-week biometry near the reference mean
Biometry of BPD 7.24 cm, HC 26.8 cm, AC 23.9 cm, and FL 5.3 cm at 28 weeks: the Hadlock formula gives an EFW of about 1,211 g. Hadlock's 1991 reference mean EFW at 28 weeks is 1,210 g, so this estimate sits almost exactly on the reference mean — z ≈ 0.01, essentially the 50th percentile.
Compare smaller-than-typical 32-week biometry of BPD 7.8 cm, HC 28.5 cm, AC 25.5 cm, and FL 5.8 cm: the Hadlock formula gives an EFW of about 1,523 g, well below the 1,953 g reference mean at 32 weeks. That gap gives z ≈ −1.73, landing around the 5th percentile — the range where a sonographer would flag possible fetal growth restriction and typically recommend closer monitoring, not a result to interpret from a single scan alone.
Now typical 36-week biometry of BPD 9.0 cm, HC 33.0 cm, AC 31.5 cm, and FL 7.0 cm: the Hadlock formula gives an EFW of about 2,794 g against a reference mean of 2,813 g for that week — z ≈ −0.05, landing almost exactly at the 48th percentile, right at the center of the reference population.
Questions
How accurate is an ultrasound-estimated fetal weight?
Even before this calculator's percentile step, the Hadlock formula's own published accuracy runs on the order of a 7-8% standard deviation between the ultrasound estimate and the baby's actual birth weight — meaning a single EFW can reasonably be off by several hundred grams either way, especially at the extremes of the weight range. Treat the EFW and its percentile as an informed estimate to track over a pregnancy, not a precise number.
Why does the calculator only use 14 reference weeks instead of every week?
Hadlock et al. (1991) published a mean-EFW-by-week reference curve, and this calculator uses the 14 even-numbered weeks (14 through 40) from that paper's table verbatim, matching the entered gestational age to the nearest one rather than interpolating between weeks or estimating odd weeks. Odd-week values could be reconstructed from the same paper's other table but couldn't be independently cross-validated to this site's sourcing standard, so they were deliberately left out rather than guessed at.
Where does the 12.7% variance figure come from?
It's taken directly from Hadlock et al. (1991), which states that fetal weight variance is uniform at approximately ±12.7% (one standard deviation) throughout gestation when expressed as a percentage of predicted weight. This calculator uses that directly quoted, constant figure to convert the gap between an estimate and the reference mean into a z-score, rather than reconstructing separate percentile columns from the same paper that couldn't be independently verified.
Can this calculator diagnose fetal growth restriction or macrosomia?
No. This is a screening and reference tool that combines a regression-based weight estimate with a statistical comparison against a reference curve — it cannot diagnose fetal growth restriction, macrosomia, or any other condition. Those diagnoses depend on a clinician reviewing serial growth measurements, Doppler studies where indicated, maternal history, and the overall clinical picture, not a single EFW percentile.
Why might my ultrasound report show a different percentile for the same measurements?
Ultrasound machines and clinics often use a different EFW formula (there are several besides Hadlock) or a different, sometimes more clinically current, reference population and percentile table than the one this calculator uses. Small differences in input measurement rounding also compound through the formula. Your clinical report's percentile, generated by equipment calibrated and validated for that purpose, should be treated as more authoritative for your specific pregnancy than this general calculator.
Does the mother's or baby's specific risk factors change this calculation?
No — this calculator applies one fixed reference curve regardless of individual risk factors like maternal diabetes, hypertension, multiple gestation, or known fetal conditions, all of which a clinician would factor into interpreting any EFW percentile. It's a general population comparison, not an individualized risk assessment.
References
- Hadlock FP et al. — In Utero Analysis of Fetal Growth: A Sonographic Weight Standard. Radiology. 1991;181(1):129-133
- Hadlock FP et al. — Estimation of Fetal Weight, Head/Body/Femur Measurements. Am J Obstet Gynecol. 1985;151:333-337
Read this first: This instrument computes a screening figure from population formulas — it is not a diagnosis, and it cannot see the whole picture a clinician can. Use it to inform a conversation, not to replace one.