When the Ultrasound And The Doctor Were Wrong. And the Patient Was Right.

What a 2-pound miss teaches us about fetal weight estimates, VBAC counseling, and humility in obstetrics.

FREE FOR ALL SUBSCRIBERS

Estimating fetal weight late in pregnancy is imprecise. Both clinical exams and ultrasound rely on assumptions, formulas, and averages. When those estimates are used to counsel a patient about VBAC versus repeat cesarean, the margin of error matters ethically, not just statistically.

The case.

A patient with a prior cesarean was counseled toward a VBAC. Ultrasound suggested an average-sized fetus. The clinician reassured her that the baby was not large and that VBAC was reasonable. The patient disagreed. She believed the baby felt much larger than predicted and requested a repeat cesarean. At delivery, the newborn weighed roughly 2 pounds more than the ultrasound estimate.

This is not a rare story. It is simply one we do not talk about honestly enough.

Why fetal weight estimation is inherently unreliable.

Late-pregnancy estimated fetal weight, whether by hands or by machine, is an inference problem, not a measurement. Ultrasound does not weigh babies. It measures dimensions and converts them into weight using population-based regression equations.

The standard biometric inputs are four measurements:

1. Biparietal diameter. A transverse skull width.

2. Head circumference. A traced ellipse of the fetal head.

3. Abdominal circumference. A single axial slice through the fetal liver and stomach.

4. Femur length. The longest straight bone that can be visualized.

These values are entered into formulas such as Hadlock I–IV. Each formula weights the measurements differently. None were designed to predict extreme size accurately at term.

BLACK BOX–STYLE WARNING
WARNING: ESTIMATED FETAL WEIGHT IS INHERENTLY IMPRECISE.

Clinical and ultrasound-based estimates of fetal weight are not exact measurements. Under routine conditions, both methods commonly deviate by ±10–15%, and errors of 20% or more are well documented, particularly at the upper and lower extremes of fetal size. For a term fetus, this can translate into errors of 1 to 2 pounds or more in either direction.

Clinical estimation based on Leopold maneuvers, fundal height, and maternal habitus is influenced by operator experience, amniotic fluid volume, fetal position, placental location, and abdominal wall thickness. Precision decreases in obesity, polyhydramnios, oligohydramnios, and prior uterine surgery.

The math behind the problem.

Most commonly used formulas have a mean absolute percentage error of about 8–12 percent at term. That means a fetus estimated at 8 pounds can reasonably weigh anywhere from about 6 to 10 pounds. The error widens at the extremes. Large babies are more often underestimated than overestimated. This is not speculation. It is repeatedly shown in validation studies.

A 2-pound error is not an outlier when the true birth weight is high. It is a predictable failure mode of the model.

Clinical estimates are no better.

Leopold maneuvers and fundal height are even less precise. Studies consistently show clinical estimates perform similarly or worse than ultrasound, with wide inter-observer variability. Experience does not eliminate bias. Confidence often exceeds accuracy

Why this matters for VBAC (or for that matter any other) counseling.

VBAC success and risk discussions frequently reference fetal size, shoulder dystocia risk, and labor progress. When those discussions rely on estimated fetal weight, clinicians must acknowledge uncertainty explicitly.

Presenting an ultrasound number as 100% reassurance, without stating the likely error range, crosses an ethical line. It transforms probabilistic information into apparent fact. That undermines informed refusal.

In this case, the patient did not reject evidence. She rejected false precision. She integrated her own bodily perception with uncertainty better than the counseling did.

The ethical lesson.

Respect for autonomy does not require agreement. It requires epistemic humility. When a patient declines VBAC because she distrusts fetal weight estimates, she is not being irrational. She may be responding appropriately to known limitations of our tools.

Professional responsibility means stating clearly:
“We could be off by a pound or more. Large babies are often underestimated. This uncertainty may matter for your decision.”

Anything less is incomplete consent.

What clinicians should take away.

The central lesson is not technical. It is moral.

Obstetrics demands epistemic humility. Estimated fetal weight is not a diagnosis. It is a probabilistic guess derived from population averages applied to an individual pregnancy. When clinicians speak with unwarranted confidence, they convert uncertainty into authority and authority into pressure.

Humility requires naming what we do not know, not minimizing it. It requires admitting that our tools systematically fail at the extremes, that large fetuses are often underestimated, and that reassurance based on a single number can mislead rather than inform. Humility also means recognizing that a patient’s embodied knowledge, her sense of size, movement, and prior birth experience, may be as valid as our algorithms. The ethical obligation is not to persuade patients to accept our preferred mode of delivery, but to ensure that decisions are made with an honest appraisal of uncertainty. In this frame, a patient declining VBAC is not resisting evidence. She is responding rationally to its limits. The clinician’s role is not to defend the model. It is to respect the decision that emerges when uncertainty is acknowledged rather than concealed.

The ultrasound was wrong here. So was the doctor to a certain extend. The patient was not.

Reflection.
If our models are imperfect, and they are, whose judgment should carry more weight when the consequences are borne by the patient’s body?