Fetal Monitoring Failures

Continuous electronic fetal monitoring produces two simultaneous traces: the fetal heart rate, and uterine activity. It runs for the length of the labor. Where the hospital archives it properly, it is a minute-by-minute record of what the people in the room could see.

That is unusual. Most of what a medical negligence case examines is a narrative written afterwards by the people whose decisions are under examination. Progress notes are composed at intervals, often hours later, by someone summarizing. The strip is not composed at all. It is machine output, generated continuously, and it does not know what anyone later concluded.

This is why a monitoring case tends to be built from the outside in. The tracing establishes what was on the screen. The notes establish what was recorded as having been seen. Where those two diverge, the divergence is itself the finding.

In 2008 ACOG, the Society for Maternal-Fetal Medicine and the National Institute of Child Health and Human Development convened a workshop to standardize how tracings are described. The result was a three-tier system, published for practicing obstetricians in ACOG Practice Bulletin No. 106.

Category I is normal. Baseline rate within range, moderate variability, no late or variable decelerations. A Category I tracing is strongly predictive of normal acid-base status at the time it is observed, and it calls for no particular intervention.

Category III is abnormal. Absent baseline variability with recurrent late decelerations, recurrent variable decelerations or bradycardia, or a sinusoidal pattern. Category III is associated with abnormal acid-base status at the time observed, and it requires prompt evaluation and action.

Category II is everything else. It is not a middle severity. It is a residual category, defined by not being either of the other two.

Two things follow from that structure and both matter in litigation. The categories describe the tracing at the time of observation, not the pregnancy and not the outcome. And a labor moves between categories, sometimes repeatedly, so a case is about a sequence rather than a snapshot.

Category III tracings are uncommon. Category I is unremarkable by definition. The overwhelming majority of the disputed hours in a birth injury case sit in Category II, and Category II is where the standard of care is genuinely contested rather than merely argued.

The reason is that Category II does not resolve to a single instruction. It is indeterminate by design: it groups patterns that carry very different implications, and the guidance is that it requires evaluation, continued surveillance, and reassessment — not that it requires delivery.

A plaintiff's case in this territory therefore does not usually say "the tracing was Category II, so a cesarean was required." That argument fails, and it fails for a good reason. What a serious case says is narrower and harder to answer: that within Category II the tracing was deteriorating in a describable way, that the recorded response did not change as it deteriorated, and that the interventions the guidance actually calls for — repositioning, fluids, stopping augmentation, oxygen, escalation to a physician — either were not tried or were not documented.

That distinction is worth stating plainly because it is where most weak claims in this area come from, and where the defense expects to win.

Monitoring cases resolve into three questions, and they are asked in order.

Was the pattern recognized? Nurses are the people watching the strip for most of a labor. The record shows what was charted, how often, and in what terms. A tracing that a reviewing expert reads as recurrent late decelerations, charted for six hours as "reassuring," is a recognition problem.

Was it escalated? Labor and delivery units run chains of command precisely because the person at the bedside is frequently not the person with authority to order an operative delivery. Most hospitals have a written escalation policy. Whether it was followed is a documentary question, and the policy itself is discoverable.

Was there a response? Intrauterine resuscitation is a defined set of measures. Whether they were attempted, and what the tracing did afterwards, is recorded on the same strip. A pattern that improves after repositioning tells one story. A pattern that does not, followed by another two hours of the same, tells another.

None of the three questions is answered by the tracing alone. All three are answered by the tracing read against the notes, the policy and the order entries, which is why these cases require the complete record rather than the summary a hospital produces on request.

Electronic fetal monitoring is a screening tool with a high false-positive rate for the outcome anyone actually cares about. Its widespread adoption did not produce the reduction in cerebral palsy that was expected of it, and this is not a fringe view — it is the mainstream understanding, and the defense will present it, correctly, in every case.

Two consequences follow.

An abnormal tracing does not establish that an injury occurred, and it does not establish that an earlier delivery would have prevented one. Those are causation questions. They are answered, if at all, by cord blood gases drawn at delivery, by neonatal imaging, by the placental pathology, and by the pattern and timing of the injury on MRI — not by the strip.

And a normal tracing does not exclude injury. A substantial share of neonatal encephalopathy is attributable to events before labor began. A case built on the assumption that the delivery must explain the outcome is a case built backwards, and it does not survive contact with a competent defense expert.

A monitoring failure is worth pursuing when the record shows both a departure from the standard of care and a mechanism by which timing changed the outcome. One without the other is not a case.

The tracing is generated by a system, archived by a system, and produced in response to a request by a person. Those are three separate opportunities for it to arrive incomplete.

What matters in practice: whether the archived data is the full continuous record or a decimated version, whether the annotations entered at the bedside came through, whether the clock on the monitor agrees with the clock on the chart and the clock on the order entries, and whether central-station archiving captured the same interval the bedside unit did.

Time synchronization sounds like a technicality and is not. A monitoring case is an argument about when. Two systems in the same room disagreeing by fifteen minutes changes what the sequence appears to show.

The practical step is a written preservation request that asks for the native archived data rather than a printed or exported summary, sent early enough that ordinary retention cycles have not run. That request costs nothing and cannot be made retroactively.