NICU Negligence

A neonatal intensive care record is the densest document in a birth injury file, and its density changes what can be established.

Vital signs are recorded continuously rather than at intervals. Ventilator settings are logged with every change. Every medication has an administration timestamp. Laboratory results carry collection and resulting times separately. Nursing assessments are performed on a fixed schedule rather than when someone thinks to.

More usefully still, the standard of care is frequently written down by the defendant. Units operate published protocols — for sepsis evaluation, for oxygen saturation targeting, for glucose screening, for feeding advancement, for central line insertion and maintenance — and those documents are discoverable.

That is an unusual position in medical negligence. Most cases require an expert to establish what should have happened. In a NICU case the institution has often already said what should have happened, and the question narrows to whether it did.

Neonatal sepsis is the most common subject of these cases, and it is a timing problem rather than a diagnostic one.

A newborn with sepsis deteriorates fast and the early signs are non-specific: temperature instability, poor feeding, lethargy, respiratory changes, apnea. Because the signs are non-specific and the deterioration is fast, the standard approach is to screen and treat empirically rather than to wait for certainty — a blood culture drawn, antibiotics started, and the treatment stopped later if the cultures are negative.

The examinable questions are correspondingly concrete. Were maternal risk factors — prolonged rupture of membranes, maternal fever, known colonization — carried forward into the newborn's plan? When the infant's observations changed, how long until a culture was drawn? How long between the order and the first dose? Antibiotic administration times are recorded to the minute and the gap between order and administration is a frequent and well-documented failure.

Meningitis in particular carries consequences — hearing loss, hydrocephalus, developmental impairment — that are strongly related to how early treatment began.

Oxygen in the newborn is a therapy with a therapeutic window, and both ends of it cause harm.

Too little produces hypoxia. Too much, particularly in premature infants, is implicated in retinopathy of prematurity, which can lead to visual impairment and blindness, and in lung injury. Modern units therefore target a saturation range rather than a floor, and the monitoring generates continuous data on time spent inside and outside it.

Ventilation carries its own risks. Excessive pressures and volumes injure the developing lung; hypocapnia from over-ventilation is associated with brain injury in preterm infants. Blood gas results and ventilator settings are both logged, so the relationship between them is reconstructable.

Retinopathy screening is itself protocol-driven — infants below defined gestational age and birth weight thresholds require examination on a schedule, and missed or late screening is a distinct failure from the oxygen management that preceded it.

Neonatal hypoglycemia is common, is screened for in identified at-risk groups, and can cause permanent brain injury if it is prolonged or severe.

The at-risk categories are known in advance: infants of diabetic mothers, those small or large for gestational age, preterm infants, and those who are unwell for any other reason. Screening protocols specify who is tested, at what intervals, and what value triggers what response.

These cases are unusually documentary because the whole pathway is numbers with times attached. Was the infant in a screening category? Were the measurements taken at the specified intervals? What was the value, and what happened next — feeding, intravenous dextrose, escalation? How long did any low value persist before it was corrected?

Hypoglycemic brain injury has a recognizable pattern on MRI, which means the causation question here is often less contested than in other birth injuries.

Necrotizing enterocolitis is a serious inflammatory condition of the bowel affecting premature infants, and it can be catastrophic.

Feeding advancement protocols exist partly to reduce its incidence — the rate at which feeds are increased, and the type of milk used, are both governed by written policy in most units. Human milk is protective relative to formula, and where donor milk availability or feeding choice was part of the picture, the discussion with the parents is part of the record.

When the condition develops, the examinable questions are recognition and escalation: abdominal distension, feeding intolerance and blood in the stool are the early signs, and the interval between those appearing in the nursing notes and imaging being obtained, feeds stopped, antibiotics started and surgical opinion sought is recorded.

As elsewhere in the unit, the protocol is the institution's own and the compliance data is timestamped.

The last category cuts across all the others, and it is frequently where the real answer sits.

Neonatal units run to nurse-to-patient ratios that vary with acuity, and those ratios are recorded in staffing assignments for each shift. Medical coverage differs overnight and on weekends. Escalation policies define who is called, in what circumstances, and how quickly a response is expected.

When a case involves a deterioration that was recognized late, the reason is often not that nobody noticed but that the person who noticed was one of two nurses covering more infants than the ratio contemplates, and the attending physician was in the operating room.

Those facts are documented — in rosters, assignment sheets, acuity records and escalation logs — and they are not produced in response to a general request for the medical record. They have to be asked for specifically, and they are frequently the difference between a case that describes an individual's error and one that describes a system that made the error likely.