Perinatal asphyxia: biomarkers and neuroimaging in neonatal care
Perinatal asphyxia describes an interruption of oxygen delivery or blood flow around birth that may lead to neonatal encephalopathy, hypoxic-ischaemic injury, and long-term developmental impairment. The clinical picture can evolve rapidly, which makes early recognition and reliable assessment essential in delivery rooms, neonatal intensive care units, and follow-up services.
Modern care combines neurological examination with biochemical markers, continuous physiological monitoring, electroencephalography, and advanced imaging. These tools do not replace careful clinical judgment. Instead, they help clinicians identify infants who may benefit from therapeutic hypothermia, estimate the severity of brain injury, and communicate a more evidence-based prognosis to families.
The subject was central to the scientific interests represented by FAOPS 2020, the planned Federation of Asian and Oceania Perinatal Societies congress in Tokyo. Although the meeting was canceled in April 2020 because of the COVID-19 pandemic and international travel restrictions, its themes remain highly relevant to perinatal and neonatal medicine.
Defining the injury behind the diagnosis
Perinatal asphyxia is often used broadly, but a diagnosis of hypoxic-ischaemic encephalopathy requires evidence that impaired oxygenation or cerebral perfusion has affected brain function. Typical clues include a low Apgar score, the need for prolonged resuscitation, metabolic acidosis in umbilical cord blood, abnormal consciousness, altered tone, seizures, and disturbances in breathing or feeding.
Cord blood gases are especially useful because they provide an immediate biochemical record of the infant’s condition at birth. A low pH, raised base deficit, and increased lactate support the presence of metabolic stress, although none of these values alone determines neurological outcome. Timing, sampling quality, maternal factors, and the infant’s subsequent examination must all be considered.
The placenta can add important context. Maternal vascular malperfusion, fetal vascular malperfusion, inflammation, infection, and chronic placental insufficiency may explain why an infant was vulnerable before an acute intrapartum event. A review of placental pathology helps connect microscopic findings with fetal growth, Doppler abnormalities, abnormal fetal heart rate patterns, and neonatal condition.
What biomarkers can reveal
Biomarkers are valuable because neurological examination may be affected by sedation, hypothermia, seizures, prematurity, or systemic illness. Blood and urine tests can provide an additional view of cellular injury, energy failure, inflammation, and disruption of the blood-brain barrier. Commonly studied candidates include lactate, neuron-specific enolase, S100B, glial fibrillary acidic protein, ubiquitin carboxyl-terminal hydrolase L1, and inflammatory cytokines.
Lactate is widely available and rises when anaerobic metabolism increases. It can support recognition of perinatal metabolic stress, but it is not specific to brain injury. S100B and neuron-specific enolase have been investigated as indicators of neuronal or glial damage, while GFAP and UCH-L1 may offer greater specificity for astroglial and neuronal injury. Their clinical usefulness depends on sampling time, gestational age, assay methods, renal clearance, and whether the sample is contaminated by hemolysis.
No single biomarker currently provides a complete answer. The most promising approach is multimodal: combine cord gases and early blood tests with serial neurological examinations, amplitude-integrated EEG or conventional EEG, and imaging. A changing biomarker profile may also be more informative than one isolated measurement, especially when clinicians need to distinguish transient depression from evolving hypoxic-ischaemic injury.
Reading the brain with neuroimaging
Cranial ultrasound is accessible and can be performed at the bedside, making it useful for detecting major hemorrhage, ventricular abnormalities, cerebral edema, or structural conditions that mimic encephalopathy. Its ability to define subtle cortical or deep gray matter injury is limited, particularly during the first hours after birth.
Magnetic resonance imaging provides a much more detailed assessment. Diffusion-weighted imaging can show restricted water movement associated with acute cellular injury, often before conventional T1- and T2-weighted changes become obvious. Patterns involving the basal ganglia and thalami may suggest an acute, severe insult, while watershed-predominant injury is often associated with more prolonged or partial hypoperfusion. These patterns are guides rather than absolute predictions.
Magnetic resonance spectroscopy adds metabolic information. Reduced N-acetylaspartate or an elevated lactate-to-N-acetylaspartate ratio may correlate with adverse neurodevelopmental outcomes. Imaging should be interpreted alongside the clinical course because therapeutic hypothermia, seizures, infection, congenital abnormalities, and the timing of the insult can alter appearances.
| Assessment method | Main information | Useful timing | Important limitation |
|---|---|---|---|
| Umbilical cord blood gas | Acidosis and metabolic stress at birth | Immediately after delivery | Does not measure brain injury directly |
| Lactate and injury proteins | Systemic stress and possible cellular damage | Serial samples in the first days | Assay and timing affect interpretation |
| Neurological examination | Consciousness, tone, reflexes, seizures | Repeated from birth onward | Influenced by medication and cooling |
| aEEG or conventional EEG | Background activity and electrographic seizures | Early and continuous monitoring | Requires expertise; aEEG is less detailed |
| Cranial ultrasound | Hemorrhage, edema, major structural changes | Bedside, early and serially | Limited sensitivity for subtle injury |
| MRI with diffusion and spectroscopy | Injury pattern, extent, and tissue metabolism | Commonly during the first week | Transport and sedation may be difficult |
Timing matters in therapeutic decisions
Therapeutic hypothermia is an established treatment for eligible term and near-term infants with moderate or severe hypoxic-ischaemic encephalopathy. The greatest benefit is associated with beginning cooling within six hours of birth, so early triage must rely on information that is available quickly: birth history, cord or early blood gas, resuscitation details, and standardized neurological assessment.
Biomarkers and imaging support this process but should not delay cooling when eligibility criteria are met. In an infant with possible encephalopathy, clinicians need to stabilize ventilation, circulation, glucose, temperature, and electrolytes while assessing for seizures and other causes of altered responsiveness. Hypoglycemia, infection, intracranial hemorrhage, metabolic disease, and medication exposure can produce overlapping signs.
EEG is particularly important because many neonatal seizures are electrographic without obvious clinical movements. Seizure burden may be linked to poorer outcomes, while an improving background pattern can provide reassurance. Continuous monitoring is therefore useful during cooling and rewarming, when seizure activity may emerge or change.
Building a reliable prognosis
Prognosis should be based on converging evidence rather than a single abnormal result. Serial examination can show whether alertness, spontaneous movement, suck, tone, and reflexes are recovering. Persistent severe abnormalities, a markedly suppressed EEG background, extensive diffusion restriction, and unfavorable spectroscopy findings may indicate a high risk of death or major disability, but interpretation must account for gestational age and the clinical context.
MRI timing also matters. Early diffusion abnormalities can evolve, and pseudonormalization may reduce the visibility of restricted diffusion later in the course. For that reason, the imaging protocol and day of life should be recorded clearly. Follow-up imaging may be needed when the first scan is technically limited or when the clinical trajectory does not match the initial findings.
Long-term assessment is just as important as neonatal survival. Motor impairment, epilepsy, hearing or visual problems, language delay, executive dysfunction, and difficulties with learning may appear at different ages. Structured developmental surveillance allows early referral for physiotherapy, occupational therapy, speech and language support, vision care, and family-centered intervention.
Improving interpretation across services
A consistent pathway reduces delays and prevents isolated test results from driving major decisions. Delivery-room teams should document the timing and duration of resuscitation, cord gas values, first neurological findings, glucose levels, suspected sentinel events, and the start of any cooling treatment. Neonatal units should then maintain a shared record of EEG, laboratory, imaging, and neurological trends.
Families need clear explanations that distinguish risk from certainty. Terms such as “abnormal MRI” or “elevated biomarker” can sound definitive when they are actually one part of a probability assessment. Conversations should explain what is known, what remains uncertain, why repeat examinations are necessary, and how follow-up will identify emerging needs.
Practical recommendations for clinical teams include:
- Use standardized criteria for suspected hypoxic-ischaemic encephalopathy and therapeutic hypothermia eligibility.
- Obtain and interpret cord gases, lactate, and other biomarkers in relation to timing and clinical findings.
- Start EEG monitoring early when encephalopathy or seizures are suspected, including during cooling and rewarming.
- Schedule MRI with diffusion-weighted sequences and, where available, spectroscopy according to local neonatal protocols.
- Link neonatal findings with placental review and structured neurodevelopmental follow-up.
Translating evidence into care
Research continues to focus on biomarkers that are rapid, brain-specific, and reliable across gestational ages and treatment conditions. Multi-marker panels may eventually distinguish reversible physiological depression from established neuronal injury more accurately than current tests. Artificial intelligence may also assist with MRI pattern recognition and EEG interpretation, provided that models are validated across diverse hospitals and populations.
The strongest clinical framework remains integrated assessment. A cord gas can establish the seriousness of the birth event; a neurological examination can identify encephalopathy; EEG can uncover silent seizures; biomarkers can track biological stress; and MRI can describe the location and extent of injury. Each method answers a different question, and their agreement is more meaningful than any isolated abnormality.
Resources preserved through the FAOPS 2020 website reflect the international collaboration that supports progress in this field. Perinatal centers can turn that shared scientific knowledge into action by standardizing early assessment, expanding access to EEG and MRI, reviewing placental findings, and maintaining long-term developmental pathways for affected infants. Build a coordinated local protocol that brings these tools together before the next high-risk delivery.
