Hypoxic ischemic encephalopathy (HIE) is global brain injury from failure of oxygen delivery, and on imaging it is recognised by symmetrical, non-vascular injury of the most metabolically active grey matter rather than by any single territory. The pattern is decided by the rate of the insult, not its cause: a partial, prolonged insult injures the parasagittal watershed cortex and subcortical white matter, while an acute, profound insult injures the basal ganglia, ventrolateral thalami, hippocampi, perirolandic cortex and brainstem [1][2]. Diffusion-weighted imaging turns positive within hours and pseudonormalises around days 7 to 10, so the sequence you scan on and the day you scan decide what you see.
This article covers both populations that get scanned for HIE and are usually written up separately: the comatose adult after cardiac arrest, where CT and MRI now sit inside a formal neuroprognostication algorithm, and the encephalopathic newborn, where MRI timing and MR spectroscopy drive counselling. Anchors are the 2025 ERC and ESICM post-resuscitation care guidelines [3] and the 2026 American Academy of Pediatrics clinical report on therapeutic hypothermia [8].
- Key facts for practice and radiology board exams
- Terminology: HIE, neonatal encephalopathy and HIBI
- Why the injury is symmetrical: selective vulnerability
- Hypoxic ischemic encephalopathy on MRI in adults
- CT findings after cardiac arrest
- Where imaging fits in post-cardiac-arrest neuroprognostication
- Neonatal hypoxic ischemic encephalopathy: patterns by gestational age
- When to scan a cooled newborn
- NICHD MRI pattern and what it predicts
- Therapeutic hypothermia: what the radiologist needs to know
- Mimics and pitfalls
- Reporting checklist
- DICOM scrollable case
- Frequently asked questions
- References
Key facts for practice and radiology board exams
- Two patterns, one variable. Partial prolonged insult equals parasagittal watershed cortex plus subcortical white matter, deep grey spared. Acute profound insult equals basal ganglia, ventrolateral thalami, hippocampi, perirolandic cortex and brainstem, remaining cortex relatively spared. Mixed patterns are common.
- Timing beats sequence choice. DWI is positive within hours, peaks at days 3 to 5 and pseudonormalises at roughly days 7 to 10. T2 and FLAIR change from about 24 hours and persist. Never call a brain normal on DWI alone inside the pseudonormalisation window.
- Adults, CT. Loss of grey-white differentiation, sulcal and cisternal effacement, a reduced grey-white matter ratio (normal above 1; reported 100 percent specificity thresholds cluster at about 1.10 to 1.23), and the pseudo-subarachnoid haemorrhage, reversal and white cerebellum signs in severe cases [3][5][7].
- Adults, guideline. The 2025 ERC and ESICM guidelines use generalised oedema with a marked grey-white ratio reduction on CT, or extensive diffusion restriction on MRI at 2 to 7 days after return of spontaneous circulation, as one of at least two concordant predictors, assessed no earlier than 72 hours. New in 2025: repeat the CT at 72 to 96 hours if the first CT was clean and the patient is still comatose [3].
- Neonates, timing. DWI on days 2 to 4, conventional imaging on days 4 to 8, with the second week giving the fullest picture of pattern and outcome. Deep grey matter lactate to N-acetylaspartate ratio on MR spectroscopy is the single most accurate biomarker: pooled sensitivity 82 percent, specificity 95 percent, against 91 percent and 51 percent for conventional MRI [11].
- Cooling. Therapeutic hypothermia at 33.5 to 34.5 degrees Celsius, started within 6 hours and continued for 72 hours, for infants at 36 0/7 weeks or more with moderate to severe HIE [8]. It reduces the burden of injury on MRI but does not degrade MRI’s predictive accuracy [13].
- The absent posterior limb sign on neonatal T1 imaging, meaning loss of the normal high signal in the posterior limb of the internal capsule, is one of the strongest single predictors of an abnormal motor outcome in a term infant [10].
Terminology: HIE, neonatal encephalopathy and HIBI
Three overlapping labels appear in reports and in the literature, and mixing them up causes real confusion at multidisciplinary meetings.
| Term | Population | What it means |
|---|---|---|
| Hypoxic ischemic encephalopathy (HIE) | Neonates, and by convention adults too | A clinical encephalopathy with evidence of a hypoxic-ischaemic cause. In the newborn it is the subset of neonatal encephalopathy attributable to a perinatal hypoxic-ischaemic event. |
| Neonatal encephalopathy (NE) | Neonates | The umbrella clinical syndrome of disturbed neurological function in the first days of life, regardless of cause. Sepsis, stroke, metabolic disease, genetic disease and channelopathies all sit inside it, and imaging is often what separates them. |
| Hypoxic-ischaemic brain injury (HIBI) | Adults | The term used in resuscitation and critical care guidelines for post-cardiac-arrest brain injury. It is the leading cause of death among patients who reach an intensive care unit after out-of-hospital cardiac arrest [3]. |
| Anoxic brain injury | Any | A loose clinical synonym. True pure anoxia without any ischaemic component is rare, so most imaging literature prefers hypoxic-ischaemic. |
Why the injury is symmetrical: selective vulnerability
Global hypoperfusion does not damage the brain uniformly. Two properties decide which structures fail first: the local metabolic rate, and the density of glutamate and other excitatory amino acid receptors, which drive excitotoxic calcium influx once ATP-dependent pumps fail. Superimposed on that is the diving reflex, in which sympathetic redistribution preserves flow to the brainstem, cerebellum and deep grey matter at the expense of the cortex.
Put those together and the imaging pattern becomes predictable from the physiology of the arrest.
| Insult | Physiology | Injured on imaging | Relatively spared |
|---|---|---|---|
| Partial, prolonged (mild to moderate; for example prolonged hypotension, chronic placental insufficiency, slow asphyxia) | Blood flow is redistributed away from the cortex to the deep grey and brainstem, so the terminal arterial watersheds fail first | Parasagittal cortex and subcortical white matter at the anterior and posterior interarterial watersheds; in the preterm neonate the periventricular white matter instead | Basal ganglia, thalami, brainstem, cerebellum |
| Acute, profound (near-total; for example cardiac arrest, cord prolapse, uterine rupture, abruption, strangulation) | Redistribution is overwhelmed, so the structures with the highest metabolic rate fail first | Ventrolateral thalami, posterior putamina and globus pallidus, hippocampi, perirolandic cortex, dorsal brainstem, and the posterior limb of the internal capsule in neonates | The remaining cortex and subcortical white matter, at least early |
The practical consequence is that the pattern is a clue to the mechanism, and a mismatch is worth flagging. A neonate with an isolated deep grey pattern points to an acute sentinel event; extensive watershed injury points to a longer, partial insult that may predate labour.
Hypoxic ischemic encephalopathy on MRI in adults
MRI is more sensitive than CT and is the study that defines the extent of injury, but it is harder to obtain in a ventilated, unstable patient, so in practice it is usually performed later than the first CT. The findings evolve through a predictable sequence.
| Phase | Time after the insult | What you see |
|---|---|---|
| Hyperacute to acute | Minutes to 24 hours | Restricted diffusion (high DWI, low ADC) in cortex, basal ganglia and thalami. Conventional sequences may be normal. CT is frequently normal at this stage. |
| Early subacute | 24 hours to about 1 week | Diffusion abnormality peaks around days 3 to 5. T2 and FLAIR hyperintensity develops in the same regions. Cortical swelling with sulcal effacement. |
| Pseudonormalisation | About days 7 to 10 | ADC returns through normal as extracellular water increases. DWI can look deceptively unremarkable. T2 and FLAIR abnormalities persist and become the reliable finding. |
| Late subacute | 1 to 4 weeks | Cortical laminar necrosis: gyriform T1 shortening, best seen on unenhanced T1, from about 2 weeks and peaking at 1 to 3 months. Gyriform contrast enhancement may appear. |
| Chronic | Beyond 1 to 2 months | Volume loss, ex vacuo ventricular dilatation, ulegyria in survivors of childhood insults, and delayed post-hypoxic leukoencephalopathy in a small minority after carbon monoxide poisoning or opioid overdose. |
Adult MRI patterns and what they predict
In a retrospective series of 64 adults with global hypoxic-ischaemic injury on MRI, only 5 had a relatively favourable outcome, and both favourable patterns were the limited ones [2]. The message is blunt: once an adult brain shows an established hypoxic-ischaemic pattern on MRI, the outcome is usually poor whichever pattern it is, and the pattern refines rather than reverses that expectation.
| Pattern | Typical outcome in the reference series [2] |
|---|---|
| Watershed (border zone) cortical and subcortical injury | Relatively favourable |
| Basal ganglia involvement without cortical involvement | Relatively favourable |
| Diffuse cortical plus deep grey matter injury, with or without perirolandic sparing | Poor |
| Medial occipital involvement with perirolandic involvement | Poor |
| Precentral gyrus involvement | Poor |
| Diffuse white matter involvement | Poor |
| Brainstem involvement | Poor |
| Cerebellar involvement | Poor |
| Hippocampal involvement | Poor |
CT findings after cardiac arrest
CT is almost always the first study, because it is fast, it excludes haemorrhage as a cause of the arrest, and it can be done on a ventilated patient. It is insensitive early: a normal CT in the first hours after return of spontaneous circulation does not exclude severe injury, which is exactly why the 2025 guidelines now ask for a repeat scan [3].
- Loss of grey-white matter differentiation. Cytotoxic oedema lowers grey matter attenuation towards that of white matter. Look first at the insular ribbon, the lentiform nuclei and the cortex of the high convexity.
- Effacement of sulci, basal cisterns and the perimesencephalic cistern from vasogenic oedema and diffuse swelling.
- Reduced grey-white matter ratio (GWR). The quantitative version of the same finding, sampled in Hounsfield units at the basal ganglia level and at the high convexity.
- Pseudo-subarachnoid haemorrhage sign. Diffusely low-attenuation swollen brain plus engorged superficial and deep veins makes the cisterns and vessels look falsely hyperdense. Attenuation is usually well below that of true subarachnoid blood, and the distribution follows vessels rather than filling the sulci. Calling this true subarachnoid haemorrhage is the classic trap.
- Reversal sign. White matter attenuation exceeds grey matter attenuation, an inversion of the normal relationship. Predominantly described in children and it carries a dismal prognosis.
- White cerebellum sign. The cerebellum and brainstem look strikingly hyperdense against diffusely hypodense cerebral hemispheres, because the posterior fossa is relatively spared rather than because it is abnormal [7].
Grey-white matter ratio: how it is measured and why there is no single cut-off
Grey matter is normally denser than white matter, so a normal GWR is above 1, and the lower the ratio the worse the oedema. Regions of interest are usually placed at the basal ganglia level, sampling the caudate head and putamen against the posterior limb of the internal capsule and the corpus callosum, and at the high convexity, sampling cortex against the centrum semiovale. Most studies then average them.
Reported thresholds giving 100 percent specificity for a poor neurological outcome sit at roughly 1.10 to 1.23, but they shift with scanner, reconstruction kernel, software and region-of-interest placement, which is why guidelines describe a “marked reduction” rather than endorsing a number [3][5]. Automated GWR software has been developed and performs comparably to manual measurement, but it does not remove the calibration problem.
Timing matters more than the threshold. In a substudy of the Targeted Temperature Management trial, oedema was present on 10 percent of CTs performed within 24 hours of return of spontaneous circulation compared with 46 percent of those performed between 24 hours and 7 days, with sensitivity rising from 14 to 57 percent and specificity from 97 to 100 percent. In a separate comparison, the area under the curve for GWR rose from 0.70 for scans within 6 hours to 0.92 for scans at 72 to 96 hours [3].
Where imaging fits in post-cardiac-arrest neuroprognostication
This is the part radiologists most often get wrong, usually by over-reading a single scan into a prognosis. The 2025 ERC and ESICM guidelines, and the 2023 Neurocritical Care Society guidelines before them, are explicit that no single predictor is accurate enough to be used alone [3][4].
- Exclude confounders first, above all residual sedation, neuromuscular blockade and hypothermia.
- Consider prognostication only in a patient who is not awake and not obeying commands, defined in 2025 as a Glasgow Coma Scale motor score below 6, at 72 hours or later after return of spontaneous circulation. This is a change from the 2021 threshold of a motor score of 3 or less.
- Poor outcome is likely when two or more of the following are present: absent pupillary and corneal reflexes at 72 hours or later; bilaterally absent N20 somatosensory evoked potential at 24 hours or later; a highly malignant EEG beyond 24 hours; neuron-specific enolase above 60 micrograms per litre at 48 and/or 72 hours; status myoclonus within 72 hours; or diffuse and extensive anoxic injury on brain CT or MRI.
- Imaging specifics: a marked reduction of the grey-white matter ratio on brain CT within 72 hours after return of spontaneous circulation, or extensive diffusion restriction on brain MRI at 2 to 7 days after return of spontaneous circulation.
- New in 2025: repeat the brain CT if the patient remains unconscious at the time of prognostication, 72 to 96 hours after return of spontaneous circulation, and the first CT showed no sign of hypoxic-ischaemic brain injury.
- The guidelines also formalise favourable predictors for the first time, including the absence of diffusion restriction on MRI, a motor score of 4 to 5 at 72 to 96 hours, neuron-specific enolase below 17 micrograms per litre, and a continuous EEG background.
Two practical points for the report. First, imaging is one of the few prognostic modalities that sedative drugs cannot confound, which is precisely why it is valuable in a patient whose examination is uninterpretable. Second, the guidelines ask that these studies be read by someone with specific experience, and suggest telemedicine neuroradiology consultation where that is not available locally [3]. Describe the extent and distribution of injury; do not write the prognosis.
Neonatal hypoxic ischemic encephalopathy: patterns by gestational age
The same rate-of-insult logic applies, but the vulnerable structures shift with brain maturation, because the regions that are actively myelinating and metabolically busiest change through gestation.
| Gestation | Partial, prolonged insult | Acute, profound insult |
|---|---|---|
| Preterm, under about 32 to 34 weeks | Periventricular white matter injury, historically periventricular leukomalacia; germinal matrix and intraventricular haemorrhage as a separate but frequently coexisting process | Thalami, basal ganglia, anterior cerebellar vermis and dorsal brainstem, often with relative cortical sparing |
| Term, 36 weeks and above | Parasagittal watershed cortex and subcortical white matter, in the anterior and posterior interarterial border zones | Ventrolateral thalami, posterior putamina, hippocampi, perirolandic cortex, corticospinal tracts and dorsal brainstem, with loss of the normal posterior limb of the internal capsule signal |
Signs to look for on a neonatal scan
- Absent posterior limb sign. On T1 the posterior limb of the internal capsule is normally myelinated and bright by term. Loss of that high signal, or T2 hyperintensity there, is among the strongest single predictors of an abnormal motor outcome [10].
- Restricted diffusion in the ventrolateral thalami and posterior putamina, the signature of the acute profound pattern, often more conspicuous than any T1 or T2 change in the first days.
- Parasagittal cortical highlighting and loss of cortical ribbon definition in the watershed zones on T2, easily missed against the high water content of the immature brain.
- T1 hyperintensity in the deep grey matter, appearing after the first week, and later loss of the normal T2 hypointensity of the myelinating thalamus.
- Cerebral oedema on cranial ultrasound with slit-like ventricles, increased periventricular echogenicity and, on Doppler, a falling resistive index. Ultrasound is useful for screening and for haemorrhage, but it is not a substitute for MRI in HIE.
When to scan a cooled newborn
A single well-timed MRI answers more questions than two badly timed ones, and the two most useful sequences peak at different times.
| Window | What it is good for | Caveat |
|---|---|---|
| Days 1 to 2 | Planning immediate management if the baby is unstable | Underestimates or misses injury. If the scan is normal in a clinically encephalopathic infant, be prepared to repeat it. |
| Days 2 to 4 | Diffusion-weighted imaging is at its most informative | Conventional T1 and T2 changes may still be minimal. |
| Days 4 to 8 | Conventional T1 and T2 define the pattern; the standard single-scan window | Diffusion begins to pseudonormalise from about day 7. |
| Days 5 to 15 | Proton MR spectroscopy of the deep grey matter, best acquired with the structural scan | Requires a spectroscopy-capable protocol and someone able to quantify the peaks. |
| Second week | Fullest picture of the final pattern and of outcome | Later than many units discharge, so it needs planning. |
On the numbers, conventional MRI performed anywhere in the neonatal period has a pooled sensitivity of 91 percent for adverse outcome but a specificity of only 51 percent, so a lot of abnormal-looking scans belong to children who do well. Deep grey matter lactate to N-acetylaspartate peak-area ratio on MR spectroscopy performs far better, with a pooled sensitivity of 82 percent and specificity of 95 percent, and beat conventional MRI on head-to-head analysis. Apparent diffusion coefficient of brain water alone and the posterior limb sign in isolation had poor discriminatory power in that meta-analysis, which is an argument for reading them together rather than dropping them [11].
NICHD MRI pattern and what it predicts
The National Institute of Child Health and Human Development Neonatal Research Network pattern is the classification most often quoted in trials and in follow-up clinics, and it maps cleanly onto outcome at school age. In 124 children from the Network hypothermia trial followed to 6 or 7 years, death or an IQ below 70 tracked the pattern almost linearly [12].
| NICHD pattern | Definition | Death or IQ below 70 at 6 to 7 years |
|---|---|---|
| 0 | Normal MRI | 4 of 50 (8 percent) |
| 1A | Minimal cerebral lesions, no involvement of basal ganglia, thalamus, internal capsule or watershed | 1 of 6 (17 percent) |
| 1B | Extensive cerebral lesions | 1 of 4 (25 percent) |
| 2A | Basal ganglia and thalamic, anterior or posterior limb of internal capsule, or watershed infarction | 3 of 8 (38 percent) |
| 2B | Pattern 2A plus cerebral lesions | 32 of 49 (65 percent) |
| 3 | Hemispheric devastation | 7 of 7 (100 percent) |
Mean IQ was 90 plus or minus 13 in children with a normal neonatal MRI and 69 plus or minus 25 in those with an abnormal one. A normal neonatal MRI predicted a normal outcome with 92 percent specificity and a 92 percent positive predictive value; patterns 2B and 3 combined predicted death or an IQ below 70 with 81 percent sensitivity, 78 percent specificity and an 87 percent negative predictive value [12]. The older Barkovich scoring systems, which grade basal ganglia and watershed injury separately, remain useful and were the first to show that the deep grey score predicts motor outcome better than the watershed score [10].
Therapeutic hypothermia: what the radiologist needs to know
The 2026 American Academy of Pediatrics clinical report sets the current parameters: cooling to 33.5 to 34.5 degrees Celsius, started within 6 hours of birth and continued for 72 hours, in infants born at 36 0/7 weeks or later with moderate to severe HIE, reduces death or moderate-to-severe neurodevelopmental impairment. The report stresses that cooling is a package that includes neuromonitoring, neuroimaging and structured developmental follow-up, and that referring units need transfer pathways because most affected babies are not born in cooling centres [8].
- Cooling changes what you see, not how you read it. In the MRI substudy of the TOBY trial, cooled infants had fewer basal ganglia and thalamic lesions (odds ratio 0.36), fewer white matter lesions (0.30) and fewer abnormal posterior limbs (0.38), and were more likely to have a normal scan (2.81). Predictive accuracy for death or disability at 18 months was 0.84 in cooled infants versus 0.81 in non-cooled infants [13].
- Cooling protocols do not transfer automatically. The HELIX trial randomised 408 infants across seven tertiary units in India, Sri Lanka and Bangladesh. Death or moderate or severe disability at 18 to 22 months occurred in 50 percent of cooled infants and 47 percent of controls (risk ratio 1.06), while mortality was higher with cooling, 42 percent versus 31 percent. Cooling is not recommended outside neonatal intensive care units able to deliver the whole package [14].
- Adjuncts have not delivered. The HEAL trial found that erythropoietin given alongside hypothermia did not reduce death or neurodevelopmental impairment and was associated with more serious adverse events, so MRI remains the outcome measure rather than a therapeutic target [15].
- Expect a request for an MRI before discharge. Neuroimaging is part of the standard package for any cooled infant, which is why timing the scan into the days 4 to 8 window matters operationally, not just academically [8][16].
Mimics and pitfalls
| Entity | How it differs from hypoxic ischemic encephalopathy |
|---|---|
| Acute ischaemic stroke | Follows an arterial territory, is usually unilateral, and often has a visible occlusion or perfusion deficit. HIE is symmetrical and crosses territories. |
| Neonatal hypoglycaemia | Parieto-occipital predominant white matter and cortical injury, often with sparing of the deep grey matter that HIE targets. The two can coexist. |
| Kernicterus | Symmetrical T1 hyperintensity of the globus pallidus and subthalamic nuclei in the acute phase, with the clinical and biochemical context of severe hyperbilirubinaemia. |
| Mitochondrial and organic acid disorders (Leigh syndrome, maple syrup urine disease) | Symmetrical deep grey or brainstem signal change, but usually with lactate on spectroscopy in the absence of an acute event, an atypical distribution, and a progressive rather than a single-insult course. |
| Carbon monoxide poisoning | Globus pallidus predilection with later confluent white matter change; delayed post-hypoxic leukoencephalopathy is a recognised sequela. |
| Status epilepticus | Cortical, hippocampal and pulvinar restricted diffusion, frequently unilateral or lobar and at least partly reversible. |
| Pseudo-subarachnoid haemorrhage on CT | Not a mimic of HIE but a consequence of it. Do not report true subarachnoid haemorrhage in a diffusely swollen post-arrest brain without checking attenuation values and distribution. |
| Normal neonatal myelination | The posterior limb of the internal capsule is normally T1 bright at term. Judging it against an incorrect gestational age is a common source of both false positives and false negatives. |
Reporting checklist
- State the modality and, critically, the interval from the insult or from return of spontaneous circulation to the scan. Every finding below is time-dependent.
- Describe the distribution: watershed, deep grey, cortical, brainstem, cerebellar, or a combination, and say whether it is symmetrical.
- On CT, comment on grey-white differentiation, sulcal and cisternal effacement, and note explicitly if apparent hyperdensity in the cisterns is a pseudo-subarachnoid haemorrhage appearance.
- On MRI, report DWI and ADC together and state whether the study falls inside the pseudonormalisation window.
- In a neonate, state the posterior limb of the internal capsule signal and, where relevant, assign the NICHD pattern and report the deep grey lactate to N-acetylaspartate ratio.
- In an adult, describe extent and distribution and stop there. Prognostication is multimodal and belongs to the treating team.
- Recommend the next scan and when: a repeat CT at 72 to 96 hours in a persistently comatose adult with a normal first CT, or a repeat MRI in an encephalopathic neonate scanned in the first 48 hours.
DICOM scrollable case
Scroll through the full study below.
Frequently asked questions
References
- Huang BY, Castillo M. Hypoxic-ischemic brain injury: imaging findings from birth to adulthood. Radiographics. 2008;28(2):417-439.
- Muttikkal TJ, Wintermark M. MRI patterns of global hypoxic-ischemic injury in adults. J Neuroradiol. 2013;40(3):164-171.
- Nolan JP, Sandroni C, Cariou A, Cronberg T, Friberg H, Genbrugge C, et al. European Resuscitation Council and European Society of Intensive Care Medicine guidelines 2025: post-resuscitation care. Intensive Care Med. 2025;51(12):2213-2288.
- Rajajee V, Muehlschlegel S, Wartenberg KE, Alexander SA, Busl KM, Chou SHY, et al. Guidelines for neuroprognostication in comatose adult survivors of cardiac arrest. Neurocrit Care. 2023;38(3):533-563.
- Lopez Soto C, Dragoi L, Heyn CC, Kramer A, Pinto R, Adhikari NKJ, et al. Imaging for neuroprognostication after cardiac arrest: systematic review and meta-analysis. Neurocrit Care. 2020;32(1):206-216.
- Beekman R, Hirsch KG. Brain imaging after cardiac arrest. Curr Opin Crit Care. 2023;29(3):192-198.
- Lai YN, Chung JY. The white cerebellum sign. J Acute Med. 2024;14(1):48-49.
- Zanelli SA, Wusthoff CJ, Lucke AM, Kaufman DA. Therapeutic hypothermia for neonatal hypoxic-ischemic encephalopathy: clinical report. Pediatrics. 2026;157(2):e2025073627.
- Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress. A clinical and electroencephalographic study. Arch Neurol. 1976;33(10):696-705.
- Barkovich AJ, Hajnal BL, Vigneron D, Sola A, Partridge JC, Allen F, et al. Prediction of neuromotor outcome in perinatal asphyxia: evaluation of MR scoring systems. AJNR Am J Neuroradiol. 1998;19(1):143-149.
- Thayyil S, Chandrasekaran M, Taylor A, Bainbridge A, Cady EB, Chong WK, et al. Cerebral magnetic resonance biomarkers in neonatal encephalopathy: a meta-analysis. Pediatrics. 2010;125(2):e382-e395.
- Shankaran S, McDonald SA, Laptook AR, Hintz SR, Barnes PD, Das A, et al. Neonatal magnetic resonance imaging pattern of brain injury as a biomarker of childhood outcomes following a trial of hypothermia for neonatal hypoxic-ischemic encephalopathy. J Pediatr. 2015;167(5):987-993.
- Rutherford M, Ramenghi LA, Edwards AD, Brocklehurst P, Halliday H, Levene M, et al. Assessment of brain tissue injury after moderate hypothermia in neonates with hypoxic-ischaemic encephalopathy: a nested substudy of a randomised controlled trial. Lancet Neurol. 2010;9(1):39-45.
- Thayyil S, Pant S, Montaldo P, Shukla D, Oliveira V, Ivain P, et al. Hypothermia for moderate or severe neonatal encephalopathy in low-income and middle-income countries (HELIX): a randomised controlled trial in India, Sri Lanka, and Bangladesh. Lancet Glob Health. 2021;9(9):e1273-e1285.
- Wu YW, Comstock BA, Gonzalez FF, Mayock DE, Heagerty PJ, Hamrick SEG, et al. Trial of erythropoietin for hypoxic-ischemic encephalopathy in newborns. N Engl J Med. 2022;387(2):148-159.
- Machie M, de Vries LS, Inder T. Advances in neuroimaging biomarkers and scoring. Clin Perinatol. 2024;51(3):629-647.
Improve Content And Help Your Colleagues!
Found an error in the post? Please let us know using our contact page and we will update it with due credits!
If you wish to contribute radiological images for this case (or any other article on the website), you can submit them here and you will be duly credited: Submit radiology case
Wow