The PATTERN depends on the severity and the maturity of the brain, and the TIMING of the scan changes what diffusion shows.
Orient first
- Two patterns, determined by the severity of the insult. A PROFOUND acute insult injures the metabolically active deep grey structures — basal ganglia, thalami, perirolandic cortex and the hippocampi. A PROLONGED PARTIAL insult injures the watershed cortex and subcortical white matter.
- MATURITY changes the target: the preterm brain injures the periventricular white matter and germinal matrix; the term brain injures the deep grey nuclei and watershed zones.
- TIMING IS CRITICAL AND COUNTERINTUITIVE. Diffusion restriction appears early, PEAKS at roughly days 2–4, and then PSEUDONORMALISES around days 7–10. A scan in that window can look falsely reassuring on diffusion alone.
- Loss of the normal T1-bright signal in the posterior limb of the internal capsule is an important prognostic sign at term.
- Therapeutic hypothermia alters the evolution and may delay the appearance of changes, so the report must state whether cooling was given.
Acquire the study
- T1 and T2 in at least two planes, DWI with ADC, and susceptibility-weighted imaging for haemorrhage; add spectroscopy where available.
- Record the exact POSTNATAL AGE at scanning and whether therapeutic hypothermia was administered — both change the interpretation entirely.
- Feed-and-wrap rather than sedation where possible; motion is the main cause of a non-diagnostic neonatal study.
- Confirm which sequence you are on before judging any signal — see the MRI sequence primer.
- Use neonatal-appropriate windowing; adult settings make the unmyelinated brain look uniformly abnormal.
The manoeuvre
- State the gestational age at birth, the postnatal age at scanning, and whether cooling was given.
- Assess the DEEP GREY structures: basal ganglia, thalami and the perirolandic cortex, on T1, T2 and diffusion.
- Assess the posterior limb of the INTERNAL CAPSULE for its normal T1-bright signal.
- Assess the WATERSHED zones — cortex and subcortical white matter between arterial territories.
- Review DWI and the ADC map, and interpret them against the postnatal age, explicitly considering pseudonormalisation.
- Assess the periventricular white matter, particularly in a preterm infant, and look for germinal matrix haemorrhage on susceptibility imaging.
- Look for haemorrhage, sinus thrombosis and any structural malformation, which change the diagnosis altogether.
- If spectroscopy is available, assess for a lactate peak, which is an early and prognostically useful marker.
What confirms it
- A recognised injury pattern in the appropriate distribution for maturity and severity, on a scan whose timing supports the finding.
What licenses you to exclude it
- ⚠️ A NORMAL-LOOKING DIFFUSION SCAN IN THE PSEUDONORMALISATION WINDOW DOES NOT EXCLUDE INJURY. If the scan falls around days 7–10, say so and interpret conventional sequences rather than reassuring on diffusion.
- A very early scan may precede visible change; state when a repeat would be informative.
- Normal imaging does not exclude a metabolic or genetic cause of encephalopathy — those are different investigations.
The classic misread
- Reassuring on diffusion in the pseudonormalisation window.
- Applying term patterns to a preterm brain.
- Reading the unmyelinated neonatal brain at adult window settings.
- Not recording postnatal age or cooling status, without which the study cannot be interpreted.