MANUSCRIPT CITATION

Li Y, Wisnowski JL and Chalak L, et al. Mild hypoxic-ischemic encephalopathy (HIE):
timing and pattern of MRI brain injury. Pediatr Res. 2022 Dec;92(6):1731-1736. doi:
10.1038/s41390-022-02026-7. Epub 2022 Mar 30. PMID: 35354930; PMCID: PMC9771796.

REVIEWED BY

Mahmoud Abdelreheem, MBBCh, MSc, MRCPCH
ST4 Paediatric trainee, West Yorkshire, UK
Paediatrics department
Calderdale Royal Hospital NHS trust
Mahmoud.abdelreheem@nhs.net

Hassanein Moustafa, MBBCh, MSc (Paed), MRCPCH, PGCME
Neonatal GRID ST5
Department of neonatology
Liverpool Women’s Hospital
United Kingdom
hassanein.moustafa@doctors.org.uk

TYPE OF INVESTIGATION

Diagnosis

QUESTION

In neonates with mild HIE, does performing brain MRIs within the first 8 days of age detect
and classify the brain injury severity and pattern.

METHODS

• Design: observational retrospective cohort study
• Allocation: term and near-term infants born at ≥36 weeks gestational age with mild HIE using a modified Sarnat examination with evidence of perinatal depression
• Blinding: Unblinded
• Follow-up period: not applicable
• Setting: nine hospitals in the United States in between 2013–2019
• Patients: 142 infants with mild HIE stratified by presence of brain injury on MRI.
• Inclusion criteria: infants ≥36 weeks gestational age with mild HIE using a modified Sarnat examination (at least one Sarnat exam abnormality of any severity in any of the six categories (i.e., consciousness, activity, tone, posture, primitive reflexes, or autonomic nervous system) documented at any time between 1- 6 hours of age. Additionally, infants had evidence of perinatal depression with at least one of the following: 10-minute Apgar score <5, need for resuscitation at 10 min, pH < 7.00 in an arterial or venous cord or infant gas performed by 60 min of age, or base deficit ≥15 mmol/L in a cord or infant gas performed by 60 min of age
• Exclusion criteria: any patient with any of the following:
– birthweight <1800 g (i.e., intrauterine growth restriction)
– head circumference <30 cm
– encephalopathy occurring after birth (i.e., postnatal collapse)
– unavailable MRI
• Intervention: brain MRI at or before 8 days of age, which was scored by 2 reviewers using a validated HIE severity classification system, with discrepancies were resolved by consensus. Severity and timing of MRI brain injury (i.e., acute, subacute, chronic) was scored on the subset of MRIs. A global injury severity score was calculated only in patients imaged within 8 days of age, because diffusion-weighted signal abnormalities, which constitute a component of the scoring system
• Outcomes: timing and pattern of brain injury
• Sample Size and analysis: 142 infants were eligible for this study. Each patient was classified as having one or more of the following patterns of brain injury: none, watershed, deep gray, punctate white matter lesions, arterial ischemic stroke, focal parenchymal lesions, hippocampal injury, and atypical lesions. Atypical lesions were defined as any abnormality not fitting into the previously described categories and thus are atypical findings in the setting of HIE. Intraparenchymal and subdural haemorrhages were scored as trace, mild/moderate if no mass effect, and severe if mass effect was present. Intraventricular haemorrhage was similarly classified as trace, mild/moderate if no ventricular dilation, and severe if ventricular dilation was present. Subarachnoid haemorrhage was scored as either trace or focal.
The time window of brain injury was assessed in all brain MRI studies performed within 8 days of age. Acute injury was defined by the presence of restricted diffusion. Subacute injury was defined by presence of an abnormality on T1 and T2-weighted imaging without accompanying diffusion abnormality or volume loss. Chronic injury was defined by the presence of parenchymal volume loss.
The authors examined how brain injury timing and severity varied by clinical factors such as antenatal complications and treatment with TH using Wilcoxon Rank-Sum test for continuous variables and Pearson’s chi-squared test for categorical variables. The authors also examined whether the frequency of brain injury varied by the number of Sarnat exam abnormalities. The authors compared the rate of MRI brain injury in subjects who met criteria for mild HIE based on more restrictive definitions used in two clinical trials: (1) MEND (NCT03071861): mild HIE = one or two moderate or severe modified Sarnat abnormality in any of the six categories; and (2) TIME (NCT04176471): mild HIE = at least 2 Sarnat abnormalities of any severity, but not qualifying for moderate/severe encephalopathy.
• Patient follow-up: not applicable

MAIN RESULTS

• Brain parenchymal injury was present in 87 (61%) infants with mild HIE
• No clinical factors were significantly associated with the presence of brain injury
• The most common brain injury patterns were watershed (22%), deep gray nuclei (20%), punctate white matter (18%), and atypical lesions (18%). 22% had more than one injury pattern
• Intracranial hemorrhage was present in 60 (42%) infants, including 21 (15%) with moderate or severe hemorrhage. The most common location of hemorrhage was subdural (34%), followed by intraventricular (12%), intraparenchymal (6%), and subarachnoid (3%). Three subdural hemorrhages were classified as large enough to exhibit mass effect and one large intraventricular hemorrhage led to non-communicating hydrocephalus. There were no cases of large intraparenchymal hemorrhage
• The majority (85%) of infants in the cohort were treated with Therapeutic Hypothermia (TH). The 21 infants with mild HIE who did not receive TH exhibited no difference in mean lowest cord pH when compared to the treated group. Infants who did not receive TH had a higher median 5-min Apgar than those who did receive TH (7 vs. 4, p < 0.0001). Infants who underwent TH did not differ significantly from the untreated group in frequency of brain injury on MRI (60% vs. 71%, p = 0.30). However, watershed injury was less common in those who received TH (18% vs. 48%, p = 0.003)
• Among the 125 (88%) of infants with mild HIE who received an early brain MRI at ≤8 days of age, 75 (60%) had evidence of parenchymal brain injury. The median injury score in this group with early MRI was 5 (IQR 2–9, range 0–84) and when present, the brain injury was predominantly mild. A greater percentage of infants exhibited subacute (37%) than acute (32%) or chronic (1%) injury
• There were no clinical factors that were significantly associated with either severity or timing of brain injury in the subgroup of infants with early MRI.

CONCLUSION

This large multicenter study evaluating the timing and pattern of brain injury in infants with mild HIE found that approximately two-thirds of infants exhibited evidence of brain injury on MRI, with a notably high rate of subacute lesions. Novel neuroprotective treatments for mild HIE will ideally target both subacute and acute injury mechanisms. The predominance of subacute injury suggests that mild HIE may result from injuries occurring hours or even days before delivery, potentially placing these infants beyond the critical six-hour neuroprotective window for therapeutic hypothermia. Consequently, the study raises the possibility that TH might be less effective for infants with mild HIE. This finding underscores the urgent need for clinical trials to assess the efficacy of TH and other neuroprotective therapies, as well as to determine which patient subgroups may benefit most from such interventions. Additionally, the study observed that variations in the definition of mild HIE have a minimal impact on the frequency and severity of brain injury.

COMMENTARY

Neonatal hypoxic-ischaemic encephalopathy (HIE) remains one of the leading causes of neonatal mortality and long-term disability worldwide. Infants with mild HIE, representing 50% of all HIE cases, are often perceived as low risk, with excellent prognosis and no long-term disability. However, recent studies have shown that one-quarter of infants with mild HIE have abnormal outcomes defined as death, motor impairment, or developmental delay at follow-up up to 18 months [1].
This study examines the evolution and spatial distribution of brain injuries on MRI in neonates with mild HIE. It addresses a significant gap in literature and has important implications for early diagnosis, prognosis, and treatment strategies. Brain parenchymal injury was present in 87 (61%) infants, and intracranial hemorrhage was observed in 60 (42%) infants.

 

The authors recruited 142 neonates, although only 125 were included in certain analyses, likely due to incomplete imaging. This transparent reporting enhances the credibility of the findings by ensuring that the data analyzed were of high quality. This study highlights how subtle brain injuries evolve over time, confirming that early detection may allow for timely interventions [2, 3]. Moreover, this study explores brain injury patterns, revealing that even mild HIE can cause noticeable changes in brain structure. Identifying these patterns is vital, as it helps clinicians differentiate HIE-related changes from other neonatal brain pathologies. This informs clinical decisions, including whether to consider therapeutic hypothermia (TH) in borderline cases [4]. Additionally, the study’s use of advanced MRI techniques underscores the potential of neuroimaging as both a diagnostic and prognostic tool in neonatal care.

 

However, the study has limitations, notably the small number of infants who did not receive TH, which may limit the generalizability of the findings. The authors also acknowledge variability in MRI protocols across institutions, the lack of EEG data, and potential selection bias toward initiating TH in more severely affected infants within the mild HIE spectrum. Notably, there have been no randomized controlled trials evaluating the efficacy of TH in mild HIE, and many centers continue to question its utility in this population.

 

Clinically, the implications of this work are twofold. First, it emphasizes the need to monitor neonates with mild HIE, as subtle injuries may have long-term neurodevelopmental consequences. Second, it supports incorporating MRI into routine evaluation, facilitating identification of infants who could benefit from neuroprotective interventions. The findings also advocate for further studies to evaluate the efficacy of TH and other therapeutic agents, leading to individualized management strategies for affected infants [5, 6].

 

In summary, this study delivers a well-designed investigation that enhances our understanding of mild HIE. It lays the groundwork for future research aimed at optimizing the timing of neuroimaging and improving outcomes through early intervention. The study serves as a reminder that even mild insults to the neonatal brain warrant careful attention due to their potential to impact long-term development [7]. Future studies with larger cohorts and standardized imaging protocols are needed to confirm these observations and refine MRI timing in mild HIE cases.

REFERENCES

1. Conway JM, Walsh BH, Boylan GB, Murray DM. Mild hypoxic ischaemic
encephalopathy and long term neurodevelopmental outcome – A systematic review.
Early Hum Dev. 2018 May;120:80-87. doi: 10.1016/j.earlhumdev.2018.02.007. Epub
2018 Feb 26. PMID: 29496329.

2. Rutherford M, Malamateniou C, McGuinness A, Allsop J, Biarge MM, Counsell S.
Magnetic resonance imaging in hypoxic-ischaemic encephalopathy. Early Hum Dev.
2010 Jun;86(6):351-60. doi: 10.1016/j.earlhumdev.2010.05.014. Epub 2010 Jun 12.
PMID: 20541877.

3. Barkovich AJ. MR imaging of the neonatal brain. Neuroimaging Clin N Am. 2006 Feb;16(1):117-35, viii-ix. doi: 10.1016/j.nic.2005.10.003. PMID: 16543088.

4. Singla M, Chalak L, Kumar K, Hayakawa M, Mehta S, Neoh SH, Kitsommart R, Yuan
Y, Zhang H, Shah PS, Smyth J, Wandita S, Yeo KT, Lim G, Oei JL. “Mild” Hypoxic-
Ischaemic Encephalopathy and Therapeutic Hypothermia: A Survey of Clinical Practice
and Opinion from 35 Countries. Neonatology. 2022;119(6):712-718. doi:
10.1159/000526404. Epub 2022 Oct 6. PMID: 36202069.

5. Robertson NJ, Wyatt JS.The magnetic resonance revolution in brain imaging:
impact on neonatal intensive care.Archives of Disease in Childhood – Fetal and
Neonatal Edition 2004;89:F193-F197.

6. Wu YW, Monsell SE, Glass HC, Wisnowski JL, Mathur AM, McKinstry RC, Bluml S, Gonzalez FF, Comstock BA, Heagerty PJ, Juul SE. How well does neonatal neuroimaging correlate with neurodevelopmental outcomes in infants with hypoxic-ischemic encephalopathy? Pediatr Res. 2023 Sep;94(3):1018-1025. doi: 10.1038/s41390-023-02510-8. Epub 2023 Mar 1. PMID: 36859442; PMCID: PMC10444609.

7.Walsh BH, Inder TE. MRI as a biomarker for mild neonatal encephalopathy. Early
Hum Dev. 2018 May;120:75-79. doi: 10.1016/j.earlhumdev.2018.02.006. Epub 2018
Feb 17. PMID: 29463417.