Background: There is increasing concern that infants with mild hypoxic-ischaemic encephalopathy (HIE) may develop seizures and progress to moderate HIE beyond the therapeutic window for cooling. Objective: The aim of this study was to examine the effect of therapeutic hypothermia on magnetic resonance imaging (MRI) biomarkers and neurological outcomes in infants with mild HIE and seizures within 24 h after birth. Methods: This study shows an observational cohort study on 366 (near)-term infants with mild HIE and normal amplitude-integrated electroencephalography background. Results: Forty-one infants showed progression (11.2%); 29/41 (70.7%) were cooled. Infants with progression showed cerebral metabolite perturbations and higher white matter injury scores compared to those without in both cooled and non-cooled groups (p = 0.001, p = 0.02). Abnormal outcomes were seen in 5/12 (42%) non-cooled and 7/29 (24%) cooled infants with progression (p = 0.26). Conclusions: Early biomarkers are needed to identify infants with mild HIE at risk of progression. Mild HIE infants with progression showed a higher incidence of brain injury and abnormal outcomes.

© 2022 The Author(s). Published by S. Karger AG, Basel

Background

Term infants with mild hypoxic-ischaemic encephalopathy (HIE), representing almost half of those with HIE, had long been considered a low-risk group. However, recent evidence shows that a significant proportion of these infants will develop significant disability including behaviour problems and neurodevelopmental impairment [1].

The current management of mild HIE is controversial. Despite the lack of clear evidence [2], therapeutic hypothermia (TH) is increasingly offered to neonates with mild HIE and nearly a third of the neonates referred for assessment for TH have now mild HIE [3]. The reasons behind this therapeutic creep may be the concern that seizures may develop after 6 h of age and these infants may progress to a moderate stage of encephalopathy beyond the therapeutic window for cooling. In this study, we aimed to examine the clinical features, brain injury, and neurological outcomes in (near-)term infants with mild HIE who developed seizure within 24 h after birth and the effect of whole-body hypothermia on them.

Methods

We retrospectively identified infants born at ≥36 weeks gestational age with perinatal asphyxia [4] and mild HIE admitted to the level III neonatal intensive care units of the University of Campania network and to the IRCCS “Burlo Garofolo,” Italy between January 2016 and October 2020. As part of the local clinical protocol, all infants received a standardized neurological examination [5] soon after birth and in case of any abnormality; amplitude-integrated electroencephalography (aEEG) monitoring was performed for at least 24 h. Infants were included if there was evidence of mild encephalopathy and showed aEEG background with upper margin of aEEG activity >10 μV and lower margin >5 μV (online suppl. material; for all online suppl. material, see www.karger.com/doi/10.1159/000527871) [6].

Mild HIE was defined as such if any of the six categories [5] was abnormal regardless of the severity: the definition still remained mild HIE as long as no more than two abnormalities were moderate/severe [7]. The decision to offer TH was based on the attending clinician’s preference. Regardless of whether they were treated with TH, these infants also received a brain magnetic resonance imaging (MRI) as part of their routine care.

Only infants with seizure onset in the 24 h after birth and confirmed on aEEG were considered as part of the progression group. Cranial MRI and thalamic proton MR spectroscopy (MRS) were performed on 1.5 (Siemens) or 3T (GE) scanners between 4 and 14 days of age. Conventional MRI data were reported using a validated scoring system [8]. All MR data were analysed centrally while masked to the clinical outcomes. Infants with mild HIE and progression had also an 18- to 24-month neurodevelopmental assessment (Bayley-III) based on which the presence of neurological disability was defined [9]. Further details on the neurological examination and neurodevelopmental assessment are in the online supplementary material.

Statistical Analysis

Categorical variables were compared using a χ2 test or Fisher’s exact test. Analysis of variance and Mann-Whitney U tests were used to compare the means among the different groups. Analyses were performed with SPSS Statistics V.24 (IBM). A p value <0.05 was considered statistically significant.

ResultsStudy Population and Neonatal Course

From a subgroup of 366 infants with mild encephalopathy, 264 underwent TH and 102 did not. Forty-one infants progressed to a moderate stage of HIE (11.2%) (Fig. 1A). Median [IQR] age at seizure onset was 10 [7] hours. Median [IQR] age at start of aEEG registration was 2.8 [2.4] hours. Two hundred thirty-five out of 325 (72.3%) mild HIE infants without progression were cooled.

Fig. 1.

A Study profile. Patient recruitment and retention flow chart. B Cranial magnetic resonance (MR) imaging of three infants. a T2-weighted axial MR shows arterial ischaemic stroke in the territory of the left posterior cerebral artery and middle cerebral artery with loss of the cortical ribbon in the left parietal and occipital lobes. a1 The corresponding T1-weighted sagittal MR shows involvement of both the parietal and occipital lobes. Neurodevelopmental outcome at 24 months showed cognitive delay with speech difficulties (Bayley Scales of Infant and Toddler Development–Third Edition cognitive composite score 80, language composite score 74 and motor composite score 82). b T2-weighted axial MR shows mild diffused white matter injury. b1 The corresponding T1-weighted sagittal MR shows a punctate lesion with involvement of the cortex (white arrow). Neurodevelopmental outcome at 24 months was normal (Bayley Scales of Infant and Toddler Development–Third Edition cognitive composite score 100, language composite score 100 and motor composite score 91). c T2-weighted axial shows moderate diffused white matter injury with multiple punctate lesions (white arrows) and intraventricular bilateral haemorrhage (black arrows). c1 The same punctate lesions in the white matter are shown on T1-weighted sagittal MR (white arrows). Neurodevelopmental outcome at 18 months was abnormal (Bayley Scales of Infant and Toddler Development–Third Edition cognitive composite score 80, language composite score 71 and motor composite score 97).

In the progression group, 25 out of 41 (61%) infants were already undergoing TH at the time of progression whereas 16 (39%) were not. Of these latter, 12 infants were not cooled whereas 4 infants were cooled after 6 h after birth when seizures occurred (median [IQR], 8.5 [2] hours). The incidence of seizures was not significantly different between the infants who were undergoing TH within 6 h after birth (25/260, 9.6%) and those who were not (16/106, 15%) when seizures occurred (p = 0.13).

The baseline characteristics of progression and non-progression infants were similar apart from significantly lower birth weight and longer hospital stay in the infants with progression. The lactate dehydrogenase activity was significantly higher in the infants with progression. Sleep-wake cycling onset occurred later in infants with progression (Table 1). In 12/41 (29%) of the infants with progression, aEEG background had become moderately abnormal before seizure onset. The number of abnormal categories on the modified Sarnat score increased from admission to before seizure onset (online suppl. Fig. 1).

Table 1.

Clinical characteristics, MRI and MRS of the infants with and without progression: neurodevelopmental outcomes at 18–24 months in the infants with progression

Magnetic Resonance Imaging and MRS

Thirty-five infants (85%) with progression showed brain injury on conventional MRI. Isolated white matter (WM) injury was seen in 28/35 (80%), with focal punctate lesions in 61% (17/28) of the infants. In 7 infants (20%), there was also cortical involvement together with WM injury (one with arterial ischaemic stroke), and in one (3%), there was basal ganglia/thalami involvement alongside WM injury. One infant had an intraventricular haemorrhage (Fig. 1B).

The WM lesion scores in the cooled infants with progression (mean 1.6, SD 0.7) were significantly higher than in the ones without progression (mean 0.5, SD 0.6; p = 0.001). This difference remained significant after the exclusion of the 4 infants who were cooled after 6 h after birth. The same occurred in the non-cooled infants who progressed (mean 1.5, SD 0.8) compared with those who did not (mean 0.8, SD 0.9; p = 0.02). In the non-cooled infants, cortex lesion scores were also significantly higher in the progression group (mean 0.33, SD 0.5 vs. mean 0.08, SD 0.26; p = 0.008). Within the progression group, there was no difference in injury scores between the cooled and non-cooled infants in the deep nuclei, cortex, or WM.

One hundred fifty-three out of 325 infants (47%) without progression showed brain injury on conventional MRI. WM injury was seen in 145 of them, 112 infants (73%) had isolated WM injury whereas 33 (21.6%) showed WM alongside cortex or basal ganglia/thalami injury. Eight infants (5.2%) had basal ganglia/thalami injury alone. WM lesion scores were significantly higher in those infants without progression who were not cooled compared with those cooled (mean 0.88, SD 0.91 vs. mean 0.52, SD 0.66; p = 0.02). No significant difference between cooled and non-cooled infants without progression was found in any of the other MRI or MRS biomarkers (Table 1).

Infants with progression had worse thalamic metabolites on MRS (Table 1). Two non-cooled infants with progression showed an elevated (≥0.22) lactate/NAA ratio, which suggests injury to the deep brain nuclei [10]. The mean thalamic ratios of NAA/choline were significantly higher in cooled than non-cooled infants with progression (p = 0.04). Infants with progression with adverse outcomes had significantly higher mean WM injury scores and thalamic ratios of NAA/choline, lactate/NAA (online suppl. Table 1).

Neurodevelopmental Outcome

Twelve out of 41 (29.2%) infants with progression had disability at 18–24 months. All had mild neurological disability. There was no significant difference in the rate of adverse outcomes between non-cooled and cooled infants with progression [5/12 (42%) vs. 7/29 (24%), respectively, p = 0.26] (Table 1). Any Bayley-III cognitive, motor, or language score <85 was present overall in 21/41 (51%) of the infants with progression.

Discussion

In this multicentre observational study on infants with mild HIE, neonates with progression were shown to have thalamic metabolite perturbation on MRS and more severe WM injury in comparison to the infants without progression. We found an overall rate of disability of 29.2% and any Bayley-III cognitive, motor, or language score <85 was noted in 51% of the infants who progressed to moderate HIE after birth.

There has been only limited research thus far describing the nature of injury in mild HIE infants with progression and the effect of TH on this. Previous studies only focussed on the infants with progression who were not cooled. Recently, Parmentier et al. [11] showed in a cohort of 39 infants with perinatal asphyxia and seizures within 24 h after birth that all of them had brain injury on MRI and over a half had an adverse outcome. However, the aEEG background before seizure onset was either abnormal or unknown in these infants. Therefore, we do not know whether these patients had a more severe brain injury to start with.

We found that the number of abnormalities on neurological examination increased overtime in infants with progression together with a progressive worsening aEEG background before seizure onset in some of them. These results indicate that these infants may have shown a worsening encephalopathy, and they underline the importance of serial neurological examination and continuous aEEG for the identification of these high-risk infants.

Our data show that still a significant percentage of infants with mild HIE go on to develop mild neurological disability and language delay after disease progression despite TH. This incidence is higher than what has been previously reported in mild HIE without progression. In a secondary analysis of the magnetic resonance biomarkers in neonatal encephalopathy (MARBLE) study, adverse neurodevelopmental outcomes were seen in 2/14 (14.3%) of non-cooled infants with mild HIE and in none of those who were cooled [12]. In neonates with moderate and severe HIE treated with TH, seizures have been associated with brain injury and abnormal outcome [13]. However, similar data in mild HIE are lacking. Preclinical evidence suggests that even after a mild hypoxic-ischemic insult there is exacerbation of damage by neonatal seizures, particularly in the hippocampus [14].

Even though we did find improved thalamic metabolites on MRS in the cooled infants with mild HIE who progressed, the neurodevelopmental outcome at 18–24 months was not significantly different between cooled and non-cooled infants. The severity and predominance of the WM injury in these infants who later progressed to moderate HIE may reflect a partial or repeated hypoxic ischaemia. Therefore, the efficacy of TH might be not certain [15].

In agreement with previous evidence [5], a high percentage of infants with mild HIE without progression showed brain injury on MRI. Despite the decreased WM injury in the cooled infants with mild HIE, no other difference was found in any of the other MRI biomarkers. Previous research showed that cooled infants with mild HIE had reduced WM injury and improved thalamic MRS biomarkers. However, the 2-year outcome was similar between cooled and non-cooled infants [12]. Adequately powered trials are required to examine the efficacy of TH in this population.

Our study has strengths and limitations. Firstly, this is one of the largest studies in mild HIE infants who progressed to more severe stages: it also includes both infants who underwent TH and those who did not and with long-term outcome available. As part of their routine care, all the neonates recruited had aEEG monitoring and a standardized neurological examination soon after birth. Secondly, a normal aEEG and a standardized neurological examination soon after birth as inclusion criteria make it unlikely that these infants had an under-recognized encephalopathy instead of a true progression. The main limitation of this study is its retrospective design and the lack of longer neurological outcome data. Given the predominance of WM injury in neonates with progression, implications on later cognitive and behavioural outcomes during childhood are likely.

Conclusion

This study shows in a large cohort of infants with mild HIE that only a small proportion of them may develop seizures and progress to moderate HIE. We found a high rate of brain injury on MRI and adverse outcomes among the infants with progression. These data underline the need to develop early biomarkers to promptly identify the infants who are at higher risk of progression from mild to moderate encephalopathy.

Acknowledgments

The authors would like to acknowledge the Italian Society of Neonatology, Campania Section for the support and network coordination, which made this study possible.

Statement of Ethics

The study protocol was approved by the Ethics Committee of the University of Campania “Luigi Vanvitelli,” which waived the requirement for informed consent with anonymized clinical data (protocol number 416).

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.

Author Contributions

Paolo Montaldo conceived the idea, analysed the data, and wrote the original draft of the manuscript. Simona Puzone, Francesca Galdo, and Umberto Pugliese supported the data acquisition and interpretation. Elisabetta Caredda, Anna Maietta, Serena Ascione, Mario Diplomatico, Sabino Moschella, Ferdinando Spagnuolo, Vincenzina Roma, Massimiliano De Vivo, and Mauro Carpentieri assisted in the data collection in different sites. Lucio Giordano, Laura Travan, Carlo Capristo, Giovanni Chello, and Emanuele Miraglia del Giudice assisted in designing the study and interpretation of the data. Alessandra D'Amico assisted in the MRI data analysis and interpretation. Mario Cirillo designed the study, interpreted the data, and supervised all aspects of the work.

Data Availability Statement

All data generated or analysed during this study are included in this article and its online supplementary material files. Further enquiries can be directed to the corresponding author.

This article is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC). Usage and distribution for commercial purposes requires written permission. Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.