Brain segmentation in patients with perinatal arterial ischemic stroke.

BACKGROUND: Perinatal arterial ischemic stroke (PAIS) is associated with adverse neurological outcomes. Quantification of ischemic lesions and consequent brain development in newborn infants relies on labor-intensive manual assessment of brain tissues and ischemic lesions. Hence, we propose an automatic method utilizing convolutional neural networks (CNNs) to segment brain tissues and ischemic lesions in MRI scans of infants suffering from PAIS. MATERIALS AND METHODS: This single-center retrospective study included 115 patients with PAIS that underwent MRI after the stroke onset (baseline) and after three months (follow-up). Nine baseline and 12 follow-up MRI scans were manually annotated to provide reference segmentations (white matter, gray matter, basal ganglia and thalami, brainstem, ventricles, extra-ventricular cerebrospinal fluid, and cerebellum, and additionally on the baseline scans the ischemic lesions). Two CNNs were trained to perform automatic segmentation on the baseline and follow-up MRIs, respectively. Automatic segmentations were quantitatively evaluated using the Dice coefficient (DC) and the mean surface distance (MSD). Volumetric agreement between segmentations that were manually and automatically obtained was computed. Moreover, the scan quality and automatic segmentations were qualitatively evaluated in a larger set of MRIs without manual annotation by two experts. In addition, the scan quality was qualitatively evaluated in these scans to establish its impact on the automatic segmentation performance. RESULTS: Automatic brain tissue segmentation led to a DC and MSD between 0.78-0.92 and 0.18-1.08 mm for baseline, and between 0.88-0.95 and 0.10-0.58 mm for follow-up scans, respectively. For the ischemic lesions at baseline the DC and MSD were between 0.72-0.86 and 1.23-2.18 mm, respectively. Volumetric measurements indicated limited oversegmentation of the extra-ventricular cerebrospinal fluid in both the follow-up and baseline scans, oversegmentation of the ischemic lesions in the left hemisphere, and undersegmentation of the ischemic lesions in the right hemisphere. In scans without imaging artifacts, brain tissue segmentation was graded as excellent in more than 85% and 91% of cases, respectively for the baseline and follow-up scans. For the ischemic lesions at baseline, this was in 61% of cases. CONCLUSIONS: Automatic segmentation of brain tissue and ischemic lesions in MRI scans of patients with PAIS is feasible. The method may allow evaluation of the brain development and efficacy of treatment in large datasets.

Early predictors of neurodevelopment after perinatal arterial ischemic stroke: a systematic review and meta-analysis.

BACKGROUND AND AIMS: Perinatal arterial ischemic stroke (PAIS) often has lifelong neurodevelopmental consequences. We aimed to review early predictors (<4 months of age) of long-term outcome. METHODS: We carried out a systematic literature search (PubMed and Embase), and included articles describing term-born infants with PAIS that underwent a diagnostic procedure within four months of age, and had any reported outcome parameter ≥12 months of age. Two independent reviewers included studies and performed risk of bias analysis. RESULTS: We included 41 articles reporting on 1395 infants, whereof 1255 (90%) infants underwent follow-up at a median of 4 years. A meta-analysis was performed for the development of cerebral palsy (n = 23 studies); the best predictor was the qualitative or quantitative assessment of the corticospinal tracts on MRI, followed by standardized motor assessments. For long-term cognitive functioning, bedside techniques including (a)EEG and NIRS might be valuable. Injury to the optic radiation on DTI correctly predicted visual field defects. No predictors could be identified for behavior, language, and post-neonatal epilepsy. CONCLUSION: Corticospinal tract assessment on MRI and standardized motor assessments are best to predict cerebral palsy after PAIS. Future research should be focused on improving outcome prediction for non-motor outcomes. IMPACT: We present a systematic review of early predictors for various long-term outcome categories after perinatal arterial ischemic stroke (PAIS), including a meta-analysis for the outcome unilateral spastic cerebral palsy. Corticospinal tract assessment on MRI and standardized motor assessments are best to predict cerebral palsy after PAIS, while bedside techniques such as (a)EEG and NIRS might improve cognitive outcome prediction. Future research should be focused on improving outcome prediction for non-motor outcomes.

Feasibility and safety of intranasally administered mesenchymal stromal cells after perinatal arterial ischaemic stroke in the Netherlands (PASSIoN): a first-in-human, open-label intervention study.

BACKGROUND: Perinatal arterial ischaemic stroke (PAIS) is an important cause of neurodevelopmental disabilities. In this first-in-human study, we aimed to assess the feasibility and safety of intranasally delivered bone marrow-derived allogeneic mesenchymal stromal cells (MSCs) to treat PAIS in neonates. METHODS: In this open-label intervention study in collaboration with all neonatal intensive care units in the Netherlands, we included neonates born at full term (≥36 weeks of gestation) with MRI-confirmed PAIS in the middle cerebral artery region. All eligible patients were transferred to the neonatal intensive care unit of the Wilhelmina Children's Hospital. Neonates received one dose of 45-50 × 106 bone-marrow derived MSCs intranasally within 7 days of presenting signs of PAIS. The primary endpoints were acute and subacute safety outcomes, including vital signs, blood markers, and the occurrence of toxicity, adverse events, and serious adverse events. The occurrence of unexpected cerebral abnormalities by a repeat MRI at 3 months of age was a secondary endpoint. As part of standard clinical follow-up at Wilhelmina Children's Hospital, we assessed corticospinal tract development on MRI and performed motor assessments at 4 months of age. This study is registered with ClinicalTrials.gov, NCT03356821. FINDINGS: Between Feb 11, 2020, and April 29, 2021, ten neonates were enrolled in the study. Intranasal administration of MSCs was well tolerated in all ten neonates. No serious adverse events were observed. One adverse event was seen: a mild transient fever of 38°C without the need for clinical intervention. Blood inflammation markers (C-reactive protein, procalcitonin, and leukocyte count) were not significantly different pre-administration versus post-administration and, although thrombocyte levels increased (p=0·011), all were within the physiological range. Follow-up MRI scans did not show unexpected structural cerebral abnormalities. All ten patients had initial pre-Wallerian changes in the corticospinal tracts, but only four (40%) patients showed asymmetrical corticospinal tracts at follow-up MRI. Abnormal early motor assessment was found in three (30%) infants. INTERPRETATION: This first-in-human study demonstrates that intranasal bone marrow-derived MSC administration in neonates after PAIS is feasible and no serious adverse events were observed in patients followed up until 3 months of age. Future large-scale placebo-controlled studies are needed to determine the therapeutic effect of intranasal MSCs for PAIS. FUNDING: Netherlands Organization for Health Research and Development (ZonMw).