Size and Location of Preterm Brain Injury and Associations With Neurodevelopmental Outcomes.

BACKGROUND AND OBJECTIVES: We examined associations of white matter injury (WMI) and periventricular hemorrhagic infarction (PVHI) volume and location with 18-month neurodevelopment in very preterm infants. METHODS: A total of 254 infants born <32 weeks' gestational age were prospectively recruited across 3 tertiary neonatal intensive care units (NICUs). Infants underwent early-life (median 33.1 weeks) and/or term-equivalent-age (median 41.9 weeks) MRI. WMI and PVHI were manually segmented for quantification in 92 infants. Highest maternal education level was included as a marker of socioeconomic status and was defined as group 1 = primary/secondary school; group 2 = undergraduate degree; and group 3 = postgraduate degree. Eighteen-month neurodevelopmental assessments were completed with Bayley Scales of Infant and Toddler Development, Third Edition. Adverse outcomes were defined as a score of less than 85 points. Multivariable linear regression models were used to examine associations of brain injury (WMI and PVHI) volume with neurodevelopmental outcomes. Voxel-wise lesion symptom maps were developed to assess relationships between brain injury location and neurodevelopmental outcomes. RESULTS: Greater brain injury volume was associated with lower 18-month Motor scores (ß = -5.7, 95% CI -9.2 to -2.2, p = 0.002) while higher maternal education level was significantly associated with higher Cognitive scores (group 3 compared 1: ß = 14.5, 95% CI -2.1 to 26.9, p = 0.03). In voxel-wise lesion symptom maps, brain injury involving the central and parietal white matter was associated with an increased risk of poorer motor outcomes. DISCUSSION: We found that brain injury volume and location were significant predictors of motor, but not cognitive outcomes, suggesting that different pathways may mediate outcomes across domains of neurodevelopment in preterm infants. Specifically, assessing lesion size and location may allow for more accurate identification of infants with brain injury at highest risk of poorer motor outcomes. These data also highlight the importance of socioeconomic status in cognitive outcomes, even in preterm infants with brain injury.

Pain Exposure and Brain Connectivity in Preterm Infants.

Importance: Early-life exposure to painful procedures has been associated with altered brain maturation and neurodevelopmental outcomes in preterm infants, although sex-specific differences are largely unknown. Objective: To examine sex-specific associations among early-life pain exposure, alterations in neonatal structural connectivity, and 18-month neurodevelopment in preterm infants. Design, Setting, and Participants: This prospective cohort study recruited 193 very preterm infants from April 1, 2015, to April 1, 2019, across 2 tertiary neonatal intensive care units in Toronto, Canada. Structural connectivity data were available for 150 infants; neurodevelopmental outcomes were available for 123 infants. Data were analyzed from January 1, 2022, to December 31, 2023. Exposure: Pain was quantified in the initial weeks after birth as the total number of invasive procedures. Main Outcome and Measure: Infants underwent early-life and/or term-equivalent-age magnetic resonance imaging with diffusion tensor imaging to quantify structural connectivity using graph theory measures and regional connection strength. Eighteen-month neurodevelopmental outcomes were assessed with the Bayley Scales of Infant and Toddler Development, Third Edition. Stratifying by sex, generalized estimating equations were used to assess whether pain exposure modified the maturation of structural connectivity using an interaction term (early-life pain exposure × postmenstrual age [PMA] at scan). Generalized estimating equations were used to assess associations between structural connectivity and neurodevelopmental outcomes, adjusting for extreme prematurity and maternal education. Results: A total of 150 infants (80 [53%] male; median [IQR] gestational age at birth, 27.1 [25.4-29.0] weeks) with structural connectivity data were analyzed. Sex-specific associations were found between early-life pain and neonatal brain connectivity in female infants only, with greater early-life pain exposure associated with slower maturation in global efficiency (pain × PMA at scan interaction P = .002) and local efficiency (pain × PMA at scan interaction P = .005). In the full cohort, greater pain exposure was associated with lower global efficiency (coefficient, -0.46; 95% CI, -0.78, to -0.15; P = .004) and local efficiency (coefficient, -0.57; 95% CI, -1.04 to -0.10; P = .02) and regional connection strength. Local efficiency (coefficient, 0.003; 95% CI, 0.001-0.004; P = .005) and regional connection strength in the striatum were associated with cognitive outcomes. Conclusions and Relevance: In this cohort study of very preterm infants, greater exposure to early-life pain was associated with altered maturation of neonatal structural connectivity, particularly in female infants. Alterations in structural connectivity were associated with neurodevelopmental outcomes, with potential regional specificities.

Automatic Prediction of Paediatric Cardiac Output From Echocardiograms Using Deep Learning Models.

Background: Cardiac output (CO) perturbations are common and cause significant morbidity and mortality. Accurate CO assessment is crucial for guiding treatment in anaesthesia and critical care, but measurement is difficult, even for experts. Artificial intelligence methods show promise as alternatives for accurate, rapid CO assessment. Methods: We reviewed paediatric echocardiograms with normal CO and a dilated cardiomyopathy patient group with reduced CO. Experts measured the left ventricular outflow tract diameter, velocity time integral, CO, and cardiac index (CI). EchoNet-Dynamic is a deep learning model for estimation of ejection fraction in adults. We modified this model to predict the left ventricular outflow tract diameter and retrained it on paediatric data. We developed a novel deep learning approach for velocity time integral estimation. The combined models enable automatic prediction of CO. We evaluated the models against expert measurements. Primary outcomes were root-mean-squared error, mean absolute error, mean average percentage error, and coefficient of determination (R2). Results: In a test set unused during training, CI was estimated with the root-mean-squared error of 0.389 L/min/m2, mean absolute error of 0.321 L/min/m2, mean average percentage error of 10.8%, and R2 of 0.755. The Bland-Altman analysis showed that the models estimated CI with a bias of +0.14 L/min/m2 and 95% limits of agreement -0.58 to 0.86 L/min/m2. Conclusions: Our model estimated CO with strong correlation to ground truth and a bias of 0.17 L/min, better than many CO measurements in paediatrics. Model pretraining enabled accurate estimation despite a small dataset. Potential uses include supporting clinicians in real-time bedside calculation of CO, identification of low-CO states, and treatment responses.

Contexte: Les perturbations du débit cardiaque sont fréquentes et associées à des taux élevés de morbidité et de mortalité. Une évaluation juste du débit cardiaque est essentielle pour orienter le choix du traitement anesthésique et des soins critiques. Or, il est difficile de mesurer le débit cardiaque, même pour les experts. Les méthodes fondées sur l'intelligence artificielle semblent toutefois prometteuses pour évaluer le débit cardiaque avec exactitude et rapidité. Méthodologie: Nous avons analysé des échocardiogrammes pédiatriques chez des personnes dont le débit cardiaque est normal ainsi que chez des patients qui étaient atteints d'une cardiomyopathie dilatée et dont le débit cardiaque était réduit. Des experts ont mesuré le diamètre de la voie d'éjection ventriculaire gauche, l'intégrale de la vitesse par rapport au temps (IVT), le débit cardiaque et l'index cardiaque. L'outil EchoNet-Dynamic est un modèle d'apprentissage profond qui donne une estimation de la fraction d'éjection chez les adultes. Nous avons modifié ce modèle afin qu'il puisse prédire le diamètre de la voie d'éjection ventriculaire gauche et l'avons entraîné à l'aide de données pédiatriques. Nous avons également mis au point une nouvelle approche d'apprentissage profond pour l'estimation des valeurs d'IVT. La combinaison de ces modèles a permis de prédire de façon automatique le débit cardiaque, et nous avons évalué les résultats obtenus par rapport à ceux des experts. Les principaux critères d'évaluation étaient l'erreur moyenne quadratique (EMQ), l'erreur moyenne absolue (EMA), le pourcentage d'erreur moyen (PEM) ainsi que le coefficient de détermination (R2). Résultats: Dans un ensemble d'essais n'ayant pas été utilisé au cours de l'entraînement du modèle, l'index cardiaque a été estimé avec une EMQ de 0,389 L/min/m2, une EMA de 0,321 L/min/m2, un PEM de 10,8 % et un R2 de 0,755. Selon l'analyse de Bland-Altman, le biais pour les estimations de l'index cardiaque était de + 0,14 L/min/m2, et les limites de concordance à 95 % étaient de ­0,58 à 0,86 L/min/m2. Conclusions: Les estimations générées par le modèle pour le débit cardiaque montraient une forte corrélation avec les valeurs de référence et un biais à 0,17 L/min, ce qui est mieux que bien des mesures du débit cardiaque utilisées en pédiatrie. Malgré un petit ensemble de données, le modèle entraîné a permis de produire une estimation juste. Les utilisations potentielles comprennent l'aide aux cliniciens dans le calcul du débit cardiaque en temps réel et au chevet du patient, le dépistage d'un faible débit cardiaque et l'évaluation de la réponse au traitement.

Retraining an Artificial Intelligence Algorithm to Calculate Left Ventricular Ejection Fraction in Pediatrics.

OBJECTIVES: Identifying patients with low left ventricular ejection fraction (LVEF) and monitoring LVEF responses to treatment are important clinical goals. Can a deep-learning algorithm predict pediatric LVEF within clinically acceptable error? DESIGN: The study authors wanted to fine-tune an adult deep-learning algorithm to calculate LVEF in pediatric patients. A priori, their objective was to refine the algorithm to perform LVEF calculation with a mean absolute error (MAE) ≤5%. SETTING: A quaternary pediatric hospital PARTICIPANTS: A convenience sample (n = 321) of echocardiograms from newborns to 18 years old with normal cardiac anatomy or hemodynamically insignificant anomalies. Echocardiograms were chosen from a group of healthy controls with known normal LVEF (n = 267) and a dilated cardiomyopathy patient group with reduced LVEF (n = 54). INTERVENTIONS: The artificial intelligence model EchoNet-Dynamic was tested on this data set and then retrained, tested, and further validated to improve LVEF calculation. The gold standard value was LVEF calculated by clinical experts. MEASUREMENTS AND MAIN RESULTS: In a random subset of subjects (n = 40) not analyzed prior to selection of the final model, EchoNet-Dynamic calculated LVEF with a MAE of 8.39%, R2 = 0.47 without, and MAE 4.47%, R2 = 0.87 with fine-tuning. Bland-Altman analysis suggested that the model slightly underestimates LVEF (bias = -2.42%). The 95% limits of agreement between actual and calculated values were -12.32% to 7.47%. CONCLUSIONS: The fine-tuned model calculates LVEF in a range of pediatric patients within clinically acceptable error. Potential advantages include reducing operator error in LVEF calculation and supporting independent LVEF assessment by inexperienced users.

Location and Size of Preterm Cerebellar Hemorrhage and Childhood Development.

OBJECTIVE: To examine the association between cerebellar hemorrhage (CBH) size and location and preschool-age neurodevelopment in very preterm neonates. METHODS: Preterm magnetic resonance images of 221 very preterm neonates (median gestational age = 27.9 weeks) were manually segmented for CBH quantification and location. Neurodevelopmental assessments at chronological age 4.5 years included motor (Movement Assessment Battery for Children, 2nd Edition [MABC-2]), visuomotor integration (Beery-Buktenica Developmental Test of Visual-Motor Integration, 6th Edition), cognitive (Wechsler Primary and Preschool Scale of Intelligence, 3rd Edition), and behavioral (Child Behavior Checklist) outcomes. Multivariable linear regression models examined the association between CBH size and 4.5-year outcomes accounting for sex, gestational age, and supratentorial injury. Probabilistic maps assessed CBH location and likelihood of a lesion to predict adverse outcome. RESULTS: Thirty-six neonates had CBH: 14 (6%) with only punctate CBH and 22 (10%) with ≥1 larger CBH. CBH occurred mostly in the inferior aspect of the posterior lobes. CBH total volume was independently associated with MABC-2 motor scores at 4.5 years (ß = -0.095, 95% confidence interval = -0.184 to -0.005), with a standardized ß coefficient (-0.16) that was similar to that of white matter injury volume (standardized ß = -0.22). CBH size was similarly associated with visuomotor integration and externalizing behavior but not cognition. Voxelwise odds ratio and lesion-symptom maps demonstrated that CBH extending more deeply into the cerebellum predicted adverse motor, visuomotor, and behavioral outcomes. INTERPRETATION: CBH size and location on preterm magnetic resonance imaging were associated with reduced preschool motor and visuomotor function and more externalizing behavior independent of supratentorial brain injury in a dose-dependent fashion. The volumetric quantification and localization of CBH, even when punctate, may allow opportunity to improve motor and behavioral outcomes by providing targeted intervention. ANN NEUROL 2020;88:1095-1108.

Mechanical Ventilation Duration, Brainstem Development, and Neurodevelopment in Children Born Preterm: A Prospective Cohort Study.

OBJECTIVES: To determine, in children born preterm, the association of mechanical ventilation duration with brainstem development, white matter maturation, and neurodevelopmental outcomes at preschool age. STUDY DESIGN: This prospective cohort study included 144 neonates born at <30 weeks of gestation (75 male, mean gestational age 27.1 weeks, SD 1.6) with regional brainstem volumes automatically segmented on magnetic resonance imaging at term-equivalent age (TEA). The white matter maturation was assessed by diffusion tensor imaging and tract-based spatial statistics. Neurodevelopmental outcomes were assessed at 4.5 years of age using the Movement Assessment Battery for Children, 2nd Edition, and the Wechsler Primary and Preschool Scale of Intelligence, 4th Edition, full-scale IQ. The association between the duration of mechanical ventilation and brainstem development was validated in an independent cohort of children born very preterm. RESULTS: Each additional day of mechanical ventilation predicted lower motor scores (0.5-point decrease in the Movement Assessment Battery for Children, 2nd Edition, score by day of mechanical ventilation, 95% CI -0.6 to -0.3, P < .0001). Prolonged exposure to mechanical ventilation was associated with smaller pons and medulla volumes at TEA in 2 independent cohorts, along with widespread abnormalities in white matter maturation. Pons and medulla volumes at TEA predicted motor outcomes at 4.5 years of age. CONCLUSIONS: In neonates born very preterm, prolonged mechanical ventilation is associated with impaired brainstem development, abnormal white matter maturation, and lower motor scores at preschool age. Further research is needed to better understand the neural pathological mechanisms involved.