Interoperable and explainable machine learning models to predict morbidity and mortality in acute neurological injury in the pediatric intensive care unit: secondary analysis of the TOPICC study.

Background: Acute neurological injury is a leading cause of permanent disability and death in the pediatric intensive care unit (PICU). No predictive model has been validated for critically ill children with acute neurological injury. Objectives: We hypothesized that PICU patients with concern for acute neurological injury are at higher risk for morbidity and mortality, and advanced analytics would derive robust, explainable subgroup models. Methods: We performed a secondary subgroup analysis of the Trichotomous Outcomes in Pediatric Critical Care (TOPICC) study (2011-2013), predicting mortality and morbidity from admission physiology (lab values and vital signs in 6 h surrounding admission). We analyzed patients with suspected acute neurological injury using standard machine learning algorithms. Feature importance was analyzed using SHapley Additive exPlanations (SHAP). We created a Fast Healthcare Interoperability Resources (FHIR) application to demonstrate potential for interoperability using pragmatic data. Results: 1,860 patients had suspected acute neurological injury at PICU admission, with higher morbidity (8.2 vs. 3.4%) and mortality (6.2 vs. 1.9%) than those without similar concern. The ensemble regressor (containing Random Forest, Gradient Boosting, and Support Vector Machine learners) produced the best model, with Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.91 [95% CI (0.88, 0.94)] and Average Precision (AP) of 0.59 [0.51, 0.69] for mortality, and decreased performance predicting simultaneous mortality and morbidity (0.83 [0.80, 0.86] and 0.59 [0.51, 0.64]); at a set specificity of 0.995, positive predictive value (PPV) was 0.79 for mortality, and 0.88 for mortality and morbidity. By comparison, for mortality, the TOPICC logistic regression had AUROC of 0.90 [0.84, 0.93], but substantially inferior AP of 0.49 [0.35, 0.56] and PPV of 0.60 at specificity 0.995. Feature importance analysis showed that pupillary non-reactivity, Glasgow Coma Scale, and temperature were the most contributory vital signs, and acidosis and coagulopathy the most important laboratory values. The FHIR application provided a simulated demonstration of real-time health record query and model deployment. Conclusions: PICU patients with suspected acute neurological injury have higher mortality and morbidity. Our machine learning approach independently identified previously-known causes of secondary brain injury. Advanced modeling achieves improved positive predictive value in this important population compared to published models, providing a stepping stone in the path to deploying explainable models as interoperable bedside decision-support tools.

Limbic pathway vulnerability associates with neurologic outcome in children after cardiac arrest.

AIM: To analyze whether brain connectivity sequences including diffusion tensor imaging (DTI) and resting state functional magnetic resonance imaging (rsfMRI) identify vulnerable brain regions and networks associated with neurologic outcome after pediatric cardiac arrest. METHODS: Children aged 2 d-17 y with cardiac arrest were enrolled in one of 2 parent studies at a single center. Clinically indicated brain MRI with DTI and rsfMRI and performed within 2 weeks after arrest were analyzed. Tract-wise fractional anisotropy (FA) and axial, radial, and mean diffusivity assessed DTI, and functional connectivity strength (FCS) assessed rsfMRI between outcome groups. Unfavorable neurologic outcome was defined as Pediatric Cerebral Performance Category score 4-6 or change > 1 between 6 months after arrest vs baseline. RESULTS: Among children with DTI (n = 28), 57% had unfavorable outcome. Mean, radial, axial diffusivity and FA of varying direction of magnitude in the limbic tracts, including the right cingulum parolfactory, left cingulum parahippocampal, corpus callosum forceps major, and corpus callosum forceps minor tracts, were associated with unfavorable neurologic outcome (p < 0.05). Among children with rsfMRI (n = 12), 67% had unfavorable outcome. Decreased FCS in the ventromedial and dorsolateral prefrontal cortex, insula, precentral gyrus, anterior cingulate, and inferior parietal lobule were correlated regionally with unfavorable neurologic outcome (p < 0.05 Family-Wise Error corrected). CONCLUSION: Decreased multimodal connectivity measures of paralimbic tracts were associated with unfavorable neurologic outcome after pediatric cardiac arrest. Longitudinal analysis correlating brain connectivity sequences with long term neuropsychological outcomes to identify the impact of pediatric cardiac arrest in vulnerable brain networks over time appears warranted.

Quantitative Electroencephalography (EEG) Predicting Acute Neurologic Deterioration in the Pediatric Intensive Care Unit: A Case Series.

Introduction: Continuous neurologic assessment in the pediatric intensive care unit is challenging. Current electroencephalography (EEG) guidelines support monitoring status epilepticus, vasospasm detection, and cardiac arrest prognostication, but the scope of brain dysfunction in critically ill patients is larger. We explore quantitative EEG in pediatric intensive care unit patients with neurologic emergencies to identify quantitative EEG changes preceding clinical detection. Methods: From 2017 to 2020, we identified pediatric intensive care unit patients at a single quaternary children's hospital with EEG recording near or during acute neurologic deterioration. Quantitative EEG analysis was performed using Persyst P14 (Persyst Development Corporation). Included features were fast Fourier transform, asymmetry, and rhythmicity spectrograms, "from-baseline" patient-specific versions of the above features, and quantitative suppression ratio. Timing of quantitative EEG changes was determined by expert review and prespecified quantitative EEG alert thresholds. Clinical detection of neurologic deterioration was defined pre hoc and determined through electronic medical record documentation of examination change or intervention. Results: Ten patients were identified, age 23 months to 27 years, and 50% were female. Of 10 patients, 6 died, 1 had new morbidity, and 3 had good recovery; the most common cause of death was cerebral edema and herniation. The fastest changes were on "from-baseline" fast Fourier transform spectrograms, whereas persistent changes on asymmetry spectrograms and suppression ratio were most associated with morbidity and mortality. Median time from first quantitative EEG change to clinical detection was 332 minutes (interquartile range: 201-456 minutes). Conclusion: Quantitative EEG is potentially useful in earlier detection of neurologic deterioration in critically ill pediatric intensive care unit patients. Further work is required to quantify the predictive value, measure improvement in outcome, and automate the process.

New Cluster of Acute Flaccid Myelitis in Western Pennsylvania.

Acute flaccid myelitis is a debilitating illness characterized by acute onset of limb weakness, with one or more spinal segments displaying magnetic resonance imaging-confirmed gray matter lesions. Since the first outbreak in 2014, tracking by the Centers for Disease Control and Prevention has demonstrated biennial epidemics in the United States, with a current outbreak occurring in 2018. The cases of 3 children with acute flaccid myelitis who were initially thought to have common nonneurologic diagnoses are presented. Emergency physicians need to be vigilant to recognize the subtleties of acute flaccid myelitis because the illness progression is rapid and therapy is nuanced.

Pericarditis as the Presenting Feature of Graves Disease in a Pediatric Patient.

Pericarditis is a rare presentation of thyrotoxicosis associated with Graves disease. This association has not been previously described in the pediatric literature. We report a 17-year-old male patient who presented with chest pain, dyspnea, and tachycardia. He was found to have diffuse ST-segment elevation consistent with pericarditis. At presentation, he was noted to have bilateral proptosis. Abnormal thyroid function studies and an elevated thyroid-stimulating hormone receptor antibody level confirmed the diagnosis of Graves thyrotoxicosis. The patient was treated with anti-inflammatory and antithyroid agents and improved in time. We discuss previously reported cases of Graves disease-associated pericarditis in adults, potential etiologies, and management strategies.

Synchronized Current Oscillations of Formic Acid Electro-oxidation in a Microchip-based Dual-Electrode Flow Cell.

We investigate the oscillatory electro-oxidation of formic acid on platinum in a microchip-based dual-electrode cell with microfluidic flow control. The main dynamical features of current oscillations on single Pt electrode that had been observed in macro-cells are reproduced in the microfabricated electrochemical cell. In dual-electrode configuration nearly in-phase synchronized current oscillations occur when the reference/counter electrodes are placed far away from the microelectrodes. The synchronization disappears with close reference/counter electrode placements. We show that the cause for synchronization is weak albeit important, bidirectional electrical coupling between the electrodes; therefore the unidirectional mass transfer interactions are negligible. The experimental design enables the investigation of the dynamical behavior in micro-electrode arrays with well-defined control of flow of the electrolyte in a manner where the size and spacing of the electrodes can be easily varied.