Bedside Neuromonitoring in Pediatric ECMO Patients Using High-Density Diffuse Optical Tomography
Poster No:
2612
Submission Type:
Abstract Submission
Authors:
Sophia McMorrow1, Tessa George1, Chloe Sobolewski1, Dalin Yang1, Sung Min Park1, Kelsey King2, Ahmed Said1, Adam Eggebrecht1
Institutions:
1Washington University School of Medicine in St. Louis, St. Louis, MO, 2Roosevelt University, St. Louis, MO
First Author:
Co-Author(s):
Introduction:
Extracorporeal membrane oxygenation (ECMO) is a form of life support for patients with severe, reversible respiratory or cardiac failure. ECMO patients are susceptible to neurological complications including seizures, hemorrhage, and infarction, resulting in an increased risk of neurodevelopmental, behavioral, and functional deficits1,2. Conventional neuroimaging techniques suffer limited sensitivity to detect early signs of brain injury or pose additional risks such as transportation to imaging facilities3. Thus, there is a clear and present need for continuous bedside neuromonitoring with high sensitivity and specificity. Herein, we establish the feasibility of high-density diffuse optical tomography (HD-DOT) for continuous neuromonitoring in pediatric ECMO patients with image quality comparable to functional magnetic resonance imaging4. Specifically, we assess the feasibility of HD-DOT for bedside assessment of high bandwidth cerebral oxygenation and functional connectivity.
Methods:
We used an HD-DOT system optimized for pediatric patients with 80 dual-wavelength sources (685nm & 830nm) and 78 avalanche-photodiode detectors that together support over 1200 usable measurements per wavelength. Data were collected from four patients (age 2 weeks, 9 months, 28 months, and 16 years), with multi-day longitudinal data in three patients. Data collection periods were 1-3 hours. Raw data quality metrics, including mean light levels and temporal variance in HD-DOT measurements, were analyzed using NeuroDOT data processing pipelines5 to identify high-quality4,6, low-motion7 data for further analysis. Patient-specific facial landmarks and an age-appropriate anatomical atlas were used to model light propagation and depth sensitivity of the HD-DOT array (Fig. 1A-1D). Unfiltered 10 Hz HD-DOT measurements were converted to relative hemoglobin concentrations and aligned with 0.2 Hz physiology measurements for analysis of high bandwidth cerebral oxygenation (Fig. 1). The HD-DOT array field of view was intersected with the Gordon parcellation8, and HD-DOT data was bandpass filtered at low bandwidths (0.009-0.08 Hz) for analysis of low frequency temporal correlation (functional connectivity; Fig. 2).
Results:
We present example results from one neonatal patient on veno-arterial ECMO that included a clamp trial. Clamp trials are pre-planned periods of temporary separation from ECMO support to test the patient's ability to come off ECMO. HD-DOT measurements sampling superficial tissue reveal notable changes in concentrations of deoxygenated (HbR) and oxygenated hemoglobin (HbO) and their difference (HbD) in response to clamp trial events (Fig. 1E). Concurrently recorded physiology data changed similarly (Fig. 1F). After superficial signal regression, relative hemoglobin concentrations in cerebral tissue reveal temporal variation distinct from the response in superficial signals in response to clamp trial events. (Fig. 1G). Analysis of low frequency temporal correlation relative to parcels in the motor and visual cortex reveal consistent bilateral correlation during baseline periods and disrupted correlation during the clamp trial (Fig. 2E). These changes were quantified via test-retest analyses relative to an initial baseline period, which reveal consistent spatial patterns during quiescence and strong disruptions during the clamp trial (Fig. 2F).
Conclusions:
We establish feasibility of HD-DOT for bedside neuromonitoring for pediatric ECMO patients, a valuable step towards understanding brain health during ECMO support. Future analyses will investigate spatial-temporal variation in cortical hemoglobin changes and the relationship between cerebral hemodynamics and physiological parameters and outcomes. Overall, these investigations hold potential for illuminating associations between brain health in pediatric patients on ECMO and outcomes.
Lifespan Development:
Early life, Adolescence, Aging
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Anatomy and Functional Systems
Novel Imaging Acquisition Methods:
NIRS 2
Imaging Methods Other
Physiology, Metabolism and Neurotransmission :
Cerebral Metabolism and Hemodynamics 1
Keywords:
Blood
Congenital
Cortex
Development
Neurological
Optical Imaging Systems (OIS)
PEDIATRIC
Pediatric Disorders
1|2Indicates the priority used for review
Provide references using author date format
Ferradal, S. L. et al. (2016). "Functional Imaging of the Developing Brain at the Bedside Using Diffuse Optical Tomography." Cerebral Cortex vol. 26, no. 4, pp. 1558-1568.
Gordon, E. M. et al. (2016). "Generation and Evaluation of a Cortical Area Parcellation from Resting-State Correlations." Cerebral Cortex vol. 26, no. 1, pp. 288-303.
https://www.nitrc.org/projects/neurodot
Lin, N. et al. (2018). "Neuromonitoring in the neonatal ECMO patient." Seminars in Perinatology 42(2): 111-121.
Sadhwani, A. et al. (2019). "Early Neurodevelopmental Outcomes in Children Supported with ECMO for Cardiac Indications." Pediatric Cardiology vol. 40, no. 5, pp. 1072-1083.
Said, A. S. et al. (2020). "Neurological Monitoring and Complications of Pediatric Extracorporeal Membrane Oxygenation Support." Pediatric Neurology vol. 108, 31-39.
Sherafati, A. et al. (2020). "Global motion detection and censoring in high-density diffuse optical tomography." Hum Brain Mapping vol. 41, no, 14, pp. 4093-4112.