Journal Article Epublish
PubMed: 40150711 DOI: 10.3390/bioengineering12030247
Commercial NIRS May Not Detect Hemispheric Regional Disparity in Continuously Measured COx/COx-a: An Exploratory Healthy and Cranial Trauma Time-Series Analysis.
Bioengineering (Basel) 12(3)
Authors: Amanjyot Singh Sainbhi, Logan Froese, Kevin Y Stein, Nuray Vakitbilir, Alwyn Gomez, Abrar Islam, Tobias Bergmann, Noah Silvaggio, Mansoor Hayat, Frederick A Zeiler
Affiliations
- Department of Biomedical Engineering, Price Faculty of Engineering, University of Manitoba, Winnipeg, MB R3T 5V6, Canada.
- Department of Clinical Neurosciences, Karolinska Institutet, 171 77 Stockholm, Sweden.
- Undergraduate Medicine, Rady Faculty of Health Sciences, University of Manitoba, Winnipeg, MB R3E 3P5, Canada.
- Section of Neurosurgery, Department of Surgery, Rady Faculty of Health Sciences, University of Manitoba, Winnipeg, MB R3A 1R9, Canada.
- Department of Human Anatomy and Cell Science, Rady Faculty of Health Sciences, University of Manitoba, Winnipeg, MB R3E 0J9, Canada.
- Pan Am Clinic Foundation, Winnipeg, MB R3M 3E4, Canada.
- Division of Anaesthesia, Department of Medicine, Addenbrooke's Hospital, University of Cambridge, Cambridge CB2 0QQ, UK.
Abstract
Continuous metrics of cerebral autoregulation (CA) assessment have been developed using various multimodal cerebral physiological monitoring devices. However, CA regional disparity remains unclear in states of health and disease. Leveraging existing archived data sources, we preliminarily evaluated regional hemispheric disparity in CA using the near infrared spectroscopy (NIRS)-derived cerebral oximetry index (COx/COx-a). Along with bilateral NIRS, regional cerebral oxygen saturation, arterial blood pressure, cerebral perfusion pressure, and bilateral COx/COx-a were derived using three different temporal resolutions-10 s, 1 min, and 5 min-based on non-overlapping mean values. The regional disparity between hemispheres was evaluated based on median and median absolute deviation. Further, patient-level autoregressive integrative moving average models were calculated for each signal stream and used to generate personalized vector autoregressive models. Multi-variate cerebral physiologic relationships between hemispheres were assessed via impulse response functions and Granger causality analyses. Data from 102 healthy control volunteers, 27 spinal surgery patients, and 95 TBI patients (varying in frontal lobe pathology impacting the optode path; 64 without bifrontal lobe pathology, 15 without left frontal lobe pathology, 11 without right frontal lobe pathology, and 5 with bifrontal lobe pathology) were retrospectively analyzed. For subjects with or without cranial pathology, no difference in COx/COx-a was found between hemispheres regardless of the analytic method. In TBI patients without pathology underneath the NIRS sensor, distant parenchymal injury does not seem to have an effect on the CA of uninjured frontal lobes. Further work is required to characterize regional disparities with multi-channel CA measurements in healthy and disease states.
MeSH
Frontal Lobe, Pathology, Multiple Chronic Conditions, Wounds And Injuries