Send to

Choose Destination
J Neurosurg. 2016 Sep;125(3):648-60. doi: 10.3171/2015.6.JNS15972. Epub 2015 Dec 18.

Brain magnetic resonance imaging CO2 stress testing in adolescent postconcussion syndrome.

Author information

Departments of 1 Anesthesia and Perioperative Medicine.
Health Sciences Centre.
University of Manitoba, Winnipeg, Manitoba; and.
Pediatrics and Child Health, and.
Section of Neurosurgery.
Pan Am Concussion Program.
Children's Hospital Research Institute of Manitoba.
Diagnostic Imaging.
Community Health Sciences.
Departments of 11 Anesthesia.
Physiology, and.
University of Toronto, Ontario, Canada.
Medical Imaging.


OBJECT A neuroimaging assessment tool to visualize global and regional impairments in cerebral blood flow (CBF) and cerebrovascular responsiveness in individual patients with concussion remains elusive. Here the authors summarize the safety, feasibility, and results of brain CO2 stress testing in adolescents with postconcussion syndrome (PCS) and healthy controls. METHODS This study was approved by the Biomedical Research Ethics Board at the University of Manitoba. Fifteen adolescents with PCS and 17 healthy control subjects underwent anatomical MRI, pseudo-continuous arterial spin labeling MRI, and brain stress testing using controlled CO2 challenge and blood oxygen level-dependent (BOLD) MRI. Post hoc processing was performed using statistical parametric mapping to determine voxel-by-voxel regional resting CBF and cerebrovascular responsiveness of the brain to the CO2 stimulus (increase in BOLD signal) or the inverse (decrease in BOLD signal). Receiver operating characteristic (ROC) curves were generated to compare voxel counts categorized by control (0) or PCS (1). RESULTS Studies were well tolerated without any serious adverse events. Anatomical MRI was normal in all study participants. No differences in CO2 stimuli were seen between the 2 participant groups. No group differences in global mean CBF were detected between PCS patients and healthy controls. Patient-specific differences in mean regional CBF and CO2 BOLD responsiveness were observed in all PCS patients. The ROC curve analysis for brain regions manifesting a voxel response greater than and less than the control atlas (that is, abnormal voxel counts) produced an area under the curve of 0.87 (p < 0.0001) and 0.80 (p = 0.0003), respectively, consistent with a clinically useful predictive model. CONCLUSIONS Adolescent PCS is associated with patient-specific abnormalities in regional mean CBF and BOLD cerebrovascular responsiveness that occur in the setting of normal global resting CBF. Future prospective studies are warranted to examine the utility of brain MRI CO2 stress testing in the longitudinal assessment of acute sports-related concussion and PCS.


ASL = arterial spin labeling; AUC = area under the curve; BOLD = blood oxygen level–dependent; CBF = cerebral blood flow; CVR = cerebrovascular reactivity; DTI = diffusion tensor imaging; ET = end-tidal; GRE = gradient recalled echo planar; MNI = Montreal Neurological Institute; MPET = model-based prospective end-tidal; MPRAGE = magnetization-prepared rapid gradient-echo; PCD = postconcussion disorder; PCS = postconcussion syndrome; PCSS = Post-Concussion Symptom Scale; PaCO2 = arterial partial pressure of carbon dioxide; ROC = receiver operating characteristic; SPM = statistical parametric mapping; SRC = sports-related concussion; TBI = traumatic brain injury; adolescent; blood oxygen level–dependent imaging; dmIC = dorsal mid–insular cortex; fMRI = functional MRI; magnetic resonance imaging; pCASL= pseudo-continuous arterial spin labeling; postconcussion syndrome; sports-related concussion; trauma

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Sheridan PubFactory
Loading ...
Support Center