Table 3

Studies assessing risk of prolonged deficits following head trauma

PaperStudy typeSubject characteristicsInclusion criteriaOutcome measuresFindings/results
A. Concussion and prolonged deficits
Chen et al104Prospective cohort9 Concussions, 6 healthy controlsPersistent symptoms following concussionSymptom reporting and fMRI (working memory task)Significantly reduced task-related BOLD changes in the prefrontal cortex in athletes with prolonged symptoms following concussion. Activation patterns improved as symptoms improved
Vagnozzi et al105Prospective (multicentre) cohort40 Concussions (assessed at day 3, 15, 22 and 30 days postinjury), 30 healthy controlsRecent concussion
Age 16–35
Symptom reporting and MRS
Used single voxel (ROI: right frontal lobe), and chemical shift techniques to analyse data
Self-reported symptoms recovered within 3–15 days
Significant differences between concussed and control groups were observed in metabolite ratios at day 3 postinjury. Metabolite changes gradually recovered to control levels within 30 days of injury
Cubon et al106Cross-sectional10 Collegiate students with prolonged symptoms (9 injured during sport 1 injured in a fall), 10 healthy controls, 5 moderate-to-severe TBI patientsPersistent symptoms 1 month postinjuryDTI (MD and FA analysed using tract-based spatial statistics)Significant increase in MD in concussed compared to controls. Similar results were observed in the moderate but not severe TBI patients when compared to controls
Gosselin et al107Cross-sectional14 Patients with mild TBI (injuries sustained in recreational activity, work and motor vehicle accidents), 23 controlsPersistent symptoms, mild TBI, (recruited from 2 tertiary trauma centres)Postconcussion symptom scale, Beck depression inventory, ERP, fMRI (working memory task)Attenuated BOLD signal changes were found in the mild TBI group compared with the control group.
Symptom severity and BOLD signal changes were correlated (where patients with more severe symptoms had lower BOLD signal changes)
Mild TBI subjects also had reduced N350 amplitude for the working memory task compared to controls
Henry et al108Prospective cohortCollege athletes
10 concussions (scanned at 5 days and 6 months post injury)
10 controls (scanned at time 0 and 18 months)
Recent concussionSymptom reporting and MRS (ROI: prefrontal and primary motor cortex)Neurometabolic differences between concussed and control were observed in the acute phase (lower N-acetylaspartate : creatine levels in the prefrontal cortex and lower glutamate  :creatine levels in the motor cortex) as well as the delayed phase (increase in the myoinositol levels in the motor cortex)
Slobounov et al109Cross-sectionalCollege athletes
17 concussions (scanned day 10 postinjury)
17 controls
Recent concussion, clinically recoveredSymptom checklist, NP testing, rsFMRIAll athletes had recovered clinically within 10 days
rsFMRI revealed disrupted functional network both at rest and in response to a graded physical test (reduced interhemispheric connectivity in the primary visual cortex, hippocampus and dorsolateral prefrontal cortex)
Baillargeon et al110Cross-sectional48 Concussions
48 Controls
(9–12 years (n=32), 13–16 years (n=34), adults (n=30))
Concussion assessed >6 months postinjuryPostconcussion symptom scale, NP tests, EEG (visual 3-stimulus oddball paradigm)Concussed athletes had lower P3b amplitudes than the control athletes
Adolescent athletes showed persistent deficits in working memory
Johnson et al111Cross-sectionalCollegiate athletes
14 Concussions
15 Controls
9 Additional concussions
Recent concussion, recovered clinicallyrsFMRI (measured the default mode network)Significant default mode network connectivity differences were observed between concussed and control groups
Regression analysis revealed a significant reduction in magnitude of connection between various structures in the brain as a function of the number of concussions
Tallus et al112Cross-sectional19 Individuals with mild TBI (No data provided on the mechanism of mild TBI)—11 persistent symptoms, 8 clinically recovered), 9 healthy controlsInjury sustained 5 years earlier, GCS 13–15 on admission, normal MRIUsed navigated transcranial magnetic stimulation and electromyography to measure MTMT was higher in some (but not all) mild TBI subjects than in controls. Changes were observed even in subjects who had recovered their symptoms
Authors suggested that subtle prolonged changes may exist in some patients following mTBI and that in a proportion of these patients the changes may be ‘compensated’
B. Repeated head impact (subconcussive blows) and prolonged deficits
Bazarian et al113Prospective cohort9 High school athletes, 6 controls1 Concussion, others suffered between 26 and 399 subconcussive blowsDTI at baseline and postseasonPercentage of voxels with significant prepost FA or MD changes was highest for the concussion subject, intermediary for those with subconcussive head blows and lowest for controls
Breedlove et al114Prospective cohort (2 seasons)24 Male high school football players followed in season 1, 28 players followed in season 2 (including 14 from previous season)All players monitored, concussions diagnosed by clinicianHelmet telemetry system, computerised NP testing, fMRI Divided groups into concussed, not concussed and not concussed but demonstrable cognitive or functional deficit. Found ‘substantial’ portion of the cohort without concussion demonstrated neurophysiological changes on fMRIThere was also a relationship between number of head impacts and changes on NP testing
Gysland et al115Prospective cohort46 Male collegiate football playersNo concussions in the 3 months before baseline testingComputerised NP test battery, SAC, SOT, BESS and graded symptom checklist, self-reported concussion history and Head impact telemetryChanges in symptoms, balance function or NP function were independent of prior concussion history and total, number magnitude and location of impacts over 1 season
Neselius et al116Prospective cohort30 Boxers and 25 non-boxing controlsHigh-level amateur boxers (national or international level) with a minimum of 45 boutsCSF samples collected by lumbar puncture (1–6 days after a bout and after at least 14 days of rest)Increase in biomarkers for acute brain injury in the boxing group (NFL, GFAB, T-Tau and SB-100) compared to controls. NFL and T-Tau concentrations remained elevated even after a period of rest
  • BESS, balance error scoring system; BOLD, blood oxygenation level-dependent; CSF, cerebrospinal fluid; DTI, Diffusion tensor imaging; ERP, event-related brain potential; FA, fractional anisotropy; fMRI, functional magnetic resonance imaging; EEG, electroencephalogram; GCS, Glasgow Coma Score; GFAB, glial fibrillary acidic protein; MD, mean diffusivity; MRS, MR spectroscopy; MT, motor threshold; NFL, neurofilament light protein; NP, neuropsychological; ROI, regions of interest; rsFMRI, resting state fMRI; SAC, Standardised assessment of concussion; SOT, Sensory organisation test (computerised test of balance function); T-Tau, total Tau.TBI, traumatic brain injury.