Diffusion Tensor Imaging in Sport-Related Concussion: A Systematic Review Using an a priori Quality Rating System

Abstract

Diffusion tensor imaging (DTI) of brain white matter (WM) may be useful for characterizing the nature and degree of brain injury after sport-related concussion (SRC) and assist in establishing objective diagnostic and prognostic biomarkers. This study aimed to conduct a systematic review using an a priori quality rating strategy to determine the most consistent DTI-WM changes post-SRC. Articles published in English (until June 2020) were retrieved by standard research engine and gray literature searches (N = 4932), using PRISMA guidelines. Eligible studies were non-interventional naturalistic original studies that conducted DTI within 6 months of SRC in current athletes from all levels of play, types of sports, and sex. A total of 29 articles were included in the review, and after quality appraisal by two raters, data from 10 studies were extracted after being identified as high quality. High-quality studies showed widespread moderate-to-large WM differences when SRC samples were compared to controls during the acute to early chronic stage (days to weeks) post-SRC, including both increased and decreased fractional anisotropy and axial diffusivity and decreased mean diffusivity and radial diffusivity. WM differences remained stable in the chronic stage (2–6 months post-SRC). DTI metrics were commonly associated with SRC symptom severity, although standardized SRC diagnostics would improve future research. This indicates that microstructural recovery is often incomplete at return to play and may lag behind clinically assessed recovery measures. Future work should explore interindividual trajectories to improve understanding of the heterogeneous and dynamic WM patterns post-SRC.

Introduction

Sport-related concussion (SRC) is a relatively common form of traumatic brain injury (TBI). It can occur in any sport when there is a blow to the head, neck, or body that sends a strong force to the head, with diagnosis based on international consensus guidelines.¹ Compared to other forms of TBI, SRC is characterized by a high chance of repetitive concussive blows and contact given that playing sport is often an ongoing activity. SRC is an important public health issue, given that it impacts all levels of play (amateur to professional) across a variety of sports.^2,3 It is estimated that 1.6–3.8 million incidences occur every year in the United States.⁴ After an SRC, persons may experience rapid-onset short-lived neurological impairments and a range of clinical signs and symptoms, including possible loss of consciousness, headaches, balance problems, dizziness, and cognitive changes, which usually subside within the first 2 weeks post-injury. Persistent post-concussive clinical symptoms can occur for some athletes,¹ with ∼15% of athletes returning to sport after >30 days.⁵

The neuropathophysiology of SRC is not well understood.^6,7 Typically, there appears to be no gross brain pathology, but a range of biochemical and neurophysiological changes that cause dysfunction at the cellular level, including transient neurotransmitter dysregulation, metabolic changes, and neuroinflammation.⁸ Additionally, abnormal brain perfusion and microstructural damage can occur post-SRC, particularly in deep subcortical white matter (WM).⁹ WM sequences (such as T1W, T2W, and T2 fluid-attenuated inversion recovery) are not sufficiently sensitive to reliably detect WM microstructural changes. In contrast, more advanced brain imaging techniques, such as diffusion tensor imaging (DTI), can provide information about changes in WM microstructural integrity. DTI changes can reflect myelin alteration, differences in WM fiber density, alterations of axonal membrane permeability, and changes in axon density.¹⁰ In the past decade, diagnostic guidelines have recommended further studies to determine the clinical utility of DTI for improving SRC diagnosis and prognosis.¹¹ This has resulted in a rapidly expanding SRC-DTI literature in recent years.

There have been six systematic reviews between 2012 and 2019 investigating the characteristic patterns of WM post-SRC,^9,12

–16 with the most recent review including peer-reviewed literature up to January 2019.⁹ The reviews have reported evidence for widespread WM-DTI metric changes after SRC, with major inconsistencies in the direction and magnitude of these changes. In these reviews, higher, lower, and no change in magnetic resonance (MR) diffusion metrics (i.e., axial diffusivity [AD], fractional anisotropy [FA], mean diffusivity [MD], and radial diffusivity [RD]) have been reported post-SRC when compared to controls. Thus far, no clear conclusions regarding DTI's potential clinical utility in SRC populations have been reported.

Divergent findings within the literature relate to variability in study characteristics (e.g., SRC diagnosis, participant demographics, and sports played), study design (e.g., cross-sectional, longitudinal, or period of study post-injury), substantial interstudy variability in DTI methodology (e.g., b-value, number of gradient directions), reported DTI parameters (e.g., AD, FA, MD, and RD), the analytical protocol (e.g., brain areas studied, tractography), as well as other underlying pathology. Further, divergent conclusions across previous systematic reviews relate to disparities in study eligibility criteria, including measurement at different time points post-SRC (e.g., acute vs. chronic outcomes^{9,12,13,15,16}), eligibility of retired athletes,^12,13,16 eligibility of participants with subconcussive impacts rather than formally diagnosed SRC,^9,15,16 and eligibility of various study designs (e.g., prospective, longitudinal studies⁹). These methodological differences contribute to large variations in study quality.

In light of these disparities, this systematic review aimed to synthesize the rapidly expanding literature and determine the clinical utility of DTI in the acute (0–6 days), subacute (7–30 days), and chronic stages (1–6 months) post-SRC when applying strict eligibility criteria and quality assessment measures to overcome previous study and review limitations. The current review focused on studies performing assessments within 6 months of formally diagnosed SRCs in current athletes. This time frame reflects the most beneficial diagnosis period. The exclusion of undiagnosed SRC avoids confounds of other sport-related pathology or other forms of mild TBI,¹⁷ and the exclusion of retired athletes avoids WM variance related to older age.¹⁵

Additionally, and most important, because previous reviews have shown heterogeneous methodologies and major inconsistencies in the direction/presence of effects for DTI metrics after SRC, the current review used an a priori study-quality strategy that only extracted and compared results from high-quality studies. This approach has not been used in previous systematic reviews of this literature. Importantly, this strategy aligns with current efforts in the field to improve the use of DTI in research (e.g., TRACK-TBI group¹⁸ and National Institute of Neurological Disorders and Stroke [NINDS] Common Data Elements [CDEs]¹⁹) and address current gaps that limit clinical translation.¹⁴

Methods

Methods for analysis and inclusion criteria were registered in the PROSPERO (the International Prospective Register of Systematic Reviews) database for systematic reviews (protocol ID: CRD42018094340). The methodology for the current systematic review uses PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) guidelines.²⁰

Search methods

Study retrieval occurred by searching the following databases (from inception to June 1, 2020): Medline, EMBASE, PsycInfo, PsycExtra, PsycArticles, Cumulative Index to Nursing and Allied Health Literature, Academic Search Premier, SPORTdiscus, PubMed, Scopus, Informit, Web of Science, and multiple Cochrane databases (i.e., Reviews, DARE, and Centre for Controlled Trials). Search strategies were created with the assistance of Research Librarians (see Supplementary Table S1). Search terms (n = 56) were generated from the keywords “diffusion tensor imaging,” “concussion,” and “sport” (see Supplementary Table S2 for further details).

To cover the gray literature, the first 200 articles (with no date restriction) from a manual Google Scholar search titled “diffusion tensor imaging in sports concussion” were searched. A snowballing technique was also applied with the retrieval of references from citation by the Google Scholar citation option from all eligible studies. All processes and outcomes are outlined in Figure 1.

FIG. 1.

PRISMA flow chart.

Eligibility criteria

Studies were eligible for inclusion in the review if they met the following pre-specified criteria: 1) included only sports-related TBI; 2) examined current human athletes of either sex from any level of play and types of sports; 3) DTI conducted within 6 months of injury to ensure that any SRC-related brain injury was captured within a clinically meaningful period; 4) reported on cross-sectional, observational, cohort, correlational, or longitudinal study design to allow a comprehensive overview of naturalistic articles; and 5) the study was original research published in a peer-reviewed journal. Studies were excluded if participants included retired athletes, if the study reported on non-naturalistic study designs (e.g., conference abstracts, reviews, commentary articles, or randomized controlled trials), or if the study was not available in English.

Study screening

One reviewer (N.E.) screened all titles and abstracts to identify studies that were potentially relevant for inclusion in the review (k = 491). All decisions were checked by a second reviewer (L.C.). If needed, study authors were contacted by e-mail to obtain further information to determine the eligibility of the study (authors of 15 studies were contacted; nine authors did not respond, four provided the requested information, and two did not provide the requested information). The snowballing technique was then applied where the reference lists of identified articles were screened for suitable studies, and one additional eligible record was obtained. Next, full-text versions were screened to further determine the eligibility for inclusion.

Data extraction process

Information was extracted regarding general study characteristics (i.e., authors, publication details, funding, study design, and inclusion/exclusion criteria), sample characteristics, SRC diagnosis, clinical assessment, DTI acquisition and analyses, statistical analyses, study findings, and study strengths and weaknesses. Study findings of interest included conventional DTI metrics (i.e., AD, FA, MD, and RD) and associations between clinical or cognitive assessments and DTI results. A quantitative synthesis was conducted by directly extracting and/or computing effect sizes for all available data in each individual study. Effect sizes were computed from group means and standard deviations (SDs), t values, or p values depending on what data were available. Average effect sizes for each DTI metric reported in each study were calculated by averaging all regional effects (Figs. 2 –4). Cohen's d was used as the measure of effect, where 0.2 = small, 0.5 = medium, and 0.8 = large. Group-level regional effects are provided in Supplementary Figures S1–S3.

FIG. 2.

Acute effects of sports-related concussions from high-quality studies (0–6 days).

FIG. 3.

Subacute effects of sports-related concussions from high-quality studies (7–30 days).

FIG. 4.

Chronic effects of sports-related concussions from high-quality studies (1–6 months).

Risk of bias assessment and optimization of the National Institutes of Health quality assessment tool

Extracted studies were rated to assess the risk of bias attributable to flaws in study methodology. The National Institutes of Health (NIH) Quality Assessment Tool for Observational and Cohort and Cross-Sectional Articles²¹ was modified to suit the context of SRC and DTI research (see Supplementary Table S3). Details of the modified tool are provided in the Supplementary Materials. Briefly, key modifications included: 1) in-depth quality assurance analysis of the SRC definitions, the DTI methods and analytical protocol (given that these aspects were central to the primary aim of this review); 2) subdividing quality assessment questions into multiple subdomains with independent scores; 3) including subsections applicable to specific study designs (e.g., attrition rate in longitudinal design); 4) implementation of a percentage-weighted scoring system; and 5) grouping of studies into three quality categories based on their weighted ratings and quartile. (i.e., low [quartile 1], medium [quartile 2], and high [quartiles 3–4]).

The modified quality tool included 14 questions related to study population, sample size, study design (e.g., cross-sectional vs. longitudinal), validity of measures, analysis procedure, and appropriate consideration of confounding factors. Question 11 was used to assess the DTI protocol and DTI analyses in detail (see Supplementary Table S3 and legend).

Quality ratings were completed by two independent raters (N.E. and L.J.). Agreement between raters was adequate with an intraclass correlation of 0.74 (p = 0.004). This was followed by a final consensus discussion between raters to fully review the unmatched low-, medium-, and high-quality classifications. After this discussion (with L.C.), the absolute agreement intraclass correlation coefficient on the quality rating scores was excellent (0.96; p < 0.001). The reviewers who completed the rating had expertise in neuropsychology (N.E., L.C.), neuroimaging (L.C., L.J.), and TBI (N.E.).

Results

Studies' demographic, sampling, ascertainment, timeline, clinical, and diffusion tensor imaging characteristics

There were 29 eligible studies (20 longitudinal, nine cross-sectional), published between 2010 and 2020 (Fig. 1; Table 1). Studies report on athletes in the United States (16 of 29 studies) and Canada (13 of 29 studies). There was considerable growth in the number of published studies, with 12 of 29 published since the last systematic review (2019–2020).

Table 1.

Summary of Study Characteristics (k = 29)

Study authors, date, location design	N		% Male		Age (years)		MRI scans	SRC diagnosis	Scanner			Clinical and cognitive variables	Exclusion criteria
Study authors, date, location design	SRC	Control	SRC	Control	SRC	Control	MRI scans	SRC diagnosis	dir	B (s/mm²)	DTI metrics	Clinical and cognitive variables	Neurological	Substance	Mental	Cognition	Other medical	Other injury	Medication	Unrelated TBI	MRI related	Other
Baker et al. (2020)⁴³ USA	26	13	58		15.1 ± 1.1		∼1 week post-SRC (scan 1), ∼18–30 days post-SRC (scan 2)	By physician; consistent with current guidelines	Not provided		FA	Post-Concussion Symptom Scale, SCAT-5 (18–30 days post-injury)	x		x			x		x		Exclusion: GCS <13
Borich et al. (2013)⁴⁸ Canada	12	10	83	90	15.5 ± 1.2	15.7 ± 0.9	≤2 months post-SRC	By team coach or physician; no information on guidelines used	60	700	FA, MD, RD, AD, trace	No. of SRCs sustained, time since last diagnosed SRC, SCAT-2 score	x						x
Brett et al. (2020)³⁹ USA	69		83		19.7 ± 1.1		∼1–2 days post-SRC (scan 1), 9 ± 6 days post-SRC (scan 2), 7 days post-RTP (scan 3)	Consistent with current guidelines	30	1000	MD	Previous concussion, GPA, reading score, health problems, mental disorders, amnesia, reinjury, SRC symptom score (pre-season, 1–2 days post SRC, 9 ± 6 days post-SRC, 7 days post-RTP)						x
Chamard et al. (2016)⁴⁵ Canada	10	8	0	0	21.4 ± 1.1	21.0 ± 1.4	∼6 months post-SRC	Standardized SRC history form	64	700	Trace	No. of previous SRCs, approximate date of each SRC, mechanism of injury, and nature and duration of relevant post-SRC symptoms	x	x	x	x			x	x		Controls: no history of TBI
Churchill et al. (2017)²² Canada	26	16	46	46	17–24	18–23	1–7 days post-SRC	Consistent with current guidelines	30	700	FA, MD	SCAT-3 (baseline, pre-season) and post-injury										SRC: diagnosis by physician required; controls: no concussion in past 6 months;
Churchill et al. (2019)⁴⁰ Canada	24	122	46	51	20.0 ± 1.9	20.3 ± 2.0	1–6 days post-SRC (scan 1), median 27 days post-SRC (scan 2), 1 year after return to play (scan 3)	By physician	30	700	FA, MD	SCAT-3 (baseline, pre-season) and post-injury (1–6 days post-SRC (27 days post-SRC)										no criteria reported
Churchill et al. (2019)⁴¹ Canada	33	33	48	48	20.3 ± 2.0	20.5 ± 1.7	Pre-season (scan 1), 1 to 7 days (scan 2), median 19 days post-SRC (scan 3)	By physician	30	700	FA, AD, RD	SCAT-3 (baseline, pre-season) and post-injury										Significant SRC-related deficits at baseline (SCAT3);
Churchill et al. (2020)³³ Canada	12	44	58	45	20.3 ± 1.5	20.0 ± 1.8	Median 4 days (scan 1), 38 days (scan 2) post-SRC	By physician	30	700	FA, MD	SCAT-3 (baseline, pre-season) and post-injury										no criteria reported
Fuchs et al. (2015)²³ USA	3	8	100	100	19–23	19–23	1 day post-SRC, 1 and 2 week post-injuryControls: one scan, no follow-up imaging	SRCs were defined using the AAN guidelines.	25	100	FA, MD, RD, AD	N/A										No criteria reported
Henry et al. (2011)²⁴ Canada	16	8	100	100	22.1 ± 1.7	22.8 ± 1.5	0–5 days post-SRC, 6 months laterControls: baseline, 18 months post-baseline	SRCs were defined using the AAN guidelines.	64	700	FA, MD, AD	Concussive symptomatology assessed using PCSS at same time as MRI scans	x	x	x	x			x	x		Controls: no history of TBI
Hirad et al. (2019)³⁸ USA(mTBI cohort only)	29	58	52	72	19.5	21.6	0–4 days of SRC.	By on-field certified athletic trainer; consistent with Sport Concussion Assessment Tool 2	60	1000	FA	Blood samples (peripheral tau)										Controls: no history of SRC
Koerte et al. (2012)⁴² Canada	5	20	100	100	22.2 ± 1.6		Pre-season, post-season	SRC defined using 2008 Zurich consensus statement	60	700	FA, RD, AD, trace	Concussion incidence and neuropsychological testing not reported in this study	x		x
Lancaster et al. (2016)²⁵ USA	26	26	100	100	17.6 ± 1.5	18.0 ± 1.5	24 h post-SRC, 1 week later	SRC defined using U.S. Department of Defense's definition for mTBI	30	1000	FA, MD, AD, RD	Baseline (pre-season) = self-reported health history, WTAR, SCAT-3, SAC, BESSSRC athletes + matched-controls: SCAT-3, Alternate SAC, BESS within 24 h and at 8 days post SRC			x		x	x	x	x
Mac Donald et al. (2019)⁴⁷ USA	22	24	41	46	13.3 ± 1.6	12.9 ± 1.7	4–6 weeks post-SRC, 6–12 months post-SRC	By physician	32	1000	FA	Neuropsychological evaluation: Immediate Post-Concussion Assessment and Cognitive Test, King Devick, Weschler Intelligence Scale for Children 5th Ed., Delis-Kaplan Executive Function System Test, Symbol Digit Modalities Test, Grooved Peg Board TestHealth: Health Behavior Inventory, Pediatric Quality of Life InventoryMental health: Patient Health Questionnaire, Generalized Anxiety Disorder, Adolescent Sleep-Wake ScaleTests completed at each visit	x		x					x	x
Manning et al. (2017)²⁶ Canada	17	26	100	100	13.3 ± 0.6	13.0 ± 0.1	24–72 h post-SRC, 3 months post-SRCControls: one scan, no follow-up imaging	By physician	64	1000	FA, MD, AD, RD, trace	Past SRC history, SCAT-3, ImPACT, and balance testing, at the same time as MRI scans	x									Controls: no concussion in past 6 months
Manning et al. (2019)³⁷ Canada	21		0		20.1 ± 1.0		1–3 days (scan 1), 3 months (scan 2), 6 months (scan 3) post-SRC	By a sports medicine physician	64	1000	FA, MD	SCAT3, time since the most recent concussion, self-reported total number of concussions (1–3 days post-SRC)					x			x	x
Maugans et al. (2012)²⁷ USA	12	12	75	75	13.4 ± 1.2		2 days, 2 weeks, 1+ month post-SRCControls: one scan, no follow-up imaging	SRC defined using 2008 Zurich consensus statement	32	1000	FA, AD, RD, trace	Past SRC history and ImPACT at the same time as MRI scans	x					x	x		x
Meier et al. (2016)²⁸ USA	40	46	75	65	20.1 ± 1.4	20.3 ± 1.5	2 days, 1 week, 1 month post-SRCControls: one scan, no follow-up imaging	Consistent with current guidelines	30	1000	FA, AD, RD	Past SRC history, SRC information, and HAM-D/HAM-A/ANAM self-reported and cognitive battery at the same time as MRI scans		x	x
Muftuler et al. (2020)³⁶ USA	96	82	100	100	18.1 ± 1.5	18.4 ± 1.7	2 days (scan 1), 8 days (scan 2), 15 days (scan 3), 45 days (scan 4) post-SRC	By certified athletic trainer or team physicians	30	1000, 2000	FA, MD, RD, AD	SCAT3, Standardized Assessment of Concussion, Balance Error Scoring System (pre-season baseline and post-injury at each scan)			x			x	x	x	x
Murugavel et al. (2014)²⁹ USA	21	16	100	100	20.2 ± 1.0	19.90 ± 1.67	2 days, 2 weeks, 2 months post-SRCControls: one scan, no follow-up imaging	By physician; consistent with consensus guidelines	64	1000	FA, MD, AD, RD	Baseline: Past SRC history, SCAT2, and ImPACTPost-SRC: SCAT2, PHQ-9, GAD-7, hybrid tests from ImPACT and paper/pencil tests at the same time as MRI scansNot reported in this study	x		x		x				x	Controls: no history of TBI;excluded if left-handed
Niogi et al. (2019)³⁴ USA	7	25	100	100	23.8 ± 2.1	20–35	1–5 days post-SRC	By team physician	55	Not provided	FA	Immediate Post-Concussion Assessment and Cognitive Test within 3 days of SRC	x							x	x
Pasternak et al. (2014)³⁰ Canada	11	27	55		17–26	17–26	Pre-season, 72 h post-SRC	SRC defined using 2008 Zurich consensus statement	60	700	FA, FA_t, RD_t, AD_t	SCAT-2 and ImPACT at the same time as MRI scans, but not reported in this study	x			x					x
Satchell et al. (2019)⁶⁹ USA	17		65		13		14–78 days post-SRC	By sports medicine provider	10	1000	FA, MD	Concussion symptom inventory										No criteria reported
Schranz et al. (2018)³¹ Canada	15	33	0	0	21,0 ± 1.5		Pre-season, 72 h, 3 months, 6 months post-SRC, post-season	Consistent with consensus guidelines	64	1000	FA, MD, AD, RD	SCAT-3, ImPACT at the same time as the MRI scans										Controls: no concussions in past 10 months;
Virji-Babul et al. (2013)⁷⁰ Canada	12	10	83	90	15.5 ± 1.2	15.7 ± 0.9	2 months post-SRC	Not specified	60	700	FA, MD	SCAT-2, within 2 months post-SRC	x						x			controls: no SRC
Wilde et al. (2020)³⁵ USA	18	13	44	46	19.7 ± 1.8	19.9 ± 1.1	2 days post-SRC	By athletic trainer or team physician	64	1300	FA, MD, RD, AD	SCAT-3, Automated Neuropsychological Assessment Metric battery	x		x				x		x	Controls: no past year concussion, no contact sport;SRC inclusion: ≥14 days to return to play;
Wu et al. (2018)³² USA	10	12	60		14.6 ± 1.6		3 days, <3 months post-SRCControls: two scans with the same interval	SRC diagnosis based on ACRM criteria	30	1000	FA, AD	GCS, PTA, LOC, and sports mechanism of injury	x	x	x					x	x	not fluent English or Spanish speakers, left-handed
Zhang et al. (2010)⁴⁶ USA	15	15	70		20.8 ± 1.7	21.3 ± 1.5	∼30 days post-SRC	SRC defined using Cantu Data Driven Revised Concussing Grading Guidelines (2006)	20	Not Provided	FA, AD, no. of fibers	PCSC, ImPACT, neuropsychological battery (HVLT, Stroop, TMTA and B, SDMT), and Beatty test fatigue rating, previous scanning										Controls: no concussions
Zhu et al. (2015)⁴⁴ USA	8	11	100	100	20 ± 1.3	20.5 ± 1.8	24 h, 7 days, 30 days post-SRCControls: three scans with the same interval	Consistent with consensus guidelines	25	1000	FA, MD, AD, RD, trace	SRC groups: baseline pre-season ImPACT and then on day 1 post-SRC and every 2–3 days until the athlete returned to their baseline scores (around day 7)	x									Controls: did not meet ACSM guidelines for physical activity, resistance, or neuromotor exercise.

AAN, American Academy of Neurology; ACRM, American Congress of Rehabilitation Medicine; AD, axial diffusivity; ADC, apparent diffusion coefficient; ADHD, attention-deficit hyperactivity disorder; AIS, Abbreviated Injury Scale; ANAM, Automated Neuropsychological Assessment Metrics; BESS, Balance Error Scoring System; CT scan, computed tomography scan; dir, directions; FA, fractional anisotropy; GAD-7, Generalized Anxiety Disorder 7-item; GCS, Glasgow Coma Scale; HAM-A, Hamilton Anxiety Rating Scale; HAM-D, Hamilton Depression Rating Scale; ImPACT, Immediate Post-Concussion Assessment and Cognitive Testing; Kax, axial kurtosis; Krad, radial kurtosis; LD, learning disorder; LOC, loss of consciousness; MD, mean diffusivity; MK, mean kurtosis; MRI, magnetic resonance imaging; mTBI, mild traumatic brain injury; N/A, not applicable; PCSC, Post-Concussion Symptom Checklist; PCSS, Post-Concussion Symptom Scale; PTA, post-traumatic amnesia; PTSD, post-traumatic stress disorder; RD, radial diffusivity; SAC, Standardized Assessment of Concussion; SCAT, Sport Concussion Assessment Tool; SDMT, Symbol Digit Modality Test; SRC, sports-related concussion; T, Tesla; TBI, traumatic brain injury; TMT, Trail Making Test; PHQ-9, Patient Health Questionnaire- 9; WM, white matter; WTAR, Wechsler Test of Adult Reading.

The timing of the DTI post-SRC ranged from acute (0–6 days), subacute (7–30 days), to chronic (>1 month to ∼6 months) periods, with 20 studies conducting a DTI scan during the acute phase.^{22

–41} Two studies did not provide information regarding timing of DTI post-SRC^42,43; subsequently, study authors clarified that it was within the eligible 6-month time frame.

All but three studies^23,37,38 conducted clinical assessments post-SRC, with 11 studies also conducting further follow-up assessments^{24

–29,31,36,39,40,44} and 11 conducted pre-SRC clinical assessment.^{22,25,30,31,33,36,39

–42,44} Clinical assessment was primarily conducted through brief screening assessment tools. One study used a qualitative measure of concussive symptomatology.⁴⁵ Traditional neuropsychological measures were rarely used (6 of 29 studies^{29,35,39,40,46,47}).

FA was quantified in all but one study,³⁹ MD in 20 of 29 studies, AD in 15 of 29 studies, and RD in 14 of 29 studies. DTI analytical techniques were varied and included whole-brain or ROI (region of interest) voxel-based analyses, ROI-based analyses, and whole-brain or ROI tractography.

Quality ratings

Ten of 29 studies (35%) were rated high quality, 14 of 29 (48%) were rated medium quality, and 5 of 29 (17%) were rated low quality (see Table 2). Mean weighted quality score was 66% (SD = 10%) for both raters. A small, non-significant positive correlation was observed between mean weighted consensus quality scores and year of publication (r = 0.19, t = 1.03, p = 0.31).

Table 2.

Quality-Assurance–Weighted Scores Rating after Consensus

Article	Weighted score % (rater #1)	Weighted score % (rater #2)	Rating of weighted scores
Borich et al. (2013)⁴⁸ (45)	74	74	H
Churchill et al. (2019)³⁷	72	72	H
Churchill et al. (2020)³⁰	72	72	H
Henry et al. (2011)²¹	72	77	H
Lancaster et al. (2016)²²	72	72	H
Mac Donald et al. (2019)⁴⁴	86	86	H
Manning et al. (2017)²³	72	72	H
Meier et al. (2016)²⁵	72	77	H
Murugavel et al. (2014)²⁶	72	72	H
Muftuler et al. (2020)³³	81	81	H
Baker et al. (2020)⁴⁰	67	67	M
Brett et al. (2020)³⁶	62	62	M
Chamard et al. (2016)⁴²	69	69	M
Churchill et al. (2017)¹⁹	64	64	M
Churchill et al. (2019)³⁸	62	62	M
Hirad et al. (2019)³⁵	67	67	M
Manning et al. (2019)³⁴	67	67	M
Maugans et al. (2012)²⁴	67	58	M^a
Niogi et al. (2019)⁶⁴	67	67	M
Pasternak et al. (2014)²⁷	62	62	M
Virji-Babul et al. (2013)⁷⁰ (66)	69	69	M
Wilde et al. (2020)³²	69	69	M
Wu et al. (2018)²⁹	62	58	M^a
Zhang et al. (2010)⁴³	58	58	L
Zhu et al. (2015)⁴¹	58	62	M^a
Fuchs et al. (2015)²⁰	39	48	L
Koerte et al. (2012)³⁹	48	48	L
Satchell et al. (2019)⁶⁵	48	48	L
Schranz et al. (2018)²⁸	53	53	L

Studies were rated as medium quality if only one rater rated them as low quality.

H, high (≥Q3, i.e., 72% for both raters); M, medium (Q2 and Q1); L, low (<Q1, i.e., 62% for rater #1 and 60% for rater #2).

High-quality studies were those that reported demographic information^{28,29,40,47,48} and inclusion/exclusion criteria^{24
–26,28,29,36,47,48} and where DTI results were reported accurately based on a DTI protocol that was able to assess the stated study aims. These studies also adjusted for key confounding variables and were all longitudinal except for one,⁴⁸ although only three studies collected DTI data pre- and post-SRC.^33,36,47 Six did not have significant attrition.^{24
–26,36,40,47} Nevertheless, the high-quality studies had some limitations. Only three studies used control participants with no previous concussion history,^24,47,48 and one provided no information about control participants' concussion history.²⁹ The SRC diagnosis method was limited in another two studies,^47,48 and two had suboptimal reliability and validity of clinical assessment.^29,47 Only one study reported that imaging analysis was conducted blinded of clinical status.⁴⁷ Finally, none of the high-quality–rated studies reported sample-size justification, power description, or variance and effect estimates relevant for populations using DTI.

Diffusion tensor imaging findings in high-quality studies

Computed average effect sizes from the 10 high-quality studies are presented in Figures 2 (acute effects), 3 (subacute effects), and 4 (chronic effects). Only conventional DTI metrics (i.e., AD, FA, MD, and RD) from high-quality studies are synthesized in the text because there remain too few high-quality studies reporting on diffusional kurtosis imaging metrics to compare findings across studies (k = 2) adequately.

Group differences in AD were examined in seven studies. Compared to controls post-SRC, three of seven studies reported significantly higher AD in SRC groups,^24,26,28 three of seven studies reported significantly lower AD in SRC groups,^25,26,48 and two of seven studies found no significant group differences in AD.^29,36 Significant moderate-to-large effect sizes were observed across the acute (k = 4) and subacute stages (k = 2), and large effect sizes were observed in the chronic stage (k = 2; see Figs. 1–3). Of the studies that investigated within- or between-group changes over time,^24,25,36 no significant longitudinal changes were identified. However, cross-sectional group differences observed in the acute stage of one study were no longer significant in the subacute or chronic stages.²⁸

FA was examined in 10 studies; group differences were explored in 9 of 10 studies, and individual-level differences were explored in 1 of 10 studies. Compared to controls post-SRC, four of nine studies reported significantly higher FA in SRC groups,^24,26,28,48 three of nine studies reported significantly lower FA in SRC groups,^29,40,47 and two of nine studies found no group differences in FA at any time points.^25,36 Significant moderate-to-large effect sizes were observed at the acute (k = 4), subacute (k = 3), and chronic stages (k = 4). Of the studies that investigated within- or between-group changes over time,^24,29,36,47 three of four studies reported no significant longitudinal changes,^24,29,36 and one study showed further reductions in FA over 6 months for the SRC versus control group (d = 1.18; large effect).⁴⁷ At the individual level, abnormal voxels for FA were observed among the SRC group, with limited spatial overlap observed between the acute and chronic time points.³³

MD was examined in seven studies; group differences were explored in six of seven studies, and individual-level differences were explored in one of seven studies. Compared to controls post-SRC, four of six studies reported significantly lower MD in SRC groups,^24
–26,48 one of six studies reported significantly higher MD in the SRC group,⁴⁰ and one of six studies found no group differences in MD.³⁶ Significant moderate-to-large effect sizes were observed across the acute (k = 4) and subacute stages (k = 3), and large effect sizes were observed in the chronic stage (k = 3). Of the studies that investigated within- or between-group changes over time,^24,25,36 no significant longitudinal changes were identified. At the individual level, abnormal voxels for MD were observed among the SRC group, with limited spatial overlap observed between the acute and chronic time stages.³³

Group differences in RD were examined in six studies. Compared to controls post-SRC, three of six studies reported significantly lower RD in SRC groups,^25,26,28 two of six studies reported no significant group differences,^36,48 and one of six studies reported significantly higher RD in the SRC group.²⁹ Significant moderate-to-large effect sizes were observed in the acute stage (k = 4), moderate effect sizes were observed in the subacute stage (k = 1), and large effect sizes were observed in the chronic stage (k = 1). Of the studies that investigated within- or between-group changes over time,^25,29,36 two of three studies reported no significant longitudinal changes,^25,36 and one study found that the significantly higher RD in the acute stage for the SRC group was no longer significantly different from controls in the subacute or chronic stages.²⁹

Clinical and cognitive assessment correlations with diffusion tensor imaging metrics in high-quality studies

Seven of 10 high-quality studies reported on correlations between DTI metrics and clinical assessment measures (i.e., symptom severity, duration, and SRC frequency; see Fig. 5). Associations between symptom severity and DTI metrics were investigated in six studies, with 19 of 27 correlations (70%) showing a significant relationship (2 of 19 small effect; 11 of 19 moderate effect; and 6 of 19 large effect). The largest correlations were observed between symptom severity with AD and FA, albeit an inconsistent direction of effects was reported. Symptom duration and SRC frequency were investigated less frequently against DTI metrics (k = 4), and when this was examined, significant correlations were rarely observed (3 of 22 correlations were significant; 14%). No high-quality studies reported on correlations between cognitive task performance and DTI metrics.

FIG. 5.

Correlations between DTI metrics and clinical assessments in high-quality studies. A = acute; SA = subacute; C = chronic; ns = non-significant (value not provided). Green values = small effect size; blue values = moderate effect size; purple values = large effect size; uncolored values = non-significant correlations.

Discussion

Main findings of high-quality studies

Moderate-to-large effects on DTI metrics were typically observed post-SRC in current athletes, with WM abnormalities usually showing stable trajectories throughout the acute, subacute, and chronic stages. The evidence therefore suggests that microstructural recovery is often incomplete at return to play, which typically occurs after recovery of clinical measures. Without regional specificity, decreased MD and RD were most commonly observed after an SRC, whereas the FA and AD findings were inconclusive (i.e., both higher and lower FA and AD observed). WM abnormalities observed using conventional DTI metrics were generally associated with greater severity of clinical SRC symptoms, but less frequently associated with symptom duration or SRC frequency, suggesting that microstructural recovery may lag behind clinically assessed recovery measures.

Diffusion tensor imaging changes after sport-related concussion

The results from high-quality studies suggest that some DTI metrics may be sensitive to acute and early chronic microstructural WM changes after an SRC. Regional effects were reported in group-level analyses in high-quality studies, indicating that there is common WM alteration attributable to SRC (Supplementary Figs. S1–S3). The lack of reproducible regional effects across studies may be attributable to SRCs exerting broadly distributed biochemical sequelae on the brain, variability of athlete characteristics and concussive injury mechanisms, as well as methodological limitations such as small sample size and differences in DTI parameters.⁴⁹ Four studies observed increased FA post-SRC (and decreased MD/RD, which are inversely associated with FA⁵⁰). This is thought to reflect diffuse axonal injury, with subtle cytotoxic or vasogenic edema and axonal swelling increasing anisotropic diffusion and causing increases in FA.^51,52 However, it is important to note that whereas four studies reported increased FA post-SRC, three studies reported decreased FA when comparing SRC athletes to controls. Decreased FA post-SRC is thought to reflect tract disruption, progressive Wallerian degeneration, and loss of myelin integrity.^53,54

Although it has been theorized that FA is increased acutely after SRC and decreased thereafter,⁵⁵ the current findings reveal a more complex picture whereby studies have reported both broadly distributed increases and decreases in FA (and inverse association with MD/RD) in the acute, subacute, and chronic stages, with limited evidence of direction of effects being tied to temporality (at least over a ∼6-month period). Similarly, a recent meta-analysis examining tract-based spatial statistics in SRC did not find evidence to support a simple time effect, but more dynamic effects likely to reflect interindividual variability in brain recovery processes.⁵⁶ Decreased FA and increased MD/RD could reflect chronic and cumulative physiological injuries subsequent to repetitive subconcussive impacts often observed among athletes in contact sports.⁵⁷ In contrast, increased FA and decreased MD/RD may reflect specific microstructural neuroinflammatory injuries post-SRC.⁵⁷ Of relevance, Churchill and colleagues found that although the SRC group reported decreased FA compared to controls, elevated FA was associated with greater SRC symptom severity,⁴⁰ which could reflect both SRC neuroinflammatory injuries and cumulative impacts of subconcussive knocks.

Although this review attempted to limit the effect of subconcussive impacts on DTI outcomes by only including studies with formally diagnosed SRC, it is impossible to completely parse out these effects among athletes in contact sports. Further, it may be the case that both increased and decreased FA coexist in the same person depending on the region of the brain. Indeed, when comparing the abnormal voxels of individual athletes, Churchill and colleagues identified both abnormally elevated and depressed FA and MD voxels post-SRC.³³ Future longitudinal work into individual-level differences post-SRC can provide greater clarification on diverging direction of effects in DTI metrics and their timeline. It has also been suggested that FA directions are ambiguous, with differences simply reflecting alterations in microstructure, and further interpretations must be backed by strong theoretical foundations.¹⁰

There was no evidence of further DTI metric changes in studies with follow-up assessments over 1–6 months, meaning that the DTI differences did not significantly recover or worsen across time. This is consistent with recent reports of sustained neuroanatomical differences years after concussive hits.⁵⁸ Future longer-term studies with larger sample sizes are needed to confirm this conclusion.⁵⁹

Methodological considerations

The validity of DTI findings in SRC remains limited by a number of methodological issues. Besides observing that DTI SRC research has typically been exploratory in nature (with no power analyses), the most significant quality issues noted included limited demographic information and limited assessment validity (i.e., examination of associations between clinical variables, SRC, and DTI), among other issues affecting risk of bias (i.e., absence of assessor blinding and significant attrition in some longitudinal studies). In high-quality–rated studies, extracted WM areas were reported either by ROI, voxel-based WM, or tractography with clear exclusion of gray matter. In contrast, several low- to medium-quality studies included gray matter, which fundamentally affects the magnitude and nature of DTI parameters.⁶⁰

Methodological issues were also present in high-quality studies. There was an unequal representation of sex and sport type, which are known to impact SRC recovery and outcome.^61,62 Critically, SRC diagnostic methods also varied between studies (including limited or absent reporting in some instances), which is concerning given that this diagnosis is central to the study of SRC-related DTI changes. The presence of previous concussion history (i.e., ranging from 1 to 4 across athletes with available information) may be contributing to DTI directionality differences and impacting recovery and outcome post-SRC.⁵⁹ Yet, only two studies examined correlations between SRC frequency and DTI metrics. This makes it difficult to ascertain whether the etiology of observed DTI-WM changes were from a singular concussion during the study or cumulative change from multiple previous hits.^9,13

Methodological limitations were also observed with the control groups among high-quality studies. Ideally, the control group should be well matched demographically, and information regarding concussion history and sport involvement during the study should be clearly reported, given that interpretation of the extent of WM changes could be further complicated by any subconcussive hits experienced.⁶³ In this regard, two control groups may be required: one with no SRC history that does not play sport and another with no SRC history that regularly engages in sport. Finally, the ability to comment on the extent to which DTI changes reflect the severity of SRC is limited because of heterogeneous cognitive measures used, the common lack of injury severity information (i.e., Glasgow Coma Scale, loss of consciousness, retrograde amnesia, and duration of post-traumatic amnesia), or tests sensitive to subtle cognitive deficits and their changes over time.⁶⁴ Standardized SRC diagnosis and better neuropsychological measures would improve future imaging research and study replication.⁶⁵

Common among both high- and low- to medium-quality studies was the relatively small sample size and limited neuroimaging before SRC. Larger sample sizes may be achievable through collaborative cohorts and consortiums. Greater representation of female athletes is also necessary. Additionally, longitudinal study design is a fundamental requirement given that SRC is a dynamic process, which requires serial evaluation and extraction of individual trajectories. Future work should include imaging pre-SRC and examine DTI changes over a longer follow-up period to determine microstructural recovery time.³⁴

Strengths and limitations

This is the first systematic review of the SRC and DTI literature to apply an a priori study-quality strategy that only extracted and compared results from high-quality studies. This approach aligns with current efforts in the field to improve the use of DTI in research (e.g., TRACK-TBI group¹⁸ and NINDS CDEs¹⁹) and address current gaps that limit clinical translation.¹⁴ The standard NIH quality rating tool was adapted (Supplementary Material) for use in this review and can be used in future reviews of neuroimaging studies. Despite adequate inter-rater reliability, it is recommended that two raters, with cross-disciplinary expertise, independently perform the quality assessment followed by an inter-rater discussion to achieve 100% concordance for high-, medium-, and low-quality ratings.

Several limitations of this review warrant consideration. First, this review focused on the effect of SRC on WM microstructure, whereas the impact on white and gray matter volume was not evaluated. Integration of all three outcomes in future reviews of high-quality studies will yield a more complete understanding of the effect of SRC. Second, the variability in magnitude and direction of DTI metrics observed in this review may relate to continued maturation of white matter microstructure throughout adolescence and young adulthood.⁶⁶ Some high-quality studies reported on in this review recruited participants as young as 13 years; thus, variation in neurodevelopmental stage could be an important moderating factor for microstructural outcomes, although this should be controlled by age- and sex-matched controls.

Conclusion

Decreased MR diffusivities (MD and RD) in SRC samples (maximum of 6 months after SRC) compared to controls are the most consistent DTI-WM changes, as extracted by an a priori quality review strategy. Future DTI SRC research should aim to further standardize the SRC diagnosis protocols and neuropsychological measures used, conduct longitudinal studies that capture pre-SRC DTI metrics, and ensure analyses are adequately powered. The development of a “standard” diffusion protocol that performs equally across different scanner types and sites, and clearly specifies all diffusion acquisition parameters (e.g., ABCD imaging protocol for young people,⁶⁷ ENIGMA DTI harmonization processing steps⁶⁸), would benefit DTI SRC research.

Footnotes

Acknowledgments

We thank the library staff at Macquarie University (Sydney, New South Wales, Australia) who provided valuable input into the review process.

Funding Information

This work was supported by the NHMRC Career Development Fellowship with UNSW/NeuRA extension to 2019 (APP1045400; CIA/PI, Cysique) and an NHMRC Scholarship (GNT1169377; to Lees).

Author Disclosure Statement

No competing financial interests exist.

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2

Supplementary Figure S3

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

References

McCrory

, Meeuwisse

, Dvorak

, Aubry

, Bailes

, Broglio

, Cantu

R.C.

, Cassidy

, Echemendia

R.J.

, and Castellani

R.J.

(2017). Consensus statement on concussion in sport—the 5th International Conference on Concussion in Sport held in Berlin, October 2016. Br. J. Sports Med. 51, 838–847.

Giza

C.C.

, and Kutcher

J.S.

(2014). An introduction to sports concussions. Continuum (Minneap. Minn), 20, 6 Sports Neurology, 1545–1551.

Muth

C.C.

(2018). Sport-related concussion. JAMA, 319, 840–840.

Langlois

J.A.

, Rutland-Brown

, and Wald

M.M.

(2006). The epidemiology and impact of traumatic brain injury: a brief overview. J. Head Trauma Rehabil. 21, 375–378.

Kerr

Z.Y.

, Zuckerman

S.L.

, Wasserman

E.B.

, Covassin

, Djoko

, and Dompier

T.P.

(2016). Concussion symptoms and return to play time in youth, high school, and college american football athletes. JAMA Pediatr. 170, 647–653.

Manley

, Gardner

A.J.

, Schneider

K.J.

, Guskiewicz

K.M.

, Bailes

, Cantu

R.C.

, Castellani

R.J.

, Turner

, Jordan

B.D.

, Randolph

, Dvořák

, Hayden

K.A.

, Tator

C.H.

, McCrory

, and Iverson

G.L.

(2017). A systematic review of potential long-term effects of sport-related concussion. Br. J. Sports Med. 51, 969–977.

Giza

C.C.

, and Difiori

J.P.

(2011). Pathophysiology of sports-related concussion: an update on basic science and translational research. Sports Health, 3, 46–51.

Bigler

E.D.

(2018). Structural neuroimaging in sport-related concussion. Int. J. Psychophysiol. 132, 105–123.

Schneider

D.K.

, Galloway

, Bazarian

J.J.

, Diekfuss

J.A.

, Dudley

, Leach

J.L.

, Mannix

, Talavage

T.M.

, Yuan

, and Myer

G.D.

(2019). Diffusion tensor imaging in athletes sustaining repetitive head impacts: a systematic review of prospective studies. J. Neurotrauma, 36, 2831–2849.

10.

Jones

D.K.

, Knösche

T.R.

, and Turner

(2013). White matter integrity, fiber count, and other fallacies: the do's and don'ts of diffusion MRI. Neuroimage, 73, 239–254.

11.

Hobbs

J.G.

, Young

J.S.

, and Bailes

J.E.

(2016). Sports-related concussions: diagnosis, complications, and current management strategies. Neurosurg. Focus, 40, E5.

12.

Kamins

, Bigler

, Covassin

, Henry

, Kemp

, Leddy

J.J.

, Mayer

, McCrea

, Prins

, Schneider

K.J.

, Valovich McLeod

T.C.

, Zemek

, and Giza

C.C.

(2017). What is the physiological time to recovery after concussion?. A systematic review. Br. J. Sports Med. 51, 935–940.

13.

Gardner

, Kay-Lambkin

, Stanwell

, Donnelly

, Williams

W.H.

, Hiles

, Schofield

, Levi

, and Jones

D.K.

(2012). A systematic review of diffusion tensor imaging findings in sports-related concussion. J. Neurotrauma, 29, 2521–2538.

14.

McCrea

, Meier

, Huber

, Ptito

, Bigler

, Debert

C.T.

, Manley

, Menon

, Chen

J.-K.

, and Wall

(2017). Role of advanced neuroimaging, fluid biomarkers and genetic testing in the assessment of sport-related concussion: a systematic review. Br. J. Sports Med. 51, 919–929.

15.

Chamard

, and Lichtenstein

J.D.

(2018). A systematic review of neuroimaging findings in children and adolescents with sports-related concussion. Brain Inj. 32, 816–831.

16.

Asken

B.M.

, DeKosky

S.T.

, Clugston

J.R.

, Jaffee

M.S.

, and Bauer

R.M.

(2018). Diffusion tensor imaging (DTI) findings in adult civilian, military, and sport-related mild traumatic brain injury (mTBI): a systematic critical review. Brain Imaging Behav. 12, 585–612.

17.

Rabinowitz

A.R.

, Li

, and Levin

H.S.

(2014). Sport and nonsport etiologies of mild traumatic brain injury: similarities and differences. Annu. Rev. Psychol. 65, 301–331.

18.

International Traumatic Brain Injury Research Initiative. (2014). Transforming Research and Clinical Knowledge in Traumatic Brain Injury (Track-TBI). The Regents of the University of California: University of California, San. Francisco.

19.

Grinnon

S.T.

, Miller

, Marler

J.R.

, Lu

, Stout

, Odenkirchen

, and Kunitz

(2012). National Institute of Neurological Disorders and Stroke Common Data Element Project—approach and methods. Clin. Trials, 9, 322–329.

20.

Liberati

, Altman

D.G.

, Tetzlaff

, Mulrow

, Gøtzsche

P.C.

, Ioannidis

J.P.

, Clarke

, Devereaux

P.J.

, Kleijnen

, and Moher

(2009). The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med. 6, e1000100.

21.

National Institutes of Health. (2014). Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies. National Institutes of Health: Bethesda, MD

22.

Churchill

N.W.

, Hutchison

M.G.

, Richards

, Leung

, Graham

S.J.

, and Schweizer

T.A.

(2017). The first week after concussion: Blood flow, brain function and white matter microstructure. Neuroimage Clin. 14, 480–489.

23.

Fuchs

, Hotiu

, Jantzen

K.J.

, Steinberg

, and Kelso

J.S.

(2015). Diffusion tensor imaging in mild traumatic brain injuries—acute state and short-term recovery. Medical Research Archives. https://esmed.org/MRA/mra/article/view/320 (Last accessed August 10, 2021 ).

24.

Henry

L.C.

, Tremblay

, Lee

, Brun

, Lepore

, Theoret

, Ellemberg

, and Lassonde

(2011). Acute and chronic changes in diffusivity measures after sports concussion. J. Neurotrauma, 28, 2049–2059.

25.

Lancaster

M.A.

, Olson

D.V.

, McCrea

M.A.

, Nelson

L.D.

, LaRoche

A.A.

, and Muftuler

L.T.

(2016). Acute white matter changes following sport-related concussion: a serial diffusion tensor and diffusion kurtosis tensor imaging study. Hum. Brain Mapp. 37, 3821–3834.

26.

Manning

K.Y.

, Schranz

, Bartha

, Dekaban

G.A.

, Barreira

, Brown

, Fischer

, Asem

, Doherty

T.J.

, Fraser

D.D.

, Holmes

, and Menon

R.S.

(2017). Multiparametric MRI changes persist beyond recovery in concussed adolescent hockey players. Neurology, 89, 2157–2166.

27.

Maugans

T.A.

, Farley

, Altaye

, Leach

, and Cecil

K.M.

(2012). Pediatric sports-related concussion produces cerebral blood flow alterations. Pediatrics, 129, 28–37.

28.

Meier

T.B.

, Bergamino

, Bellgowan

P.S.

, Teague

, Ling

J.M.

, Jeromin

, and Mayer

A.R.

(2016). Longitudinal assessment of white matter abnormalities following sports-related concussion. Hum. Brain Mapp. 37, 833–845.

29.

Murugavel

, Cubon

, Putukian

, Echemendia

, Cabrera

, Osherson

, and Dettwiler

(2014). A longitudinal diffusion tensor imaging study assessing white matter fiber tracts after sports-related concussion. J. Neurotrauma, 31, 1860–1871.

30.

Pasternak

, Koerte

I.K.

, Bouix

, Fredman

, Sasaki

, Mayinger

, Helmer

K.G.

, Johnson

A.M.

, Holmes

J.D.

, Forwell

L.A.

, Skopelja

E.N.

, Shenton

M.E.

, and Echlin

P.S.

(2014). Hockey Concussion Education Project, Part 2. Microstructural white matter alterations in acutely concussed ice hockey players: a longitudinal free-water MRI study. J. Neurosurg. 120, 873–881.

31.

Schranz

A.L.

, Manning

K.Y.

, Dekaban

G.A.

, Fischer

, Jevremovic

, Blackney

, Barreira

, Doherty

T.J.

, Fraser

D.D.

, Brown

, Holmes

, Menon

R.S.

, and Bartha

(2018). Reduced brain glutamine in female varsity rugby athletes after concussion and in non-concussed athletes after a season of play. Hum. Brain Mapp. 39, 1489–1499.

32.

, Merkley

T.L.

, Wilde

E.A.

, Barnes

, Li

, Chu

Z.D.

, McCauley

S.R.

, Hunter

J.V.

, and Levin

H.S.

(2018). A preliminary report of cerebral white matter microstructural changes associated with adolescent sports concussion acutely and subacutely using diffusion tensor imaging. Brain Imaging Behav. 12, 962–973.

33.

Churchill

N.W.

, Hutchison

M.G.

, Graham

S.J.

, and Schweizer

T.A.

(2020). Baseline vs. cross-sectional MRI of concussion: distinct brain patterns in white matter and cerebral blood flow. Sci. Rep. 10, 1643.

34.

Niogi

S.N.

, Luther

, Kutner

, Shetty

, McCrea

H.J.

, Barnes

, Weiss

, Warren

R.F.

, Rodeo

S.A.

, Zimmerman

R.D.

, Tsiouris

A.J.

, and Härtl

(2019). Increased sensitivity to traumatic axonal injury on postconcussion diffusion tensor imaging scans in National Football League players by using premorbid baseline scans. J. Neurosurg. doi: 10.3171/2019.3.JNS181864.

35.

Wilde

E.A.

, Goodrich-Hunsaker

N.J.

, Ware

A.L.

, Taylor

B.A.

, Biekman

B.D.

, Hunter

J.V.

, Newman-Norlund

, Scarneo

, Casa

D.J.

, and Levin

H.S.

(2020). Diffusion tensor imaging indicators of white matter injury are correlated with a multimodal electroencephalography-based biomarker in slow recovering, concussed collegiate athletes. J. Neurotrauma, 37, 2093–2101.

36.

Muftuler

L.T.

, Meier

T.B.

, Keith

, Budde

M.D.

, Huber

D.L.

, and McCrea

M.A.

(2020). Serial diffusion kurtosis magnetic resonance imaging study during acute, subacute, and recovery periods after sport-related concussion. J. Neurotrauma, 37, 2081–2092.

37.

Manning

K.Y.

, Llera

, Dekaban

G.A.

, Bartha

, Barreira

, Brown

, Fischer

, Jevremovic

, Blackney

, Doherty

T.J.

, Fraser

D.D.

, Holmes

, Beckmann

C.F.

, and Menon

R.S.

(2019). Linked MRI signatures of the brain's acute and persistent response to concussion in female varsity rugby players. Neuroimage Clin. 21, 101627.

38.

Hirad

A.A.

, Bazarian

J.J.

, Merchant-Borna

, Garcea

F.E.

, Heilbronner

, Paul

, Hintz

E.B.

, van Wijngaarden

, Schifitto

, Wright

D.W.

, Espinoza

T.R.

, and Mahon

B.Z.

(2019). A common neural signature of brain injury in concussion and subconcussion. Sci. Adv. 5, eaau3460.

39.

Brett

B.L.

, Wu

Y.C.

, Mustafi

S.M.

, Saykin

A.J.

, Koch

K.M.

, Nencka

A.S.

, Giza

C.C.

, Goldman

, Guskiewicz

K.M.

, Mihalik

J.P.

, Duma

S.M.

, Broglio

S.P.

, McAllister

T.W.

, McCrea

M.A.

, and Meier

T.B.

(2020). The association between persistent white-matter abnormalities and repeat injury after sport-related concussion. Front. Neurol. 10, 1345.

40.

Churchill

N.W.

, Hutchison

M.G.

, Graham

S.J.

, and Schweizer

T.A.

(2019). Mapping brain recovery after concussion: from acute injury to 1 year after medical clearance. Neurology, 93, E1980–E1992.

41.

Churchill

N.W.

, Caverzasi

, Graham

S.J.

, Hutchison

M.G.

, and Schweizer

T.A.

(2019). White matter during concussion recovery: comparing diffusion tensor imaging (DTI) and neurite orientation dispersion and density imaging (NODDI). Hum. Brain Mapp. 40, 1908–1918.

42.

Koerte

I.K.

, Kaufmann

, Hartl

, Bouix

, Pasternak

, Kubicki

, Rauscher

, Li

D.K.B.

, Dadachanji

S.B.

, Taunton

J.A.

, Forwell

L.A.

, Johnson

A.M.

, Echlin

P.S.

, and Shenton

M.E.

(2012). A prospective study of physician-observed concussion during a varsity university hockey season: white matter integrity in ice hockey players. Part 3 of 4. Neurosurgical Focus 33, E3:. 1–7.

43.

Baker

J.G.

, Willer

B.S.

, Dwyer

M.G.

, and Leddy

J.J.

(2020). A preliminary investigation of cognitive intolerance and neuroimaging among adolescents returning to school after concussion. Brain Inj. 34, 818–827.

44.

Zhu

D.C.

, Covassin

, Nogle

, Doyle

, Russell

, Pearson

R.L.

, Monroe

, Liszewski

C.M.

, DeMarco

J.K.

, and Kaufman

D.I.

(2015). A potential biomarker in sports-related concussion: brain functional connectivity alteration of the default-mode network measured with longitudinal resting-state fMRI over thirty days. J. Neurotrauma, 32, 327–341.

45.

Chamard

, Lefebvre

, Lassonde

, and Theoret

(2016). Long-term abnormalities in the corpus callosum of female concussed athletes. J. Neurotrauma, 33, 1220–1226.

46.

Zhang

, Johnson

, Pennell

, Ray

, Sebastianelli

, and Slobounov

(2010). Are functional deficits in concussed individuals consistent with white matter structural alterations: combined FMRI & DTI study. Exp. Brain Res. 204, 57–70.

47.

Mac Donald

C.L.

, Barber

, Wright

, Coppel

, De Lacy

, Ottinger

, Peck

, Panks

, Sun

, Zalewski

, and Temkin

(2019). Longitudinal clinical and neuroimaging evaluation of symptomatic concussion in 10- to 14-year-old youth athletes. J. Neurotrauma, 36, 264–274.

48.

Borich

, Makan

, Boyd

, and Virji-Babul

(2013). Combining whole-brain voxel-wise analysis with in vivo tractography of diffusion behavior after sports-related concussion in adolescents: a preliminary report. J. Neurotrauma, 30, 1243–1249.

49.

Steenerson

, and Starling

A.J.

(2017). Pathophysiology of sports-related concussion. Neurol. Clin. 35, 403–408.

50.

Eierud

, Craddock

R.C.

, Fletcher

, Aulakh

, King-Casas

, Kuehl

, and LaConte

S.M.

(2014). Neuroimaging after mild traumatic brain injury: review and meta-analysis. Neuroimage Clin. 4, 283–294.

51.

Dimou

, and Lagopoulos

(2014). Toward objective markers of concussion in sport: a review of white matter and neurometabolic changes in the brain after sports-related concussion. J. Neurotrauma, 31, 413–424.

52.

Niogi

S.N.

, and Mukherjee

(2010). Diffusion tensor imaging of mild traumatic brain injury. J. Head Trauma Rehabil. 25, 241–255.

53.

Shenton

M.E.

, Hamoda

H.M.

, Schneiderman

J.S.

, Bouix

, Pasternak

, Rathi

, Vu

M.A.

, Purohit

M.P.

, Helmer

, Koerte

, Lin

A.P.

, Westin

C.F.

, Kikinis

, Kubicki

, Stern

R.A.

, and Zafonte

(2012). A review of magnetic resonance imaging and diffusion tensor imaging findings in mild traumatic brain injury. Brain Imaging Behav. 6, 137–192.

54.

Yin

, Li

D.-D.

, Huang

, Gu

C.-H.

, Bai

G.-H.

, Hu

L.-X.

, Zhuang

J.-F.

, and Zhang

(2019). Longitudinal changes in diffusion tensor imaging following mild traumatic brain injury and correlation with outcome. Front. Neural Circuits, 13, 28–28.

55.

Bazarian

J.J.

, Zhong

, Blyth

, Zhu

, Kavcic

, and Peterson

(2007). Diffusion tensor imaging detects clinically important axonal damage after mild traumatic brain injury: a pilot study. J. Neurotrauma, 24, 1447–1459.

56.

Hellewell

S.C.

, Nguyen

V.P.B.

, Jayasena

R.N.

, Welton

, and Grieve

S.M.

(2020). Characteristic patterns of white matter tract injury in sport-related concussion: an image based meta-analysis. Neuroimage Clin. 26, 102253.

57.

Schneider

D.K.

, Galloway

, Bazarian

J.J.

, Diekfuss

J.A.

, Dudley

, Leach

J.L.

, Mannix

, Talavage

T.M.

, Yuan

, and Myer

G.D.

(2019). Diffusion tensor imaging in athletes sustaining repetitive head impacts: a systematic review of prospective studies. J. Neurotrauma, 36, 2831–2849.

58.

Terry

D.P.

, Mewborn

C.M.

, and Miller

L.S.

(2019). Repeated sport-related concussion shows only minimal white matter differences many years after playing high school football. J. Int. Neuropsychol. Soc. 25, 950–960.

59.

McCrea

, Guskiewicz

, Randolph

, Barr

W.B.

, Hammeke

T.A.

, Marshall

S.W.

, Powell

M.R.

, Ahn

K.W.

, Wang

, and Kelly

J.P.

(2013). Incidence, clinical course, and predictors of prolonged recovery time following sport-related concussion in high school and college athletes. J. Int. Neuropsychol. Soc. 19, 22–33.

60.

Alexander

A.L.

, Lee

J.E.

, Lazar

, and Field

A.S.

(2007). Diffusion tensor imaging of the brain. Neurotherapeutics, 4, 316–329.

61.

McCrea

, Broshek

D.K.

, and Barth

J.T.

(2015). Sports concussion assessment and management: future research directions. Brain Inj. 29, 276–282.

62.

Sicard

, Moore

R.D.

, and Ellemberg

(2018). Long-term cognitive outcomes in male and female athletes following sport-related concussions. Int. J. Psychophysiol. 132, 3–8.

63.

Seifert

, and Shipman

(2015). The pathophysiology of sports concussion. Curr. Pain Headache Rep. 19, 36.

64.

Sicard

, Moore

R.D.

, and Ellemberg

(2019). Sensitivity of the cogstate test battery for detecting prolonged cognitive alterations stemming from sport-related concussions. Clin. J. Sport Med. 29, 62–68.

65.

Narayana

, Charles

, Collins

, Tsao

J.W.

, Stanfill

A.G.

, and Baughman

(2019). Neuroimaging and neuropsychological studies in sports-related concussions in adolescents: current state and future directions. Front. Neurol. 10, 538.

66.

Tamnes

C.K.

, Ostby

, Fjell

A.M.

, Westlye

L.T.

, Due-Tonnessen

, and Walhovd

K.B.

(2010). Brain maturation in adolescence and young adulthood: regional age-related changes in cortical thickness and white matter volume and microstructure. Cereb. Cortex, 20, 534–548.

67.

Hagler

D.J.

Jr. , Hatton

, Cornejo

M.D.

, Makowski

, Fair

D.A.

, Dick

A.S.

, Sutherland

M.T.

, Casey

B.J.

, Barch

D.M.

, Harms

M.P.

, Watts

, Bjork

J.M.

, Garavan

H.P.

, Hilmer

, Pung

C.J.

, Sicat

C.S.

, Kuperman

, Bartsch

, Xue

, Heitzeg

M.M.

, Laird

A.R.

, Trinh

T.T.

, Gonzalez

, Tapert

S.F.

, Riedel

M.C.

, Squeglia

L.M.

, Hyde

L.W.

, Rosenberg

M.D.

, Earl

E.A.

, Howlett

K.D.

, Baker

F.C.

, Soules

, Diaz

, de Leon

O.R.

, Thompson

W.K.

, Neale

M.C.

, Herting

, Sowell

E.R.

, Alvarez

R.P.

, Hawes

S.W.

, Sanchez

, Bodurka

, Breslin

F.J.

, Morris

A.S.

, Paulus

M.P.

, Simmons

W.K.

, Polimeni

J.R.

, van der Kouwe

, Nencka

A.S.

, Gray

K.M.

, Pierpaoli

, Matochik

J.A.

, Noronha

, Aklin

W.M.

, Conway

, Glantz

, Hoffman

, Little

, Lopez

, Pariyadath

, Weiss

S.R.

, Wolff-Hughes

D.L.

, DelCarmen-Wiggins

, Feldstein Ewing

S.W.

, Miranda-Dominguez

, Nagel

B.J.

, Perrone

A.J.

, Sturgeon

D.T.

, Goldstone

, Pfefferbaum

, Pohl

K.M.

, Prouty

, Uban

, Bookheimer

S.Y.

, Dapretto

, Galvan

, Bagot

, Giedd

, Infante

M.A.

, Jacobus

, Patrick

, Shilling

P.D.

, Desikan

, Li

, Sugrue

, Banich

M.T.

, Friedman

, Hewitt

J.K.

, Hopfer

, Sakai

, Tanabe

, Cottler

L.B.

, Nixon

S.J.

, Chang

, Cloak

, Ernst

, Reeves

, Kennedy

D.N.

, Heeringa

, Peltier

, Schulenberg

, Sripada

, Zucker

R.A.

, Iacono

W.G.

, Luciana

, Calabro

F.J.

, Clark

D.B.

, Lewis

D.A.

, Luna

, Schirda

, Brima

, Foxe

J.J.

, Freedman

E.G.

, Mruzek

D.W.

, Mason

M.J.

, Huber

, McGlade

, Prescot

, Renshaw

P.F.

, Yurgelun-Todd

D.A.

, Allgaier

N.A.

, Dumas

J.A.

, Ivanova

, Potter

, Florsheim

, Larson

, Lisdahl

, Charness

M.E.

, Fuemmeler

, Hettema

J.M.

, Maes

H.H.

, Steinberg

, Anokhin

A.P.

, Glaser

, Heath

A.C.

, Madden

P.A.

, Baskin-Sommers

, Constable

R.T.

, Grant

S.J.

, Dowling

G.J.

, Brown

S.A.

, Jernigan

T.L.

, and Dale

A.M.

(2019). Image processing and analysis methods for the Adolescent Brain Cognitive Development Study. Neuroimage, 202, 116091.

68.

Jahanshad

, Kochunov

P.V.

, Sprooten

, Mandl

R.C.

, Nichols

T.E.

, Almasy

, Blangero

, Brouwer

R.M.

, Curran

J.E.

, de Zubicaray

G.I.

, Duggirala

, Fox

P.T.

, Hong

L.E.

, Landman

B.A.

, Martin

N.G.

, McMahon

K.L.

, Medland

S.E.

, Mitchell

B.D.

, Olvera

R.L.

, Peterson

C.P.

, Starr

J.M.

, Sussmann

J.E.

, Toga

A.W.

, Wardlaw

J.M.

, Wright

M.J.

, Hulshoff Pol

H.E.

, Bastin

M.E.

, McIntosh

A.M.

, Deary

I.J.

, Thompson

P.M.

, and Glahn

D.C.

(2013). Multi-site genetic analysis of diffusion images and voxelwise heritability analysis: a pilot project of the ENIGMA-DTI working group. Neuroimage, 81, 455–469.

69.

Satchell

E.K.

, Friedman

S.D.

, Bompadre

, Poliakov

, Oron

, and Jinguji

T.M.

(2019). Use of diffusion tension imaging in the evaluation of pediatric concussions. Musculoskelet. Sci. Pract. 42, 162–165.

70.

Virji-Babel

, Borich

M.R.

, Makan

, Moore

, Frew

, Emery

C.A.

, and Boyd

L.A.

(2013). Diffusion tensor imaging of sports-related concussion in adolescents. Pediatr. Neurol. 48, 24–29.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.21 MB

0.33 MB

0.17 MB

0.02 MB

0.68 MB