Sensitivity and Specificity of Exercise Intolerance on Graded Exertion Testing for Diagnosing Sport-Related Concussion: A Systematic Review and Exploratory Meta-Analysis

Abstract

There is no single gold standard test to diagnose sport-related concussion (SRC). Concussion-related exercise intolerance, that is, inability to exercise to the individual's appropriate level due to exacerbation of concussion-like symptoms, is a frequent finding in athletes early after SRC that has not been systematically evaluated as a diagnostic test of SRC. We performed a systematic review and proportional meta-analysis of studies that evaluated graded exertion testing in athletes after SRC. We also included studies of exertion testing in healthy athletic participants without SRC to assess specificity. Pubmed and Embase were searched in January 2022 for articles published since 2000. Eligible studies included those that performed graded exercise tolerance tests in symptomatic concussed participants (> 90% of subjects had an SRC, seen within 14 days of injury), at the time of clinical recovery from SRC, in healthy athletes, or both. Study quality was assessed using the Newcastle-Ottawa Scale. Twelve articles met inclusion criteria, most of which were of poor methodological quality. The pooled estimate of incidence of exercise intolerance in participants with SRC equated to an estimated sensitivity of 94.4% (95% confidence interval [CI]: 90.8, 97.2). The pooled estimate of incidence of exercise intolerance in participants without SRC equated to an estimated specificity of 94.6% (95% CI: 91.1, 97.3). The results suggest that exercise intolerance measured on systematic testing within 2 weeks of SRC may have excellent sensitivity for helping to rule in the diagnosis of SRC and excellent specificity for helping to rule out SRC. A prospective validation study to determine the sensitivity and specificity of exercise intolerance on graded exertion testing for diagnosing SRC after head injury as the source of symptoms is warranted.

Introduction

Sport-related concussion (SRC) is a form of mild traumatic brain injury (mTBI) and a significant public health concern.^1,2 The Centers for Disease Control and Prevention estimates that up to 3.8 million concussions occur annually in the United States from sports and recreational activity, although this number may be an underestimate.³ The clinical presentation of SRC includes a wide range of physical, behavioral, cognitive, sleep, somatic, and/or emotional signs and symptoms, unexplained by another cause (e.g., medications or cervical trauma), that typically appear within minutes or hours of injury, and may spontaneously resolve within days. Hence, many SRC may not be recognized or reported.² There has been a lack of consensus on the diagnostic criteria for SRC,⁴ although the field has recently moved toward a more standardized assessment and definition.² Nevertheless, SRC diagnosis remains a clinical one with no single gold standard test to objectively confirm the diagnosis. Diagnosis is made using a combination of self-reported symptoms that can be linked to a concussive head injury,⁵ impairments on a concussion-focused physical examination,^6,7 and additional tests that may include neurocognitive testing,⁸ reaction time,⁹ and other measures.¹⁰

The ability to exercise without return/worsening of symptoms is required for athletes to successfully complete a graduated return-to-play protocol before unrestricted return to sport.² This can be conceptualized as restoration of normal exercise tolerance after SRC. Concussion-related exercise intolerance, that is, inability to exercise to the individual's appropriate level due to exacerbation of concussion-like symptoms and/or excessive fatigue,^11,12 is a frequent finding in athletes early after SRC that has been characterized as a physiological biomarker of SRC.^13,14 Data from the measurement of the degree of exercise intolerance on graded exertion testing within 10 days of injury are used to prescribe individualized sub-symptom threshold aerobic exercise as treatment after SRC,¹⁵ which has been shown to reduce the duration of symptoms and the incidence of persisting post-concussive symptoms (PPCS) beyond one month.¹⁶ Additionally, the degree of exercise intolerance in the first 10 days after SRC, that is, the lower the heart rate threshold (HRt) at which more than mild symptom exacerbation occurs, has been associated with longer duration of recovery when compared with those who have a higher HRt (i.e., better exercise tolerance).^17,18

Diagnostic measures provide information about the presence or absence of a condition, in this case SRC. Exercise intolerance as a diagnostic test of SRC has not been systematically evaluated using traditional cross-sectional studies. To assess the sensitivity and specificity of exercise intolerance for the diagnosis of SRC, we performed a systematic review of studies that evaluated graded exertion testing in symptomatic athletes within 14 days of injury. Because there were limited publications that included exercise tolerance on graded exertion testing in both injured participants and in healthy controls, we performed an exploratory proportional meta-analysis, primarily of single-arm studies with emphasis on homogeneity of included samples to estimate the sensitivity and specificity of exercise intolerance on graded exertion testing for the diagnosis of SRC.

To establish accurate and useful estimates in this context, studies must have had clear and near uniformity of definition of exercise intolerance and of SRC. We included studies of graded exertion testing in participants with SRC but excluded those that had exercise intolerance as an inclusion criterion. For example, Leddy and colleagues¹⁵ performed graded exertion testing at the initial visit in a randomized controlled trial (RCT) of adolescents after SRC. Exercise intolerance was an inclusion criterion for randomization, so this study was not included in this analysis. We hypothesized that concussion-related exercise intolerance on graded exertion testing would have excellent estimated sensitivity and specificity for the diagnosis of SRC.

Methods

This systematic review was deemed non-human subjects research and an exemption from the need for consent was granted. PRISMA guidelines for systematic reviews were followed.¹⁹

Study definitions and populations

Sensitivity

Test sensitivity is the proportion correctly identified as having the disease divided by the number of subjects in the population with disease.²⁰ For this sample, we used patients with symptomatic SRC who were within 2 weeks of injury.

Specificity

Test specificity is the probability of the test being negative when the disease screened for is absent.²⁰ For this sample, we used patients with recent SRC at the time of clinical clearance (i.e., they had recovered from their SRC and were cleared for unrestricted return to sport), or healthy controls who had performed graded exertion testing as part of a concussion-related research study.

Literature search

Senior authors (MNH, BSW, JJL) identified keywords and a research data scientist with years of experience performing systematic reviews was involved in developing the search strategy. Pubmed and Embase were searched in January 2022 for human subjects research published since 2000 using a combination of the following terms: concussion, exercise test, treadmill test, graded exertion test, exercise tolerance/intolerance, and Buffalo Concussion Treadmill Test (BCTT). Search syntaxes for both databases are included in Supplementary Appendix S1. To assess the effectiveness of our search, we confirmed retrieval of three studies that we knew assessed exercise testing in participants with SRC.^16,21,22 Duplicate articles were removed and a title/abstract screen was performed by two separate reviewers (EL, MSZN). The lead author (MNH) reviewed the results of the first 50 screens from both reviewers as a quality control measure and weekly meetings were held during the screening phase to address any concerns. Cohen's Kappa of inter-rater reliability was calculated.²³ The references of included articles were screened for possible additional articles, and authors of included articles were contacted to ask if they knew of any other articles or abstracts that should be included.

Study selection criteria

Inclusion criteria

Studies that performed graded exercise tolerance tests in participants with acute SRC (< 14 days since injury), in non-concussed athletes, or both. SRC was required to be diagnosed using standard criteria set forth by the Concussion In Sport Group (CISG)² and the study population had to be primarily sport-related injuries (> 90%). For studies that included only healthy controls, the exercise tolerance test protocol had to be identical to an exercise protocol used for concussion assessment, such as the BCTT²⁴ or a modified Balke protocol.²⁵ Non-specific concussion-like symptoms such as headaches and dizziness during graded exercise can be due to a variety of non-pathological conditions in healthy non-athletes, but athletes who play sports are typically not expected to have exercise intolerance. Hence, we included only studies that had athletic controls who participated in organized sport or an equivalent activity (e.g., recreational sport or regular exercise).

Exclusion criteria

Studies were excluded if they did not perform the first exercise tolerance assessment within 14 days of injury, did not provide results of exercise tolerance assessment or comment on the incidence of exercise intolerance, included participants with non-SRC (> 10%), or included injuries that were more severe than concussion/mTBI (e.g., intracranial bleed).

Data extraction

The full text of screened articles was read, and the following data were extracted by three reviewers (MNH, MSZN, AN) using a standard data form: study author, study year, study design, SRC sample size, healthy/control sample size, age, sex, days from injury to assessment, definition of study population, method of exercise tolerance assessment, and incidence of concussion symptom exacerbation (i.e., exercise intolerance). Two researchers (MSZN, AN) shared work and extracted data from half the articles each, and one researcher (MNH) performed the secondary extraction. The lead author reviewed all discrepancies and discussed them with the research assistants during weekly research meetings. No agreement statistic was performed for extracted data.

Risk of bias and level of evidence

Study quality assessment was performed using the Newcastle-Ottawa Scale,²⁶ a validated checklist that assesses risk of bias (ROB) in observational, case control, and cohort studies, and grades the following domains: selection, comparability, and outcome/exposure. Good quality corresponds to 3–4 points in selection, 1–2 points in comparability, and 2–3 points in outcome/exposure. Fair quality corresponds to 2 points in selection, 1–2 points in comparability, and 2–3 points in outcome/exposure. Poor quality corresponds to 0–1 points in selection or 0 points in comparability, or 0–1 points in outcome/exposure. Level of Evidence was determined using guidelines set forth by Melnyk and associates,²⁷which uses a seven-level grading system that ranges from systematic review of RCTs (Level 1) to expert opinion (Level 7). As with the data extraction process, two research assistants split the work followed by a secondary assessment by the lead author. All discrepancies were discussed, and the lead author made the final decision.

Statistical analysis

Individual studies' demographics and incidence of concussion-induced exercise intolerance on graded exertion tests were tabulated and displayed in forest plots. A meta-analysis was performed by calculating the pooled incidence estimate of exercise intolerance with 95% confidence intervals (CIs) using a random effects model with double arcsine transformation (Freeman-Tukey) in participants with and without SRC separately. To include studies with zero incidence, a continuity correction was performed by adding 0.5 to the numerator and denominator of incidence.²⁸ The pooled incidence estimate and CI for participants with SRC is the meta-analysis sensitivity estimate for exercise intolerance and, similarly, one minus the incidence estimate for participants without SRC is the meta-analysis specificity estimate. All analyses were performed using JBI System for the Unified Management, Assessment and Review of Information (JBI SUMARI) (JBI, Adelaide, Australia).²⁹

Results

The database search identified 352 unique articles, and other sources identified one medical school poster. Three hundred five articles were removed during title/abstract screen by two independent reviewers. There were 35 disputed articles among the screeners, which were reviewed by the lead author. The Cohen's Kappa of inter-rater agreement during screening was 0.673, which corresponds to substantial agreement.³⁰ The full texts of 48 articles were read and a total of 12 articles was included in the analysis. Figure 1 presents the PRISMA flowchart of study inclusion and exclusion.

FIG. 1.

PRISMA flowchart.

Table 1 presents the characteristics of the included studies: sex distribution, age, and type of graded exertion test used. Except for three studies,^31
–33 all evaluated either SRC or healthy participants but not both. The study by Cordingley and co-workers³⁴ assessed exercise tolerance in children with symptomatic concussion at their initial assessment (n = 106) and again after they had clinically recovered (n = 65). The initial assessments could not be used because they were >14 days since injury, so we used only the recovered assessment in our analysis for the control arm. The 2021 study by Leddy and colleagues³⁵ had two arms, acutely concussed and a healthy, athletic control arm, but it had exercise intolerance as an inclusion criterion for the concussed sample, so only the control arm was included. Similarly, Bogdanowicz and associates³³ studied two separate populations, an adolescent athlete group and a non-athletic adult group (medical students), so only half the sample (adolescent athletes) was included in this analysis.

Table 1.

Description of Included Studies

Study	Design	Year of publication	Graded exertion test	Has SRC arm	Has control arm	Included sample	Participant sex	Age
Bogdanowicz³³	Case-control	2022	BCMT	Yes	Yes	22	50% male	Range 15–18 years old
Chizuk²²	Cross-sectional	2021	BCTT	Yes	No	163	52% male	Mean 15.2 ± 2 years
Cordingley³⁴	Retrospective chart review	2016	BCTT	No	Yes	65	43% male	Range 11–19 years old
Gaetz³⁸	Quasi-experimental	2009	Progressive cycle ergometer	No	Yes	75	40% male	Range 18–24 years old
Haider³²	Case-control	2020	BCBT/BCTT	Yes	Yes	40	60% male	Mean 15.9 ± 1.1 years
Hall³¹	Case-control	2021	BCBT	Yes	Yes	40	50% male	Mean 15.7 ± 1.4 years
Leddy¹⁷	RCT	2018	BCTT	Yes	No	27	66% male	Mean 15.2 ± 1.5 years
Leddy³⁵	Prospective cohort	2021	BCTT	No	Yes	32	75% male	Mean 15.5 ± 1.4 years
Leddy¹⁶	RCT	2021	BCTT	Yes	No	115	63% male	Mean 15.7 ± 1.4 years
Miutz³⁹	Cross-sectional	2020	BCTT/mSRT	No	Yes	58	100% male	Range 15–16 years old
Powierza³⁷	Cross-sectional	2017	BCTT	No	Yes	22	45% male	Mean 22 ± 2 years
Prim³⁶	Cross-sectional	2021	6-min Chester Step Test	No	Yes	15	87% male	Mean 29.3 ± 6.4 years

BCBT, Buffalo Concussion Bike Test; BCMT, Buffalo Concussion March-in-Place Test; BCTT, Buffalo Concussion Treadmill Test; mSRT: modified Shuttle Run Test; RCT, randomized controlled trial; SRC, sport-related concussion.

Study population

All SRC studies evaluated adolescents, whereas the two studies^36,37 of non-SRC controls included college-aged adults. The CISG² definition of SRC was uniform across studies. The study by Chizuk and colleagues²² included some non-SRC but qualified for inclusion because >90% were SRC. The definition of controls in the studies, however, was less uniform. The Leddy group^31
–33,35 used a very homogenous definition of controls: healthy adolescents (aged 13–18 years) who played at least one organized sport, did not experience a concussion in the past year, and had no more than three life-time concussions. Studies on controls^37
–39 that reported on the normal (non-concussed) response to aerobic exercise for the purpose of sport-concussion management used athletic controls who exercised regularly. To develop an exertion test for active-duty service members, Prim and associates³⁶ studied young, healthy adults who were exercising 3 times a week; hence, they were athletic and thus included in the analysis. Cordingly and co-workers³⁴ performed a longitudinal cohort study of exercise intolerance at the time of symptomatic concussion and after clinical recovery. They did not include a non-concussed sample per se, but we included their recovered time-point exercise assessment in our analysis because these were adolescent athletes who had begun a return-to-play protocol after having been declared clinically recovered from their concussion.

Graded exertion tests

Three studies employed cycle leg ergometers,^31,32,38 two used a 6-min step test,³⁶ and one used a 15-min march-in-place test to assess exercise tolerance after concussion.³³ The most common graded exertion test used to evaluate exercise tolerance after SRC or within the non-concussed population was the BCTT.

Table 2 presents results of the study quality analysis. Except for two RCTs, the studies were either cross sectional, case-control, or cohort (Level of Evidence 4). Most studies received a poor-quality score because they either did not have a comparison arm or had one that we could not include. Thus, they were treated as an observational study in the context of this meta-analysis and scored poorly in the comparability domain.

Table 2.

Study Quality Assessment

Study	Level of Evidence	Selection				Compar-ability	Exposure			Total
Study	Level of Evidence	Q1	Q2	Q3	Q4	Q1	Q1	Q2	Q3	Total
Bogdanowicz 2022³³	4	1	1	1	1	2	1	0	0	7
Chizuk 2021²²	4	1	1	0	0	0	1	0	0	3
Cordingley 2016³⁴	4	1	0	1	1	0	1	1	1	7
Gaetz 2009³⁸	4	1	1	0	0	0	1	0	0	3
Haider 2020³²	4	1	1	1	1	2	1	1	0	8
Hall 2021³¹	4	1	1	1	1	2	1	1	0	8
Leddy 2018¹⁷	2	1	1	0	0	0	1	0	0	3
Leddy 2021a³⁵	4	1	1	0	0	0	1	0	0	3
Leddy 2021b¹⁶	2	1	1	0	0	0	1	0	0	3
Miutz 2020³⁹	4	1	1	0	0	0	1	0	0	3
Powierza 2017³⁷	4	1	1	0	0	0	1	0	0	3
Prim 2021³⁶	4	1	1	0	0	0	1	0	0	3

Level of Evidence grading system: 1—systematic review and meta-analyses of randomized controlled trial; 2—one or more randomized controlled trials; 3—controlled trial; 4—case-control or cohort study; 5—systematic review of descriptive or qualitative studies; 6—single descriptive or qualitative study; 7—expert opinion.

For the specificity determination, Figure 2 presents the forest plot of proportions of participants without SRC who had exercise intolerance. The pooled estimate of incidence of exercise intolerance in participants without SRC was 5.4% (95% CI: 2.7, 8.9). This corresponds to a specificity of 94.6% (95% CI: 91.1, 97.3).

FIG. 2.

Forest plot of proportions of participants without SRC who had exercise intolerance. CI, confidence interval; SRC, sport-related concussion.

For the sensitivity determination, Figure 3 presents the forest plot of proportions of participants with SRC who had exercise intolerance. The pooled estimate of incidence of exercise intolerance in participants with SRC was a sensitivity of 94.4% (95% CI: 90.8, 97.2).

FIG. 3.

Forest plot of proportions of SRC participants with concussion-related exercise intolerance. CI, confidence interval; SRC, sport-related concussion.

Discussion

This systematic review identified studies that allowed us to estimate the sensitivity and specificity of concussion-related exercise intolerance on graded exertion testing for the diagnosis of SRC. Because we did not find a single publication that specifically studied the sensitivity and specificity of exercise intolerance on graded exertion testing in patients acutely after SRC versus a matched control group, we performed a proportional meta-analysis to estimate the incidence of concussion-like exercise intolerance separately in patients with and without SRC. For this reason, the results should be considered exploratory until they can be validated in prospective studies of exercise testing in concussed patients compared with matched controls.

Nevertheless, we found excellent diagnostic sensitivity and specificity for early (1–14 days post-injury) exercise intolerance for SRC, with an estimated sensitivity of 94.4% and an estimated specificity of 94.6%. These values are comparable or superior to those of other common measures used to test for the diagnosis of SRC. One study of a computerized neurocognitive test, the Immediate Post-concussion Assessment and Cognitive Test (ImPACT), reported a sensitivity and specificity of 55.0% and 97.5%, respectively, within 1 day of injury,⁴⁰ whereas another study done within 3 days of injury reported 91.4% sensitivity and 69.1% specificity.⁴¹ The King-Devick (KD) Test, a rapid number reading test, has a sensitivity and specificity of 86% and 90%, respectively, for diagnosing SRC when used on the sideline, although the diagnostic accuracy declines a few days after injury.^42,43 The Sport Concussion Assessment Tool (SCAT), a multi-modal assessment including self-reported symptoms, a neurocognitive screen, a vestibular exam, and a brief oculomotor exam, is reported to have sensitivities of 80.4–89.1% and specificities of 69.0–80.9% when used for immediate assessment on sport sidelines.^44,45

Most included articles had equal sex distribution, with only three being male dominant. The incidence of exercise intolerance in male dominant studies of non-concussed controls (3.1%, 10.3%, and 3.2%) was not significantly different from the pooled mean incidence (5.4%) in the entire group without SRC, suggesting that the incidence of exercise intolerance is similar in healthy males and females. All included SRC studies involved adolescents, whereas the control studies included both adolescents and young adults. We were unable to control for the covariate of age due to the limitations of an exploratory proportional meta-analysis. Gaetz and Iverson³⁸ reported an exercise intolerance rate of 12% in a healthy young adult population (aged 18–24 years), which suggests that non-adolescent athletes may demonstrate more exercise intolerance than younger athletes.

Because all the included studies evaluated athletes, our results may not generalize to non-athletes. One potentially important study that merits further discussion did not meet inclusion criteria because 40% of the sample was non-SRC: Orr and colleagues¹⁸ performed graded exertion testing using the Bruce protocol in concussed adolescents (aged 12–16 years) 5–7 days from injury and reported that only 46% of their population was exercise intolerant, that is, 54% of their acutely concussed sample was not. Exercise tolerant participants had few clinical indicators of concussion, were safely transitioned to a return-to-activity protocol, and all recovered within 10 days. The exercise intolerant participants, however, had high clinical indicators of concussion and experienced prolonged recovery (mean recovery time of 45.6 days). The authors commented on the usefulness of graded exertion testing for concussion management. We contacted the authors to confirm the non-athlete nature of their study population before exclusion. Future studies should assess the incidence of exercise intolerance in the non-sport population.

Exercise intolerance has been considered to be a “physiological biomarker” of concussion¹⁷ because there is evidence of abnormal regulation of cerebral blood flow (CBF) during exercise that is associated with reduced exercise performance.⁴⁶ Conversely, recovery of exercise tolerance has been associated with return to normal brain function during cognitive tasks and normalization of local CBF on functional magnetic resonance imaging (MRI).⁴⁷ The diagnosis of SRC currently is made by a combination of history and physical examination findings after an appropriate mechanism of injury,² and there is a differential diagnosis for the clinical presentation after head injury because post-traumatic symptoms are non-specific and may originate from preexisting conditions and/or a concomitant cervical injury.¹¹

One goal for the field, therefore, is to find more objective means to support the clinical diagnosis of brain injury as the source of post-traumatic symptoms.² Demonstration of exercise intolerance in the days to 2 weeks after SRC may help to increase clinician confidence that the source of symptoms after head trauma is due to brain injury as opposed to other potential sources.¹⁷ Our study compared the incidence of concussion-related exercise intolerance in patients with and without a diagnosed SRC. It did not examine the sensitivity and specificity of exercise intolerance for SRC diagnosis after any head injury. To accomplish this, researchers would have to enroll concussed athletes and compare them with athletic controls with a recent head injury but who were not diagnosed with SRC.

Clinical implications

Sub-symptom threshold aerobic exercise treatment has been proven to facilitate recovery from SRC when it is based upon the individual patient's heart rate threshold identified on systematic exercise testing performed within 10 days of injury.^15,16 Clinicians without their own equipment can partner with physical therapists, athletic trainers, or exercise physiologists, practitioners with the expertise and equipment to conduct exercise testing. When the diagnosis of concussion may be in question because of non-specificity of symptoms or lack of an accurate history, clinicians can employ the principle of “exercise intolerance” to help raise or lower the pre-test probability of the diagnosis without the need for questionably reliable computerized cognitive tests,⁴⁸ sophisticated imaging, or expensive and thus far unproven blood or other fluid biomarkers.

Limitations

The major limitation of this study is that we were unable to perform a traditional comparative meta-analysis and so our results require validation. Another limitation is that our results are subject to selection bias because all the studies of exercise tolerance within 14 days of SRC were from the same group of researchers at the University at Buffalo. This is not necessarily surprising because the concept of exercise tolerance testing in SRC⁴⁹ and of using the data to prescribe sub-symptom threshold aerobic exercise to improve recovery⁵⁰ originated in Buffalo. Our clinic offers graded exertion testing within the initial clinical encounter by experts in exercise rehabilitation for SRC, so patients with noticeable exercise intolerance may preferably seek care there. Our results would not be achievable by clinics that do not have resources to conduct exercise testing. Nevertheless, centers with equipment and that use the “Buffalo protocol” for treatment based on exercise testing in SRC patients have reported good outcomes in those with prolonged symptoms.^51
–53

The generalizability of our findings is limited, and more studies of the sensitivity and specificity of exercise tolerance after SRC should be conducted at other centers. Another limitation is that there is no gold standard diagnostic test for concussion, so the determination of sensitivity and specificity is based upon the clinical diagnosis of concussion by experienced physicians. Lastly, this review is subject to publication bias. We searched two of the largest databases (PubMed and Embase), which encompass 91% of the published literature,⁵⁴ and we contacted sport-concussion researchers who performed graded exertion testing. Still, we were unable to review unpublished work or literature that was not included in these databases.

Conclusions

The results of this proportional meta-analysis suggest that exercise intolerance measured on systematic testing within 2 weeks of SRC has excellent (94%) sensitivity for the diagnosis of SRC and excellent (95%) specificity for helping to rule out SRC. When combined with a careful history and concussion-relevant physical examination, systematic determination of exercise tolerance could significantly raise or lower the pre-test probability of concussion to help clinicians manage their patients more successfully. A prospective validation study to determine the sensitivity and specificity of exercise intolerance to diagnose concussion as the source of symptoms after head injury is warranted.

Footnotes

Transparency,Rigor,and Reproducibility Summary

The methods of this exploratory meta-analysis were guided by the Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines (PRISMA). The screening of articles was facilitated by the web-based software platform JBI SUMARI. Two databases, namely, Pubmed and Embase, were initially searched from 2000 to Jan 20, 2022, with an updated search before submission to confirm no new articles of interest were published that may change our results. Following initial screen for duplicate articles, two authors (EL and MSZN) independently assessed the eligibility of retrieved articles based on titles and abstracts. Data on the 12 included articles have been comprehensively reported in the manuscript and a list of all full-text publications screened for data extraction is provided in the Supplement File. For other details concerning the search process and outcomes, please contact the corresponding author.

Acknowledgments

The authors thank the University at Buffalo Health Sciences Library for helping to develop our search strategy.

Authors' Contributions

The authors contributed as follows. MNH.: writing—original draft (lead); formal analysis (lead); writing—review and editing (equal); methodology. EL: data curation; review and editing (equal). MSZN: data curation; review and editing (equal). AN: data curation; review and editing (equal). HMC: data curation; review and editing (equal). JCM: formal analysis; review and editing (equal). JIM: formal analysis; review and editing (equal); BSW: conceptualization; supervision; review and editing (equal). JJL: conceptualization; supervision; review and editing (equal).

Funding Information

No funding was received for this work.

Author Disclosure Statement

No competing financial interests exist.

Supplementary Material

Supplementary Appendix S1

References

Roos

, Kerr

, Djoko

, et al. Concussion incidence in US high school and collegiate basketball. Brain injury, 2016; 30(5-6):531; doi: 10.3109/02699052.2016.1162060

McCrory

, Meeuwisse

, Dvorak

, et al. Consensus statement on concussion in sport—the 5th International Conference on Concussion in Sport held in Berlin, October 2016. Br J Sports Med, 2017; 51(11):838–847; doi: 10.1136/bjsports-2017-097699

Langlois

, Rutland-Brown

, Wald

. The epidemiology and impact of traumatic brain injury: a brief overview. J Head Trauma Rehabil, 2006; 21(5):375–378

Gliklich

, Leavy

, Karl

, et al. A framework for creating standardized outcome measures for patient registries. J Comp Eff Res, 2014; 3(5):473–480.

Sady

, Vaughan

, Gioia

. Psychometric characteristics of the postconcussion symptom inventory in children and adolescents. Arch Clin Neuropsychol, 2014; 29(4):348–363.

Kontos

, Deitrick

, Collins

, et al. Review of vestibular and oculomotor screening and concussion rehabilitation. J Athl Train, 2017; 52(3):256–261; doi: 10.4085/1062-6050-51.11.05

Haider

, Leddy

, Du

, et al. Practical management: brief physical examination for sport-related concussion in the outpatient setting. Clin J Sport Med, 2020; 30(5):513–517; doi: 10.1097/jsm.0000000000000687

Farnsworth

, 2nd, Dargo

, Ragan

, et al. Reliability of computerized neurocognitive tests for concussion assessment: a meta-analysis. J Athl Train, 2017; 52(9):826–833; doi: 10.4085/1062-6050-52.6.03

Vartiainen

, Holm

, Lukander

, et al. A novel approach to sports concussion assessment: computerized multilimb reaction times and balance control testing. J Clin Exp Neuropsychol, 2015; 1–15; doi: 10.1080/13803395.2015.1107031

10.

Haider

, Leddy

, Pavlesen

, et al. A systematic review of criteria used to define recovery from sport-related concussion in youth athletes. Br J Sports Med, 2018; 52(18):1179–1190; doi: 10.1136/bjsports-2016-096551

11.

Leddy

, Baker

, Merchant

, et al. Brain or strain? Symptoms alone do not distinguish physiologic concussion from cervical/vestibular injury. Clin J Sport Med, 2015; 25(3):237–242.

12.

Kozlowski

, Graham

, Leddy

, et al. Exercise intolerance in individuals with postconcussion syndrome. J Athl Train, 2013; 48(5):627–635.

13.

Leddy

, Wilber

, Willer

. Active recovery from concussion. Curr Opin Neurol, 2018; 31(6):681–686; doi: 10.1097/wco.0000000000000611

14.

Leddy

, Baker

, Kozlowski

, et al. Reliability of a graded exercise test for assessing recovery from concussion. Clin J Sport Med, 2011; 21(2):89–94.

15.

Leddy

, Haider

, Ellis

, et al. Early subthreshold aerobic exercise for sport-related concussion: a randomized clinical trial. JAMA Pediatr, 2019; 173(4):319–325; doi: 10.1001/jamapediatrics.2018.4397

16.

Leddy

, Master

, Mannix

, et al. Early targeted heart rate aerobic exercise versus placebo stretching for sport-related concussion in adolescents: a randomised controlled trial. Lancet Child Adolesc Health, 2021; 5(11):792–799.

17.

Leddy

, Hinds

, Miecznikowski

, et al. Safety and prognostic utility of provocative exercise testing in acutely concussed adolescents: a randomized trial. Clin J Sport Med, 2018; 28(1):13–20.

18.

Orr

, Bogg

, Fyffe

, et al. Graded exercise testing predicts recovery trajectory of concussion in children and adolescents. Clin J Sport Med, 2021; 31(1):23–30; doi: 10.1097/jsm.0000000000000683

19.

Moher

, Liberati

, Tetzlaff

, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med, 2009; 151(4):264–269.

20.

Parikh

, Mathai

, Parikh

, et al. Understanding and using sensitivity, specificity and predictive values. Indian J Ophthalmol, 2008; 56(1):45.

21.

Leddy

, Hinds

, Miecznikowski

, et al. Safety and prognostic utility of provocative exercise testing in acutely concussed adolescents: a randomized trial. Clin J Sport Med, 2018; 28(1):13–20; doi: 10.1097/JSM.0000000000000431

22.

Chizuk

, Willer

, Horn

, et al. Sex differences in the Buffalo Concussion Treadmill Test in adolescents with acute sport-related concussion. J Sci Med Sport, 2021; 24(9):876–880.

23.

Cohen

A coefficient of agreement for nominal scales. Educ Psychol Meas, 1960; 20(1):37–46.

24.

Leddy

, Willer

. Use of graded exercise testing in concussion and return-to-activity management. Curr Sports Med Rep, 2013; 12(6):370–376; doi: 10.1249/jsr.0000000000000008

25.

Wolthuis

, Froelicher

Jr ., Fischer

, et al. New practical treadmill protocol for clinical use. Am J Cardiol, 1977; 39(5):697–700.

26.

Stang

Critical evaluation of the Newcastle-Ottawa Scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol, 2010; 25(9):603–605.

27.

Melnyk

, Fineout-Overholt

Evidence-Based Practice in Nursing & Healthcare: A Guide to Best Practice. Lippincott Williams & Wilkins: 2011.

28.

Ross

Chapter 6–Distributions of Sampling Statistics. In: Introduction to Probability and Statistics for Engineers and Scientists Fifth Edition. Elsevier: Los Angeles, CA; 2014; pp. 218–219.

29.

Piper

System for the unified management, assessment, and review of information (SUMARI). J Med Library Assoc JMLA, 2019; 107(4):634.

30.

McHugh

ML.

Interrater reliability: the kappa statistic. Biochemia Medica, 2012; 22(3):276–282.

31.

Hall

, Haider

, Horn

, et al. Exercise-induced headache in the concussed is associated with dysregulation of cerebral blood flow. Clin J Sport Med, 2021; 31(2):164; doi: 10.1097/JSM.0000000000000912

32.

Haider

, Johnson

, Mannix

, et al. The Buffalo Concussion Bike Test for concussion assessment in adolescents. Sports Health, 2019; 1941738119870189

33.

Bogdanowicz

, MN;

Willer

, BS; Leddy

. Buffalo Concussion March-in-Place Test (BCMT). Buffalo, NY; 2022. https://medicine.buffalo.edu/departments/pharmtox/news-and-events/news.host.html/content/shared/smbs/news/2022/02/medical-student-research-forum-2022-14340.detail.html

34.

Cordingley

, Girardin

, Reimer

, et al. Graded aerobic treadmill testing in pediatric sports-related concussion: safety, clinical use, and patient outcomes. J Neurosurg Pediatr, 2016; 18(6):693–702.

35.

Leddy

, Lesh

, Haider

, et al. Derivation of a focused, brief concussion physical examination for adolescents with sport-related concussion. Clin J Sport Med, 2021; 31(1):7–14; doi: 10.1097/jsm.0000000000000686

36.

Prim

, Davila

, McCulloch

. A pilot study on exertional tasks with physiological measures designed for the assessment of military concussion. Concussion, 2021; 6(1); doi: 10.2217/cnc-2020-0018

37.

Powierza

, Clark

, Hughes

, et al. Validation of a self-monitoring tool for use in exercise therapy. PM R, 2017; 9(11):1077–1084; doi: 10.1016/j.pmrj.2017.03.012

38.

Gaetz

, Iverson

. Sex differences in self-reported symptoms after aerobic exercise in non-injured athletes: implications for concussion management programmes. Br J Sports Med, 2009; 43(7):508–13; doi: 10.1136/bjsm.2008.051748

39.

Miutz

, Black

, Jordan

, et al. Feasibility, reliability, and concurrent validity of a field test of exertion in high school students. Clin J Sport Med, 2020; 30(3):e91; doi: 10.1097/JSM.0000000000000844

40.

Resch

, Brown

, Schmidt

, et al. The sensitivity and specificity of clinical measures of sport concussion: three tests are better than one. BMJ Open Sport Exerc Med, 2016; 2(1):e000012; doi: 10.1136/bmjsem-2015-000012

41.

Schatz

, Sandel

. Sensitivity and specificity of the online version of ImPACT in high school and collegiate athletes. Am J Sports Med, 2013; 41(2):321–326.

42.

Arca

, Starling

, Acierno

, et al. Is King-Devick testing, compared with other sideline screening tests, superior for the assessment of sports-related concussion?: a critically appraised topic. Neurologist, 2020; 25(2):33–37.

43.

Galetta

, Liu

, Leong

, et al. The King-Devick test of rapid number naming for concussion detection: meta-analysis and systematic review of the literature. Concussion, 2016; 1(2):CNC8.

44.

Fuller

, Tucker

, Starling

, et al. The performance of the world rugby head injury assessment screening tool: a diagnostic accuracy study. Sports Med Open, 2020; 6(1):1–12.

45.

Downey

, Hutchison

, Comper

. Determining sensitivity and specificity of the Sport Concussion Assessment Tool 3 (SCAT3) components in university athletes. Brain Inj, 2018; 32(11):1345–1352.

46.

Clausen

, Pendergast

, Wilier

, et al. Cerebral blood flow during treadmill exercise is a marker of physiological postconcussion syndrome in female athletes. J Head Trauma Rehabil, 2016; 31(3):215–224.

47.

Leddy

, Cox

, Baker

, et al. Exercise treatment for post concussion syndrome: a placebo controlled pilot study of changes in fMRIBlood flow and symptoms. Brain Inj, 2012; 26(4-5):388–389; doi: 10.3109/02699052.2012.660091

48.

Broglio

, Ferrara

, Macciocchi

, et al. Test-retest reliability of computerized concussion assessment programs. J Athl Train, 2007; 42(4):509–514.

49.

Leddy

, Kozlowski

, Donnelly

, et al. A preliminary study of subsymptom threshold exercise training for refractory post-concussion syndrome. Clin J Sport Med, 2010; 20(1):21–27.

50.

Willer

, Leddy

, Pendergast

, et al. Use of low-level exercise to improve the physiologic condition of individuals with post concussion syndrome. J Head Trauma Rehabil, 2009; 24(5):397; doi: 10.1097/HTR.0b013e3181bd6769

51.

Chrisman

SPD

, Whitlock

, Mendoza

, et al. Pilot randomized controlled trial of an exercise program requiring minimal in-person visits for youth with persistent sport-related concussion. Front Neurol 019;10:623; doi: 10.3389/fneur.2019.00623

52.

Kurowski

, Hugentobler

, Quatman-Yates

, et al. Aerobic exercise for adolescents with prolonged symptoms after mild traumatic brain injury: an exploratory randomized clinical trial. J Head Trauma Rehabil, 2017; 32(2):79–89; doi: 10.1097/HTR.0000000000000238

53.

Chan

, Iverson

, Purtzki

, et al. Safety of active rehabilitation for persistent symptoms after pediatric sport-related concussion: a randomized controlled trial. Arch Phys Med Rehabil, 2018; 99(2):242–249; doi: 10.1016/j.apmr.2017.09.108

54.

Slobogean

, Verma

, Giustini

, et al. MEDLINE, EMBASE, and Cochrane index most primary studies but not abstracts included in orthopedic meta-analyses. J Clin Epidemiology, 2009; 62(12):1261–1267.

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

0.02 MB