Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Mar 13.
Published in final edited form as: Physiol Meas. 2021 Sep 27;42(9):10.1088/1361-6579/ac2207. doi: 10.1088/1361-6579/ac2207

Executive dysfunction after multiple concussions is not related to cerebrovascular dysfunction

Erin D Ozturk 1, Mary Alexis Iaccarino 2,3,4, Jason W Hamner 1, Stacey E Aaron 1,2, Danielle L Hunt 5,6, William P Meehan III 5,6,7, David R Howell 6,8,9, Can Ozan Tan 1,2,10,*
PMCID: PMC12980787  NIHMSID: NIHMS2147187  PMID: 34450608

Abstract

Objective.

We investigated the relation between prior concussion history and working memory (WM), self-reported cognitive symptom burden, and cerebrovascular function in adolescents and young adults (14–21 years old).

Approach.

We recruited 59 participants, 34 clinically diagnosed with a sports-related concussion and 25 controls. Concussed subjects were studied at baseline (within 28 days of their injury) and eight weeks after, while control subjects only had one assessment. We assessed WM (n-back task up to four-back), and neurovascular coupling (cerebrovascular responses at middle cerebral artery during n-back tasks) using a transcranial Doppler ultrasonograph.

Main results.

There was no significant difference in WM between controls and concussed participants (p = 0.402). However, WM capacity was lower in those who had sustained ⩾3 concussions (7.1% with WM capacity of four) compared to those with their first ever concussion (33.3%) and controls (28.0%, overall p = 0.025). At the sub-acute point (n = 24), self-reported cognitive symptom burden was mostly resolved in all but two participants. Despite the resolution of symptoms, WM performance was not different eight weeks post injury (p = 0.706). Neurovascular coupling was not different between controls and concussed participants regardless of prior concussion history.

Significance.

Up to 20% of concussed individuals experience covert sequelae lasting beyond the resolution of self-reported overt symptoms. How a prior history of concussion impacts the potential for sequelae is not well established, and the underlying mechanisms are unknown. Despite no alterations in neurovascular coupling, a history of prior concussion was associated with significant deficits in WM capacity, and lasted beyond self-reported cognitive symptom resolution.

Keywords: adolescence, neurovascular coupling, concussion

1. Introduction

Data suggests that over 2.5 million concussions of any severity and in all age groups are seen in emergency departments in the U.S. each year (Marin et al 2014) and approximately 65% of these are sustained by children and adolescents (CDC 2011). The vast majority of individuals who sustain a concussion physically recover with little or no intervention within one to two months (Meehan et al 2011, Ledoux et al 2018). Some individuals may experience unresolved cognitive sequelae, including problems with executive function (Fazio et al 2007, Howell et al 2013, McInnes et al 2017), and concussions can have long-term effects that last beyond the recovery from overt symptoms (Kamins et al 2017). However, the data remain equivocal, with earlier studies showing no effect of concussion on cognitive function (McCrea et al 2003, Collie et al 2006). The concern about long-lasting cognitive sequelae of concussion is compounded by evidence suggesting that the sequelae of repetitive injuries may be cumulative (Gronwall and Wrightson, 1975, Guskiewicz et al 2003, Nordstrom et al 2014).

We (Tan et al 2014) and others (Giza and Hovda 2014) have suggested a derangement in the relationship between metabolic demand and cerebral blood flow (CBF), termed ‘neurovascular coupling,’ may account for some of the cognitive symptoms observed after a concussion. This is primarily based on animal data, suggesting that after moderate to severe traumatic brain injuries, local CBF decreases and neurovascular ‘uncoupling’ occurs (Giza and Hovda 2014). Impaired neurovascular coupling, in turn, can have a substantial adverse impact on cognitive function (Jor’dan et al 2017). However, while there is one study that explored neurovascular coupling with a visual stimulus (reading), there are no comparable data investigating neurovascular coupling in response to an executive function task after concussion (Wright et al 2017).

Thus, we sought to investigate how a prior history of concussion impacts self-reported cognitive symptom burden, working memory (WM) function, and cerebrovascular function following a sports-related concussion in adolescents and young adults. We deliberately chose to rely on a limited WM task (n-back), which elicits well-established and well-defined cerebrovascular responses (Owen et al 2005, Ozturk and Tan 2018, Ozturk et al 2020) and hypothesized that (1) WM function would be worse in concussed individuals with a prior history of concussion compared to controls and patients who presented with their first ever concussion, and (2) cerebrovascular responses to WM engagement would be diminished in those who had a history of concussion.

2. Methods

2.1. Participants

We recruited 59 participants (34 with a clinical diagnosis of sports-related concussion, and 25 broadly age- and sex-matched healthy controls) between the ages of 14 and 21 years old (table 1). Participants in the ‘concussion’ group sustained a physician-diagnosed concussion within 28 days of their injury (12.4 ± 4.8 days) and had sustained a concussion during sports or other recreational activities. We defined concussion consistent with the Berlin guidelines, as a brain injury caused by a direct blow to the head, face, neck or elsewhere on the body, resulting in the rapid onset of impairment of neurological function (McCrory et al 2017).

Table 1.

Demographic, clinical, and resting hemodynamic variables.

Control (N = 25) Concussed (N = 34)
 p value
First lifetime concussion (12) Second lifetime concussion (8) Third or greater lifetime concussions (14)
Sex (male) 10 (40.0%) 4 (33.3%) 3 (37.5%) 6 (42.9%) 0.966
Age (years) 18.1 (2.5) 17.9 (2.4) 17.4 (2.1) 17.9 (2.3) 0.906
[14.4–21.5] [14.7–20.7] [15.0–19.8] [15.0–21.9]
Body mass index 22.3 (3.0) 22.4 (2.8) 23.8 (2.2) 23.5 (4.4) 0.544
Total PCSI 46.6 (29.9) 35.7 (17.8) 47.6 (24.4) 0.588
[4–83] [14–60] [17–114]
Cognitive PCSI score 10.3 (7.1) 6.0 (3.1) 13.5 (6.7) 0.063
[0–21] [2–9] [5–25]
Physical PCSI score 15.2 (10.7) 16.1 (8.4) 15.3 (8.7) 0.975
[4–32] [4–25] [3–36]
TSI (days) 12.750 (5.345) 11.375 (3.701) 12.643 (5.153) 0.804
CBFv (cm s−1) 58.7 (8.6) 60.7 (11.0) 61.3 (10.2) 59.6 (12.7) 0.914
Heart rate (bpm) 69.5 (6.8) 73.2 (12.3) 72.4 (12.4) 68.0 (10.0) 0.504
Diastolic BP (mmHg) 59.8 (4.6) 63.1 (6.8) 61.5 (4.1) 64.6 (8.4) 0.110
Systolic BP (mmHg) 106.9 (8.1) 109.7 (13.2) 110.6 (6.8) 107.8 (8.0) 0.697
Mean BP (mmHg) 75.5 (4.9) 78.6 (8.0) 77.9 (3.7) 79.0 (7.9) 0.311
End-tidal CO2 (mmHg) 33.4 (3.1) 33.8 (2.5) 32.7 (2.3) 33.2 (3.1) 0.879
Open in a new tab

p values are based on one-way ANOVA for continuous variables, and χ2 test for categorical variables. Values show mean (%) for categorical variables and mean (S.D.) for continuous variables. Numbers in brackets show the range. Data on self-reported post-concussion symptoms (PCSI scores) are missing for five patients (those >14 days post-injury). Numbers in brackets for PCSI scores show the range. PCSI: post concussion symptom index; TSI: time since injury; CBFv: cerebral blood flow velocity; BP: blood pressure; CO2: carbon dioxide.

All participants in the concussion group were asked for their history of prior concussion(s) (one year before the index concussion or earlier) diagnosed by a healthcare professional. Subsequently, we classified the concussed participants into three groups: first lifetime concussion (for whom this was their first ever concussion), second lifetime concussion (those who have had one diagnosed concussion prior to the index one), and third or greater lifetime concussion (those who have had two or more concussions prior to the index one). All concussed participants visited the laboratory for an identical assessment acutely or sub-acutely after injury and then during a follow-up eight weeks later.

Volunteers for the healthy control group were recruited among those who presented to a sports injury prevention clinic with no complaints. All participants were free from cardiovascular disease, diabetes, cancer, past or current neurological disease, tobacco use, current use of cardioactive drugs, epilepsy, and diagnosed history of migraine headaches. Prior to testing, all participants were asked to abstain from caffeine for 12 h before the study to avoid potential hemodynamic and cerebrovascular effects.

This study was approved by the Partners Institutional Review Board (Protocol #2017P002645). All participants (or one of their parents of those <18 years old) gave written informed consent (and assent if <18 years old). All procedures conformed with the Declaration of Helinski.

2.2. Measurements

Prior to testing, participants diagnosed with a concussion were asked to fill out the post-concussion symptom inventory (PCSI) to assess post-concussion symptoms. In addition to the total PCSI score, two different subscales of PCSI—physical and cognitive—were calculated based on previously established criteria (Sady et al 2014).

After arrival at the laboratory, all participants were instrumented with a standard five-lead ECG (Dash 2000/5000, GE) to monitor a continuous recording of heart rate throughout the study. A CO2 analyzer (VacuMed, United States of America) was used to monitor end-tidal CO2 through a nasal cannula during the baseline and executive function task. Brachial oscillometric blood pressure was used as a standard measure of arterial blood pressure. The Finapres blood pressure system (Finometer, Finapres Medical System, Netherlands), calibrated with the arterial blood pressure from the blood pressure cuff, was used to measure a beat-by-beat arterial pressure in the finger.

A transcranial Doppler ultrasonograph (2 MHz probe; MultiDop T2, DWL, Germany) placed on the right temple on the temporal window to access the middle cerebral artery (MCA) at a depth of 45–65 mm was used to record MCA velocity (MCAv). A Mueller-Moll probe fixation device was used to hold the transcranial Doppler ultrasonograph probe in place.

2.3. Experimental protocols

All protocols were performed while the participants were in a seated position. Following instrumentation, five minutes of resting measures (heart rate, arterial blood pressure, MCAv, and EtCO2) were recorded continuously. Following the resting measures, protocols to assess executive function and neurovascular coupling ensued.

All signals were digitized and stored at 1000 Hz (PowerLab, AD Instruments, Colorado Springs, CO) and stored offline for analysis (LabChart, AD Instruments, Colorado Springs, CO). Arterial pressure and MCAv were decimated to 5 Hz and low pass filtered with a cutoff of 0.4 Hz to provide mean waveforms. Data was analyzed using custom software written in MATLAB (version R2019a).

To assess executive function, we chose the n-back task, a test of WM, which is an important component of executive function. WM is conventionally defined as the ability to temporarily hold on to new information and is limited in capacity. Thus, we chose the n-back task because previous data have shown an impact on WM after concussion (McInnes et al 2017). Moreover, the n-back task has a well-established neurobiology, relies on areas supplied primarily by the MCA, and the time course of the neural activation and blood flow responses to this task are unambiguously defined (Owen et al 2005). Thus, this task is uniquely suited to gain insight on the relationship between executive function and neurovascular coupling.

In the n-back task, the number (i.e., ‘n’) of letters that the participant needs to hold (and update) in their WM progressively increases. In the first task (0-back), a series of letters appeared in succession on a computer screen. Each letter was displayed for 0.5 s, with a 2 s response interval followed by a 0.5 s inter-trial interval indicated by a blank screen. Subjects were asked to click on a button keypad (within two seconds) each time they saw a letter. For the subsequent tasks, subjects were asked to click either the match (left) button or no-match (right) button on the keypad every time they saw a letter, the following letter (1-back), or every other letter (2-back), and so on, up to 4-back. Further details and analysis of the n-back task are thoroughly described in our prior work (Ozturk et al 2020).

To assess neurovascular coupling, we examined the cerebral blood flow velocity responses at the middle cerebral artery (MCAv) to each n-back task as the task difficulty (i.e., ‘n’) progressively increased. For each ‘n’, participants’ average reaction time and accuracy (overall percent correct responses) were recorded. Peak MCAv, time to peak (from the beginning of the task) and where MCAv reached a plateau were statistically identified automatically and calculated separately for each individual (Ozturk and Tan 2018, Ozturk et al 2020). These variables identify aspects of the neurovascular coupling response consistent with previous work (Silvestrini et al 1993, Klingelhöfer et al 1997, Sorond et al 2013). We assessed neurovascular coupling using both MCAv and cerebrovascular conductance (blood flow velocity over arterial blood pressure). The latter provides a measure of changes in cerebral perfusion while accounting for any changes in arterial (thus, perfusion pressure) that may change during task performance due to stress and/or anxiety, and can, in turn, impact cerebral blood flow velocity independently.

To explore the relationship between executive function and neurovascular coupling, we matched the n-back tasks by their difficulty across all individuals. We had all participants complete the n-back up to the 4-back to ensure that all reached their maximal WM capacity. To this end, we chose the ‘n’ that corresponded with the highest (i.e., longest) reaction time for each individual. This effectively eliminates the bias due to the fact that task difficulty varies between individuals. We defined ‘working memory capacity’ and the ‘n’ with the highest reaction time (above 50% accuracy). We relied on reaction time, instead of solely relying on accuracy as an indication of difficulty, because the latter cannot differentiate between the true task failure because of task difficulty and task failure due to inattention (‘boredom’). The stimuli (i.e., letters) for every ‘n’ were randomized to provide different versions for every task to account for possible learning effects during the n-back task. We also calculated so-called d′ as a secondary approach to assess executive function. This measure is a well-known signal detection measure to assess familiarity with a cognitive task while limiting the influence of response bias (Snodgrass and Corwin 1988). Specifically, this measurement is calculated based on the hit rate (the proportion of trials where the stimulus was presented and the subject responded that the stimulus was present) and the false alarm rate (the proportion of trials where the stimulus was not presented yet the subject responded that the stimulus was present). Subsequently, d′ can be calculated by subtracting the z-transformation of the hit rate and false alarm rate (d′ = (H) − (FA)).

2.4. Statistics

All statistical analyses were performed using R Language for Statistical Computing version 4.0.0. All descriptive data was presented as mean ± standard deviation. Continuous variables were compared using one-way ANOVA followed by Tukey’s test for group-level comparisons. Categorical and ordinal variables were compared using the χ2 test. Relations between ordinal variables (e.g., WM capacity stratified by prior concussions) were compared using the Cochran–Mantel–Haenszel test, and the difference between these relations across acute post-injury and eight-week follow-up was tested for significance using the 2 × 2 Woolf test for homogeneity of odds ratios stratified across time points (Woolf, 1955). Differences were considered statistically significant at p = 0.05 level.

3. Results

There were no significant differences in demographic, clinical, or hemodynamic variables (Table 1). It is important to note that, compared to the control group, diastolic and mean arterial pressure appeared to be higher in the concussion groups, though this difference was not statistically significant (Table 1). Additionally, individuals with a third or greater lifetime concussion appeared to have a higher cognitive symptom burden, although the difference did not reach statistical significance (p = 0.063).

There were no significant differences in WM capacity between controls and all concussed participants (p = 0.402; Table 2). However, when we explored WM capacity separately across subgroups (controls and those who had first, second, or third or greater lifetime concussions), the difference approached statistical significance (p = 0.086; table 2) without any difference in age or sex (p = 0.906 and p = 0.966, respectively). This relationship was mostly driven by a lower WM capacity in those with three or greater lifetime concussions (Table 2). In fact, those who had a history of concussion had a significantly lower WM capacity compared to controls (χ2 test, p = 0.049), and the gap widened when those with three or greater lifetime concussions were compared to controls (χ2 test, p = 0.025) (Figure 1).

Table 2.

Working function capacity in controls and those who have a history of no, prior, or two or more prior concussions (diagnosed by a physician).

Control (N = 25) Concussed (N = 34)
 p value
First lifetime concussion (12) Second lifetime concussion (8) Third or greater lifetime concussions (14)
Working memory capacity 0.086
1 0 (0.0%) 0 (0.0%) 0 (0.0%) 3 (21.4%)
2 9 (36.0%) 3 (25.0%) 5 (62.5%) 4 (28.6%)
3 9 (36.0%) 5 (41.7%) 2 (25.0%) 6 (42.9%)
4 7 (28.0%) 4 (33.3%) 1 (12.5%) 1 (7.1%)
d’ 1.13 (0.53) 0.94 (0.52) 1.07 (0.45) 1.06 (0.45) 0.761
% Correct responses 73.6 (10.6) 72.2 (9.5) 73.1 (13.7) 73.9 (10.3) 0.981
Reaction time 0.95 (0.15) 0.99 (0.16) 0.87 (0.13) 0.99 (0.23) 0.469
Open in a new tab

‘Working memory capacity’ was defined as the ‘n’ (during n = 0, 1, 2, 3, and 4 n-back tasks) with the highest reaction time (above 50% accuracy). d′, % correct responses, and reaction times represent those that correspond to the n-back task when n was at maximum working capacity. See text for details. Values show mean (%) for categorical variables and mean (S.D.) for continuous variables.

Figure 1.

Figure 1.

Open in a new tab

Likert plot of percentage of participants in each group at acute post-injury with a given working memory capacity (color bars). Individuals in control group do not have any prior diagnosed concussion, per design. At acute post-injury, p = 0.086 across all groups, p = 0.049 control and none versus one and two or more, and p = 0.025 Ccontrol and none versus two or more. There was no difference in percentage distributions between acute post-injury (shown here) and follow-up (Woolf test on homogeneity of odds ratios, p = 0.706).

PCSI scores were not reported by five participants. All of the remaining participants in the concussion group (n = 29) reported some symptoms (i.e., none were symptom-free). Cognitive, but not physical or total symptom burden (PCSI) followed the same pattern as WM capacity.

24 of the 34 concussed individuals returned for their eight-week follow-up visit. At the follow-up, cognitive (as well as total) PCSI symptom burden was significantly lower compared to their acute post-injury in all but two participants, and 16 did not report any cognitive symptoms (Figure 2). However, despite this apparent improvement in self-reported cognitive symptoms, the relationship between WM capacity and history of prior concussions remained eight weeks later. This relationship was evident across both acute post-injury and follow-up assessments (Cochran–Mantel–Haenszel test, M2 = 12.2, p = 0.057), and was not different between the time points (Woolf test on homogeneity of odds ratios, x2 = 0.14, p = 0.706). Additionally, there were no significant differences in n-back performance between those who recovered at eight weeks and those who did not (p = 0.43). Within the group who did not recover (n = 8), there was no relation between cognitive scores and n-back performance (linear regression, adjusted R2 = 0.04, p = 0.30).

Figure 2.

Figure 2.

Open in a new tab

PCSI cognitive symptom score at acute post-injury and at two-month follow-up. Gray fills (violin plots) show the distribution of the data.

In contrast to our hypothesis, the cerebrovascular responses to WM engagement was not different between the groups (control versus concussed) nor impacted by prior concussion history. There were no statistically significant differences in any of the components of neurovascular coupling (Table 3).

Table 3.

Components of neurovascular coupling across the groups at the acute post-injury point.

Control (N = 25) Concussed (N = 34)
 p value
First lifetime concussion (12) Second lifetime concussion (8) Third or greater lifetime concussions (14)
Cerebral blood flow velocity
Maximum response time (s) 25.2 (12.8) 28.9 (14.7) 22.4 (5.1) 30.8 (10.0) 0.336
Maximum response (cm s−1) 5.47 (6.13) 8.74 (10.35) 4.88 (5.03) 8.2 (6.7) 0.424
Plateau response (cm s−1) 1.97 (5.31) 4.25 (9.26) 1.07 (3.86) 2.8 (5.5) 0.663
Cerebrovascular conductance
Maximum response time (s) 24.5 (11.9) 28.7 (15.4) 20.1 (5.5) 24.6 (9.2) 0.449
Maximum response (cm s−1 mmHg−1) 0.083 (0.114) 0.092 (0.092) 0.107 (0.082) 0.113 (0.164) 0.883
Plateau response (cm s−1 mmHg−1) 0.017 (0.093) 0.041 (0.079) 0.025 (0.077) 0.014 (0.094) 0.864
Open in a new tab

Values show mean (S.D.) for continuous variables.

4. Discussion

Our results demonstrate that for adolescents and young adults, a history of prior concussion has a significant impact on WM capacity—a key component of executive function. This impact on WM function is paralleled by self-reported cognitive symptoms (PCSI score in cognitive domain) but lasts beyond self-reported symptom resolution. In contrast to our hypothesis, differences in WM capacity were not related to neurovascular coupling (table 3). Thus, our results show that there may be an important, yet otherwise unnoticed, cognitive sequelae of concussion, at least in adolescents and young adults, that lasts beyond the acute phase and does not relate to cerebrovascular function.

Most self-reported symptoms resolve within the acute stage of injury. In fact, a previous study showed that cognitive symptoms returned to normal within five to seven days after a sports-related injury (McCrea et al 2003). Consistent with prior work, in our study, only 8 out of the 24 concussed individuals who completed their eight week follow-up assessment reported cognitive symptoms, and of these eight, all but two presented with lower cognitive symptom burden. However, despite the self-reported symptom resolution, some reduction in WM capacity was evident in those who had suffered their second lifetime concussions compared to those who had suffered their first lifetime concussion (figure 1).

Importantly, this reduction was clearly apparent in those who had suffered their third or greater lifetime concussion. Thus, our results add to these prior observations, suggesting a cumulative effect of concussions on WM that may be evident beyond the acute stage injury. This is consistent with an earlier report (Gronwall and Wrightson 1975) and recent systematic review of the literature (Manley et al 2017), suggesting that multiple concussions may be an independent risk factor for cognitive impairment. However, these reports, and our findings, are at odds with an earlier report that found no evidence of cognitive decline due to repetitive head injuries (Collie et al 2006). While the reasons for this discrepancy is not clear, it may be due to our narrow focus on WM instead of a broader exploration of cognitive function. Future research is warranted to further clarify the impact of repetitive concussions on cognitive function.

This cumulative impact appears to be in contrast to prior reports that repetitive head injuries may not lead to clinically meaningful changes on scores obtained from concussion assessment battery (Caccese et al 2019). However, it is important to note that while the differences in WM function we observed were paralleled by self-reported cognitive post-concussive burden acutely, these differences are evident in the sub-acute stage (eight weeks after the acute post-injury study) and despite self-reported symptom resolution. Thus, our results reinforce the notion that while most individuals who experience a concussion recover within a short period after the initial injury, subsequent cognitive sequelae might last longer.

While cognitive impairments, including failures of WM and executive function, are commonly experienced by people in their late 60s or early 70s, they are also found in younger adults after brain injuries. For example, a large cohort study of almost 30,000 adults (>18 years old) with mild traumatic brain injuries, concussions as well as unspecified mild head injuries, reported a three to four fold increase in the risk of dementia, independent of age, sex, and other comorbidities (Lee et al 2013). Our study extends these findings, suggesting that a history of prior concussions may adversely impact WM capacity. Nonetheless, this apparent impact remains to be confirmed in a larger cohort and across different domains of cognitive function.

The pathophysiology underlying these cognitive sequelae remain unclear. Conventional CT or MRI imaging does not show any abnormalities after a concussion (Mayer et al 2017). Therefore, we reasoned that other mechanisms must play a role, and that alterations in regional cerebrovascular function may be one of these mechanisms (Tan et al 2014). In contrast to our hypothesis, we did not find a difference in cerebrovascular responses to WM engagement between the groups.

Prior studies suggest a link between neurovascular coupling (assessed the same way as in the current study) and executive function (Sorond et al 2011, Sorond et al 2013). Therefore, the lack of a contribution of neurovascular coupling is surprising. It is possible that concussion may impact neural oxygen extraction and consumption (Champagne et al 2020), and this alteration in neural energetics (e.g., extraction fraction) rather than oxygen supply per se may result in a decline in executive function (Medow et al 2017). Alternatively, it may be alterations in neural structure which are not apparent in conventional CT and MR neuroimaging. In fact, studies using advanced MR imaging suggests that traumatic axonal injury (represented by damage to white matter) may represent a fundamental feature of not only moderate (MacDonald et al 2007) but also mild (Inglese et al 2005) brain injuries, although counter-evidence also exists (Shenton et al 2012). Lastly, it is possible that the peripheral impact of concussion may play an indirect role. For example, autonomic dysfunction after brain injuries has been observed (Lehrer et al 1989), which persists beyond clinical symptom resolution (Esterov and Greenwald 2017) and relates to the extent of injury exposure (Hilz et al 2017). In turn, autonomic function can directly impact cognitive function (Saint Martin et al 2015). However, we did not explore neural energetics, axonal structure, nor autonomic function, and cannot comment on these possibilities.

Our study has several other limitations. We do not have detailed history on participants’ prior concussion history beyond the number of diagnoses by a healthcare professional. Therefore, we could not account for the exact timing of prior injuries (relative to the index one) and associated symptom burden at the time of prior injury. Second, our sample size was determined to provide sufficient statistical power to compare the two groups (control versus concussed), and it is possible that the relatively small number of individuals in each subgroup (first, second, and third or more lifetime concussion) may have biased our results. Given the relatively large effect we observed in the two groups between the controls (n = 25) and those with third or more lifetime concussions (n = 14, our main result, see also figure 1) is based on 49 observations, we do not anticipate any bias due to our sample size. Nonetheless, future work is needed in bigger cohorts. Lastly, we relied on one specific test to assess a small yet fundamental component of executive function and did not use a comprehensive neurocognitive battery. This was a necessary and deliberate choice to be able to assess neurovascular coupling in tandem and without ambiguity. Thus, our results on cognitive function are limited to WM capacity only.

The varying cognitive maturation levels may impact the n-back performance, and thus, may significantly influence the overall interpretation of the results from this study. The age ranges of participants within each group were not different between the groups (p = 0.906; table 1), while the n-back responses were. Furthermore, n-back performance within the control group was not different between those who are older than 18 years and those who are 17 or younger (chi-squared = 0.91, p = 0.64). Thus, the results of the sub-analysis by age groups suggested that the n-back performances of the participants were not directly affected by age, regardless of the exact developmental mechanisms/stage. Future research should address the possible physio-pathological implications of concussion and responsiveness to challenges such as the n-back test at different stages of development in teenagers and young adults. To further rule out the possibility of bias, we repeated our main analyses with a subgroup (17 years of age or younger). This sub-analysis showed that our main result (that a history of prior concussion has a significant impact on WM capacity) is not influenced by age; data from only those 17 and under lead to a p = 0.048 (versus 0.025 with the whole cohort). These results clearly demonstrate that age is not likely to be a source of major bias in the analysis; our results and interpretation is consistent regardless of whether or not the whole cohort is used. While these findings demonstrate the rigor and consistency of our findings, we do acknowledge that age is a potential limitation of the study.

Despite its limitations, our study provides an attempt to quantify the contribution of prior concussion history to acute post-concussion symptom burden and underlying mechanisms. Our results suggest that a history of prior concussion may have a significant impact on WM capacity which lasts beyond self-reported cognitive symptom burden, but that this impact cannot be explained by an alteration in neurovascular coupling. Future studies should explore this apparent impact of prior concussions with a more comprehensive neurocognitive battery and using a broader set of tools to assess underlying physiology.

Funding Statement

This research was funded by Eunice Kennedy Shriver National Institute of Child Health & Human Development (R03HD094560, DRH) and the National Institute of Neurological Disorders and Stroke (R03NS106444, COT).

Footnotes

Author Disclosure Statement

Erin D. Ozturk, Mary Alexis Iaccarino, J. W. Hamner, Danielle L. Hunt, William P. Meehan III, and David R. Howell have no conflicts of interest to disclose. Can Ozan Tan served as a data science consultant for Lokavant Inc. and received consultancy fees for work unrelated to this research.

Data availability statement

N.A

References

  1. Caccese JB, Best C, Lamond LC, Difabio M, Kaminski TW, Watson D, Getchell N and Buckley TA 2019. Effects of repetitive head impacts on a concussion assessment battery Med. Sci. Sports Exerc. 51 1355–61 [DOI] [PubMed] [Google Scholar]
  2. CDC 2011. Nonfatal traumatic brain injuries related to sports and recreation activities among persons aged <19 years - United States, 2001 – 2009 CDCMorbidity and Mortality Weekly Report 60 1337–42 [PubMed] [Google Scholar]
  3. Champagne AA, Coverdale N, Ruis JF, Mark C and Cook DJ 2020. Compromised resting cerebral metabolism after sport-related concussion: a calibrated MRI study Neurology 95 133–146 [DOI] [PubMed] [Google Scholar]
  4. Collie A, McCrory P and Makdissi M 2006. Does history of concussion affect current cognitive status? Br. J. Sports Med. 40 550–1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Esterov D and Greenwald BD 2017. Autonomic dysfunction after mild traumatic brain injury Brain Sci. 7 100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Fazio VC, Lovell MR, Pardini JE and Collins MW 2007. The relation between post concussion symptoms and neurocognitive performance in concussed athletes NeuroRehabilitation 22 207–16 [PubMed] [Google Scholar]
  7. Giza CC and Hovda DA 2014. The new neurometabolic cascade of concussion Neurosurgery 75 S24–33 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Gronwall D and Wrightson P 1975. Cumulative effect of concussion Lancet 2 995–7 [DOI] [PubMed] [Google Scholar]
  9. Guskiewicz KM, McCrea M, Marshall SW, Cantu RC, Randolph C, Barr W, Onate JA and Kelly JP 2003. Cumulative effects associated with recurrent concussion in collegiate football players: the NCAA concussion Study JAMA 290 2549–55 [DOI] [PubMed] [Google Scholar]
  10. Hilz MJ, Wang R, Markus J, Ammon F, Hosl KM, Flanagan SR, Winder K and Koehn J 2017. Severity of traumatic brain injury correlates with long-term cardiovascular autonomic dysfunction J. Neurol. 264 1956–67 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Howell D, Osternig L, Van Donkelaar P, Mayr U and Chou LS 2013. Effects of concussion on attention and executive function in adolescents Med. Sci. Sports Exerc. 45 1030–7 [DOI] [PubMed] [Google Scholar]
  12. Inglese M, Makani S, Johnson G, Cohen BA, Silver JA, Gonen O and Grossman RI 2005. Diffuse axonal injury in mild traumatic brain injury: a diffusion tensor imaging study J. Neurosurg. 103 298–303 [DOI] [PubMed] [Google Scholar]
  13. Jor’dan AJ, Poole VN, Iloputaife I, Milberg W, Manor B, Esterman M and Lipsitz LA 2017. Executive network activation is linked to walking speed in older adults: functional MRI and TCD ultrasound evidence from the MOBILIZE boston study J. Gerontol. A Biol. Sci. Med. Sci. 72 1669–75 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kamins J et al. 2017. What is the physiological time to recovery after concussion? a systematic review Br. J. Sports Med. 51 935–40 [DOI] [PubMed] [Google Scholar]
  15. Klingelhöfer J, Matzander G, Sander D, Schwarze J, Boecker H and Bischoff C 1997. Assessment of functional hemispheric asymmetry by bilateral simultaneous cerebral blood flow velocity monitoring J.Cereb. Blood Flow Metab. 17 577–85 [DOI] [PubMed] [Google Scholar]
  16. Ledoux AA et al. 2018. Natural progression of symptom change and recovery from concussion in a pediatric population JAMA Pediatr. 173 e183820. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lee YK, Hou SW, Lee CC, Hsu CY, Huang YS and Su YC 2013. Increased risk of dementia in patients with mild traumatic brain injury: a nationwide cohort study PLoS One 8 e62422. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Lehrer PM, Groveman A, Randolph C, Miller MH and Pollack I 1989. Physiological response patterns to cognitive testing in adults with closed head injuries Psychophysiology 26 668–75 [DOI] [PubMed] [Google Scholar]
  19. Mac Donald C L, Dikranian K, Song SK, Bayly PV, Holtzman DM and Brody DL 2007. Detection of traumatic axonal injury with diffusion tensor imaging in a mouse model of traumatic brain injury Exp. Neurol. 205 116–31 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Manley G et al. 2017. A systematic review of potential long-term effects of sport-related concussion Br. J. Sports Med. 51 969–77 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Marin JR, Weaver MD, Yealy DM and Mannix RC 2014 Trends in visits for traumatic brain injury to emergency departments in the United States JAMA 311 1917–9 [DOI] [PubMed] [Google Scholar]
  22. Mayer AR, Quinn DK and Master CL 2017. The spectrum of mild traumatic brain injury: a review Neurology 89 623–32 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. McCrea M, Guskiewicz KM, Marshall SW, Barr W, Randolph C, Cantu RC, Onate JA, Yang J and Kelly JP 2003. Acute effects and recovery time following concussion in collegiate football players: the NCAA concussion study JAMA 290 2556–63 [DOI] [PubMed] [Google Scholar]
  24. McCrory P et al. 2017. Consensus Statement on Concussion in Sport-the 5th International Conference on Concussion in Sport Held in Berlin (October 2016(Br J Sports Med) [DOI] [PubMed]
  25. McInnes K, Friesen CL, MacKenzie DE, Westwood DA and Boe SG 2017. Mild Traumatic Brain Injury (mTBI) and chronic cognitive impairment: a scoping review PLoS One 12 e0174847. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Medow MS, Kothari ML, Goetz AM, O’Donnell-Smith MB, Terilli C and Stewart JM 2017. Decreasing cerebral oxygen consumption during upright tilt in vasovagal syncope Physiol. Rep. 5 e13286. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Meehan WP 3rd, d’Hemecourt P, Collins CL and Comstock RD 2011. Assessment and management of sport-related concussions in United States high schools Am. J. Sports Med. 39 2304–10 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Nordstrom P, Michaelsson K, Gustafson Y and Nordstrom A 2014. Traumatic brain injury and young onset dementia: a nationwide cohort study Ann. Neurol. 75 374–81 [DOI] [PubMed] [Google Scholar]
  29. Owen AM, McMillan KM, Laird AR and Bullmore E 2005. N-back working memory paradigm: a meta-analysis of normative functional neuroimaging studies Hum. Brain Mapp. 25 46–59 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Ozturk ED, Lapointe MS, Kim DI, Hamner JW and Tan CO 2020. Impact of 6-month exercise training on neurovascular function in spinal cord injury Med. Sci. Sports Exerc. 53 38–46 [DOI] [PubMed] [Google Scholar]
  31. Ozturk ED and Tan CO 2018. Human cerebrovascular function in health and disease: insights from integrative approaches J. Physiol. Anthropol. 37 4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Sady MD, Vaughan CG and Gioia GA 2014. Psychometric characteristics of the postconcussion symptom inventory in children and adolescents Arch. Clin. Neuropsychol. 29 348–63 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Saint Martin M, Roche F, Thomas-Anterion C, Barthelemy JC and Sforza E 2015. Eight-year parallel change in baroreflex sensitivity and memory function in a sample of healthy older adults J. Am. Geriatr. Soc. 63 270–5 [DOI] [PubMed] [Google Scholar]
  34. Shenton ME et al. 2012. A review of magnetic resonance imaging and diffusion tensor imaging findings in mild traumatic brain injury Brain Imaging Behav. 6 137–92 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Silvestrini M, Caltagirone C, Cupini LM, Matteis M, Troisi E and Bernardi G 1993. Activation of healthy hemisphere in poststroke recovery. A transcranial Doppler study Stroke 24 1673–7 [DOI] [PubMed] [Google Scholar]
  36. Snodgrass JG and Corwin J 1988. Pragmatics of measuring recognition memory: applications to dementia and amnesia J. Exp. Psychol. Gen. 117 34–50 [DOI] [PubMed] [Google Scholar]
  37. Sorond FA, Hurwitz S, Salat DH, Greve DN and Fisher ND 2013. Neurovascular coupling, cerebral white matter integrity, and response to cocoa in older people Neurology 81 904–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sorond FA et al. 2011. Neurovascular coupling is impaired in slow walkers: the MOBILIZE Boston Study Ann.Neurol. 70 213–20 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Tan CO, Meehan WP III, Iverson GL and Taylor JA 2014. Cerebrovascular regulation, exercise, and mild traumatic brain injury Neurology 83 1665–72 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Woolf B 1955. On estimating the relation between blood group and disease Ann. Hum. Genet. 19 251–3 [DOI] [PubMed] [Google Scholar]
  41. Wright AD, Smirl JD, Bryk K and van Donkelaar P 2017. A prospective transcranial doppler ultrasound-based evaluation of the acute and cumulative effects of sport-related concussion on neurovascular coupling response dynamics J. Neurotrauma 34 3097–106 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

N.A

RESOURCES