Neuropsychological Factors Related to College Ice Hockey Concussions

Abstract

We analyzed data from 74 male collegiate hockey players. Each athlete’s season began with a baseline administration of the Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT) neuropsychology test battery. Fourteen athletes sustained a sport-related head injury and were readministered the test to assess the impact of the injury. A significant decrease in performance (compared to baseline) on immediate and delayed word recall and designs followed the first concussion. Following a second sport-related concussion, the 4 affected athletes showed significant decrease in visual motor speed. Performance improved on 2 response speed measures (Ps < .01). More errors occurred during a visual processing/discrimination task and immediate recall of designs declined (Ps < .05). We discuss the results in light of recent work related to the impact of early-life concussions and head injury on late-life consequences, such as chronic traumatic encephalopathy, and more immediate issues such as return-to-play decisions for athletes.

The last decade has seen dramatic increases in research addressing the neuropsychological impacts of sport-related head injuries (especially concussions) in children and young adults. ^{1 –3} It is estimated that concussions represent 9% of all sport-related injuries. ⁴ Because they are often referred to as “mild head injury,” they often go unnoticed by medical professionals despite the fact they often show forces similar to sudden acceleration, sudden deceleration, or both. Finally, and often most devastatingly, concussions that are believed to be cumulative in their impact may affect the person for years to come. ³ These impacts include deficits in cognitive, social, and behavioral functioning. ³ Thus, obtaining a concussion (or concussions) in early life may have a cumulative effect over the course of the person’s mid- and later life. Indeed, there is now data showing the long-lasting negative effects on healthy retired athletes who sustained a sport-related concussion in early adulthood. ⁵

Athletes in several intercollegiate sports are impacted by the devastating effects of concussions. Men’s ice hockey is consistently near the top of the list, yet the effects of concussions sustained by ice hockey players have only recently garnered substantial interest. ⁶ A recent summit of neuropsychological professionals interested in finding possible solutions to reduce the risk, incidence, severity, and consequences of concussions in ice hockey players advocated for a “6-component solution” to address concussions including (1) databases and metrics, (2) recognizing, diagnosing, management, and return to play, (3) player equipment and facilities, (4) prevention and education, (5) rules and enforcement, and (6) communication of outcomes. Consistent with this “call to action,” here we present 6 years’ worth of archival data collected on a sample of men’s college hockey players. We had several predictions. First, following baseline, we expected deficits in all Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT) domains at the occurrence of a first concussion. Second, those players who sustained a second concussion would show increased deficits when compared to their initial concussion and at baseline. Additionally, we are also wanting to demonstrate how a major ice hockey program is following suggestions of the summit to set guidelines for recognition, diagnosis, management, and return to play.

Our results have potentially relevant impact given (1) the cumulative (and often negative) nature of concussions sustained early in life on cognitive and behavioral performance and (2) the continuing impact they could have on late-life functioning as it relates to recent attention given to chronic traumatic encephalopathy (CTE). Ongoing debate in this area suggests that the prevalence potential for CTE and the various vulnerable populations (including college hockey players) involved requires additional research. On a more immediate level, our results also are relevant for issues related to return-to-play decisions in college athletes.

Methods

Participants

In conjunction with our university’s men’s hockey athletic trainer (author M.P.), we were given access to archival data collected over a 6-year period. A total of 74 male collegiate (mean age = 23 years) ice hockey players were followed during this time period.

Procedure

Each player began their season with a baseline administration of the ImPACT. ^7,8 If a player had contact to the head, had concussive symptoms that were not present prior to the contact, and if these symptoms did not clear within 15 minutes, the athlete was not allowed to return to play until symptoms had cleared for a period of 24 hours. At that time, the ImPACT was readministered for a second (or third) time, either that night or the next morning. No players in the sample lost consciousness as a result of their concussion. As mentioned, these data are archival and, as such, there are limitations due to the fact that the data were not collected with the current hypothesis in mind. This archival data analysis study was approved by the Institutional Review Board of the University of North Dakota.

Materials

The measure used was the ImPACT neuropsychology test battery. ⁷ This computer-based test measures multiple aspects of cognitive function including learning, immediate and delayed memory, focused attention, concentration, working memory, reaction time, response inhibition, visual processing speed, and visual–motor speed. The ImPACT also provides composite scores assessing verbal memory, visual memory, reaction time, processing speed, and impulse control. The specific tests include Word Memory, Design Memory, Xs and Os, Symbol Match, Color Match, and Three Letters (TL; Table 1). The test takes approximately 20 minutes to administer, and the stimulus array of each test is randomly varied to minimize test–retest practice effects. The test–retest reliability of the ImPACT has recently been investigated and is good. ⁹

Table 1.

Mean (and SD) ImPACT Scores Subsequent to Concussion.

ImPACT		Baseline (n = 74)	First Concussion (n = 14)	Second Concussion (n = 4)
Visual motor speed	M	40.29	42.19	43.09b
	SD	(6.71)	(7.59)	(4.73)
WM-correct
Distractors delay	M	10.97	9.57a	10.25
	SD	(1.11)	(1.79)	(.96)
WM-Delay Mem
% Correct	M	88.26	82.64a	85.80
	SD	(8.28)	(10.28)	(8.19)
MW-total % correct	M	92.54	89.21a	90.75
	SD	(5.36)	(8.54)	(5.62)
DM-total % correct	M	80.85	76.43 a	68.75
	SD	(11.82)	(11.38)	(13.87)
DM-Delay Mem %
Correct	M	77.99	75.36b	64.75
	SD	(13.22)	(9.83)	(18.68)
DM-Learning %
Correct	M	83.74	77.71	72.75a
	SD	(11.83)	(13.99)	(10.53)
Xs/Os-Errors	M	123.08	126.79	128.50a
	SD	(16.43)	(8.39)	(6.46)
TL averg counted	M	17.10	18.83	18.20b
	SD	(3.81)	(4.36)	(3.31)
TL averg counted correctly	M	16.59	17.56	17.85b
	SD	(4.01)	(4.69)	(2.46)

Abbreviations: DM, Design Memory; ImPACT, Immediate Post-Concussion Assessment and Cognitive Testing; M, mean; SD, standard deviation; TL, Three Letters; WM, Word memory.

Significant results are indicated in boldface.

^a P < .05.

^b P < .01.

Results

Following the first sport-related concussion (n = 14), a statistically significant decrease in performance (compared to baseline) was observed for immediate and delayed memory. Specifically, immediate recall was reduced for both words, t ₁₃ = 2.24, P < .05, and designs, t ₁₃ = 2.72, P < .05. Delayed recall was also reduced. Individuals were less able to remember target words, t ₁₃ = 2.67, P < .05, and designs, t ₁₃ = 3.12, P < .01, at the delayed condition. They were also less able to identify distracter words, t ₁₃ = 2.18, P < .05, at the delayed condition. When a second sport-related concussion was sustained (n = 4), significant decreases in performance were observed on multiple cognitive domains when compared to baseline performance, including reductions in visual–motor speed (Visual Motor Composite, t ₃ = 3.12, P < .01), increased errors on tasks of visual processing discrimination (Xs and Os; t ₃ = −3.36, P < .05), and more errors on immediate recall of designs, t ₃ = 3.60, P < .05. Contrary to predictions, however, performance improved in 2 response speed measures, TL avg counted, t ₃ = − 6.86, P < .01, and TL avg counted correctly, t ₃ = −6.19, P < .01.

Discussion

This study examined the effects of concussions on neurocognitive functioning as assessed by the ImPACT in collegiate hockey players after 1 (n = 14) or 2 (n = 4) concussions. Results indicate that subsequent to the first sustained head injury, players’ performance on tests of memory functioning was reduced, when compared to baseline. Both immediate and delayed memory for verbal (ie, words) and nonverbal (ie, designs) information was affected. Following the second sport-related head injury, players’ performance on immediate recall of nonverbal information (ie, designs) was reduced, as compared to baseline. Players also experienced more difficulty in tasks requiring visual processing and visual discrimination (ie, performance was marked by an increase in errors). Visual motor performance (ie, reaction time in response to visual stimuli) was variable after sustaining a second head injury, as compared to baseline. Some measures indicated that performance decreased, while other measures showed an improvement in performance after a second concussion, contrary to our main prediction. We don’t think this specific data pattern raises issues about the validity of our data or of the validity of the instruments used. Rather, a more likely explanation is that the small number of participants who experienced a second concussion (n = 4) explains this unexpected finding.

Although our results were as predicted for the most part, it is noteworthy to discuss the observation that performance on specific visual–motor tasks may be more influenced by practice effects that may account for the increase in performance on select visual–motor tasks (but see ⁹ ). That is, mean performance actually improved on 2 response speed tasks (TL avg counted and TL avg counted correctly) in the sample of ice hockey players that experienced a second concussion. Performance variability was also impacted, which is to be expected with such a small sample (n = 4). This pattern suggests additional studies may be needed with regard to the validity of some of the ImPACT modules, especially as they relate to possible testing effects. Furthermore, it may also indicate that ImPACT performance postconcussion may be affected by regression to the mean, since this sample consisted only of 4 individuals. Such a pattern could potentially impact a decision for a player to return to playing, with increases in performance suggesting that returning to play may be warranted. It also suggests that second concussions do not always result in performance decrements.

Sport-related head injuries are a growing concern among athletes, coaches, and other professionals, especially with regard to how such injuries sustained early in life could impact late-life functioning. Research has shown that repeated concussions in various sports (eg, football, ice hockey, etc) can have lasting short-term and long-term cognitive effects. ¹⁰ In fact, research ¹⁰ has revealed that collegiate ice hockey players (of which the current sample was made up of) are at 15.5 times greater risk of having a concussion during an actual game when compared to practice. The current study found that many important cognitive functions needed continually throughout the lifespan (immediate memory, delayed memory, reaction time, and visual processing / discrimination) can all be negatively affected by concussions experienced by division 1 collegiate ice hockey players.

The devastating effects of concussions are often cumulative, with declines in cognitive, behavioral, and motor skill performance continuing and progressing over time through mid-life into old age. Thus, some in our sample may be at a greater risk of a more rapid late-life decline when compared to individuals with no such concussions. Recall that 4 of our participants did experience a repeated (second) concussion, and many of the published reports attempting to establish this early-life concussion–late-life decline link are case studies ¹¹ suggesting that follow-up of these 4 participants may be fruitful. Likewise, a history of repeated head injuries (concussive as well as nonconcussive) sustained early in life has been shown to be related to CTE, with symptoms often appearing later in life after a lag period. ¹²

Recent work ⁵ suggests that neuropsychological and motor aspects of performance were impacted in individual’s 3-plus decades after having an early-life concussion. Thus, a relationship between history of sports concussion in early life and negative effects in later life on several cognitive and motor domains exists, and in fact ¹³ suggest that some individuals who sustained early-life and late-life traumatic brain injuries were at an increased risk of dementia. The CTE has gained national and international media, scientific, and medical attention within the last decade due to reports linking concussions sustained in various professional contact sports early in life (ie, boxing, professional football) to neurodegenerative disease later in life. This risk is not limited to professional athletes and may also occur in individuals who are much younger (adolescent, college age) and less experienced (amateur, high school, college). ¹⁴ As mentioned, 4 of our patients did incur a second concussion and, thus, may be at a greater risk of CTE. Follow-up on these individuals could include, according to, ¹⁵ neurological examination, magnetic resonance imaging radiological examination, and neuropsychiatric assessment. Such assessments, as well as inclusion of noninvasive techniques, might then lead to lifetime therapies of clinical relevance which may aid in the early identification of the prevalence of CTE in athletes. Of course, there are also reports ^16,17 cautioning that the causal link between repeated concussion events and CTE is either unproven or unknown, especially in athletes.

The fact that performance on some of the various measures used in the present study got better following a second concussion in some participants speaks to this ongoing controversial nature of early-life concussions and the impacts on later life via CTE. Although we currently have no way of knowing the long-term and potentially long-lasting effects of concussion and CTE in our participants, prospective studies, greater numbers of participants, and good sampling methods will eventually understand the complex relationships between the impact of early-life concussions and late-life effects including the possible development of CTE. Our preliminary results offer an intriguing pattern of performance linking multiple concussions and decrements in some aspects of information processing and neuropsychological function and, like, ¹⁵ suggest that college hockey players are a potentially vulnerable population with regard to the development of CTE later in life. Although reproducible postmortem criteria for CTE diagnosis exist, it is critical to continue to identify risk factors, identify additional diagnostic markers, and adequately assess for CTE prior to death.

In summary, although our data are based on a relatively small samples (n = 14, n = 4), they do suggest that professionals from various areas need to be aware of such data patterns and the potential late-life consequences and the impact these consequences pose for those in the more vulnerable populations including college hockey players. ¹⁸ It is also relevant to address the more immediate issue of how our results relate to return-to-play decisions in college athletes. Our results suggest that college athletes need to be aware of how they might be affected by concussions (especially repeat concussions) and of the potential short-term and long-term (possible CTE) consequences of sport-related concussions experienced early in life.