Concussions are a common and debilitating injury for high school and college athletes.1 In the United States, children and adolescents incur up to 1.9 million sports- and recreation-related concussions every year.2 The rates of diagnosed concussions have increased in recent years in concert with awareness about the potential adverse effects of mild traumatic brain injury (mTBI).1,3,4 As a result, concussion prevention has become an important focus for many professional and amateur sports leagues.5–7 Although there have been many attempts to modify rules, improve player equipment, and find pharmaceutical solutions for concussion prevention and recovery, the effects of medication in an athlete’s system during competition has yet to be thoroughly studied.5,8,9
Athletes have a history of using anti-inflammatory medications to treat illnesses, manage injury-related pain and swelling, and improve healing time in order to accelerate return to play.10,11 As a result, athletes have been shown to have higher utilization rates of both prescription and over-the-counter (OTC) medications than the general population.12–14 Adolescents sometimes use OTC medications to treat pain without medical advice or adult supervision and therefore may subject themselves to unintended consequences.15 Previous research has notably demonstrated that chronic ibuprofen use increases cognitive deficits in rats in an mTBI paradigm, and preinjury aspirin use has been associated with an increased risk of adverse outcomes in intracranial hemorrhage after head trauma.16–18
More recently, however, studies have shown no association between preinjury nonsteroidal anti-inflammatory drug (NSAID) use and exacerbated adverse outcomes in rodent concussion models or in TBI patients.19,20 The purpose of this study was to assess the effects of nonsteroidal anti-inflammatory medications on the incidence, severity, and recovery of concussion in young, amateur athletes.
Methods
Data Collection
Data from 25,815 ImPACT (Immediate Post-Concussion Assessment and Cognitive Testing) concussion tests administered between 2009 and 2019 were obtained through a research agreement with ImPACT Applications Inc. The ImPACT tests were administered at multicenter institutions in Colorado and Florida. Subjects ranged in age from 12 to 22 years, and the sample population for incidence calculations in this study was 11,577. Only athletes with a baseline test were included in the study.
According to the study protocol, baseline ImPACT tests were administered to athletes for comparison when evaluating neurological symptoms after injury. Upon incurring head trauma, athletes were immediately evaluated on the sideline by a trainer, physician, coach, or other responsible faculty member. If concussion could not be ruled out, athletes were sent for ImPACT testing. For the purposes of this study, we define head trauma as an event of initial postinjury ImPACT evaluation (PI test).
ImPACT Testing, Head Trauma, and Concussion
As mentioned above, head trauma is defined here as an event of sideline postinjury evaluation leading to ImPACT testing (PI test). ImPACT test results were used to assess symptom severity by evaluating changes from baseline in five composite scores, as follows: Verbal Memory, Visual Memory, Reaction Time, Processing Speed, and Post-Concussion Symptom Score (Supplemental Table 1). Higher Verbal Memory, Visual Memory, and Processing Speed scores, as well as a lower Reaction Time, indicate better performance. A lower symptom score indicates fewer self-reported symptoms. Concussion was defined according to previously described ImPACT assessment guidelines; that is, head traumas met criteria for concussion if postinjury ImPACT results indicated a significant adverse change from baseline in at least two of the five composite scores. A significant change from baseline was defined as any adverse change greater than the standard error of the difference at the 80% confidence interval for healthy control subjects (Sdiff).21
There were 5211 PI tests conducted following head trauma with a corresponding baseline test to evaluate for concussion symptoms and 6366 baseline tests with no subsequent head trauma. A total of 1863 (35.75%) PI head trauma examinations met the criteria for concussion as outlined above. Of these, 1295 (69.51%) had at least one follow-up (FU) test after the PI test that was used to evaluate concussion recovery.
NSAID and Control Cohorts
All baseline tests were screened for mention of the use of the following medications: Advil, Aleve, aspirin, diclofenac, Excedrin, ibuprofen, meloxicam, Midol, Motrin, and naproxen. The anti-inflammatory agents included in this study were all NSAIDs. Patients were then grouped into one of two cohorts based on self-reported anti-inflammatory drug use at the baseline examination, referred to here as “preinjury use of NSAIDs.” Those who, under the “Current medications (please list)” section of the ImPACT Demographic Questionnaire, reported the use of any of the medications listed above were sorted into the anti-inflammatory (AI) cohort. All remaining subjects were sorted to the control (CT) cohort. The five patients reporting corticosteroid anti-inflammatory use were excluded from the study.
Severity Index
The Severity Index was used to assess injury severity as previously described.22 Changes in composite scores from baseline to the PI test were calculated as the difference between the two scores. If this difference did not exceed Sdiff, it was assumed that there was no change from baseline. To estimate the severity of each injury, the Severity Index was calculated by summing the number of Sdiff above baseline for each significant composite score.
Kaplan-Meier Plots
Kaplan-Meier plots were used to examine recovery from concussion. This analysis was restricted to the 1295 patients with a concussion at PI testing who had FU testing. The event for this survival analysis was defined as FU ImPACT results indicating a significant deviation from baseline in none or one of the composite scores.
Comparisons Within Genders
After statistical analyses comparing the AI and CT cohorts were performed, the subjects were further divided by gender into male and female AI and CT groups. Statistical analyses were repeated to compare AI and CT athletes within each gender.
Incidence and Person-Years
Incidence rates of head trauma and concussion were calculated as events per person-year. One calendar year from the date of a baseline test was considered a person-year. ImPACT tests are considered to be valid for up to 2 years. The Centers for Disease Control and Prevention (CDC) suggests using half this length when calculating person-years for patients who are lost to follow-up.23–26 Thus, all patients lost to follow-up or tested a second time outside of the 2-year window were given 1 person-year for a baseline test. Patients who had further ImPACT tests within the 2-year window were given credit for the full time between their tests plus an additional year if the subsequent test was another baseline.
Statistical Analyses
All statistical analyses were performed with Prism 8.0 (GraphPad Software). Chi-square tests were used to compare incidence rates of head trauma and concussion between the two cohorts as well as all categorical demographic and recovery variables. Unpaired t-tests were performed to compare differences in mean age between cohorts, mean days to recover between PI test and FU test, and deviation from baseline in the five composite scores of the ImPACT test at PI and FU tests. Gehan-Breslow-Wilcoxon tests were used to evaluate significant differences between Kaplan-Meier curves. Multivariable models were built to further assess the effects of NSAID use on concussion incidence and severity. Variables controlled in the models included gender, age, diagnosed learning disability, attention deficit hyperactivity disorder (ADHD), football, history of previous concussions, and autism. For all analyses, α = 0.05.
Results
The mean age (± standard error of the mean) was 15.35 ± 0.015 years for the CT cohort and 15.42 ± 0.092 years for the AI cohort (Table 1). The CT cohort comprised a higher percentage (p < 0.0001) of males (66.30%) than the AI cohort (44.16%) and a significantly greater portion of athletes who played football (p = 0.004). The two cohorts had a similar (p = 0.532) percentage of athletes with a history of at least 2 previous concussions (AI 9.64% vs CT 8.40%). There was no difference between the two groups in terms of the percentage of athletes with ADHD (p = 0.497).
CT and AI cohort characteristics among a sample population of 11,577
| Variable | CT Cohort | AI Cohort | Test Type | p Value |
|---|---|---|---|---|
| No. of subjects | 11,380 | 197 | ||
| No. of males | 7545 (66.30) | 87 (44.16) | Chi-square | <0.0001 |
| Age in yrs | 15.35 ± 0.015 | 15.42 ± 0.092 | t-test | 0.5416 |
| ADHD | 556 (4.89) | 12 (6.03) | Chi-square | 0.497 |
| Diagnosed learning disability | 312 (2.74) | 10 (5.08) | Chi-square | 0.048 |
| Autism | 5 (0.04) | 1 (0.51) | Chi-square | 0.005 |
| Football | 4499 (39.53) | 58 (29.44) | Chi-square | 0.004 |
| Previous concussion history (≥2) | 956 (8.40) | 19 (9.64) | Chi-square | 0.532 |
Values expressed as number (%) or mean ± standard error of the mean, unless indicated otherwise. Boldface type indicates statistical significance.
At baseline in the AI cohort, 42.63% of athletes reported the use of Advil, 32.99% ibuprofen, 6.09% Aleve, 5.08% naproxen, 4.57% Midol, 3.05% Motrin, 2.03% aspirin, 2.03% Excedrin, 0.51% diclofenac, and 2.03% “anti-inflammatory” medication.
There was no significant difference (p = 0.9219) between the two cohorts in the incidence rate of head trauma (Table 2 and Fig. 1 left). There was also no difference (p = 0.7467) between the two groups in the severity of head trauma (CT mean = 3.556, AI mean = 3.372) or the magnitude of change from baseline for any of the five individual ImPACT composite scores at PI testing.
Comparisons of head trauma between cohorts
| Variable | CT Cohort | AI Cohort | Test Type | p Value |
|---|---|---|---|---|
| Person-years | 10461.81 | 171.867 | ||
| Incidence rate per person-year (95% CI) | 0.489 (480.4–499.5) | 0.500 (0.425–0.575) | Chi-square | 0.9219 |
| Incidence, no. of cases | 5125 | 86 | ||
| Severity Index | 3.556 (3.413–3.700) | 3.372 (2.387–4.358) | t-test | 0.7467 |
| Symptom Score | 1.138 (1.086–1.189) | 1.140 (0.7185–1.561) | t-test | 0.9919 |
| Verbal Memory | 0.689 (0.655–0.724) | 0.708 (0.4565–0.9602) | t-test | 0.8896 |
| Visual Memory | 0.403 (0.381–0.426) | 0.3364 (0.1834–0.4894) | t-test | 0.4563 |
| Processing Speed | 0.371 (0.346–0.396) | 0.2769 (0.1176–0.4363) | t-test | 0.3484 |
| Reaction Time | 0.956 (0.898–1.01) | 0.9109 (0.5605–1.261) | t-test | 0.8445 |
ImPACT composite scores are reported as mean deviation from baseline in units of Sdiff (95% CI). All 5211 cases had PI testing.
Head trauma (left) and concussion incidence (right) per 1000 person-years. ns = not significant.
There was no significant difference (p = 0.7201) in the incidence rate of concussion between the AI and CT cohorts (Table 3 and Fig. 1 right). For patients with concussion at PI testing, there was no significant difference (p = 0.4921) between the two groups in the average severity of injury (CT mean = 8.449, AI mean = 7.668) or in the magnitude of change from baseline for any of the five individual ImPACT composite scores.
Comparisons of concussion presentation between cohorts
| Variable | CT Cohort | AI Cohort | Test Type | p Value |
|---|---|---|---|---|
| Person-years | 10461.81 | 171.867 | ||
| Incidence rate per person-year (95% CI) | 0.175 (168.2–182.7) | 0.169 (0.113–0.225) | Chi-square | 0.7201 |
| Incidence, no. of cases | 1834 | 29 | ||
| Severity Index | 8.449 (8.171–8.727) | 7.668 (5.548–9.789) | t-test | 0.4921 |
| Symptom Score | 2.593 (2.493–2.693) | 2.663 (1.752–3.574) | t-test | 0.8642 |
| Verbal Memory | 1.665 (1.593–1.738) | 1.691 (1.138–2.243) | t-test | 0.9313 |
| Visual Memory | 0.959 (0.910–1.009) | 0.954 (0.595–1.314) | t-test | 0.9806 |
| Processing Speed | 0.972 (0.913–1.032) | 0.659 (0.250–1.068) | t-test | 0.1964 |
| Reaction Time | 2.259 (2.120–2.398) | 1.701 (0.914–2.488) | t-test | 0.3237 |
ImPACT composite scores are reported as mean deviation from baseline in units of Sdiff (95% CI). All 1863 PI examinations included in this table met the criteria for concussion.
There was no difference (p = 0.6928) between cohorts in the percentage of patients presenting with concussion at PI who remained concussed according to ImPACT at FU testing (Table 4). No difference between cohorts was found in the severity of injury at FU testing (p = 0.8106, CT mean = 2.514, AI mean = 2.306), nor was there a difference in any of the five individual composite scores. There was also no significant difference (p = 0.6416) in concussion recovery time (Fig. 2). The median concussion recovery time was 8 days in both groups. These results were not influenced by differences in the number of recovery days taken between groups, as there was no significant difference (p = 0.7145) between groups in the number of days between PI and FU testing.
Comparisons of concussion recovery at FU testing
| Variable | CT Cohort | AI Cohort | Test Type | p Value |
|---|---|---|---|---|
| No. of subjects | 1273 | 22 | ||
| % of patients concussed at PI testing w/ a FU test | 69.4% (1273/1834) | 75.9% (22/29) | Chi-square | 0.4540 |
| % of patients w/ concussion at FU testing | 26.5% (337/1273) | 22.7% (5/22) | Chi-square | 0.6928 |
| Median days btwn tests (IQR) | 7 (5–12) | 7.5 (5–12) | t-test | 0.7145 |
| Median recovery time in days | 8.00 | 8.00 | Gehan-Breslow-Wilcoxon | 0.6416 |
| Severity Index | 2.514 (2.293–2.735) | 2.306 (0.014 to 4.600) | t-test | 0.8106 |
| Symptom Score | 0.452 (0.385–0.519) | 0.451 (−0.366 to 1.268) | t-test | 0.9945 |
| Verbal Memory | 0.593 (0.529–0.657) | 0.577 (0.085 to 1.068) | t-test | 0.9473 |
| Visual Memory | 0.436 (0.391–0.481) | 0.154 (−0.022 to 0.331) | t-test | 0.1045 |
| Processing Speed | 0.246 (0.205–0.287) | 0.245 (−0.1244 to 0.615) | t-test | 0.9975 |
| Reaction Time | 0.786 (0.688–0.884) | 0.879 (0.117 to 1.641) | t-test | 0.8089 |
ImPACT composite scores are reported as mean deviation from baseline in units of Sdiff (95% CI). All 1295 cases included in this table had at least one FU test.
Kaplan-Meier plot showing concussion recovery for up to 50 days after concussion incidence at PI testing 1.
In a multivariable regression analysis, NSAID use was not associated with concussion incidence (OR = 1.034, 95% CI 0.66–1.56, p = 0.88). Regression analysis also revealed no association between NSAID use and concussion severity (α = −0.36, 95% CI −2.71 to 1.98, p = 0.79).
Discussion
The present study investigated the effects of preinjury NSAID use on the incidence, severity, and recovery of concussion in young athletes. There were no differences between the AI and CT cohorts in the incidence of head trauma leading to ImPACT evaluation or in the incidence of concussion according to ImPACT results. Additionally, the severity of injury and recovery patterns in the AI cohort were the same as those in the CT group. This zero-effect remained consistent when results were compared between AI and CT athletes of the same gender (Supplemental Tables 2 and 3).
Anti-inflammatory medications showed no effect on the incidence of head trauma or concussion. Previous studies examining the effects of pre–head-injury exposure to anti-inflammatory drugs have provided mixed findings. Earlier studies in adults and rat models found that preinjury use of ibuprofen and aspirin may be associated with an increased risk of adverse outcomes in incidents of head trauma.16,17 More recently, however, studies have found no correlation between aspirin or other NSAID use and adverse outcomes following head trauma.19,20 The current study, to our knowledge, is the first to directly assess the effects of preinjury anti-inflammatory drug use on concussion incidence and severity in adolescent athletes. Its results support the findings of the more recent literature. It is possible that the analgesic properties of NSAIDs could have masked some of the concussion symptoms used to refer athletes for ImPACT testing and therefore decreasing the recorded head trauma and concussion incidences in the AI cohort. However, patients in the AI cohort at PI testing did not report lower symptom severity scores than those in controls, and there was no significant difference in reported headaches at PI or FU testing. Moreover, there were no significant differences between groups in the overall severity index nor in any of the other individual composite scores. Together, these results suggest that NSAIDs are not a modulator of the risk for concussion.
Athletes frequently use anti-inflammatory medications in order to aid recovery from illness and injury and allow for quicker return to competition.10,11 Moreover, health professionals often use anti-inflammatory medications to manage concussions in adolescent athletes.27–29 However, there is evidence that postconcussion overuse of analgesics, including NSAIDs, leads to chronic posttraumatic headaches in adolescents.30 In addition, there are many known adverse risks associated with prophylactic NSAID use, suggesting that there is reason to limit NSAID intake.31 In light of the ongoing opioid epidemic in the United States, there is an effort among healthcare providers to reduce the use of opioids in treating pain. As a result, recent publications have supported multimodal pain therapy models that involve NSAIDs, making it important to better understand the associated risks of NSAID use as treatment plans rely on them more frequently.32–34 Despite the other studied deleterious effects of NSAID use, the present study suggests that the baseline use of NSAIDs does not adversely affect the incidence or severity of concussion in adolescent athletes.
Differences in demographics between the cohorts may have influenced concussion incidence. The AI cohort comprised mostly females, whereas our CT group (and entire data set) was majority male. This discrepancy substantiates previous studies that have found that more women report regular anti-inflammatory drug use than men.35,36 However, the findings in research specific to athletes have been mixed. Some have confirmed greater NSAID use in women, whereas other studies have reported no significant relationship between sex and the use of anti-inflammatory medications.12,13,37 When cohorts in the current study were separated by gender (Supplemental Tables 2 and 3), there were still no significant differences in any of the studied outcomes between the AI and CT groups within either gender. The literature reporting on gender differences in concussion incidence is largely agreed on. Specifically, female athletes demonstrate higher concussion rates than those in males when competing in comparable sports.38–40 This suggests that the incidence of concussion in the AI cohort may be overestimated relative to that in the CT group. However, the higher percentage of females in the AI cohort leads to a significantly lower proportion of football players than in the CT cohort. Young athletes competing in high-contact sports such as football are known to be at greater risk of incurring concussions relative to lower-contact sports.38 This effect would result in an overestimation of the concussion rate in the CT group as compared to the AI group. Ultimately, the results when separating cohorts by gender were consistent with the results found in the combined cohorts, which further validates the findings.
The AI cohort included a significantly greater portion of individuals with a diagnosed learning disability or autism, which may have confounded the results of the study. The literature demonstrates that high school and collegiate athletes with learning disabilities present at baseline with a greater history of concussions according to self-reported history questionnaires.41,42 Iverson et al. speculated that this difference may be a result of slower processing speed and reaction time or differences in reporting.41 However, we found no differences between the AI and CT cohorts in the deviation from baseline of processing speed or reaction time following head trauma or concussion. Additionally, there was no significant difference between our cohorts in concussion history at baseline. Therefore, it is unlikely that differences in learning disability prevalence meaningfully altered the results. Results of a multivariable analysis further verified our findings despite baseline differences in cohort demographics.
The key limitation to this study is that the exact timing and dosage of NSAID use in our AI cohort was unclear. Therefore, we cannot be certain that athletes in the AI cohort had NSAIDs in their system while competing, nor can we ascertain differences between the acute and chronic effects of NSAID consumption. Previous research has found that 33.5% of all Fédération Internationale de Football Association (FIFA) World Cup athletes took NSAIDs before at least two of every three matches.43 Warner et al. further found that student athletes who used NSAIDs for pain prophylaxis and who believed that NSAIDs boosted athletic performance were more likely to be daily users.14 These findings indicate the tendency of athletes to take NSAIDs frequently, specifically before competing. Given that the half-lives of Advil/ibuprofen and Aleve/naproxen, the most used NSAIDs in our cohort, are 2 hours and 12–15 hours, respectively, these drugs may have been active in the athletes’ systems during competition.44,45
Because of the limitations inherent to this study, the absence of an effect of NSAIDs on susceptibility to severe head injury and concussion warrants further validation. Future prospective studies that control for the timing and dosage of NSAID consumption would provide clarity to the results found in this analysis. Exploration of the potential differences between the effects of chronic NSAID use and transient precompetition use on concussion would be beneficial.
This study has additional limitations. Concussion due to non–sports-related accidents were not included. Though concussion was suspected after examination by physicians or athletic trainers, it is not known whether a physician formally diagnosed a concussion in the athletes. The sole use of ImPACT as a proxy for concussion assessment may have decreased sensitivity for the detection of clinical concussion. Additionally, the exact timing and dosage of NSAID consumption in our AI cohort was unclear. Athletes also may have underreported their use of anti-inflammatory medications and could have started taking anti-inflammatory drugs at some point after baseline testing, thus creating potential reporting bias. For recovery analysis, treatment protocols and timing of PI testing were not standardized, which could impact results. We believe differences in timing did not significantly affect results because the timing of PI tests was not statistically different between the cohorts. However, we do not know the exact duration of concussion since the athletes could have recovered in the days prior to the test. Thus, Kaplan-Meier analysis may overestimate recovery time. Patients in both cohorts were also taking other medications, including amphetamines, which may influence concussion incidence and thus the results of this study. Furthermore, some NSAIDs included in this study, such as aspirin, have other physiological effects, including antiplatelet activity, that could influence the data. Other factors that may affect ImPACT scores, such as sleep deprivation or pain, were not controlled for and therefore may be underlying confounders to the results. Finally, estimating person-years and calculating incidence accordingly may select for an absence of effect, increasing the likelihood of finding nonsignificant results.
Missing data may have influenced our findings. Accordingly, Markov Chain Monte Carlo multiple imputation was performed with associated sensitivity analysis of NSAID use regression parameters adjusting for gender, age, diagnosed learning disability, ADHD, football, history of previous concussions, and autism (Supplemental Table 4). Findings remained consistent, demonstrating that NSAID use was not significantly associated with concussion incidence or severity.
Conclusions
This study provides evidence that the preinjury use of NSAIDs does not affect the risk of concussion in young athletes. Furthermore, preinjury NSAIDs did not affect the severity or recovery of concussion as assessed by ImPACT testing in our study population. Future studies should be prospective in order to control for demographic differences as well as drug dosage and timing of consumption.
Disclosures
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.
Author Contributions
Conception and design: Spiera, Dreher, Marayati, Shankar, Gometz, Lovell, Choudhri. Acquisition of data: Hannah, Li, Dreher, Shankar, Gometz, Lovell, Choudhri. Analysis and interpretation of data: Spiera, Hannah, Durbin, Schupper, Choudhri. Drafting the article: Spiera, Hannah. Critically revising the article: Spiera, Hannah, Li, Dreher, Marayati, Durbin, Schupper, Gometz, Lovell, Choudhri. Reviewed submitted version of manuscript: Spiera, Li, Ali. Approved the final version of the manuscript on behalf of all authors: Spiera. Statistical analysis: Hannah, Ali, Durbin. Study supervision: Schupper, Gometz, Lovell, Choudhri.
Supplemental Information
Online-Only Content
Supplemental material is available with the online version of the article.
Supplemental Tables 1–4. https://thejns.org/doi/suppl/10.3171/2021.2.PEDS2115.
Previous Presentations
Presented at the 49th Annual Meeting of the AANS/CNS Section on Pediatric Neurological Surgery, Elevating Health Through Knowledge, December 2020.
References
- 1↑
McKeithan L, Hibshman N, Yengo-Kahn AM, et al. Sport-related concussion: evaluation, treatment, and future directions. Med Sci (Basel). 2019;7(3):E44.
- 2↑
Bryan MA, Rowhani-Rahbar A, Comstock RD. Rivara F; Seattle Sports Concussion Research Collaborative. Sports- and recreation-related concussions in US youth. Pediatrics. 2016;138(1):e20154635.
- 3↑
Zuckerman SL, Kerr ZY, Yengo-Kahn A, et al. Epidemiology of sports-related concussion in NCAA athletes from 2009-2010 to 2013-2014: incidence, recurrence, and mechanisms. Am J Sports Med. 2015;43(11):2654–2662.
- 4↑
Schallmo MS, Weiner JA, Hsu W. Assessing trends in the epidemiology of concussions among high school athletes. Orthop J Sports Med. 2017;5(7 suppl 6):2325967117S2325900439.
- 5↑
Enniss TM, Basiouny K, Brewer B, et al. Primary prevention of contact sports-related concussions in amateur athletes: a systematic review from the Eastern Association for the Surgery of Trauma. Trauma Surg Acute Care Open. 2018;3(1):e000153.
- 6
Baugh CM, Shapiro ZE. Concussions and youth football: using a public health law framework to head off a potential public health crisis. J Law Biosci. 2015;2(2):449–458.
- 7
Navarro SM, Sokunbi OF, Haeberle HS, et al. Short-term outcomes following concussion in the NFL: a study of player longevity, performance, and financial loss. Orthop J Sports Med. 2017;5(11):2325967117740847.
- 8↑
Marshall CM. Sports-related concussion: a narrative review of the literature. J Can Chiropr Assoc. 2012;56(4):299–310.
- 10↑
Lippi G, Franchini M, Guidi GC, Kean WF. Non-steroidal anti-inflammatory drugs in athletes. Br J Sports Med. 2006;40(8):661–663.
- 12↑
Tscholl PM, Vaso M, Weber A, Dvorak J. High prevalence of medication use in professional football tournaments including the World Cups between 2002 and 2014: a narrative review with a focus on NSAIDs. Br J Sports Med. 2015;49(9):580–582.
- 13↑
Alaranta A, Alaranta H, Heliövaara M, et al. Ample use of physician-prescribed medications in Finnish elite athletes. Int J Sports Med. 2006;27(11):919–925.
- 14↑
Warner DC, Schnepf G, Barrett MS, et al. Prevalence, attitudes, and behaviors related to the use of nonsteroidal anti-inflammatory drugs (NSAIDs) in student athletes. J Adolesc Health. 2002;30(3):150–153.
- 15↑
Chambers CT, Reid GJ, McGrath PJ, Finley GA. Self-administration of over-the-counter medication for pain among adolescents. Arch Pediatr Adolesc Med. 1997;151(5):449–455.
- 16↑
Fabbri A, Servadei F, Marchesini G, et al. Antiplatelet therapy and the outcome of subjects with intracranial injury: the Italian SIMEU study. Crit Care. 2013;17(2):R53.
- 17↑
Browne KD, Iwata A, Putt ME, Smith DH. Chronic ibuprofen administration worsens cognitive outcome following traumatic brain injury in rats. Exp Neurol. 2006;201(2):301–307.
- 18
Ivascu FA, Howells GA, Junn FS, et al. Predictors of mortality in trauma patients with intracranial hemorrhage on preinjury aspirin or clopidogrel. J Trauma. 2008;65(4):785–788.
- 19↑
Esquivel AO, Sherman SS, Bir CA, Lemos SE. The interaction of intramuscular ketorolac (Toradol) and concussion in a rat model. Ann Biomed Eng. 2017;45(6):1581–1588.
- 20↑
Zangbar B, Pandit V, Rhee P, et al. Clinical outcomes in patients on preinjury ibuprofen with traumatic brain injury. Am J Surg. 2015;209(6):921–926.
- 21↑
Iverson GL, Lovell MR, Collins MW. Interpreting change on ImPACT following sport concussion. Clin Neuropsychol. 2003;17(4):460–467.
- 22↑
Hannah T, Dreher N, Li AY, et al. Assessing the predictive value of primary evaluation with the Immediate Post-Concussion Assessment and Cognitive Test following head injury. J Neurosurg Pediatr. 2020;26(2):171–178.
- 23
Brett BL, Solomon GS, Hill J, Schatz P. Two-year test-retest reliability in high school athletes using the four- and two-factor ImPACT composite structures: the effects of learning disorders and headache/migraine treatment history. Arch Clin Neuropsychol. 2018;33(2):216–226.
- 24
Echemendia RJ, Bruce JM, Meeuwisse W, et al. Long-term reliability of ImPACT in professional ice hockey. Clin Neuropsychol. 2016;30(2):328–337.
- 25
CDC. Principles of Epidemiology in Public Health Practice: An Introduction to Applied Epidemiology and Biostatistics. 3rd ed. CDC. 2006.Accessed April 19, 2021.https://www.cdc.gov/csels/dsepd/ss1978/SS1978.pdf
- 26
Schatz P. Long-term test-retest reliability of baseline cognitive assessments using ImPACT. Am J Sports Med. 2010;38(1):47–53.
- 27
Petrelli T, Farrokhyar F, McGrath P, et al. The use of ibuprofen and acetaminophen for acute headache in the postconcussive youth: a pilot study. Paediatr Child Health. 2017;22(1):2–6.
- 28
Kinnaman KA, Mannix RC, Comstock RD, Meehan WP III. Management of pediatric patients with concussion by emergency medicine physicians. Pediatr Emerg Care. 2014;30(7):458–461.
- 29
Kinnaman KA, Mannix RC, Comstock RD, Meehan WP III. Management strategies and medication use for treating paediatric patients with concussions. Acta Paediatr. 2013;102(9):e424–e428.
- 30↑
Heyer GL, Idris SA. Does analgesic overuse contribute to chronic post-traumatic headaches in adolescent concussion patients?. Pediatr Neurol. 2014;50(5):464–468.
- 31↑
Warden SJ. Prophylactic use of NSAIDs by athletes: a risk/benefit assessment. Phys Sportsmed. 2010;38(1):132–138.
- 32
Martinez L, Ekman E, Nakhla N. Perioperative opioid-sparing strategies: utility of conventional NSAIDs in adults. Clin Ther. 2019;41(12):2612–2628.
- 33
Wick EC, Grant MC, Wu CL. Postoperative multimodal analgesia pain management with nonopioid analgesics and techniques: a review. JAMA Surg. 2017;152(7):691–697.
- 34
Gross JL, Perate AR, Elkassabany NM. Pain management in trauma in the age of the opioid crisis. Anesthesiol Clin. 2019;37(1):79–91.
- 35↑
Davis JS, Lee HY, Kim J, et al. Use of non-steroidal anti-inflammatory drugs in US adults: changes over time and by demographic. Open Heart. 2017;4(1):e000550.
- 36↑
Zhou Y, Boudreau DM, Freedman AN. Trends in the use of aspirin and nonsteroidal anti-inflammatory drugs in the general U.S. population. Pharmacoepidemiol Drug Saf. 2014;23(1):43–50.
- 37↑
Da Silva ER, De Rose EH, Ribeiro JP, et al. Non-steroidal anti-inflammatory use in the XV Pan-American Games (2007). Br J Sports Med. 2011;45(2):91–94.
- 38↑
Scopaz KA, Hatzenbuehler JR. Risk modifiers for concussion and prolonged recovery. Sports Health. 2013;5(6):537–541.
- 39
Davis-Hayes C, Gossett JD, Levine WN, et al. Sex-specific outcomes and predictors of concussion recovery. J Am Acad Orthop Surg. 2017;25(12):818–828.
- 40
Cheng J, Ammerman B, Santiago K, et al. Sex-based differences in the incidence of sports-related concussion: systematic review and meta-analysis. Sports Health. 2019;11(6):486–491.
- 41↑
Iverson GL, Wojtowicz M, Brooks BL, et al. High school athletes with ADHD and learning difficulties have a greater lifetime concussion history. J Atten Disord. 2020;24(8):1095–1101.
- 42↑
Nelson LD, Guskiewicz KM, Marshall SW, et al. Multiple self-reported concussions are more prevalent in athletes with ADHD and learning disability. Clin J Sport Med. 2016;26(2):120–127.
- 43↑
Tscholl P, Junge A, Dvorak J. The use of medication and nutritional supplements during FIFA World Cups 2002 and 2006. Br J Sports Med. 2008;42(9):725–730.
- 44↑
Miles S. Ibuprofen. In: Enna SJ, Bylund DB, eds.xPharm: The Comprehensive Pharmacology Reference. Elsevier;2007:1–7.
- 45↑
Valentovic M. Naproxen. In: Enna SJ, Bylund DB, eds.xPharm: The Comprehensive Pharmacology Reference. Elsevier;2007:1–6.
