Key Points
Question
What are the early clinical characteristics associated with persistent symptoms 30 days after mild traumatic brain injury (mTBI)?
Findings
In this cohort study of 803 patients presenting at a median of 1.5 hours after an mTBI (Glasgow Coma Scale score ≥13), 9 characteristics were significantly associated with symptoms persisting for 30 days: female sex; elevated body mass index; mechanism of injury; a prior diagnosis of headaches or migraines, depression, or anxiety; and the presence of focal neurological deficits, headache, or multiple computed tomography scans at intake.
Meaning
These findings suggest that patients with any of these characteristics should be identified as candidates for early intervention intended to prevent persistent 30-day postconcussive symptoms and should receive follow-up for mTBI.
This cohort study examines the factors associated with symptoms that persist for 30 days in adult emergency department patients presenting shortly after a mild traumatic brain injury.
Abstract
IMPORTANCE
Traumatic brain injury (TBI) is a common presentation in which it is clinically difficult at presentation to identify patients at risk for persistent symptoms, defined as lasting longer than 30 days. Few studies report the time period from injury to presentation; thus, it may be useful to determine early findings that identify patients who are likely to have persistent 30-day symptoms.
OBJECTIVE
To identify parameters associated with persistent 30-day symptoms in adult emergency department (ED) patients presenting shortly after a mild TBI (mTBI).
DESIGN, SETTING, AND PARTICIPANTS
This cohort analysis used the large and still enrolling HeadSMART II (Head Injury Serum Markers and Multi-modalities for Assessing Response to Trauma II) dataset. HeadSMART II began enrollment in 2021, and all patients with mTBI enrolled as of July 1, 2024, were included in this analysis.
EXPOSURE
A diagnosis of mTBI, defined as a Glasgow Coma Scale score of 13 or higher.
MAIN OUTCOMES AND MEASURES
The primary outcome was mTBI symptoms persisting 30 days after presentation. The 30-day Rivermead Post Concussion Symptoms Questionnaire was used to define persistent symptoms. Wilcoxon rank-sum testing was applied to continuous variables, and the Fisher exact test was applied to categorical variables.
RESULTS
Overall, 803 HeadSMART II patients (median [IQR] age, 41.0 [27.0-57.0] years; 404 male [50.3%]) with 30-day follow-up data were assessed at a median (IQR) of 1.5 (0.7-10.6) hours after their injury. Most patients were employed (565 of 794 patients [71.2%]) and had a fall or head struck by object as the mechanism of injury (387 of 802 patients [48.3%]). Trauma above clavicle (279 of 404 men [69.1%] vs 231 of 398 women [58.0%]) and the presence of fracture (103 of 403 men [25.6%] vs 60 of 396 women [15.2%]) or laceration (156 of 404 men [38.6%] vs 94 of 398 women [23.6%]) occurred more often in men than in women, whereas headaches occurred more often in women than in men (338 of 398 women [84.9%] vs 314 of 401 men [78.3%]). Over time, TBI symptoms declined for all. The presentation characteristics associated with persistent 30-day symptoms included female sex (odds ratio [OR], 2.09; 95% CI, 1.54-2.87); elevated body mass index (ie, 32.6; OR, 1.03; 95% CI, 1.01-1.06); mechanism of injury, including fall (OR, 2.87; 95% CI, 1.28-7.67), vehicular crash (OR, 3.79; 95% CI, 1.68-10.19), and abuse (OR, 3.67; 95% CI, 1.43-10.73); a prior diagnosis of headaches or migraines (OR, 2.76; 95% CI, 1.94-3.92), depression (OR, 2.25; 95% CI, 1.64-3.10), or anxiety (OR, 2.52; 95% CI, 1.83-3.48); and the presence of focal neurological deficits (OR, 1.52; 95% CI, 1.10-2.10), headache (OR, 2.08; 95% CI, 1.35-3.32), or multiple computed tomography scans (OR, 2.22; 95% CI, 1.49-3.36) at intake.
CONCLUSIONS AND RELEVANCE
In the 803 patients with mTBI with complete 30-day data and a median initial presentation of 90 minutes after injury, those with any of the 9 identified baseline characteristics were more likely than patients without those characteristics to have persistent postconcussive symptoms at 30 days.
Introduction
In 2021 there were an estimated 139 million US emergency department (ED) visits, of which 3.8% (5 252 000 visits) were owing to traumatic brain injury (TBI),1 making TBI one of the most common of all ED presentations. Although TBI may manifest as a spectrum, ranging from asymptomatic to death, the event of an ED presentation portends a greatly increased likelihood of prolonged symptoms. In fact, in a level I trauma center study2 with a high rate of positive computed tomography (CT) findings, 53% of patients with TBI exhibited persistent symptoms 1 year later. It is the group of patients with TBI with persistent symptoms for whom identification and possible therapeutic interventions could represent a substantial benefit.3,4,5,6
Despite the gravity of the pathology, identification of acutely presenting patients who are likely to recover and be asymptomatic within 30 days, vs identifying those who will experience persistent 30-day symptoms, is extremely challenging. This was demonstrated in a study7 of ED patients with suspected TBI for whom the clinical diagnostic sensitivity of ED physicians for estimating subsequent TBI symptoms was 8%. Although a number of studies have presented data to prospectively identify patients who may experience long-term symptoms after mild TBI (mTBI),8,9,10,11,12 no current routinely applied objective testing is used as part of ED standard of care to prospectively identify this cohort. Ultimately, the inability to prospectively identify ED patients with mTBI who will experience persistent 30-day symptoms makes it very difficult to determine which patients may benefit from potential therapeutic interventions.
Time to presentation after TBI is not uniform, and early presenting patients may manifest symptoms different from those of later presenters. This variability may contribute to the poor prognostic performance of clinicians. Unfortunately, the majority of TBI studies do not document the time between injury and presentation; thus, very few studies have described the parameters associated with prolonged TBI symptoms in the early presenting cohort.
HeadSMART II (Head Injury Serum Markers and Multi-modalities for Assessing Response to Trauma II) is an ongoing prospective observational study that enrolls patients with suspected mTBI at their first ED presentation.13,14 The study is designed to obtain a neurocognitive dataset and a serum biobank for future diagnostic blood marker discovery. To date, more than 1500 patients have been enrolled from multiple US and 1 international clinical sites, with the intent of characterizing the ED TBI population. Although it is still enrolling participants, with plans to continue until the study objectives are met, this robust database has sufficient granularity to describe previously unclear aspects of the TBI population.
Despite the fact that TBI symptoms can be both persistent and disabling, few interventions are associated with improved outcomes. This suboptimal state has multifactorial causes, with a major contributor being the difficulty of prospectively identifying the cohort of patients likely to remain persistently symptomatic and, therefore, potentially benefit from an intervention. This is especially true for patients with an initial Glasgow Coma Scale score of 13 or higher, which is consistent with mTBI. Thus, our purpose was to identify presenting factors associated with persistent 30-day postinjury symptoms in adult ED patients enrolled in the HeadSMART II trial who presented with suspected mTBI.
Methods
This cohort study used data from HeadSMART II, which began enrollment in the first quarter of 2021 and is ongoing; its methods have been described elsewhere.13,14 In brief, it is a large, locally ethics committee–approved, international, multicenter, prospective evaluation of adult patients with suspected mTBI (defined as a Glasgow Coma Scale score of ≥13) who presented to the ED within 96 hours of injury. Patients were included if they were aged 18 years or older, provided written informed consent (which allowed for future undetermined analyses, including this one), and could give blood samples and perform a battery of neurocognitive tests at presentation and at revisits occurring 14, 30, and 90 days later. Exclusion criteria were the need for general anesthesia at the time of presentation; a diagnosis of dementia requiring assistance for daily living; any head trauma requiring medical attention from a physician within the prior 6 months; receiving chemotherapy or radiation within the last year; having a history of stroke with disabling outcomes, brain tumor, epilepsy or intracranial surgery or hemorrhage; a psychiatric hospitalization in the last 90 days; blood transfusion within the prior 4 weeks; not having a working telephone number; currently participating in an interventional clinical trial; unable to perform study tasks on an iPad (eg, not wearing their usual corrective lenses necessary to read, or the inability to use both hands); or if they were judged unsuitable for participation as determined by any research staff. All HeadSMART II patients with mTBI enrolled as of July 1, 2024, were included in this analysis if they had 30-day follow-up data recorded (Figure). For the current study, each participating hospital obtained local institutional review board approval. This study is reported following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Figure. Flow Diagram Demonstrating the Criteria for HeadSMART II (Head Injury Serum Markers and Multi-modalities for Assessing Response to Trauma II) Participant Selection in the Analysis.
TBI indicates traumatic brain injury.
After enrollment and collection of self-reported race and ethnicity, patients underwent a battery of neurocognitive testing at presentation, which was repeated and blood drawn at each revisit. Data on race and ethnicity are included in this analysis because they have previously been associated with mTBI prognosis.15,16,17,18,19,20 Neurocognitive testing was administered by the BrainCheck iPad21 platform (for testing of Flanker, Trail Making Tests A and B, Digit Symbol Substitution, Stroop, Immediate and Delayed Recall, and a Coordination/Balance Test). In addition, Rivermead Post Concussion Symptoms Questionnaire (RPQ),22 Headache Impact Test,23 Dizziness Handicap Inventory,24 Convergence Insufficiency Symptom Survey,25 and performance of the Balance Error Scoring System26 were administered.
A reference standard diagnosis of mTBI was determined by 2 adjudicators (who were not coauthors of this article) after 30 days, blinded to the neurocognitive battery results and each other’s diagnosis, informed by the updated 2023 American Congress of Rehabilitation Medicine TBI definition.27 In the event of diagnostic disagreement between the 2 adjudicators, a third adjudicator (who was not a coauthor of this article) served as a tiebreaker.
Statistical Analysis
The size of this analysis was chosen by the date of enrollment that exceeded a 1000-patient sample, each with complete follow-up data to day 30. Descriptive characteristics are presented using median and IQR for continuous variables, with percentages of total for categorical variables. The dataset was collected using several electronic case report forms to record the demographics, medical history, presentation evaluation, and patient-reported outcomes at post-ED follow-up visits. Wilcoxon rank-sum testing was applied to continuous variables, and the Fisher exact test was applied to categorical variables. Statistical significance was defined as 2-sided P < .05.
For this analysis, data from the RPQ at 30-day follow-up visits were used as the outcome.22 The RPQ consists of 16 questions, each answered using a numeric scale where scores indicate the following: 0, not experienced at all; 1, no more of a problem; 2, mild problem; 3, moderate problem; and 4, a severe problem. The RPQ score is the sum of the 16 responses and ranges from 0 to 64. Patients were classified as having persistent symptoms if the sum of all questions reporting a moderate problem or worse (ie, ≥3) exceeded 11.28
Because the objective of this study was to identify parameters associated with longer term (30-day) symptoms following an acute mild TBI, logistic regression was used on baseline variables to identify those presentation characteristics associated with symptoms at 30 days. For each variable, the odds ratio (OR) and the risk of having symptoms at day 30 were presented with or without the nominal 95% CI. Risk was calculated using the following formula:
 ,
|
where β0 is the intercept and β1 the coefficient of the logistic regression model of the presenting characteristic on symptoms at 30 days. A Benjamini-Hochberg multiplicity adjustment was used to control for the false discovery rate.29 Nagelkerke R2 is included for each variable as an estimate of goodness of fit.30 Data were analyzed with R statistical software version 4.4.2 (R Project for Statistical Computing).
Results
This analysis included patients enrolled in the HeadSMART II database until July 1, 2024, which included a total of 803 patients (median [IQR] age, 41.0 [27.0-57.0] years; 404 male [50.3%]) who completed the 30-day assessment. Descriptive characteristics are presented as percentages in Table 1. The population was 22.5% African American or Black (181 patients), 18.8% Hispanic (151 patients), and 57.9% White (465 patients). Most patients were employed (565 of 794 patients [71.2%]) and had a fall or head struck by object as the mechanism of injury (387 of 802 patients [48.3%]). The median (IQR) time from mTBI until ED presentation was 1.5 (0.7-10.6) hours, regardless of sex, with only 104 patients (13.0%) presenting after 24 hours. The most common presenting mTBI sign or symptom was headache in 652 of 799 patients (81.6%), followed by trauma above the clavicle in 510 of 802 patients (63.6%), dizziness in 440 of 794 patients (55.4%), and confusion in 307 of 788 patients (39.0%). Trauma above the clavicle (279 of 404 men [69.1%] vs 231 of 398 women [58.0%]), fracture (103 of 403 men [25.6%] vs 60 of 396 women [15.2%]), and laceration (156 of 404 men [38.6%] vs 94 of 398 women [23.6%]) occurred more often in men than in women, whereas headache occurred more in women than in men (338 of 398 women [84.9%] vs 314 of 401 men [78.3%]).
Table 1. Demographic and Presenting Characteristics of Participants.
| Characteristic | Participants, No./total No. (%) | P value | ||
|---|---|---|---|---|
| All (N = 803) | Male (n = 404 [50.3%]) | Female (n = 399 [49.7%]) | ||
| Demographic variables | ||||
| Age, median (IQR), y | 41.0 (27.0-57.0) | 37.0 (25.0-53.0) | 46.0 (29.0-60.0) | <.001 |
| Race | ||||
| African American or Black | 181/803 (22.5) | 76/404 (18.8) | 105/399 (26.3) | .06 |
| White | 465/803 (57.9) | 250/404 (61.9) | 215/399 (53.9) | |
| Othera | 36/803 (4.5) | 18/404 (4.5) | 18/399 (4.5) | |
| Unknown | 121/803 (15.1) | 60/404 (14.9) | 61/399 (15.3) | |
| Ethnicity | ||||
| Hispanic | 151/802 (18.8) | 77/404 (19.1) | 74/398 (18.6) | .81 |
| Not Hispanic | 632/802 (78.8) | 316/404 (78.2) | 316/398 (79.4) | |
| Unknown | 19/802 (2.4) | 11/404 (2.7) | 8/398 (2.0) | |
| Employed | ||||
| No | 143/794 (18.0) | 55/399 (13.8) | 88/395 (22.3) | <.001 |
| Yes | 565/794 (71.2) | 314/399 (78.7) | 251/395 (63.5) | |
| Retired | 86/794 (10.8) | 30/399 (7.5) | 56/395 (14.2) | |
| Married | ||||
| No | 445/723 (61.5) | 226/376 (60.1) | 219/347 (63.1) | .44 |
| Yes | 278/723 (38.5) | 150/376 (39.9) | 128/347 (36.9) | |
| Body mass index, median (IQR)b | 27.5 (23.8-32.7) | 26.6 (23.7-31.0) | 28.9 (24.1-34.2) | <.001 |
| Alcohol use | ||||
| Never | 226/803 (28.1) | 86/404 (21.3) | 140/399 (35.1) | <.001 |
| Past | 112/803 (13.9) | 55/404 (13.6) | 57/399 (14.3) | |
| Current | 465/803 (57.9) | 263/404 (65.1) | 202/399 (50.6) | |
| Tobacco use | ||||
| Never | 462/803 (57.5) | 204/404 (50.5) | 258/399 (64.7) | <.001 |
| Past | 180/803 (22.4) | 98/404 (24.3) | 82/399 (20.6) | |
| Current | 161/803 (20.0) | 102/404 (25.2) | 59/399 (14.8) | |
| Mechanism of injury | ||||
| Sports | 51/802 (6.4) | 36/403 (8.9) | 15/399 (3.8) | <.001 |
| Fall or head strike | 387/802 (48.3) | 169/403 (41.9) | 218/399 (54.6) | |
| Vehicular crash | 280/802 (34.9) | 146/403 (36.2) | 134/399 (33.6) | |
| Abuse | 67/802 (8.4) | 44/403 (10.9) | 23/399 (5.8) | |
| Other | 17/802 (2.1) | 8/403 (2.0) | 9/399 (2.3) | |
| Medical history | ||||
| Prior concussion | ||||
| No | 515/774 (66.5) | 239/393 (60.8) | 276/381 (72.4) | <.001 |
| Yes | 259/774 (33.5) | 154/393 (39.2) | 105/381 (27.6) | |
| Diabetes | ||||
| No | 701/803 (87.3) | 366/404 (90.6) | 335/399 (84.0) | .006 |
| Yes | 102/803 (12.7) | 38/404 (9.4) | 64/399 (16.0) | |
| High blood pressure | ||||
| No | 567/803 (70.6) | 309/404 (76.5) | 258/399 (64.7) | <.001 |
| Yes | 236/803 (29.4) | 95/404 (23.5) | 141/399 (35.3) | |
| Prior headache or migraine disorder diagnosis | ||||
| No | 631/803 (78.6) | 360/404 (89.1) | 271/399 (67.9) | <.001 |
| Yes | 172/803 (21.4) | 44/404 (10.9) | 128/399 (32.1) | |
| Prior depression diagnosis | ||||
| No | 556/802 (69.3) | 308/403 (76.4) | 248/399 (62.2) | <.001 |
| Yes | 246/802 (30.7) | 95/403 (23.6) | 151/399 (37.8) | |
| Prior anxiety diagnosis | ||||
| No | 563/803 (70.1) | 322/404 (79.7) | 241/399 (60.4) | <.001 |
| Yes | 240/803 (29.9) | 82/404 (20.3) | 158/399 (39.6) | |
| Presenting characteristics | ||||
| Time from injury, median (IQR), h | 1.5 (0.7-10.6) | 1.5 (0.7-11.7) | 1.4 (0.7-9.7) | .90 |
| Trauma above clavicle | ||||
| No | 292/802 (36.4) | 125/404 (30.9) | 167/398 (42.0) | .001 |
| Yes | 510/802 (63.6) | 279/404 (69.1) | 231/398 (58.0) | |
| Hematoma above clavicle | ||||
| No | 552/801 (68.9) | 285/403 (70.7) | 267/398 (67.1) | .29 |
| Yes | 249/801 (31.1) | 118/403 (29.3) | 131/398 (32.9) | |
| Fracture at intake | ||||
| No | 636/799 (79.6) | 300/403 (74.4) | 336/396 (84.8) | <.001 |
| Yes | 163/799 (20.4) | 103/403 (25.6) | 60/396 (15.2) | |
| Laceration at intake | ||||
| No | 552/802 (68.8) | 248/404 (61.4) | 304/398 (76.4) | <.001 |
| Yes | 250/802 (31.2) | 156/404 (38.6) | 94/398 (23.6) | |
| Disorientation or confusion | ||||
| No | 481/788 (61.0) | 225/396 (56.8) | 256/392 (65.3) | .02 |
| Yes | 307/788 (39.0) | 171/396 (43.2) | 136/392 (34.7) | |
| Focal neurological deficit | ||||
| No | 542/803 (67.5) | 269/404 (66.6) | 273/399 (68.4) | .81 |
| Yes | 248/803 (30.9) | 129/404 (31.9) | 119/399 (29.8) | |
| Not sure | 13/803 (1.6) | 6/404 (1.5) | 7/399 (1.8) | |
| Headache at intake | ||||
| No | 147/799 (18.4) | 87/401 (21.7) | 60/398 (15.1) | .02 |
| Yes | 652/799 (81.6) | 314/401 (78.3) | 338/398 (84.9) | |
| Dizziness at intake | ||||
| No | 354/794 (44.6) | 176/397 (44.3) | 178/397 (44.8) | .94 |
| Yes | 440/794 (55.4) | 221/397 (55.7) | 219/397 (55.2) | |
| Imaging performed | ||||
| No | 119/803 (14.8) | 63/404 (15.6) | 56/399 (14.0) | .55 |
| Yes | 684/803 (85.2) | 341/404 (84.4) | 343/399 (86.0) | |
| Computed tomography scan at intake | ||||
| Neither | 198/803 (24.7) | 101/404 (25.0) | 97/399 (24.3) | .64 |
| Head | 157/803 (19.6) | 72/404 (17.8) | 85/399 (21.3) | |
| Other | 31/803 (3.9) | 15/404 (3.7) | 16/399 (4.0) | |
| Multiple | 417/803 (51.9) | 216/404 (53.5) | 201/399 (50.4) | |
Other race includes American Indian or Alaskan Native, Asian, and multiracial.
Body mass index is calculated as weight in kilograms divided by height in meters squared.
Overall, men with mTBI were younger than women (median [IQR] age, 37.0 [25.0-53.0] years vs 46.0 [29.0-60.0] years), were more likely to be employed, and were more frequently current tobacco or alcohol users. They were also more likely than women to have sustained their mTBI by abuse, assault, or sports and to have prior concussion. Women had a higher body mass index (calculated as weight in kilograms divided by height in meters squared) than men and sustained their mTBI by fall or head struck by and/or against an object. Regarding medical history, women had higher preinjury rates of diabetes, hypertension, headaches or migraines, depression, and anxiety.
Overall, the majority of patients with mTBI had persisting symptoms at 30 days, remaining constant over time (from 32.1% [248 of 773 patients] at presentation to 35.5% [268 of 754 patients] at 14 days, and 29.3% [235 of 803 patients] at 30 days) (Table 2). The rate of headache decreased from 73.5% (568 of 773 patients) at presentation, to 62.5% (471 of 754 patients) at 14 days, and 49.8% (400 of 803 patients) by 30 days. Further analysis of 30-day follow-up data indicated that the following 9 symptoms and factors at initial presentation were significantly associated with an increased probability of symptoms at 30 days: female sex (OR, 2.09; 95% CI, 1.54-2.87); elevated body mass index (ie, 32.6; OR, 1.03; 95% CI, 1.01-1.06); mechanism of injury; history of headache or migraine (OR, 2.76; 95% CI, 1.94-3.92), depression (OR, 2.25; 95% CI, 1.64-3.10), or anxiety (OR, 2.52; 95% CI, 1.83-3.48); and the presence of focal neurological deficits (OR, 1.52; 95% CI, 1.10-2.10), headache (OR, 2.08; 95% CI, 1.35-3.32), or multiple CT scans (OR, 2.22; 95% CI, 1.49-3.36) at intake (Table 3). If the mechanism of injury was related to a fall (OR, 2.87; 95% CI, 1.28-7.67), a vehicular crash (OR, 3.79; 95% CI, 1.68-10.19), or abuse (OR, 3.67; 95% CI, 1.43-10.73), the risk of having persistent symptoms was more than twice as high compared with other mechanisms of injury.
Table 2. Summary of Rivermead Post Concussion Symptoms Questionnaire Dataa.
| Visit and estimate | Score | ||
|---|---|---|---|
| Total | Male | Female | |
| Presentation | |||
| Mean (SD) | 18.8 (13.34) | 17.1 (12.67) | 20.6 (13.77) |
| Median (IQR) | 18.0 (7.0-28.0) | 16.0 (6.0-25.0) | 20.0 (9.0-29.0) |
| ≥12 (Only symptoms ≥3), % | 32.1 | 26.7 | 37.5 |
| 14 d | |||
| Mean (SD) | 22.1 (14.33) | 19.2 (13.86) | 25.0 (14.20) |
| Median (IQR) | 21.0 (10.0-32.0) | 17.0 (7.0-28.0) | 25.0 (14.0-35.0) |
| ≥12 (Only symptoms ≥3), % | 35.5 | 28.5 | 42.7 |
| 30 d | |||
| Mean (SD) | 19.9 (14.66) | 17.1 (14.15) | 22.7 (14.65) |
| Median (IQR) | 18.0 (8.0-30.0) | 14.0 (5.0-27.0) | 21.0 (11.0-33.5) |
| ≥12 (Only symptoms ≥3), % | 29.3 | 21.8 | 36.8 |
| 90 d | |||
| Mean (SD) | 17.6 (14.49) | 14.7 (13.98) | 20.3 (14.47) |
| Median (IQR) | 15.0 (5.0-28.0) | 12.0 (2.0-22.2) | 18.0 (9.0-31.0) |
| ≥12 (Only symptoms ≥3), % | 23.5 | 18.9 | 28.0 |
The questionnaire consists of 16 questions, and the score is the sum of the 16 responses and ranges from 0 to 64. Patients were classified as having persistent symptoms if the sum of all questions reporting a moderate problem or worse (ie, ≥3) exceeded 11.
Table 3. Association of Presenting Characteristics and Symptoms With 30-Day Symptoms.
| Variable and level | Risk (95% CI), %a | OR (95% CI)a | Adjusted P valueb | Nagelkerke pseudo R2 |
|---|---|---|---|---|
| Demographic variables | ||||
| Age, y | ||||
| 26 | 27.4 (23.1-31.7) | 1.00 [Reference] | NA | 0.003 |
| 57 | 30.9 (26.7-35.0) | 1.01 (1.00-1.01) | .33 | |
| Sex | ||||
| Male | 21.8 (17.8-25.8) | 1.00 [Reference] | NA | 0.039 |
| Female | 36.8 (32.1-41.6) | 2.09 (1.54-2.87) | <.001 | |
| Race | ||||
| African American or Black | 34.3 (27.3-41.2) | 1.43 (0.99-2.07) | .13 | 0.009 |
| White | 26.7 (22.6-30.7) | 1.00 [Reference] | NA | |
| Otherc | 25.0 (10.9-39.1) | 0.92 (0.40-1.93) | .85 | |
| Unknown | 33.1 (24.7-41.4) | 1.36 (0.88-2.08) | .27 | |
| Ethnicity | ||||
| Hispanic | 33.1 (25.6-40.6) | 1.23 (0.84-1.80) | .38 | 0.003 |
| Not Hispanic | 28.6 (25.1-32.2) | 1.00 [Reference] | NA | |
| Unknown | 21.1 (2.7-39.4) | 0.66 (0.19-1.86) | .56 | |
| Employed | ||||
| No | 37.1 (29.1-45.0) | 1.00 [Reference] | NA | 0.011 |
| Yes | 28.1 (24.4-31.8) | 0.67 (0.45-0.98) | .10 | |
| Retired | 23.3 (14.3-32.2) | 0.51 (0.28-0.93) | .09 | |
| Married | ||||
| No | 28.8 (24.6-33.0) | 1.00 [Reference] | NA | 0.000 |
| Yes | 28.8 (23.5-34.1) | 1.00 (0.72-1.39) | >.99 | |
| Body mass indexd | ||||
| 23.76 | 26.0 (22.2-29.8) | 1.00 [Reference] | NA | 0.014 |
| 32.57 | 31.7 (28.0-35.4) | 1.03 (1.01-1.06) | .03 | |
| Alcohol use | ||||
| Never | 30.5 (24.5-36.5) | 1.00 [Reference] | NA | 0.009 |
| Past | 37.5 (28.5-46.5) | 1.37 (0.85-2.19) | .30 | |
| Current | 26.7 (22.6-30.7) | 0.83 (0.58-1.18) | .38 | |
| Tobacco use | ||||
| Never | 28.1 (24.0-32.2) | 1.00 [Reference] | NA | 0.002 |
| Past | 29.4 (22.8-36.1) | 1.07 (0.73-1.55) | .80 | |
| Current | 32.3 (25.1-39.5) | 1.22 (0.82-1.79) | .40 | |
| Mechanism of injury | ||||
| Sports | 11.8 (2.9-20.6) | 1.00 [Reference] | NA | 0.022 |
| Fall or head strike | 27.6 (23.2-32.1) | 2.87 (1.28-7.67) | .06 | |
| Vehicular crash | 33.6 (28.0-39.1) | 3.79 (1.68-10.19) | .02 | |
| Abuse | 32.8 (21.6-44.1) | 3.67 (1.43-10.73) | .04 | |
| Other | 29.4 (7.8-51.1) | 3.12 (0.78-12.22) | .19 | |
| Medical history | ||||
| Prior concussion | ||||
| No | 29.1 (25.2-33.1) | 1.00 [Reference] | NA | 0.000 |
| Yes | 30.5 (24.9-36.1) | 1.07 (0.77-1.48) | .77 | |
| Diabetes | ||||
| No | 28.1 (24.8-31.4) | 1.00 [Reference] | NA | 0.006 |
| Yes | 37.3 (27.9-46.6) | 1.52 (0.98-2.33) | .13 | |
| High blood pressure | ||||
| No | 27.3 (23.7-31.0) | 1.00 [Reference] | NA | 0.006 |
| Yes | 33.9 (27.9-39.9) | 1.36 (0.98-1.89) | .13 | |
| Prior headache or migraine disorder diagnosis | ||||
| No | 24.4 (21.1-27.8) | 1.00 [Reference] | NA | 0.055 |
| Yes | 47.1 (39.6-54.6) | 2.76 (1.94-3.92) | <.001 | |
| Prior depression diagnosis | ||||
| No | 23.9 (20.4-27.5) | 1.00 [Reference] | NA | 0.043 |
| Yes | 41.5 (35.3-47.6) | 2.25 (1.64-3.10) | <.001 | |
| Prior anxiety diagnosis | ||||
| No | 23.3 (19.8-26.8) | 1.00 [Reference] | NA | 0.055 |
| Yes | 43.3 (37.1-49.6) | 2.52 (1.83-3.48) | <.001 | |
| Presenting characteristics | ||||
| Time from injury, h | ||||
| 0.75 | 28.8 (25.3-32.3) | 1.00 [Reference] | NA | 0.001 |
| 10.08 | 29.3 (26.1-32.4) | 1.00 (0.99-1.01) | .64 | |
| Trauma above clavicle at intake | ||||
| No | 31.2 (25.9-36.5) | 1.00 [Reference] | NA | 0.001 |
| Yes | 28.2 (24.3-32.1) | 0.87 (0.64-1.19) | .46 | |
| Hematoma above clavicle at intake | ||||
| No | 31.0 (27.1-34.8) | 1.00 [Reference] | NA | 0.005 |
| Yes | 25.3 (19.9-30.7) | 0.75 (0.54-1.05) | .19 | |
| Fracture at intake | ||||
| No | 29.1 (25.6-32.6) | 1.00 [Reference] | NA | 0.000 |
| Yes | 30.1 (23.0-37.1) | 1.05 (0.72-1.52) | .85 | |
| Laceration at intake | ||||
| No | 31.7 (27.8-35.6) | 1.00 [Reference] | NA | 0.009 |
| Yes | 24.0 (18.7-29.3) | 0.68 (0.48-0.95) | .08 | |
| Disorientation or confusion | ||||
| No | 26.2 (22.3-30.1) | 1.00 [Reference] | NA | 0.007 |
| Yes | 32.6 (27.3-37.8) | 1.36 (0.99-1.86) | .13 | |
| Focal neurological deficit | ||||
| No | 26.2 (22.5-29.9) | 1.00 [Reference] | NA | 0.014 |
| Yes | 35.1 (29.1-41.0) | 1.52 (1.10-2.10) | .04 | |
| Not sure | 46.2 (19.1-73.3) | 2.41 (0.77-7.39) | .20 | |
| Headache at intake | ||||
| No | 18.4 (12.1-24.6) | 1.00 [Reference] | NA | 0.020 |
| Yes | 31.9 (28.3-35.5) | 2.08 (1.35-3.32) | .009 | |
| Dizziness at intake | ||||
| No | 26.6 (22.0-31.2) | 1.00 [Reference] | NA | 0.005 |
| Yes | 31.8 (27.5-36.2) | 1.29 (0.95-1.76) | .19 | |
| Imaging performed | ||||
| No | 20.2 (13.0-27.4) | 1.00 [Reference] | NA | 0.010 |
| Yes | 30.8 (27.4-34.3) | 1.77 (1.11-2.90) | .06 | |
| Computed tomography scan at intake | ||||
| Neither | 19.7 (14.2-25.2) | 1.00 [Reference] | NA | 0.031 |
| Head | 25.5 (18.7-32.3) | 1.39 (0.84-2.31) | .30 | |
| Other | 29.0 (13.1-45.0) | 1.67 (0.68-3.81) | .33 | |
| Multiple | 35.3 (30.7-39.8) | 2.22 (1.49-3.36) | <.001 |
Abbreviations: NA, not applicable; OR, odds ratio.
Adjustments for continuous variables (age, body mass index, and time from injury) are estimated at the first and third quartile, and 95% CIs are nominal values.
P values were adjusted for multiplicity with a Benjamini-Hochberg false discovery rate correction.
Other race includes American Indian or Alaskan Native, Asian, and multiracial.
Body mass index is calculated as weight in kilograms divided by height in meters squared.
Discussion
This cohort study identified patient characteristics present at initial clinical presentation following mTBI that were associated with 30-day symptoms in a cohort presenting at a median of 90 minutes after injury. The very short documented time between injury and initial ED presentation is unique to our study. This is critical, not only because initial symptoms change over time, but also because patients identified as highly likely to experience persistent 30-day postconcussive symptoms could be candidates for early intervention. Risk assessment timing is important, because early interventions are those most likely to have the greatest impact on subsequent function.3,4,5,6 Although prospective validation is needed, we suspect that patients with any of the associated characteristics and symptoms identified by our analysis may represent a cohort most likely to derive benefit from early intervention and follow-up after an ED visit for suspected mTBI. Clinically, because these symptoms will constitute a substantial portion of the population with suspected mTBI, future prospective analyses will be required to determine ideal postevaluation interventions.
To our knowledge, our study is the largest to date that reports early presentation symptoms as they relate to persistent 30-day postconcussive symptoms. In fact, very few studies report the time between TBI and evaluation, much less early presentation symptoms. When absolute timing is reported, it is usually measured in days in most studies. This was supported by a PubMed literature search performed April 19, 2025, using the search term “early post concussive symptom predictors,” which identified 46 studies. Only a single study reported the time between TBI and evaluation to be less than 24 hours. In the 1 study reporting early presentations, Ganti et al6 evaluated 412 patients within 1 hour of TBI. Of these, approximately one-half were female, although sex was not found to be associated with 30-day symptom persistence. They found that only alteration of consciousness and headache were associated with 30-day postconcussive symptoms. The fact that the study by Ganti et al6 did not find a set of variables related to persistent postconcussive symptoms similar to those identified in our analysis may be a result of the much larger size of our cohort, which allowed for numerically larger subgroups.
The early identification and division of patients into cohorts with and without persisting symptom has important clinical utility. Identifying patients who are at high risk of subsequent adverse events while they are still in the ED provides the opportunity to offer interventions that may alter the trajectory of a patient’s clinical prognosis. Interventions with a beneficial effect are likely to have a greater impact when performed earlier than if delayed. In fact, it has been demonstrated that some specific early TBI interventions have time-dependent outcomes, whereas late interventions are less effective at symptom mitigation. In a study of 162 patients presenting to a sports medicine clinic, Kontos et al4 reported that only 9.1% of those with early care (98 patients), defined as occurring within 7 days of their injury, had symptoms for longer than 100 days, whereas 17.2% of those with delayed care, defined as 8 to 21 days, had a recovery period exceeding 100 days after the injury. In that study,4 the adjusted OR for prolonged symptoms in those with delayed care was 5.8 (95% CI, 1.9-17.6) times higher than that of recovery in the early care cohort. Other studies have had similar findings. Ponsford et al3 reported that patients who received an early educational and coping intervention had numerically lower frequencies of 10 of 11 postconcussive symptoms 12 weeks after their injury. It is likely that patients at high risk for 30-day symptoms would sustain the greatest benefit from early intervention, but only if they could be prospectively identified.
Although many TBI studies have a majority of male patients, an important characteristic of the HeadSMART II database is that it is nearly half female. It is well established that symptomatic outcomes are worse in women than in men with similar injury, which is consistent with our finding of female sex being associated with persistent 30-day postconcussive symptoms.31,32,33 In an analysis of 2000 patients from the TRACK-TBI dataset,2 women were much more likely to experience persistent TBI-related cognitive and somatic symptoms than their male counterparts. Another large report using Collaborative European Neurotrauma Effectiveness Research in TBI data34 found worse functional outcomes, worse quality of life, and more-severe postconcussion, depression, anxiety, and posttraumatic stress symptoms in women than in men, especially if they were younger than 45 years. This is consistent with our analysis, which had a median female age of 46.0 years. Unfortunately, the determination of causality for sex-based physiologic differences predicting outcomes is unclear. Female sex hormones are reported to decrease oxidative stress35 and excitotoxic neuronal injury,36 as well as maintain cerebral blood flow37 after TBI. The fact that these potentially beneficial effects do not provide neuroprotection suggest the presence of yet-to-be-understood factors. Direct measures of these physiological and pathological features should not be equated to symptom endorsement as pathology and symptom manifestation are related but not equivalent.
Interestingly, 2 presenting characteristics in our dataset were associated with a decreased probability of persistent 30-day symptoms. These included TBI resulting in the context of a sports injury mechanism, compared with TBI related to abuse, and the presence of a laceration. We hypothesize a putative sports protection from prolonged postconcussive symptoms is most likely the result of headgear (eg, helmets) that is encouraged in the majority of contemporary organized sports. However, the association between a laceration and the lack of postconcussive symptoms is harder to explain, but this may be confounded by the type of weapons used in an abuse scenario (ie, a knife would not have the same TBI effect as a brick).
Limitations
Overall, the HeadSMART II study is an ongoing biomarker discovery program with the intent to identify biochemical and neuropsychiatric characteristics for diagnostic and prognostic purposes in patients presenting to the ED with mTBI. As such, our secondary analysis of this database has limitations. First we report 30-day outcomes, which is a relatively short time cut point. However, identifying patients who will be asymptomatic within 30 days would allow the selection of patients for whom there would be less therapeutic intervention benefit and thus would likely be excluded from future prospective interventional studies. Second, our study evaluates outcomes as estimated by the patient’s symptoms at presentation. Although patient-reported presentation characteristics may be influenced by various aspects of bias and confounding, there are no objective measures currently available for the mTBI cohort that identify either the presentation diagnosis or subsequent outcomes. Third, although our study is the largest early presentation of mTBI reported to date, our findings are a retrospective analysis of an observational dataset, such that our results should not be considered conclusive and still require prospective validation. Fourth, predictive modeling and multivariate analysis were not applied in the present study because the focus was solely on individual baseline characteristics. Future work will incorporate these approaches, along with biomarker and neurocognitive assessment data, as well as machine learning strategies, to develop robust predictive models. Fifth, because this study used no interventions, no recommendations as to any specific therapy to mitigate persistent 30-day symptoms can be set forth.
Conclusions
Our study identified 9 characteristics of ED patients presenting a median of 90 minutes after mTBI that are significantly associated with symptoms persisting for at least 30 days. These include female sex; mechanism of injury; history of headache or migraine, depression, or anxiety; the presence of focal neurological deficits and headache at intake; or whether multiple CT scans were performed at intake. These findings highlight the importance of early clinical assessment in identifying individuals who are at risk for prolonged symptoms and thus may guide targeted interventions and management strategies. Furthermore, presentation characteristics that are associated with persistent 30-day symptoms may be useful as potential entry criteria for future therapeutic interventional TBI trials.
Data Sharing Statement
References
- 1.National Center for Health Statistics . National Hospital Ambulatory Medical Care Survey: 2021 emergency department summary tables. October 11, 2023. Accessed September 11, 2025. https://www.cdc.gov/nchs/data/nhamcs/web_tables/2021-nhamcs-ed-web-tables-508.pdf
- 2.Levin HS, Temkin NR, Barber J, et al. ; TRACK-TBI Investigators . Association of sex and age with mild traumatic brain injury-related symptoms: a TRACK-TBI Study. JAMA Netw Open. 2021;4(4):e213046. doi: 10.1001/jamanetworkopen.2021.3046 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ponsford J, Willmott C, Rothwell A, et al. Impact of early intervention on outcome following mild head injury in adults. J Neurol Neurosurg Psychiatry. 2002;73(3):330-332. doi: 10.1136/jnnp.73.3.330 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kontos AP, Jorgensen-Wagers K, Trbovich AM, et al. Association of time since injury to the first clinic visit with recovery following concussion. JAMA Neurol. 2020;77(4):435-440. doi: 10.1001/jamaneurol.2019.4552 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Leddy JJ, Master CL, Mannix R, 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. doi: 10.1016/S2352-4642(21)00267-4 [DOI] [PubMed] [Google Scholar]
- 6.Ganti L, Khalid H, Patel PS, Daneshvar Y, Bodhit AN, Peters KR. Who gets post-concussion syndrome? an emergency department-based prospective analysis. Int J Emerg Med. 2014;7:31. doi: 10.1186/s12245-014-0031-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Korley FK, Peacock WF, Eckner JT, et al. Clinical gestalt for early prediction of delayed functional and symptomatic recovery from mild traumatic brain injury is inadequate. Acad Emerg Med. 2019;26(12):1384-1387. doi: 10.1111/acem.13844 [DOI] [PubMed] [Google Scholar]
- 8.Bogoslovsky T, Wilson D, Chen Y, et al. Increases of plasma levels of glial fibrillary acidic protein, tau, and amyloid β up to 90 days after traumatic brain injury. J Neurotrauma. 2017;34(1):66-73. doi: 10.1089/neu.2015.4333 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Boucher V, Frenette J, Neveu X, et al. Lack of association between four biomarkers and persistent post-concussion symptoms after a mild traumatic brain injury. J Clin Neurosci. 2023;118:34-43. doi: 10.1016/j.jocn.2023.10.007 [DOI] [PubMed] [Google Scholar]
- 10.Visser K, Koggel M, Blaauw J, van der Horn HJ, Jacobs B, van der Naalt J. Blood-based biomarkers of inflammation in mild traumatic brain injury: a systematic review. Neurosci Biobehav Rev. 2022;132:154-168. doi: 10.1016/j.neubiorev.2021.11.036 [DOI] [PubMed] [Google Scholar]
- 11.Czeiter E, Amrein K, Gravesteijn BY, et al. ; CENTER-TBI Participants and Investigators . Blood biomarkers on admission in acute traumatic brain injury: relations to severity, CT findings and care path in the CENTER-TBI study. EBioMedicine. 2020;56:102785. doi: 10.1016/j.ebiom.2020.102785 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yue JK, Kobeissy FH, Jain S, et al. Neuroinflammatory biomarkers for traumatic brain injury diagnosis and prognosis: a TRACK-TBI pilot study. Neurotrauma Rep. 2023;4(1):171-183. doi: 10.1089/neur.2022.0060 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Peacock WF, Kuehl D, Bazarian J, et al. Defining acute traumatic encephalopathy: methods of the “HEAD Injury Serum Markers and Multi-Modalities for Assessing Response to Trauma” (HeadSMART II) Study. Front Neurol. 2021;12:733712. doi: 10.3389/fneur.2021.733712 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.BRAINBox Solutions, Inc . HEAD Injury Serum Markers and Multi-modalities for Assessing Response to Trauma (HeadSMART II). May 2021. Accessed September 12, 2025. https://clinicaltrials.gov/study/NCT04423198 [DOI] [PMC free article] [PubMed]
- 15.Sorani MD, Lee M, Kim H, Meeker M, Manley GT. Race/ethnicity and outcome after traumatic brain injury at a single, diverse center. J Trauma. 2009;67(1):75-80. doi: 10.1097/TA.0b013e31818234e8 [DOI] [PubMed] [Google Scholar]
- 16.Nakagawa K, Hoshide RR, Asai SM, et al. Favorable outcomes for Native Hawaiians and other Pacific Islanders with severe traumatic brain injury. Hawaii J Med Public Health. 2013;72(4):129-135. [PMC free article] [PubMed] [Google Scholar]
- 17.Sander AM, Pappadis MR, Davis LC, et al. Relationship of race/ethnicity and income to community integration following traumatic brain injury: investigation in a non-rehabilitation trauma sample. NeuroRehabilitation. 2009;24(1):15-27. doi: 10.3233/NRE-2009-0450 [DOI] [PubMed] [Google Scholar]
- 18.Madhok DY, Yue JK, Sun X, et al. ; The Track-TBI Investigators . Clinical predictors of 3- and 6-month outcome for mild traumatic brain injury patients with a negative head CT scan in the emergency department: a TRACK-TBI pilot study. Brain Sci. 2020;10(5):269. doi: 10.3390/brainsci10050269 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Cuthbert JP, Corrigan JD, Harrison-Felix C, et al. Factors that predict acute hospitalization discharge disposition for adults with moderate to severe traumatic brain injury. Arch Phys Med Rehabil. 2011;92(5):721-730.e3. doi: 10.1016/j.apmr.2010.12.023 [DOI] [PubMed] [Google Scholar]
- 20.Watson JD, Perrin PB, Arango-Lasprilla JC. Disparities between Native Americans and white individuals in global outcome trajectories over the 5 years after traumatic brain injury: a model systems study. PLoS One. 2025;20(4):e0321279. doi: 10.1371/journal.pone.0321279 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Yang S, Flores B, Magal R, et al. Diagnostic accuracy of tablet-based software for the detection of concussion. PLoS One. 2017;12(7):e0179352. doi: 10.1371/journal.pone.0179352 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Mississauga Halton Acquired Brain Injury Program . Rivermead Post Concussion Symptoms Questionnaire (RPCSQ). Mississauga Halton Local Health Integration Network. Accessed September 11, 2025. https://mississaugahalton.rehabcareontario.ca/Uploads/ContentDocuments/Rivermead_Post_ConcussionSymptoms_Questionnaire_(RPC)1.pdf
- 23.Houston MN, Bookbinder HA, Roach SP, et al. Reference values for the Headache Impact Test-6 questionnaire. Arch Phys Med Rehabil. 2021;102(12):2369-2376. doi: 10.1016/j.apmr.2021.05.013 [DOI] [PubMed] [Google Scholar]
- 24.Jacobson GP, Newman CW. The development of the Dizziness Handicap Inventory. Arch Otolaryngol Head Neck Surg. 1990;116(4):424-427. doi: 10.1001/archotol.1990.01870040046011 [DOI] [PubMed] [Google Scholar]
- 25.Trbovich AM, Sherry NK, Henley J, Emami K, Kontos AP. The utility of the Convergence Insufficiency Symptom Survey (CISS) post-concussion. Brain Inj. 2019;33(12):1545-1551. doi: 10.1080/02699052.2019.1658131 [DOI] [PubMed] [Google Scholar]
- 26.Bell DR, Guskiewicz KM, Clark MA, Padua DA. Systematic review of the balance error scoring system. Sports Health. 2011;3(3):287-295. doi: 10.1177/1941738111403122 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Silverberg ND, Iverson GL; ACRM Brain Injury Special Interest Group Mild TBI Task Force members . The American Congress of Rehabilitation Medicine diagnostic criteria for mild traumatic brain injury. Arch Phys Med Rehabil. 2023;104(8):1343-1355. doi: 10.1016/j.apmr.2023.03.036 [DOI] [PubMed] [Google Scholar]
- 28.Voormolen DC, Cnossen MC, Polinder S, von Steinbuechel N, Vos PE, Haagsma JA. Divergent classification methods of post-concussion syndrome after mild traumatic brain injury: prevalence rates, risk factors, and functional outcome. J Neurotrauma. 2018;35(11):1233-1241. doi: 10.1089/neu.2017.5257 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Hochberg Y, Benjamini Y. More powerful procedures for multiple significance testing. Stat Med. 1990;9(7):811-818. doi: 10.1002/sim.4780090710 [DOI] [PubMed] [Google Scholar]
- 30.Nagelkerke NJD. A note on a general definition of the coefficient of determination. Biometrika. 1991;78(3):691-692. doi: 10.1093/biomet/78.3.691 [DOI] [Google Scholar]
- 31.Gupte R, Brooks W, Vukas R, Pierce J, Harris J. Sex differences in traumatic brain injury: what we know and what we should know. J Neurotrauma. 2019;36(22):3063-3091. doi: 10.1089/neu.2018.6171 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Cancelliere C, Donovan J, Cassidy JD. Is sex an indicator of prognosis after mild traumatic brain injury: a systematic analysis of the findings of the World Health Organization Collaborating Centre Task Force on Mild Traumatic Brain Injury and the International Collaboration on Mild Traumatic Brain Injury Prognosis. Arch Phys Med Rehabil. 2016;97(2)(suppl):S5-S18. doi: 10.1016/j.apmr.2014.11.028 [DOI] [PubMed] [Google Scholar]
- 33.Bazarian JJ, Blyth B, Mookerjee S, He H, McDermott MP. Sex differences in outcome after mild traumatic brain injury. J Neurotrauma. 2010;27(3):527-539. doi: 10.1089/neu.2009.1068 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Mikolić A, van Klaveren D, Groeniger JO, et al. ; CENTER-TBI Participants and Investigators . Differences between men and women in treatment and outcome after traumatic brain injury. J Neurotrauma. 2021;38(2):235-251. [DOI] [PubMed] [Google Scholar]
- 35.Behl C. Oestrogen as a neuroprotective hormone. Nat Rev Neurosci. 2002;3(6):433-442. doi: 10.1038/nrn846 [DOI] [PubMed] [Google Scholar]
- 36.Mendelowitsch A, Ritz MF, Ros J, Langemann H, Gratzl O. 17β-Estradiol reduces cortical lesion size in the glutamate excitotoxicity model by enhancing extracellular lactate: a new neuroprotective pathway. Brain Res. 2001;901(1-2):230-236. doi: 10.1016/S0006-8993(01)02359-9 [DOI] [PubMed] [Google Scholar]
- 37.Roof RL, Hall ED. Estrogen-related gender difference in survival rate and cortical blood flow after impact-acceleration head injury in rats. J Neurotrauma. 2000;17(12):1155-1169. doi: 10.1089/neu.2000.17.1155 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Sharing Statement
