Prevalence of Incomplete Functional and Symptomatic Recovery among Patients with Head Injury but Brain Injury Debatable

Introduction

At least 38 different clinical case definitions have been used in mild traumatic brain injury (mTBI) studies. ¹ Although the International and Interagency Initiative toward Common Data Elements for Research on Traumatic Brain Injury and Psychological Health established a definition for TBI in 2010, ² there is substantial controversy regarding how mTBI diagnosis should be operationalized for emergency department (ED)-based TBI studies. Unambiguous diagnostic criteria are especially problematic in cases where signs and symptoms at presentation are subtle. ³ The lack of clear criteria for enrolling patients in TBI studies also limits the ability to compare findings from different studies. Developing a standard criteria for mTBI is one of the research priorities identified by the International Collaboration on Mild Traumatic Brain Injury Prognosis. ⁴

The most frequently cited definition of mTBI is from the American Congress of Rehabilitation Medicine (ACRM). ⁵ The ACRM defines mTBI as a traumatically induced physiological disruption of brain function as a consequence of the head being struck, striking an object, or undergoing an acceleration/deceleration movement without direct external head trauma and resulting in at least one of the following: any period of loss of consciousness (LOC); any loss of memory for events immediately before or after the injury; any alteration in mental state at the time of the injury (e.g., feeling dazed, disoriented, or confused); and focal neurological deficit(s) that may or may not be transient. Therefore, based on this definition, a patient with physical evidence of head trauma, headache, and vomiting, but without LOC, amnesia, alteration in mental status, or focal neurologic deficit, would not be classified as having mTBI.

In contrast to the ACRM criteria, the joint guidelines by American College of Emergency Physicians (ACEP) and Centers for Disease Control and Prevention (CDC) recommend consideration for a cranial computed tomography (CT) scan in patients who have suffered head trauma without LOC or amnesia, but who had a dangerous mechanism of injury, such as ejection from a motor vehicle, a pedestrian struck, or fall from a height of greater than 3 feet or 5 stairs; are older than 65 years; or post-injury have vomiting, severe headache, physical signs of a basilar skull fracture, or coagulopathy. ⁶ The ACEP/CDC criteria are aimed at identifying individuals who have suffered head trauma and warrant neuroimaging for further risk stratification. Consequently, there is evidence to suggest that predictors used in the ACEP/CDC criteria, such as age ≥65 years at time of injury, pedestrian struck/ejected from vehicle, vomiting post-head injury, and evidence of trauma above the clavicle, are important independent determinants of the risk of traumatic intracranial abnormality post-head injury. ^{7 –9} Additionally, some studies of patients with head trauma without LOC have shown that there is structural evidence of brain injury, ¹⁰ and these individuals are at risk for having persistent cognitive deficits. ¹¹ Therefore, novel approaches are needed to improve the identification of head injury patients who may not meet the ACRM criteria for TBI, but may have underlying brain injury. We hereafter refer to patients with head injury who meet the ACEP/CDC criteria for receiving a cranial CT, but who do not meet the ACRM criteria for TBI as HIBRID (Head Injury BRain Injury Debatable).

Whereas some studies have included subjects whom we label as HIBRID, ^12,13 others have not. ^14,15 There is a gap in current knowledge regarding the prognostic importance of being classified as HIBRID. It is unclear whether this group is at sufficient risk of adverse consequences from TBI to merit inclusion in prospective studies. For these reasons, we sought to determine the prevalence of incomplete functional and symptomatic recovery from TBI at 1 month post-injury and determine whether there are differences in the risk for incomplete functional and symptomatic recovery among HIBRID patients, patients meeting the ACRM criteria, and control subjects.

Methods

We analyzed data accrued from participants enrolled between April 2014 and April 2016 in a prospective, observational cohort consisting of three arms: consecutive ED patients evaluated for TBI; ED patients evaluated for traumatic injuries without evidence of head injury (trauma control); and healthy control subjects (the Head Injury S erum M arkers for A ssessing R esponse to T rauma; HeadSMART study). A detailed description of this cohort has been published previously. ¹⁶ The study was approved by the Johns Hopkins University Institutional Review Board (Johns Hopkins University, Baltimore, MD). Participants were recruited from two academic EDs located in Baltimore, Maryland (one designated a level one trauma center and the other a level two trauma center).

Briefly, eligible TBI participants were 18 years or older and presented to the ED within 24 h of blunt head injury, met the ACEP/CDC criteria for evaluation with a cranial CT scan, and consented to research blood draws. We excluded subjects who received a cranial CT, but did not meet the ACEP/CDC criteria for receiving one, and subjects with brain tumor, severe dementia, previous history of intracranial hemorrhage or intracranial surgery, seizure-induced head injury, pregnancy, no working telephone number, inability to communicate in English, or blood transfusion before initial research blood draw was obtained. Age- and sex-matched trauma control participants were 18 years or older, presented to the ED within 24 h of traumatic injury, but had no evidence of head injury, and consented to research blood draws. Exclusion criteria for trauma control participants were the same as those for TBI participants. Healthy control participants were 18 years or older, had no history of TBI, and consented to research blood draws. In addition to not meeting the exclusion criteria for TBI participants and trauma control participants, healthy control participants were excluded if they had a history of previous TBI, kidney failure, stroke, previous or active diagnosed psychiatric or neurological disease, smoking cigarettes, or recreational drug use within 2 weeks of blood draw, or blood pressure >140/80.

Demographic and clinical data were collected by trained research coordinators using structured data collection tools recommended by the National Institute of Neurological Disorders and Stroke Common Data Elements. ¹⁷ Data were collected and managed using REDCap electronic data capture tools hosted at the Johns Hopkins School of Public Health. Cranial CT images were read by one board-certified neuroradiologist. TBI patients were categorized into three groups: I) HIBRID (ACRM–); II) patients with TBI based on the ACRM criteria who had no traumatic abnormality on cranial CT (ACRM+ CT−); and III) patients with TBI who had a traumatic abnormality on cranial CT, including isolated skull fracture (CT+). Extracranial injuries are defined as noncranial fractures; or dislocations or thoracic injury (pneumothorax, hemothorax, and pulmonary contusion), abdominal injury (splenic or liver or intestinal or rectal injury), or injury to a major artery or vein.

Outcome data from the TBI groups were collected at 1 month after enrollment (time of injury) either by telephone interview (62.0% [233 of 376]) or in-person assessment (38.0% [143 of 376]). Functional outcome was ascertained using the Extended Glasgow Outcome Scale (GOSE), which categorizes recovery on a scale of 1 (dead) to 8 (upper good recovery). Incomplete functional recovery at 1 month was defined as GOSE <8. As per the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) criteria, ¹⁸ subjects were deemed to have mild post-concussion syndrome (PCS) if they reported a mild/moderate/severe problem in two of the following ICD-10 symptoms categories as measured by the Rivermead Post-Concussion Questionnaire (RPQ) ¹⁹ : 1) headaches, dizziness, general malaise, excessive fatigue, or noise intolerance; 2) irritability, emotional labiality, depression, or anxiety; 3) subjective complaints of concentration or memory difficulty; 4) insomnia; 5) reduced tolerance to alcohol; and 6) preoccupation with these symptoms and fear of permanent brain damage. Similarly, TBI subjects were deemed to have at least moderate/severe PCS if they reported mild/moderate/severe problems in at least three of the ICD-10 symptom categories listed above. ²⁰ Moderate/severe depressive symptoms was defined as a total score of 10 or more on the Patient Health Questionnaire 9 (PHQ9). ²¹

Results

Between April 2014 and May 2016, 4452 patients who were evaluated for TBI were screened for study eligibility, 1610 were deemed eligible for enrollment, 500 were enrolled, and 376 (75.2%) completed 1-month follow-up (Fig. 1). Compared to TBI subjects who completed 1-month follow-up, TBI subjects who were lost to follow-up were younger and more likely to be male, unmarried, and have used alcohol within 24 h of enrollment. However, they were not different from those who completed follow-up in terms of injury characteristics and severity (Supplementary Table 1) (see online supplementary material at http://www.liebertpub.com)

OPEN IN VIEWER

FIG. 1.

Study flow diagram. This figure describes participant sampling and how the study population was derived from the source population. ACEP, American College of Emergency Physicians; ACRM, American Congress of Rehabilitation Medicine; CT, computed tomography; HIBRID, head injury but brain injury debatable.

There was no statistically significant difference in the lost-to-follow-up rate among the different subgroups of patients evaluated for TBI (HIBRID [n = 35; 26.1%], ACRM+ CT− [n = 75; 26.7%], and ACRM+ CT+ groups [n = 14; 16.5%]; p = 0.15). We also evaluated 78 age- and sex-matched control subjects (trauma controls, n = 54; healthy control subjects, n = 24). Demographic characteristics of TBI subjects, trauma controls, and healthy controls were similar (Table 1). Among TBI subjects, 99 (26.3%) were classified into the HIBRID group, 206 (54.8%) into the ACRM+ CT− group, and 71 (18.9%) into the ACRM+ CT+ group. Subjects in the HIBRID group were older than ACRM+ CT− subjects and were more likely to be female. Fall was the most common mechanism of injury for HIBRID subjects. Although HIBRID subjects did not have obvious symptoms of impaired brain function, the majority had physical evidence of trauma above the clavicle (75.8%) and a headache (73.7%) and a few vomited post-injury (6%; Table 2). Among ACRM+ CT+ subjects, 6 (7.5%) had isolated skull fractures. Nine subjects in the HIBRID group (9.1%) had a traumatic abnormality on head CT scan (4 subjects with a subdural hematoma, 4 with a subarachnoid hemorrhage, and 1 with both). Among trauma control subjects, 37 (68.5%) had fractures (17 leg/ankle/foot, 10 wrist, 4 arm, 3 shoulder, 2 pelvic, and 2 rib fractures). Of the remaining, 13 (24%) had a sprain/strain/contusion and 4 (7.4%) had a laceration.

With regard to functional recovery at 1 month post-injury, the HIBRID group had a lower incidence of delayed functional recovery (40.4% [44 of 99]) than ACRM+ CT− subjects (60.7% [125 of 206]; p < 0.01) and ACRM+ CT+ subjects (78.9% [56 of 71]; p < 0.01). However, the incidence of delayed functional recovery within the HIBRID group was higher than the incidence in trauma controls (9.3% [5 of 54]; p < 0.01) and healthy controls (0% [0 of 24]; p < 0.01; Fig. 2). Relative to the trauma control group, the attributable risk percent of delayed functional recovery attributed to head injury in the HIBRID group was 77.1% (95% confidence interval [CI], 39.7–100).

OPEN IN VIEWER

FIG. 2.

Outcomes at 1 month post-injury. This figure is a representation of the proportions of subjects with functional disability (Glasgow Outcome Scale Extended <8); moderate/severe post-concussive symptoms (PCS), as measured by the Rivermead Post-Concussion Questionnaire, and depressive symptoms, as measured by the Patient Health Questionnaire 9, in the different study groups. ACRM, American Congress of Rehabilitation Medicine; CT, computed tomography; HIBRID, head injury but brain injury debatable.

With regard to PCS at 1 month post-injury, the HIBRID group had a lower incidence of moderate/severe PCS (18.4% [18 of 98]) than the ACRM+ CT− group (30.2% [62 of 206]; p = 0.04) and the ACRM+ CT+ group (37.1% [26 of 70]; p < 0.01). Although there was a trend toward higher incidence of moderate/severe PCS in the HIBRID group (18.4% [18 of 98]) compared to the trauma controls (9.3% [5 of 54]; p = 0.13) and healthy controls (4.2% [1 of 24]; p = 0.08), this difference was not statistically significant (Table 3). Relative to the trauma control group, the attributable risk percent of moderate/severe PCS attributed to head injury in the HIBRID group was 49.6% (95% CI, −15.2 to 100).

With regard to depressive symptoms at 1 month post-injury, the HIBRID group had a lower incidence of moderate/severe depressive symptoms (12.2% [12 of 98]) compared to ACRM+ CT− subjects (20.1% [41 of 204]; p = 0.01), but similar to ACRM+ CT+ subjects (17.6% [12 of 68]; p = 0.32). However, the incidence of moderate/severe depressive symptoms was higher than in trauma controls (0% [0 of 54]; p = 0.02) and healthy controls (0% [0 of 24]; p = 0.02).

Discussion

The diagnosis of TBI is based on evidence of alteration in brain function. mTBI especially often causes neurophysiological dysfunction with minimal or no detectable anatomic pathology. The majority of studies evaluating mTBI patients use patient or bystander report of LOC or amnesia and evidence of altered mental status and focal neurological deficit as the basis for selecting patients for enrollment (ACRM criteria). The accuracy of patient or bystander report of these signs and symptoms is suspect at best. Additionally, this approach excludes patients with head injury who may not have these signs and symptoms and yet may have an underlying brain injury. Given that LOC or alteration of consciousness requires dysfunction of the brainstem reticular activating system or bilateral limbic structures, ²² it is expected that patients who sustain injuries limited to the frontal lobes or other neocortical structures do not meet ACRM criteria for mTBI. Additionally neuropsychological decline may occur in the absence of LOC. ²³ The primary finding of this study is that patients with blunt head injury who do not meet the ACRM criteria for TBI (HIBRID group) constitute 24% of patients evaluated for TBI in the ED with a cranial CT. HIBRID patients were older than ACRM+ CT− patients, more likely to be female, and the majority had physical evidence of injury above the clavicle and a headache. The most prevalent injury mechanism in this group is falls. HIBRID patients are clinically important because although their risk of delayed functional and symptomatic recovery is lower than TBI patients meeting the ACRM criteria, it is much higher than the risk in traumatically injured control and healthy control subjects. Accordingly, this group of patients represents, although typically excluded from TBI studies, an at-risk population that deserves further research.

Estimates of incomplete functional recovery vary markedly in the literature attributed to differences in the definition of mTBI and how functional outcome is assessed. Studies examining 1-month functional outcome post-mTBI have found rates of incomplete functional recovery in 18–32% of subjects. ²⁰ There are many more studies examining 3-, 6-, and 12-month outcomes in mTBI, the largest of which demonstrated incomplete functional recovery in approximately 33% of individuals at 3 and 6 months, declining to 22% by 1 year post-injury. ²⁴ In our study, 58% of subjects had delayed functional recovery at 1 month; however, smaller proportions had at least moderate PCS (28%) and moderate/severe depression (18%). It should be noted that the high prevalence of incomplete functional recovery may be attributed, in part, to the fact that the current study excluded subjects from whom blood samples could not be obtained. Therefore, very-low-risk patients who are typically discharged from the ED shortly after receiving a negative cranial CT were often missed. Additionally, the prevalence of abnormal cranial CT scans (21%) was higher than the national estimate of 9%, ²⁵ suggesting that our cohort may have been different than the typical ED populations across the country.

In our study population, 40% of subjects in the HIBRID group experienced adverse consequences from head injury; therefore, 60% had upper good functional recovery at 1 month. Therefore, we hypothesize that the HIBRID group consists of patient head injury patients with or without TBI. The development of objective tools for distinguishing between head injury and mTBI remains an unmet clinical need. Promising technological advances that may be useful for meeting this need include: diffusion-tensor imaging (DTI) ^26,27 and blood-based biomarkers. ²⁸ Early DTI is able to distinguish between mTBI subjects and orthopedic injury controls. ²⁹ A number of blood-based biomarkers, including total Tau, ^30,31 brain-derived neurotrophic factor, ³² αII-spectrin N-terminal fragment, ³³ glial fibrillary acidic protein, ³⁴ and ubiquitin carboxy-terminal hydrolase L1, ¹⁴ have been reported to have diagnostic and prognostic value in mTBI. Additional studies are needed to determine whether these tools, in addition to known clinical predictors, can be used to further subcategorize HIBRID patients into mTBI versus head injury without TBI.

Conclusions

A significant proportion of patients evaluated for TBI do not meet the ACRM criteria for mTBI; however, they remain at high risk for experiencing adverse consequences from TBI and deserve inclusion in prospective TBI studies. Additional diagnostic approaches are needed to further subcategorize this group into mTBI versus head injury without TBI.