Rees PM. Contemporary issues in mild traumatic brain injury. Arch Phys Med Rehabil 2003;84:1885–94.
To determine (1) minimum criteria in adults for clinical diagnosis of mild traumatic brain injury (TBI) and (2) whether persistent postconcussive syndrome exists as a nosologic entity.
PubMed search by MEDLINE of head injuries from January 1977 to July 2002.
All reviews and studies of mild TBI with special reference to those on persistent postconcussive syndrome having a general trauma cohort as a control comparison.
Review of design and other methodologic issues. Studies dependent on superior strength of evidence (as defined by the American Academy of Neurology) concerning the biologic nature of persistent postconcussive syndrome.
A period of altered awareness with amnesia brought on by a direct craniofacial blow is the starting point in determining whether diffuse mild TBI has occurred. An amnestic scale is more helpful than Glasgow Coma Scale score in grading mild injury and in formulating minimum inclusion criteria for mild TBI. Neuropsychologic test results coupled with self-reported symptoms should not be taken as the primary source of evidence for mild TBI. Prolonged cognitive impairment after injury is not unique to brain trauma.
Persistent postconcussive syndrome after mild brain trauma, uncomplicated by focal injury, is biologically inseparable from other examples of the posttraumatic syndrome. To account for the persistent cognitive and behavioral sequelae of posttraumatic states, including persistent postconcussive syndrome, we need further studies on the emerging concept of limbic neuronal attrition occurring as a maladaptive response to pain and stress.
A FINE LINE EXISTS between a trivial head blow and one that affects the brain to produce “concussion” or mild traumatic brain injury (TBI). The widely varying clinical effects of subtle brain injury provide grounds for substantial physical, cognitive, and psychosocial disability. To this end, an opinion may be requested of a rehabilitation specialist, neurologist, or other physician seeing the patient for the first time months, even years, after the traumatic event. Ideally for an evidence-based diagnosis of mild TBI to be made, the following 4 factors will have to be present: (1) a credible mechanistic force applied to the brain, sufficient to cause microstructural or at least molecular injury to the brain; (2) acute clinical effects that are both recognizable and verifiable; (3) partitioning of nonspecific or confounding symptoms and findings arising independently of the brain injury; and (4) a discernible endpoint of recovery or disability. A strong need exists for the creation of minimum objective requirements and guidelines based on these precepts to determine whether mild brain injury has occurred. Each of these precepts is critically evaluated in turn in the present review of mild TBI in adults. A second main purpose of this review was to consider whether persistent postconcussive syndrome exists as a biologic entity distinct from other examples of posttraumatic syndrome.
A computer-aided English-language search of head injuries by PubMed, indexed for MEDLINE, was undertaken for the 25-year period January 1977–July 2002. The study was supplemented by cross-indexing of searched bibliographies without date restriction and reviewing of monographs and standard texts on head injuries. Requested search intersections were between mild TBI set against the postconcussive syndrome, clinical diagnostic criteria, pathophysiology, biomechanics, neuropsychologic assessment, and neuroimaging.
Biomechanics and pathophysiology
Mild brain injury or concussion can be defined as a trauma-induced, pathophysiologic alteration in mental status that may or may not invoke loss of consciousness (LOC). A force applied to the skull sufficient to cause altered mentation is the usual starting point in determining that brain injury has occurred. Deceleration brain injury without head contact has been shown experimentally in comatosed nonhuman primates
1and in harnessed pilots crashing in military aircraft in which the rapidly decelerated cranium may never contact a solid object, yet the brain is irreversibly damaged.
2The noncontact forces acting on the brain under these extreme conditions are of much greater magnitude than those occurring in commonplace jolting injuries without a head blow as in sports, roller coaster rides, or vehicular whiplash. Lower velocity rear-end vehicle collisions produce acceleration forces to the human head (from both the whiplash motion and head restraint contact
3), far less than the threshold 100 to 300g of acceleration thought necessary for cerebral injury to occur.
4Before this threshold is reached, cervical fractures will occur in human cadavers exposed to rear impacts.
5Rare instances of delayed subdural hematoma sustained in rear-end vehicle crashes have been documented,
6ostensibly from tearing of bridging veins. However, there are no histologic or biochemical data that would support a concept of primary injury to the brain parenchyma occurring without a direct head blow in mature humans in everyday life. In a series of neuropathologic examinations of more than 400 road users fatally injured without receiving head impact, McLean
7found no examples of brain injury.
In its familiar context of acceleration and deceleration contact between the head and an object, brain injury is more likely to occur if the applied forces are in a rotary (angular) direction: shear strains are then set up diffusely through the brain substance. Shear that expands into the reticular activation pathways will impair the consciousness level.
2Typically, this sequence happens when the head strikes a relatively yielding object such as a windshield of a vehicle or is hit by a fist. If, on the other hand, the impact is against a completely nonyielding surface such as concrete, as in a fall from one’s own height,
4the force to the head will be applied over a much shorter duration (<3ms) and is focally concentrated to the skull and brain surface without necessarily implicating the reticular pathways. Accordingly, depressed skull fracture with a small focal cortical contusion, may not alter consciousness.
In 1968, Oppenheimer
9examined the brains of mildly concussed patients dying of complications of multiple trauma. He described axonal disruption and separation indicative of irreversible brain damage. However, on perusal of Oppenheimer’s clinical case reports (also those of Blumbergs et al
10), systemic anoxia-ischemia may have rendered the damaged axons more vulnerable to mechanical injury, for it is recognized that neurons exposed to mechanical trauma—even mild trauma
11—depend for their survival on a milieu of adequate oxygenation and blood supply. Other evidence
- Jenkins L.W.
- Moszynski K.
- Lyeth B.G.
- et al.
Increased vulnerability of the mildly traumatized rat brain to cerebral ischemia; the use of controlled secondary ischemia as a research tool to identify common or different mechanisms contributing to mechanical and ischemic brain injury.
Brain Res. 1989; 447: 211-224
12would suggest that neuronal injury in mild TBI may remain at a biochemical level without microscopic damage to the axolemmal envelope. Biochemical changes at and within the axolemma include heightened ionic flux, glutamate transmitter surges, lactate and nitric oxide accumulation.
2Greater shear forces from mild TBI will encourage glutamate-induced neurotoxic cascade and finally disruptive membrane injury to the axolemma.
13Thus, at the severer extremes of diffuse mild TBI typified by axonal disruption, it is reasonable to apply the controversial term diffuse axonal injury, which is generally taken to denote permanent damage, visible to neuroimaging as punctate zones of shear.
Essential clinical precepts
Mild brain injury is most accurately evaluated by the acute injury characteristics, less so by perceived outcome or the self-reporting of symptoms at random points after the trauma.
2and the limbic connections subserving the Papez circuit (hippocampus, fornices, mamillary bodies, mamillothalamic tracts).
Glasgow coma scale
Originally devised in 1974
17as a clinical flow sheet and as a way to compare research data, the GCS was later adapted by others to classify degree of head injury, a score of 13 to 15 signifying mild injury. The scale was not intended by its original authors to distinguish milder types of injury, and in this context the scale shows a number of weaknesses. First, the level of alertness can be so rapidly improving that a concussed and unconscious individual may be initially comatose and “severely” injured by GCS definition (GCS score, <8), yet may quickly attain a “mild” GCS score of 13 to 15 within minutes of injury. To meet this objection, classification by standardized timing of GCS score at 30 minutes postinjury has been suggested,
186 hours postinjury in states of intoxication.
19Second, the GCS is not sensitive enough in mild injury—a perfect score of 15 will be of no help in determining whether brain injury has occurred. Last, the GCS does not take into account the presence or absence of focal neurologic injury, which leads to a common misconception that a score of 15 means a normal examination—it may not.
The clinical hallmark of diffuse injury is retrograde amnesia. Retrograde loss begins with and includes the head blow
8—a boxer glazed in a standing count or a knock-out to the floor will not afterward remember the delivering punch. Typically, in mild injury, the retrograde amnesia lasts for a period of minutes, shrinking later to seconds. More valuable than retrograde amnesia as a measure of severity of injury is PTA.
8Unlike retrograde amnesia in which there is impaired retrieval, the postevent amnesia does not later shrink, because the memory dysfunction is not a dysfunction of retrieval but encoding.
16Islands of recollection may exist within PTA, which is defined as the interval between injury and return of continuous recall. Until out of PTA, the individual is confused, but to a bystander can appear normal and be able to continue a sporting activity. From the patient’s perspective, the time spent unconscious is equal to the PTA.
Clinical outcome of milder injuries tends to correlate better with PTA than GCS.
21The Glasgow Institute of Neurological Sciences has considered brain injury to be “very mild” when there is PTA of less than 5 minutes and to be “mild” with PTA between 5 minutes and 1 hour.
22PTA of up to 24 hours was included in mild brain injury criteria adopted in 1993 by the American Congress of Rehabilitation Medicine
18(ACRM). ACRM’s other standards for mild injury were GCS score of 13 to 15, any LOC less than 30 minutes, any memory loss pre- or postinjury, or any alteration in mental state at the time of the accident.
18In sports medicine, mild TBI is further subdivided into concussion grades 1 to 3.
A revision of the World Health Organization’s Standards for the Surveillance of Neurotrauma proposes an Extended Glasgow Coma Scale (GCS-E), which blends amnesia with GCS criteria in the identification of mild TBI.
24Because the GCS-E does not fully acknowledge the limitations of the GCS, it may seem more straightforward simply to advocate the use of PTA as a severity measure for the milder degrees of injury. Finally, it should be recognized that in an injury that is mild by GCS or PTA criteria there still may be skull fractures, focal cortical contusion, or minor degrees of localized intracranial hemorrhage. To paraphrase Jennett,
8no one scale is appropriate for all types of injury, or in all circumstances.
Further clinical evaluation
The first priority in the acute-phase assessment of mild TBI is to identify the 1%
25of patients who will later require neurosurgical intervention for life-threatening complications, notably delayed epidural or subdural hematoma. By arrival time in hospital or clinic, the routine examination of a person with mild brain injury is frequently normal and the GCS score is a perfect 15. Cognitive function, including short-term memory (working memory), is commonly intact as assessed conversationally and by bedside mental status examination—the priority, then, is to deliberately enquire about amnesia in what the patient remembers or does not remember about the injury. Physically, 2 of the more commonplace findings after mild brain trauma are, first, a positive Dix-Hallpike test
26to signify traumatic vestibulopathy; and, second, anosmia, which occurs in approximately 13% of cases of mild TBI,
27the mechanism being contrecoup injury to the olfactory apparatus in some or most instances. Given that basal skull fractures can be difficult to detect by imaging methods, clinical signs of basal fracture—hemotympanum, intraorbital bruising (“raccoon” or “panda-bear” eyes), and retroauricular bruising (Battle’s sign)—are specifically sought. Of note, this seepage of bruising, from basal fracture sites to the orbital and retroauricular areas, can be delayed for up to 48 hours.
In most instances of mild injury, imaging by computed tomography (CT) or conventional magnetic resonance imaging (MRI) discloses no abnormality. Published CT positivity rates (which are as low as 9.4%)
29are generally inflated because they are derived from trauma centers or neurosurgical care units. Typical imaging findings when present are (1) hemorrhagic cortical contusion, or (2) white matter shear injury shown on MRI as small foci of altered signal or as petechial hemorrhages visualized both by CT and MRI.
14Detection rates for shear are considerably higher with MRI, more so by conventional T2-weighted gradient echo, which reveals shear injury in up to 25% to 30% of CT-negative instances of mild TBI
14and double the incidence compared with CT for cortical contusions.
MRI may also identify areas of focal cortical edema that clear in 1 to 3 months.
31Petechiae, after they disappear, will leave a permanent imprint of hemosiderin deposition on MRI.
32Except in certain locations (eg, corpus callosum), nonhemorrhagic white matter foci may be unconnected to trauma. Abnormal white matter signals can be shown in up to 30% of healthy middle-aged individuals, depending on MRI weighting.
Positive findings in mild injury are more readily depicted by using newer, more sophisticated but less readily available MRI technologies. These include diffusion-weighted and fluid-attenuated inversion recovery MRI,
34magnetization transfer imaging,
35magnetic source imaging,
36and magnetic resonance spectroscopy,
37all of which currently fall far short of 100% sensitivity in detection of mild injury.
Functional MRI (fMRI) within 1 month of injury shows different patterns of regional brain activation in response to working memory loads in patients with mild TBI compared with controls.
38Likewise, positron emission tomography (PET) will disclose abnormalities specific to the trauma within the first month.
39Those PET abnormalities (cortical and of global distribution) are not present in chronic postconcussional states in which regional frontotemporal hypometabolism exists,
40a finding not unique to brain injury and more consistent with depression.
Functional blood flow studies by single photon emission computed tomography
43(SPECT) and transcranial Doppler
44may be abnormal after mild injury. However, abnormal SPECT findings do not always relate to injury and can be caused by vascular headache states, altered mood, and other variables.
42The present searches identified only 1 SPECT study,
43which was blinded for interpretation as an extra assurance of quality: the authors found that SPECT abnormalities correlated poorly with focal CT or MRI findings as well as with the clinical outcome. The American Academy of Neurology (AAN) does not support the routine use of SPECT in the evaluation of patients with closed head injury or postconcussion syndrome.
Abnormal findings by evoked potential testing are fleeting.
46Electroencephalogram (EEG) topographic brain mapping has been used to support insurance claims, but AAN has determined that mapping has insufficient sensitivity and specificity to substantiate a role in individual patients.
47Routine EEGs are of little practical and of no prognostic value in mild head injury and do not accurately predict the risk of posttraumatic epilepsy.
48This risk after uncomplicated mild TBI is very low, showing a standardized incidence ratio of 1.5.
Neuropsychologic testing carries greatest precision in severe injury and in gauging serial progress. It is also sensitive to the effects of mild TBI and is of value in obtaining a more objective evaluation of subjective cognitive complaints. However, brain injury does not produce a unique pattern of neuropsychologic deficits.
50Similar results may occur because of cognitive difficulty from other injury-related factors, including insomnia, stress, pain, and mood disturbances.
52Test results can also be confounded by premorbid psychologic or learning difficulties or by previous head injuries, issues that all have to be taken into account in the final analysis. A major drawback is potential lack of motivation and effort on the part of the patient who essentially has control of the test outcome and may not have the incentive to perform well. Promising methods for detecting nonneurologic factors influencing psychometric test results are being developed, including TOMM
53(Test of Memory Malingering) and other methods of validity testing.
What are the minimum criteria for TBI?
Because imaging, other technologies, and neuropsychologic testing continue to show substantial limitations in diagnostic sensitivity and specificity, the decision on whether mild TBI has occurred is generally a clinical one based on the acute injury characteristics.
In accordance with best observational evidence,
51suggested core features supporting a clinical diagnosis of mild TBI are those outlined in the first and second section headings of table 1. The third section of table 1 includes self-reported symptoms: these symptoms are nonspecific, showing overlap with those of comorbid musculoskeletal injury, traumatic stress, and depression.
51The tabulated items marked by an asterisk, when taken together, will reasonably qualify as minimum post hoc criteria for mild TBI to have occurred.
Table 1Data Helpful in Clinical Diagnosis of Diffuse Mild TBI
|A credible mechanism of injury|
|Amnesia for blow|
|GCS score <15|
|Impact seizure or seizure during first week|
|Initial vomiting with headache|
|Focal CNS or cranial nerve signs on exam|
|Clinical signs of basal fracture|
|Perception of dazing at injury|
|Headache, dizziness, blurred vision, ringing in ears, increased sensitivity to noise and light, diminished libido, fatigue, and disordered sleep|
|Cognitive, affective, and behavioral symptoms|
|Neuropsychologic test findings|
∗ Minimum requirements for retrospective diagnosis.
Persistent postconcussive syndrome
Early organic postconcussive effects of headaches, vomiting, dizziness, blurred vision, tinnitus, and hypersomnolence, plus cognitive disturbances consistent with a mild subcortical encephalopathy, wear off for the most part within a matter of days or weeks after mild TBI. Overall incidence of persistent postconcussive syndrome at 12 months may be less than the commonly reported figure of about 15%
55because this figure is selectively derived from inpatient populations. Nevertheless, persons having chronic symptoms contribute a disproportionately large caseload to clinics that provide rehabilitation after mild head injury. There is an understandable tendency for the patient, family, and caregivers to wrongfully attribute all posttraumatic symptoms to the brain injury. Headaches are commonly from bruxism or comorbid neck sprain,
26Familiar during persistent postconcussive syndrome are the additive symptoms of dysphoria, negative thinking, somatic hypervigilance, dependency, and lost motivation
57—opinion has remained polarized as to whether these prolonged mental and physical symptoms relate to the brain injury or are a result of “traumatic neurosis” colored by behavior during the adjustment period that follows injury. Although disability in the acute phase of brain injury correlates quite well with its severity, late disability by self-report may be disproportionate and may even possess an inverse correlation.
50Base rates for headache and dizziness and for concentration, memory, and word-finding difficulties are high in uninjured populations.
58Iverson and McCracken
59evaluated a large sample of non-head-injured persons with chronic pain and found that 42% had 1 or more cognitive symptoms of disturbed memory or concentration and difficulty maintaining attention; moreover, 80.6% of those subjects endorsed 3 or more symptoms belonging to the research criteria for postconcussional disorder in the Diagnostic and Statistical Manual of Mental Disorders (DSM), 4th edition. Further, a study from Denmark
60found that patients with chronic daily headache not caused by concussion had a very high prevalence of persistent postconcussive syndrome symptoms.
Published observational work on the nature and etiology of persistent postconcussive syndrome and, more particularly, its cognitive sequelae has been characterized by an unfortunate lack of data, errors in sampling, and insecure methodology. After evaluating the strength of scientific evidence
61pertaining to brain concussion, AAN adopted 3 strata of evidence under classes I through III.
23Exploring the biologic makeup of persistent postconcussive syndrome does not lend itself to class I evidence (evidence provided by controlled clinical trials). In circumstances other than randomized trials, overwhelming evidence from class II studies will directly address the question.
23Class II evidence, that is, evidence provided by well-designed clinical studies, such as that corresponding to level II-2 evidence of the US Preventive Services Task Force on development of clinical practice guidelines,
62is sparse in the published literature on the late effects of concussion. Much of the published research on persistent postconcussive syndrome is dependent on nonsystematic narrative reviews of varying partisanship or on class III evidence (evidence provided by expert opinion, nonrandomized historical controls, case reports). Commonly used in formulations of persistent postconcussive syndrome are nonspecifically abnormal neuropsychologic test results
63“proving” brain injury when there were either no controls or only normative control comparisons, and without consideration paid to other system injury.
64Such results cannot be taken as the primary source of evidence or be accorded class II status, because confounding variables other than brain trauma may initiate or perpetuate cognitive impairment after injury (table 2).
Table 2Confounding Variables Contributing to Neurocognitive Disturbances After Mild Brain Injury
|Sleep deprivation in relation to pain, altered mood, accident nightmares|
|Pain-induced mental distractibility|
|Medications, notably opiates and benzodiazepines|
|Poor motivation, compensation issues|
|Anxiety, posttraumatic stress|
|Other life stressors|
|Premorbid personality, somatoform disorders|
|Past injuries and pain experiences|
|Expectation of symptoms, iatrogenesis|
Identified in the present review was a total of 10 adult studies,
61in formulating class II evidence
62concerning the biologic nature of persistent postconcussive syndrome. Those key studies are itemized in table 3 (studies using trauma controls having no brain injury)
69and table 4 (other, miscellaneous analytic control comparisons).
Table 3Evidence on the Biologic Nature of Persistent Postconcussive Syndrome: Studies With General Trauma Controls
|Hanks et al|
|157 consecutively hospitalized adults (78% mild TBI, 10% moderate, 12% severe brain injury)||125 trauma controls with other system injuries; 450 normative controls||KAS at 1mo and 12mo||KAS scores for the brain injury and trauma control groups did not differ significantly; difficulties with emotional and behavioral maladjustment not mediated by the brain injury|
|Masson et al|
|176 head-injury inpatients (mild TBI, n=114)||64 inpatients with LLI||GOS, self-rated physical, cognitive, and behavioral status, and clinical assessment at 5y||At 5y, 39.5% of mild TBI and 9.5% of LLI subjects had symptoms consistent with PPCS; in mild TBI, most disabilities were related to associated injuries|
|Dikmen et al||436 head-injury inpatients (mild TBI, n=243)||121 general trauma control inpatients with no head injury||Comprehensive battery of neuropsychologic measures at 1y||Neuropsychologic outcome comparable between injured controls and mild TBI patients|
|Dikmen et al|
|Same sample as Dikmen||Same sample as Dikmen||Psychosocial outcome at 1y: GOS, dependent living, employment and income status, SIP including physical score||Psychosocial morbidity greater in mild TBI than normative controls but comparable between mild TBI and other system injuries|
|Minderhoud et al|
|Nonhospitalized mild TBI patients receiving a structured support program (n=180) vs no formal program (n= 352) in a university-based referral program||193 injured patients without a cerebral concussion||Clinical examination; number and frequency of physical and mental sequelae at 4wk and 6mo||All physical symptoms, including headaches, neck pain, and dizziness, were shared by the mild TBI and the injured patients without concussion; those mild TBI patients receiving a positive support program did not have any mental sequelae that differed from the general trauma controls|
Abbreviations: GOS, Glasgow Outcome Scale; KAS, Katz Adjustment Scale; LLI, lower-limb injury; PPCS, persistent postconcussive syndrome, SIP, Sickness Impact Profile.
Table 4Evidence on the Biologic Nature of Persistent Postconcussive Syndrome: Miscellaneous Analytic Control Comparisons
|Binder et al|
|Meta-analysis of 314 mild TBI patients from 8 neuropsychology articles 1980–95||302 normative control subjects within the 8 articles||Effect sizes of specific neuropsychologic domains, prevalence of impairment, PPV of neuropsychologic testing||With prevalence of persistent brain injury set at 5%, PPV of neuropsychologic testing was <50% even with estimates of sensitivity and specificity kept artificially high|
|Binder and Rohling|
|Meta-analysis of 2353 persons from 18 study groups 1945–94 containing reference to financial incentives||Persons without financial incentives||Financial vs no financial incentive, effect size values averaged and weighted by group size||Late-onset symptoms were disproportionately frequent among patients seeking compensation; compared with more severely injured patients, those with shorter amnesias and mild injuries were more likely to seek compensation and fail to return to work within 18mo|
|23 patients with mild TBI referred to an outpatient head injury program||13 patients with moderate-to-severe head injuries||Clinical neurologic and mental status examination psychiatric evaluation by DSM-III criteria||Mild TBI compared with severer brain injuries (% in parentheses) showed 87% frequency of pain or headaches (38%), 26% feigned neurologic deficits (0%), 39% psychomotor slowing (0%), 87% major depression or dysthymia (31%); frequency of premorbid psychiatric history did not differ between mild and severer injury groups|
|Levin et al|
|57 admissions with mild TBI to the neurosurgery service of 3 different medical centers||56 healthy controls studied prospectively||Neuropsychologic and neurobehavioral evaluation at baseline (<7d), 1mo, and 3mo||A single uncomplicated minor head injury produces no permanent disabling neurobehavioral impairment in the great majority of patients who are free of preexisting neuropsychiatric disorder and substance abuse|
|Dencker||36 monozygotic twins and 86 same-sex dizygotic pairs||Uninjured twin||Comparison between twin pairs of symptom profile 3–25y after injury||The same late symptom profile was shown within each monozygotic pair regardless of which twin had been injured; within the dizygotic pairs, the twin having postconcussional symptoms showed a tendency to inherent behavioral traits (accident proneness, antisocial behavior) present before, and augmented by, the injury|
Abbreviation: PPV, positive predictive value.
The cited class II studies (see Table 3, Table 4) do not support the notion that persistent postconcussive syndrome after mild TBI is mediated by the brain injury. Table 3 contains 5 articles that compare the cognitive and psychosocial outcomes of mild cranial injury with those same outcomes after extracranial injuries, thereby encompassing some of the confounding variables listed in table 2, including pivotal disturbances of pain and mood that may influence mental performance. Chronic pain was reported in 95% of patients with mild TBI in a case series by Uomoto and Esselman
73; of their sample, 89% had headaches, 51% neck and shoulder pain, and 45% back pain (see also Alexander
51). Cranial and extracranial injury comparisons by Schwartz et al
74and by Gasquoine
75(who drew similar conclusions to the studies in table 3) were excluded from present consideration because of flawed selection of head injury case material
74and sole dependency on subjective self-report for outcome measures.
75Also discarded was a 1991 article by Dacey et al
76because its case material was later duplicated by the same research group in 1995 (Dikmen et al
68; see table 3). Last, a study by Taylor et al
77was excluded solely because their case material did not include mild TBI but moderate-to-severe head injuries and 2 further cohort groups, namely, chronic whiplash and other chronic pain states. Taylor found that tests of mental efficiency previously considered highly sensitive and specific for the effects of brain trauma were comparably abnormal in all 3 cohorts.
The data in Table 3, Table 4 are drawn from case samples showing mixed heterogeneity of severity within mild TBI. Poorer outcome at the severest extremes of mild TBI has been linked to CT or MRI abnormality and to subnormal admission GCS score of 13 or 14 in studies using class III levels of evidence.
80This subset of mild TBI was labeled by Hsiang et al
79as “high-risk mild head injury” or as “complicated mild” by Williams et al
78who considered the clinical outcome of this group to be nearer that usually seen in moderate brain injury and GCS scores of 9 to 12. Poorer outcome has been specifically connected to residual focal atrophy of a frontal or a temporal lobe identified by CT or MRI.
20Personality change with executive dysfunction from frontal lobe injury may or may not resolve in the first few months.
80Executive dysfunction in the presence of normal neuroimaging would usually favor depression
52over frontal lobe injury as its cause. Others have found no correlation between positive imaging and early
81neurocognitive outcome. All the cited studies had problems with methodology: some were not blinded to CT and MRI findings and in others the sample base was small
80or skewed to include moderate as well as mild injuries
80as defined clinically. Intuitively, one would expect a wide difference in prognosis between these complicated mild injuries and minimalist ones, but if this proves to be not always the case, it would again raise the issue of confounding factors.
Published evidence makes it clear that the diagnosis of mild TBI is founded on the acute injury characteristics. Diagnosis is manifestly more difficult when an injury is unwitnessed. A remembered head blow and continuous memory for all events will strongly suggest that there was no concussion to the brain. Dazing or a stunned sensation may be wrongly attributed to brain trauma when brought about by fear and being startled at the scene of an accident. Panic (“going into shock”) augmented by hyperventilation and hypocapnia
82and consequent fainting at the scene can be readily misconstrued as traumatic LOC. Injury sustained in an assault or a road crash is always unanticipated and with a strong element of startle and mental arousal. It is therefore difficult to accept the statement within the ACRM criteria for mild TBI
18(p86) that “any [emphasis added] alteration in mental state at the time of the accident” is enough by itself to establish that brain injury has occurred. Rehabilitation efforts will be misdirected if, by way of nonspecific dazing and over-labeling, a false positive diagnosis of mild TBI is reached; conversely, a gap in management will be created in overlooking diminutive brain trauma, notably when overshadowed by more dramatic multiple internal or orthopedic injuries or spinal cord injury.
A Lancet editorial
84in 1961 concluded that the PTA is “the best yardstick we have” in clinically assessing severity of head injury. Jennett has identified as a hallmark of concussion “amnesia for the impact and immediately after it.”
8(p650) PTA duration by self-report is necessarily subjective and is not always reliable. However, the volunteered duration of PTA is likely to be accurate if it is corroborated by an initially recorded GCS score of 13 or 14, the ambulance or hospital documents are examined, and the timing of PTA endpoint is objectively matched by disappearance of mental confusion and restoration of GCS score to a full value of 15. When a perfect GCS score is regained, the patient usually can be regarded as out of PTA. Both the GCS and PTA may be tainted by alcohol levels exceeding 42.5mmol/L (.20%)
85or by sedative medications given in hospital. Exaggeration or simulation of amnesia will be suspected if the posttraumatic and retrograde amnesias with respect to each other, or to the duration of LOC, are highly disproportionate. Discrepancies are not always willful but may occur at a subconscious level in the specific instance of dissociative (psychogenic) amnesia.
In their recent monograph,
87(p8) Wrightson and Gronwall offered 2 defining criteria of minimal brain injury:
- 1.The history from the patient or an observer indicates that there has been an injury to the head resulting from physical force.
- 2.The injury disturbed neurological function, with one or more of the symptoms of confusion, amnesia, or alteration of consciousness, either immediate or delayed; or there may have been other events of neurological significance such as severe or persistent headache or vomiting without other explanation.
Wrightson and Gronwall’s criteria are expanded in table 1 of the present review to include a collective series of minimum requirements that will reasonably meet a diagnosis of diffuse mild TBI. The minimum requirements do not apply to focal cortical injury, which may happen without amnesia and can be safely diagnosed only by neuroimaging. Assumed in the requirements is a credible mechanism of injury. The main area of controversy here concerns whiplash brain injury. Some practitioners would advocate a concept of brain injury without head contact as being a routine hazard of civilian life.
89However, brain injury without head contact is so rare that it is almost never seen in a clinical setting in adults.
90CT or MRI correlates of whiplash brain injury comparable to those found in mild contact brain trauma have not been published. PET scan findings in late whiplash syndrome are those of depression.
91Available evidence suggests that the persistent (and undoubted) neuropsychologic deficits that follow whiplash rest squarely on a combination of mood disturbances,
77and psychosocial vulnerability.
Corroborative forensic data of a nonmedical nature may contribute to a diagnosis of mild TBI—for example, crash dynamics or the finding of a windshield “star”—the detailing of which is beyond the scope of this review. However, attention is drawn to the role of airbags in their potential to exacerbate as well as prevent cerebral injury. Front seat airbags have a deployment velocity of up to 320km/h (200mph).
94When the occupant is behind the path of a deploying bag, the crash energy is dissipated and the occupant is protected. Conversely, if the occupant is seated too far forward (or becomes projected forward when not wearing a seatbelt), he/she will enter the path of the bag as it discharges, and the deployment energy added to the crash energy can then lead to brain injury and skull fracture even in low speed collisions.
Currently, there is no good published evidence that an uncomplicated concussion will lead to permanent neurologic sequelae. Functional neuroimaging by PET
38and to some extent evoked potential testing,
46show temporary abnormalities that offer unarguable support to an organic basis for neurocognitive disability early in the postconcussive period. Although they overlap, early and late postconcussive states are regarded by many as different processes.
56In their Belfast studies of 131 patients with mild concussion, Rutherford et al
55found that half the patients had symptoms at 1 year of which they had not complained at 6 weeks postinjury. The onset of new symptoms of persistent postconcussive syndrome after 6 weeks has no biologic precedent in uncomplicated mild brain injury and would point to other factors (see table 2). Almost ubiquitous in persistent postconcussive syndrome is depression,
51which by itself will augment headaches, poor concentration, and memory and sleep disturbances. In summarizing his well-rounded paper of 1992, Alexander wrote: “this study and much of the existing literature supports the conclusion that patients with persistent postconcussive syndrome are likely to suffer from the interaction of chronic headache and depression, rather than a specific neurologic cognitive disability.
51(p60) A role for iatrogenesis in persistent postconcussive syndrome has also been advanced.
96Giving a patient a discouraging prognosis, excessive testing, and unnecessary therapy will increase anxiety and illness-related behavior.
97In reducing a patient’s incentive to get better, a system of legal tort can equate to iatrogenesis, noting that, in some jurisdictions, the legal costs of vehicle crashes exceed the medical and hospital costs combined.
Advocates of a causal connection between persistent postconcussive syndrome and brain injury have relied heavily on neuropsychologic data.
100Such data by themselves have no diagnostic specificity.
50In studies that match persistent postconcussive syndrome to brain damage, the neuropsychologic data have invariably used normative controls (eg, Leininger et al
100) or no control comparisons at all (eg, Rimel et al
99). Included in the presently assembled data on mild TBI (see table 3) are injured controls having no head trauma. Patients in this group, compared with persistent postconcussive syndrome, showed similar neuropsychologic and neurobehavioral findings, reflecting the nonspecificity of these findings and casting doubt on a direct connection between persistent postconcussive syndrome and the brain injury. The cited studies would support a trend of recent published opinion sharing the same conclusions.
101Because of the heterogeneity of the case material (including sparse neuroimaging data), the comparison studies in table 3 do not address whether there may be 2 (or more) different persistent postconcussive syndrome populations according to CT and MRI positivity rates or to clinical severity within the mild TBI spectrum.
If the construct of persistent postconcussive syndrome is eventually to be abandoned, what terminology should take its place? Underpinning the alternative (and noncommittal) term posttraumatic syndrome, which is already in general use, more clinical similarities than differences exist between persistent postconcussive syndrome and other posttraumatic states not connected to brain trauma.
90Many questions remain, not least the part played by neuroendocrine factors in the coping mechanism after injury. That the long-term sequelae of mild TBI may be not wholly “in the mind” is suggested by preliminary work on measurable changes that occur to the physiologic stress pathways under conditions of physical and mental duress including injury and posttraumatic stress disorder
103The damaging effects of the neuroendocrine stress mediators may act as a template for the commonality of symptoms seen both in the posttraumatic syndrome and certain stress illnesses in which injury is not even a prerequisite.
103Of considerable interest is that persons with abnormal HPA reactivity related to depression or PTSD and having had no physical trauma to the brain may show MRI evidence of neuronal atrophy and attrition of dendrites within the hippocampus mediated by glucocorticoid toxicity and persistent release of hippocampal glutamate.
104Uninjured persons with depression, matched to normal controls, have significantly smaller left and right hippocampal volumes with no difference in total cerebral volumes.
103Because the dendritic atrophy does not extend to the entire neuron, it may be reversible.
Accordingly, it is feasible to envisage in mild TBI a double insult to the limbic circuitry of the hippocampus, the first insult provided by mechanical trauma, the second by maladaptive neuroendocrine stress responses acting as a wild card in reinforcing disability. Taken at face value, this new knowledge concerning neuronal attrition by plasticity in response to stress might help bridge the present divergence of opinion as to why a small minority of concussed persons continue to have severe difficulties with memory and attentiveness. Recently, Bowman
105has found that members of a sample of PTSD patients experienced as much distress because of differences in psychologic make-up as from the type or severity of the traumatic event.
Further studies are invited on the role of HPA axis dysfunction in the specific context of mild TBI by longitudinal evaluation of neuroendocrine function and hippocampal volume into the late postconcussional phase compared with head injury without symptoms of persistent postconcussive syndrome.
I thank Dr. Rosemary Basson for her suggestions in preparing the manuscript.
- Axonal degeneration induced by experimental noninvasive minor head injury.J Neurosurg. 1985; 62: 96-100
- Injury and cell function.in: Reilly P. Bullock R. Head injury. Chapman & Hall, London1997: 121-141
- Society of Automotive Engineers, Warrendale (PA)1997: 357-385 (In: Proceedings of the 41st Stapp Car Crash Conference; 1997 Nov 13–14; Lake Buena Vista (FL)) Head/neck kinematic response of human subjects in low-speed rear-end collisions.
- Biomechanics of closed head injury.in: Reilly P. Bullock R. Head injury. Chapman & Hall, London1997: 25-37
- Response of the head, neck and torso to pendulum impacts on the back.J Crash Prev Inj Control. 2001; 2: 289-306
- Subdural hematoma after whiplash injury.Lancet. 1969; 2: 237-239
- Brain injury without head impact.J Neurotrauma. 1995; 12: 621-625
- Assessment of the severity of head injury.J Neurol Neurosurg Psychiatry. 1976; 39: 647-655
- Microscopic lesions in the brain following head injury.J Neurol Neurosurg Psychiatry. 1968; 31: 299-306
- Staining of amyloid precursor protein to study axonal damage in mild head injury.Lancet. 1994; 344: 1055-1061
- Increased vulnerability of the mildly traumatized rat brain to cerebral ischemia; the use of controlled secondary ischemia as a research tool to identify common or different mechanisms contributing to mechanical and ischemic brain injury.Brain Res. 1989; 447: 211-224
- Diffuse axonal injury and traumatic coma in the primate.Ann Neurol. 1982; 12: 564-575
- Traumatically induced altered membrane permeability.J Neurotrauma. 1994; 11: 507-520
- Prevalence of MR evidence of diffuse axonal injury in patients with mild head injury and normal head CT findings.AJNR Am J Neuroradiol. 1994; 15: 1583-1589
- Mild traumatic brain injury.Neurology. 1995; 45: 1253-1260
- Neurological examination of head-injured patients.in: Becker D.P. Gudeman S.K. Textbook of head injury. WB Saunders, Philadelphia1989: 82-101
- Assessment of coma and impaired consciousness. A practical scale.Lancet. 1974; 2: 81-84
- Definition of mild traumatic brain injury.J Head Trauma Rehabil. 1993; 8: 86-87
- Methodological issues in the study of mild head injury.J Head Trauma Rehabil. 1993; 8: 30-37
- Computed tomography and magnetic resonance imaging in mild to moderate head injury.Ann Neurol. 1999; 46: 70-78
- One year outcome in mild to moderate head injury.J Neurol Neurosurg Psychiatry. 1999; 66: 207-213
- An introduction to neurosurgery. 3rd ed. William Heinemann, London1977
- Practice parameter.Neurology. 1997; 48 (Report of the Quality Standards Committee, American Academy of Neurology): 581-585
- An extended Glasgow Coma Scale (GCS-E) with enhanced sensitivity to mild brain injury.Arch Phys Med Rehabil. 2000; 81: 614-617
- Clinicopathological heterogeneity in the classification of mild head injury.Neurosurgery. 1996; 38: 245-250
- Benign paroxysmal positional vertigo.N Engl J Med. 1999; 341: 1590-1596
- Epidemiology and pathophysiology of olfactory and gustatory dysfunction in head trauma.J Head Trauma Rehabil. 1992; 7: 15-24
- Mohr J.P. Gautier J.C. Guide to clinical neurology. Churchill Livingstone, New York1995
- Clinical predictors of abnormality disclosed by computed tomography after mild head trauma.Neurosurgery. 1993; 32 (discussion 15–6): 9-15
- Brain lesions detected by magnetic resonance imaging in mild and severe head injuries.Lancet. 1986; 2: 445-446
- Mild head injury. Oxford Univ Pr, New York1989
- Imaging the injury.in: Reilly P. Bullock R. Head injury. Chapman & Hall, London1997: 167-207
- Diagnostic methods and investigation.in: Compston A. Ebers G. Lassmann H. McDonald I. Matthews B. Wekerle H. McAlpines multiple sclerosis. 3rd ed. Churchill Livingstone, London1998: 251-279
- MRI of head injury using FLAIR.Neuroradiology. 1997; 39: 239-242
- Magnetization transfer imaging in the detection of injury associated with mild head trauma.AJNR Am J Neuroradiol. 2000; 21: 875-880
- Neuromagnetic assessment of pathophysiologic brain activity induced by minor head trauma.AJNR Am J Neuroradiol. 1999; 20: 857-866
- Proton magnetic resonance spectroscopy for detection of axonal injury in the splenium of the corpus callosum of brain-injured patients.J Neurosurg. 1998; 88: 795-801
- Brain activation during working memory 1 month after mild traumatic brain injury.Neurology. 1999; 53: 1300-1308
- Dissociation of cerebral glucose metabolism and level of consciousness during the period of metabolic depression following human traumatic brain injury.J Neurotrauma. 2000; 17: 389-401
- Selected cases of poor outcome following a minor brain trauma.Brain Inj. 1994; 8: 297-308
- Neuropsychological dysfunction in depression; the relationship to regional cerebral blood flow.Psychol Med. 1994; 24: 847-857
- Assessment of brain SPECT. Report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology.Neurology. 1996; 46: 278-285
- SPECT, CT, and MRI in head injury.J Neurol Neurosurg Psychiatry. 1997; 62: 633-636
- Sounding out the brain with transcranial doppler.Lancet. 1998; 351: 654
- Evoked potential correlates of posttraumatic amnesia after closed head injury.Neurosurgery. 1984; 14: 676-678
- The postconcussional syndrome revisited.J R Soc Med. 1994; 87: 508-510
- Assessment: EEG brain mapping. Report of the American Academy of Neurology, Therapeutics and Technology Assessment Subcommittee.Neurology. 1989; 39: 1100-1101
- Medical management of head injuries.in: Becker D.P. Gudeman S.K. Textbook of head injury. WB Saunders, Philadelphia1989: 192-220
- A population-based study of seizures after traumatic brain injuries.N Engl J Med. 1998; 338: 20-24
- A review of mild head trauma. Part II: clinical implications.J Clin Exp Neuropsychology. 1997; 19: 432-457
- Neuropsychiatric correlates of persistent postconcussive syndrome.J Head Trauma Rehabil. 1992; 7: 60-69
- Depression following traumatic brain injury.Arch Phys Med Rehabil. 1998; 79: 90-103
- Test of memory malingering (TOMM). Multi Health Systems, New York1996
- Symptom validity testing.Clin Psychologist. 2001; 15: 19-45
- Symptoms at one year following concussion from minor head injuries.Injury. 1978; 10: 225-230
- Physiogenesis and psychogenesis in the ‘post-concussional syndrome’.Br J Psychiatry. 1988; 153: 460-469
- Mild traumatic brain injury.in: Grant I. Adams K.M. Neuropsychological assessment of neuropsychiatric disorders. 2nd ed. Oxford Univ Pr, New York1996: 529-551
- Symptoms following mild head injury.J Neurol Neurosurg Psychiatry. 1992; 55: 200-204
- “Postconcussive” symptoms in persons with chronic pain.Brain Inj. 1997; 11: 783-790
- Headache in 40-year-old persons in Glastrup, Denmark.Ugeskr Laeger. 1980; 142: 3071-3077
- Systems to rate the strength of scientific evidence. Evidence Report/Technology Assessment No. 47. Agency for Healthcare Research and Quality, Rockville2002 (AHRQ Publication No. 02-E016)
- Current methods of the US Preventive Services Task Force.Am J Prev Med. 2001; 20: 21-35
- A review of mild head trauma. Part I: Meta- analytic review of neuropsychological studies.J Clin Exp Neuropsychol. 1997; 19: 421-431
- Persistent Post-Concussive Syndrome. A proposed methodology and literature review to determine the effects, if any, of mild head and other bodily injury.J Clin Exp Neuropsychol. 1999; 21: 620-628
- Emotional and behavioral adjustment after traumatic brain injury.Arch Phys Med Rehabil. 1999; 80: 991-997
- Prevalence of impairments 5 years after a head injury, and their relationship with disabilities and outcome.Brain Inj. 1996; 10: 487-497
- Neuropsychological outcome at 1-year post head injury.Neuropsychology. 1995; 9: 80-90
- One year psychosocial outcome in head injury.J Int Neuropsychol Soc. 1995; 1: 67-77
- Treatment of minor head injuries.Clin Neurol Neurosurg. 1980; 82: 127-140
- Money matters.Am J Psychiatry. 1996; 153: 7-10
- Neurobehavioral outcome following minor head injury.J Neurosurg. 1987; 66: 234-243
- Closed head injury in twins.AMA Arch Gen Psychiatry. 1960; 2: 569-575
- Traumatic brain injury and chronic pain.Arch Phys Med Rehabil. 1993; 74: 61-64
- Cognitive deficits in chronic pain patients with and without history of head/neck injury.Clin J Pain. 1987; 3: 94-101
- Postconcussional symptoms in chronic back pain.Appl Neuropsychol. 2000; 7: 83-89
- Relative effects of brain and non-brain injuries on neuropsychological and psychosocial outcome.J Trauma. 1991; 31: 217-222
- Persistent neuropsychological deficits following whiplash.Arch Phys Med Rehabil. 1996; 77: 529-535
- Mild head injury classification.Neurosurgery. 1990; 27: 422-428
- High-risk mild head injury.J Neurosurg. 1997; 87: 234-238
- Magnetic resonance imaging and computerized tomography in relation to the neurobehavioral sequelae of mild and moderate head injuries.J Neurosurg. 1987; 66: 706-713
- Effects of acute injury characteristics on neuropsychological status and vocational outcome following mild traumatic brain injury.Brain Inj. 1999; 13: 873-887
- Hypocapnia.N Engl J Med. 2002; 347: 43-53
- Terminology of post-concussion syndrome.Phys Med Rehabil State Art Rev. 1992; 6: 1-7
- The best yardstick we have.Lancet. 1961; ii ([editorial]): 1445-1446
- Effect of alcohol intoxication on the diagnosis and apparent severity of brain injury.Neurosurgery. 1984; 15: 303-306
- Diagnostic and statistical manual of mental disorders (DSM-IV). 4th ed. APA, Washington (DC)1994
- Mild head injury. Oxford Univ Pr, New York1999
- Brain injury without head injury. Some physics of automobile collisions with particular reference to brain injuries occurring without physical head trauma.Appl Neuropsychol. 1995; 2: 47-62
- Handicap after acute whiplash injury.Neurology. 2002; 58 ([letter]) (discussion 158–9): 158
- In the pursuit of proof of brain damage after whiplash injury.Neurology. 1998; 51: 336-340
- PET with 18fluorodeoxyglucose and hexamethylpropylene amine oxime SPECT in late whiplash syndrome.Neurology. 1998; 51: 345-350
- Relation between neuropsychological and neuroimaging findings in patients with late whiplash syndrome.J Neurol Neurosurg Psychiatry. 1999; 66: 485-489
- Psychiatric consequences of road traffic accidents.BMJ. 1993; 307: 647-651
- Patterns of injury associated with automobile airbag use.Postgrad Med J. 1998; 74: 455-458
- Airbag associated fatal head injury.J Accid Emerg Med. 2000; 17: 139-142
- Minor head injury.J Neurol Neurosurg Psychiatry. 1994; 57: 709-716
- Pain and public policy.Lancet. 2000; 342: 1211-1213
- 1994 Claims expenditures for injuries, Insurance Company of British Columbia, Vancouver, Canada.Recovery. 1995; 6: 23
- Disability caused by minor head injury.Neurosurgery. 1981; 9: 221-228
- Neuropsychological deficits in symptomatic minor head injury patients after concussion and mild concussion.J Neurol Neurosurg Psychiatry. 1990; 53: 293-296
- The postconcussion syndrome after mild head trauma: is brain damage overdiagnosed? Part I.J Clin Neurosci. 2000; 7: 400-408
- Protective and damaging effects of stress mediators.N Engl J Med. 1998; 338: 171-179
- Hippocampal atrophy in recurrent major depression.Proc Natl Acad Sci U S A. 1996; 93: 3908-3913
- Stress effects on morphology and function of the hippocampus.Ann N Y Acad Sci. 1997; 821: 271-284
- Individual differences in posttraumatic distress.Can J Psychiatry. 1999; 44: 21-33
© 2003 American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved.