Arch Phys Med Rehabil. 2014 (Mar); 95 (3 Suppl): S132–151 ~ FULL TEXT
J. David Cassidy, PhD, DrMedSc, Carol Cancelliere, DC, MPH, Linda J. Carroll, PhD,
Pierre Côté, DC, PhD, Cesar A. Hincapié, DC, MHSc, Lena W. Holm, DrMedSc, Jan Hartvigsen, PhD,
James Donovan, BSc, DC, Catharina Nygren-de Boussard, MD, PhD, Vicki L. Kristman, PhD,
Jörgen Borg, MD, PhD
Institute of Sports Science and Clinical Biomechanics,
Faculty of Health, University of Southern Denmark,
Division of Health Care and Outcomes Research,
Toronto Western Research Institute,
University Health Network,
University of Toronto, Toronto, Ontario, Canada.
OBJECTIVE: To update the mild traumatic brain injury (MTBI) prognosis review published by the World Health Organization Task Force in 2004.
DATA SOURCES: MEDLINE, PsycINFO, Embase, CINAHL, and SPORTDiscus were searched from 2001 to 2012. We included published, peer-reviewed studies with more than 30 adult cases.
STUDY SELECTION: Controlled trials and cohort and case-control studies were selected according to predefined criteria. Studies had to assess subjective, self-reported outcomes. After 77,914 titles and abstracts were screened, 299 articles were eligible and reviewed for scientific quality. This includes 3 original International Collaboration on MTBI Prognosis (ICoMP) research studies.
DATA EXTRACTION: Eligible studies were critically appraised using the Scottish Intercollegiate Guidelines Network criteria. Two reviewers independently reviewed each study and tabled data from accepted articles. A third reviewer was consulted for disagreements.
DATA SYNTHESIS: Evidence from accepted studies was synthesized qualitatively into key findings, and prognostic information was prioritized according to design as exploratory or confirmatory. Of 299 reviewed studies, 101 (34%) were accepted and form our evidence base of prognostic studies. Of these, 23 addressed self-reported outcomes in adults, including 2 of the 3 original ICoMP research studies. These studies show that common postconcussion symptoms are not specific to MTBI/concussion and occur after other injuries as well. Poor recovery after MTBI is associated with poorer premorbid mental and physical health status and with more injury-related stress. Most recover over 1 year, but persistent symptoms are more likely in those with more acute symptoms and more emotional stress.
CONCLUSIONS: Common subjective symptoms after MTBI are not necessarily caused by brain injury per se, but they can be persistent in some patients. Those with more initial complaints and psychological distress recover slower. We need more high-quality research on these issues.
KEYWORDS: Craniocerebral trauma; Prognosis; Recovery of function; Rehabilitation
From the FULL TEXT Article:
Mild traumatic brain injury (MTBI) is a common injury after falls and traffic collisions.  It represents 70% to 90% of all TBI and has been estimated to affect more than 600 adults per 100,000 each year.  MTBI or concussion has received increasing attention mostly because of contact sports, especially American football and ice hockey.  As a result, there is more public attention and concern about potential long-lasting effects. Clinicians must deal with concerned patients who want to know how long their symptoms might last and what to expect in the future. These concerns can only be addressed by high-quality prognostic studies that follow up defined cohorts of injured subjects and use valid measures of prognostic factors and outcomes.
In 2004, the World Health Organization (WHO) Collaborating Centre Task Force on MTBI published the first systematic review  of the literature on the course and prognosis after MTBI. They searched MEDLINE, PsycINFO, CINAHL, and Embase up to the year 2000 and found 427 research articles on prognosis. After critically reviewing these studies, 120 (28%) were found to be of sufficient scientific quality to be included in their best-evidence synthesis. Of these studies, 16 focused on subjective symptoms in adults. The Task Force concluded that self-reported symptoms such as headache, fatigue, self-perceived cognitive deficits, and other symptoms reported after concussive events are also common in the acute stage of other injuries, and they are not specific to MTBI. Furthermore, these subjective symptoms are commonly associated with pain, depression, anxiety, posttraumatic stress, litigation, and other injury-related factors. Therefore, the Task Force recommended that postconcussion symptoms be assessed in the light of all contributing psychosocial factors and not be automatically attributed to brain injury per se.
In addition, the use of terms such as postconcussion syndrome (PCS) might be misleading because of doubts about the etiology of some subjective postconcussion symptoms. The Task Force found that most patients recover within 3 months to a year but that compensation-related litigation can prolong recovery. Other factors associated with prolonged symptoms were preexisting physical limitations, prior brain injury, prior neurologic problems, psychiatric problems, stress, being a student, sustaining an MTBI in a motor vehicle collision, and age >40 years. They found evidence to suggest that premorbid personality and prior psychiatric history contribute to post-MTBI stress and psychological problems, which in turn are associated with more self-reported symptoms. Also, those with more severe MTBI (eg, Glasgow Coma Scale [GCS] score of 14 or 13 and MTBI complicated by intracranial lesions and/or depressed skull fracture) have more disability than those with a GCS score of 15. Overall the Task Force concluded that self-reported symptoms were common, but there was a need for more high-quality studies on their cause, course, and prognosis.
The International Collaboration on MTBI Prognosis (ICoMP) is a team of clinicians and scientists assembled to update the WHO Collaborating Centre Task Force findings on MTBI.  ICoMP includes many of the same members who served on that task force, and were selected for their expertise in epidemiology of MTBI, clinical management of MTBI, or both. Our purpose here is to update the WHO findings on course and prognosis in adults with respect to self-reported outcomes.
The literature search and critical review strategy are outlined in detail elsewhere. Briefly, the electronic databases MEDLINE, PsycINFO, Embase, CINAHL, and SPORTDiscus were systematically searched from January 1, 2001, to June 30, 2011.  These searches were updated on February 10, 2012. The reference lists of all reviews and meta-analyses related to MTBI, and articles meeting the eligibility criteria were screened for additional studies. ICoMP members also provided studies they had knowledge about. Articles were screened for eligibility according to predefined criteria. Included were original, published, peer-reviewed research reports in English, French, Swedish, Norwegian, Danish, and Spanish, and human participants of all ages. Studies had to have a minimum of 30 MTBI cases, and for this report, had to assess self-reported outcomes after adult MTBI. The definition of MTBI had to fall within the WHO Collaborating Centre Task Force  or the Centers for Disease Control and Prevention definitions.  Excluded were publication types other than systematic reviews and meta-analyses that included an assessment of the methodological quality of the included studies, randomized controlled trials, cohort studies, and case-control studies. We also excluded basic science, animal, cadaveric, biomechanical, and laboratory studies. Although we screened systematic review reference lists for primary studies, we did not include systematic reviews in our critical review.
All eligible articles were critically appraised using a modification of the Scottish Intercollegiate Guidelines Network criteria.  Two reviewers performed independent, in-depth methodological reviews of each eligible study, and a third reviewer was consulted for disagreements. Two reviewers independently extracted data from accepted articles into evidence tables, and this evidence was synthesized to provide clear and useful conclusions linked to the evidence tables. ICoMP members also undertook 3 original research projects, and 2 are included in the results of this article.
We prioritized the evidence on prognostic factors using the framework described by Côté et al. 
Phase I studies are hypothesis generating and explore associations between potential prognostic factors and disease outcomes in a descriptive, or crude univariate way.
Phase II studies are exploratory analyses that focus on sets of prognostic factors or markers to discover which have the highest independent prognostic value.
Phase III studies are confirmatory studies with explicit hypotheses and focused examination of the strength, direction, and independence of proposed causal relationships.
Phase III studies are considered the strongest evidence for prognostic factors followed by phase II studies. Phase I studies are considered more preliminary.
Our review was conducted and is reported in compliance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.  Our protocol was registered with the International Prospective Register of Systematic Reviews (registration no. CRD42011001410) and published in Systematic Reviews. 
After applying the inclusion and exclusion criteria to 77,914 titles and abstracts, 2,170 full-text articles were assessed for eligibility. There were 173 eligible articles that assessed MTBI prognosis in adults, excluding studies of sport and military injuries, which are reported elsewhere. [11, 12] Of the 173 articles on prognosis, 51 (29%) were evaluated as having a low risk of bias, and 21 of these included subjective, self-reported outcomes. The other studies reported on objective outcomes and are reported elsewhere.  We also include 2 original studies addressing subjective outcomes in adult prognosis done by ICoMP members. [14, 15] In total, 23 studies with self-reported outcomes, including 22 cohort studies and 1 nonrandomized experimental study, form the basis of this report (Figure 1). Of the cohort studies, 1 is phase III, 16 are phase II, and 5 are phase I studies. All are English publications. We report our findings according to the length of follow up in these studies, including 1, 3, 6, 12, or more than 12 months of follow up.
Up to 1–month follow up
Table 1 A
Table 1 B
We accepted 3 cohort studies from the United States that followed up patients for up to 1 month (Table 1). Outcomes included self-reported irritability and executive functions, neurobehavioral function and symptoms, bodily pain, mental health, and postconcussion symptoms. One is a phase I study  and the other 2 are phase II studies. [17, 18] All 3 studies recruited patients from hospitals, and 2 included an orthopedic injury group. [17, 18]
With respect to recovery, Brewer et al  found that 30% of patients continued to complain of irritability and 20% continued to complain of concentration problems at 1 month. Landre et al  found that patients with MTBI had similar Medical Outcomes Study 36–Item Short-Form Health Survey (SF–36) scores (bodily pain and mental health subscales) and frequency and intensity of postconcussion symptoms as patients with other traffic injuries. Rush et al  reported a similar result comparing neurobehavioral function and symptoms at discharge between those with MTBI and those with orthopedic injuries. These results indicate that postconcussion symptoms, pain, and mental health are similar across acute injuries and not unique or specific to MTBI.
With respect to prognosis, Brewer  suggests that the presence of loss of consciousness (LOC) does not impact self-reported irritability or concentration problems measured 1 month postinjury but might impact other executive functions. However, these findings are preliminary phase I findings. Landre  found that postconcussion symptoms reported within the first week after injuries are correlated to mental health but not bodily pain. This suggests these symptoms are related to emotional distress but not to pain severity. Rush  reported that neurobehavioral function and symptoms were not associated with self-ratings of personality. However, both studies are phase II and exploratory with respect to prognosis.
Up to 3 months' follow up
The 3–month follow-up period is important because the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision stipulates that the criteria for a diagnosis of PCS include objective evidence of declines on neuropsychological testing, including difficulty in attention or memory, and 3 or more subjective symptoms present for at least 3 months (ie, fatigue; disordered sleep; headache; vertigo or dizziness; irritability or aggression with little or no provocation; anxiety, depression, or affective liability; changes in personality [eg, social or sexual inappropriateness]; or apathy or lack of spontaneity). 
Table 2 A
Table 2 B
Table 2 C
Table 2 D
Table 2 E
We accepted 5 studies [20–24] that followed up patients for up to 3 months. Outcomes included various subjective symptoms (Table 2). Two studies [20, 24] were from the United States, 2 [21, 22] were from Canada, and 1 study  was from Sweden. All are phase II cohort studies except for the study by Nygren-de Boussard et al,  which is a phase I cohort study.
Four of these studies [20–23] compared patients with MTBI to uninjured controls, and Davis  also included a second control group of patients with other injuries. Four studies [20, 21, 23, 24] recruited acute patients seen at hospitals, but Lange et al  recruited patients referred to an early intervention clinic. Two of these studies [20, 21] found little difference in postconcussion symptoms reported after injury when comparing patients with MTBI and controls, although Davis  found that patients with MTBI tended to underreport existing symptoms before their injury when compared with healthy controls. They also found that patients with MTBI attribute more of their preinjury somatic symptoms to the injury and endorse more memory symptoms. Kashluba et al  found that patients with MTBI improved substantially and did not endorse significantly more postconcussion symptoms than controls at 3 months postinjury. However, they did have a higher incidence of doing things more slowly, fatiguing quickly, and having poor balance compared with controls. Also, more patients with MTBI endorsed at least 1 symptom in the severe range compared with controls (39% vs 15%). However, Lange  showed that even though patients with MTBI endorse more symptoms than healthy controls, they recall fewer symptoms than controls before injury. They concluded that patients with MTBI misperceive their preinjury status as better than the average, and they called this the “good-old-days” bias because of the potential of misattributing symptoms to the injury. The weight of this evidence suggests that postconcussion symptoms are not specific to MTBI, and clinicians should be cautious about attributing common postinjury symptoms to the MTBI. This calls into question the validity of diagnosing postconcussion syndrome (PCS).
Nevertheless, postconcussion symptoms are troublesome for patients. Dischinger et al  followed up 180 patients with MTBI to see how many developed PCS 3 months after the injury.
They defined PCS as having 4 or more symptoms that could include any of 6 physical symptoms
(headache, dizziness, blurry/double vision, fatigue, sensitivity to light, sensitivity to noise),
3 cognitive symptoms
(difficulty concentrating, memory problems, trouble thinking),
or 3 emotional symptoms
(anxiety, depression, irritability).
At baseline, 84.2% had 4 or more postconcussion symptoms. At 3 months, 41.4% had PCS, and it was associated with female sex, baseline noise sensitivity, and baseline anxiety in women only. Furthermore, Lange et al  found that patients in litigation report more postconcussion symptoms than MTBI nonlitigants. All this highlights the complexity of symptom attribution after MTBI and the interactions of biopsychosocial issues.
Finally, Nygren-de Boussard et al  reported a phase I study of the relationship between the serum concentrations of proteins S100A1B and S100B and prognosis. S100 proteins are biochemical markers of acute brain injury, and if present and associated with cognitive impairment, might be used as a prognostic marker. Baseline S100A1B and S100B serum concentrations were elevated in 48% and 31% of the patients, respectively, and 44% reported at least 1 cognitive symptom at baseline and 26% at 3 months. However, there was no association between elevated S100 levels and cognitive symptoms at any time point. This suggests that these markers are not useful in MTBI prognosis with respect to subjective outcomes.
Up to 6 months' follow up
Six studies were accepted that reported follow up of patients for up to 6 months, including 2 from the United Kingdom, [25, 26] and 1 each from Israel,  New Zealand,  The Netherlands,  and Canada. 
Table 3 A
Table 3 B
Table 3 C
Five are phase II studies and 1 is a phase I study. Outcomes included functional disability measured by the Glasgow Outcome Scale, posttraumatic symptoms, posttraumatic stress disorder (PTSD), PCS (ie, defined by ≥3 symptoms on the Rivermead Postconcussion Symptoms Questionnaire [RPSQ]), self-reported symptoms, and community integration (Table 3). None of these studies included control groups, and they all included acute patients recruited from hospitals.
With respect to course and prognosis, Gil et al  showed that by 6 months, 14% of patients with MTBI had developed posttraumatic stress disorder (PTSD). It was more prevalent in those with a memory of the injury event (23%) than those without memory (6%). This difference was primarily due to the “reexperiencing” cluster of symptoms. PTSD was also associated with acute posttraumatic symptoms, anxiety, depression, and a history of psychiatric disorder. Hou et al  found that PCS was present in 22% of patients at 3 months and 21% at 6 months. Fatigue, forgetfulness, and sleep disturbance were most commonly reported at 3 months, and PCS was associated with self-reported activity levels. Headache, fatigue, and sleep disturbance were most commonly reported at 6 months, and PCS at 6 months was associated with negative head injury perceptions. These results suggest that cognitive and behavioral responses to MTBI might be more important in the development of PCS than demographics, injury severity, and other emotional and social factors. Norrie et al  found that fatigue prevalence diminished from 67.3% at 1 week to 29.6% at 3 months and 26.4% at 6 months. Further, fatigue severity and depression measured at 3 months were associated with fatigue prevalence at 6 months. All of these studies indicate that postconcussion symptoms continue to persist in 14% to 26% of patients with MTBI at 6 months. However, all of these studies are phase II and require confirmation.
Three other studies followed up patients for up to 6 months. De Silva et al  compared functional outcomes after more severe MTBI (GCS score, 13–14) across low-, middle-, and high-income countries using the Glasgow Outcome Scale. In this phase II study they found that 6–month recovery was better in patients from low-income countries compared with high-income countries (78% vs 60%). They speculate that sociocultural or environmental factors are responsible. Tellier et al  compared 6–month outcomes in patients with MTBI with GCS scores of 13 or 14 with those with a GCS score of 15 and found no difference in postconcussion symptom prevalence. However, they did find that those with longer posttraumatic amnesia (PTA) showed greater aggressive and disinhibited behaviors. This is a phase I study, and the results should be viewed as exploratory.
Finally, the study by Stulemeijer et al  deserves special attention. They developed and internally validated a clinical prediction rule for good recovery, defined as a score of 0 (no problem), 1 (not a problem anymore), or 2 (mild problem but not interfering with daily activities) on 13 of 16 postconcussion symptoms measured by the Rivermead Postconcussion Symptoms Questionnaire (RPSQ). If the patient did not have any preinjury comorbid physical problems, had low levels of early postconcussion symptoms (ie, a score of 0 [no problem], 1 [not a problem anymore], or 2 [mild, but not interfering with daily activities] on at least 13 of the 16 postconcussion symptoms measured by the RPSQ), and had low levels of early posttraumatic stress, they had a 90% chance of a good recovery. LOC, GCS, PTA, and abnormal computed tomography findings did not predict recovery. These results show that early identification of patients with MTBI who are likely to have good recovery is feasible, but Stulemeijer's prediction rule needs to be validated in another setting before it can be recommended for widespread use.
Up to 1–year follow up
Table 4 A
Table 4 B
Table 4 C
Table 4 D
Table 4 E
Table 4 F
Six accepted cohort studies reported 1–year follow up of patients, including 3 studies from Canada, [14, 15, 31] 2 from the United States, [32, 33] and 1 study from Lithuania  (Table 4). One is a phase III study,  3 are phase II studies, [14, 33, 34] and 2 are phase I studies. [15, 31] Outcomes include self-reported recovery, levels of fatigue, perceived activities and behaviors, satisfaction with reintegration to normal living, posttraumatic stress, psychiatric impairment, postconcussion symptoms, depressive symptomatology, health care use, PCS as defined by >3 symptoms on the RPSQ, and the prevalence of posttraumatic headache (PTHA).
Two studies report on a population-based cohort of MTBI after traffic collisions from the Canadian province of Saskatchewan. Cassidy et al  found that the median time to self-reported recovery was 100 days in this cohort, and about 23% reported not being recovered by 1 year. Hartvigsen et al  found that the most common symptoms reported by those not recovered at 1 year were sleep disturbances (44%), tiredness (39%), forgetfulness (27%), dizziness (25%), neck pain (25%), and low back pain (19%). Some of these symptoms might be due to coexisting whiplash injuries to the spine. They also reported that more than 50% of these symptomatic participants reported more than 3 symptoms. Most who continued to seek care for their symptoms at 1 year postinjury were seeing medical doctors, although a substantial number were also seeking care from physical therapists, registered massage therapists, and chiropractors. In a phase II analysis of prognostic factors from the same cohort, Cassidy  found that prolonged recovery was associated with age >50 years, less education, poor expectations for recovery, depressive symptomatology, hearing problems, arm numbness, confusion, headache intensity, low back pain intensity, and mid-back pain intensity. Sex, loss of consciousness (LOC), and posttraumatic amnesia (PTA) were not associated with recovery. Overall, these results suggest that traffic-related MTBI occurs with other injuries to the neck and back, and expectation for recovery, depression, and somatic complaints determine the outcome.
The remaining 4 studies all had control groups to compare to patients with MTBI recruited from emergency departments (EDs). One phase III prognostic study  focused on fatigue and compared 173 patients with MTBI and no PTA or LOC, with 58 patients with LOC ≤30 minutes and/or PTA <24 hours and 128 patients with other mild nonhead injuries. By 1–year follow up, levels of fatigue were slightly higher in the group with MTBI and no LOC or PTA, but all groups were still within population norms indicating low levels of fatigue. Worse fatigue at 1–year follow up was associated with preinjury fatigue, marital status, lawyer involvement, and baseline poor medical and mental health, but not with type of injury. These results confirm that postinjury fatigue is no worse in MTBI than in other injuries and is associated with psychosocial factors.
Using the same injury cohort as de Leon et al,  McLean et al  examined prognostic factors associated with persistent PCS 1 year after injury. They compared 251 patients with MTBI to 256 patients with minor nonhead injuries. Outcomes included PCS defined as ≥3 symptoms rated as at least mild on the RPSQ, mental and physical health measured by the SF–36, and level of cognitive symptoms measured by the Sickness Impact Profile–Alertness Behavior subscale. Compared with non–head-injured patients, those with MTBI had slightly worse mental and physical health at 1 year. They also reported more postconcussion symptoms (RPSQ, 13.9 vs 3.7) and had a higher incidence of PCS (≈56% vs ≈28%) at 1 year postinjury. In the combined cohort of 507 patients, baseline mental and physical health was associated with PCS and cognitive symptoms, but having an MTBI was not. The findings of this phase II study are in agreement with those of previous studies with short-term outcomes that suggest that the development of PCS and cognitive symptoms are not specific to head injury.
In another study, Friedland and Dawson  came to a similar conclusion after comparing 64 patients with MTBI to 35 non–head-injured patients and following them up to between 6 and 9 months postinjury. In this phase I study, patients with symptoms of posttraumatic stress did not do well in terms of functional outcome regardless of injury type. The patients with MTBI were not particularly worse off compared with those with other injuries, but they did have lower psychosocial scores on the Sickness Impact Profile but no other significant difference on outcomes.
Finally, Stovner et al  used historical and prospective cohort designs to measure the prevalence of posttraumatic headache (PTHA) in patients with MTBI and patients with orthopedic injuries. Both studies indicate that PTHA prevalence is similar in patients with MTBI and in orthopedic-injured patients. In the historical cohort, more than 90% of all patients had recovered from their PTHA by 1 month. In the prospective cohort, 10% of patients with MTBI and 12% of orthopedic-injured patients complained of persistent headache (ie, >15d/mo) after 1 year. Although photophobia was more common in patients with MTBI, there were no other differences among groups with respect to frequency or types of symptoms at 1 year. The authors conclude that headache occurring more than 3 months after MTBI is unlikely caused by brain injury per se. The results of this phase II study should be interpreted with caution, since other authors have found a very low rate of expectation of any chronic sequelae after MTBI in Lithuania.  These results suggest that PTHA is not a problem in Lithuania after MTBI, but studies from other jurisdictions do not necessarily agree.
More than 1–year follow up
Table 5 A
Table 5 B
We accepted 2 phase II cohort studies [38, 39] and 1 nonrandomized experimental study  with follow up for more than 1 year postinjury, including 1 study each from Brazil, Canada, and Sweden (Table 5). Outcomes included postconcussion symptoms measured by the RPSQ, health-related quality of life, anxiety, depression, self-reported cognitive function, self-reported memory, fatigue, sleep disturbance, and loneliness.
The 2 phase II studies included uninjured controls to look at the prognostic value of S100B36 and the apolipoprotein E (APOE) ε4 genotype.  De Almeida Lima et al  followed up 38 cases of MTBI treated at an ED for 18 months and compared them with 39 household controls. They found no correlation between S100B protein levels and abnormal findings on a computed tomography scan, or between S100B and health-related quality of life or depression at follow up, confirming the results of Nygren-de Boussard et al  that S100B is not a useful prognostic marker in patients with MTBI.
Sundström et al  looked at the prognostic value of APOE in 31 patients with MTBI and compared them with matched controls. Outcomes included simple questions about various postconcussion symptoms. Postinjury fatigue was more common in MTBI cases with APOE ε4 than without it (58% vs 32%). Among carriers of APOE ε4, those with MTBI had more fatigue than controls without MTBI (58% vs 17%). These results are preliminary and need to be confirmed in a phase III study.
Finally, Ozen and Fernandes  conducted a nonrandomized experiment with undergraduate university students to determine whether expectations of MTBI symptoms influence self-reported symptoms. Students were initially surveyed about past head injuries, and then a subset of those with and without head injury were surveyed again under 2 separate scenarios. Under a “diagnosis threat” scenario, 22 students with and 21 students without past head injury were tested with the knowledge that the tests were focused on comparing outcomes between those with and without past MTBI. Under the “neutral” scenario, 21 students with and 23 students without past head injury were tested without knowledge that the tests were focused on past MTBI status. The diagnosis threat group with past MTBI reported more cognitive errors and memory failures than all others. The neutral scenario group with past MTBI reported more anxiety than others. These results suggest that expectations influence self-reported cognitive and memory results.
Our results support the previous finding of the WHO Collaborating Centre Task Force on MTBI that self-reported symptoms such as headache, fatigue, self-perceived cognitive deficits and other so-called postconcussion symptoms are common in the acute stage of injury but are not specific to MTBI.  When compared with uninjured controls, patients with MTBI do report more postconcussion symptoms at 3 months [21, 22] and at 1 year. [36, 37] However, postconcussion symptoms are equally prevalent in those with other nonhead injuries. [17, 18, 20, 31, 32, 34]
Most of the postconcussion symptoms could be viewed as common reactions to the stress of injury, or other mental or physical health stressors. For example, Landre et al  showed that acute postconcussion symptoms are associated with emotional distress, but not type of injury. De Leon et al  found that fatigue severity at 1–year follow up was associated with baseline fatigue, past mental health issues, past medical disability, marital status, and being involved in litigation, but not the type of injury (ie, MTBI vs nonhead injury). All this evidence calls into question the validity of the PCS as a specific diagnosis and sequelae of MTBI. These symptoms are common in the general population,  in patients with chronic pain,  and after whiplash injury to the neck. [41, 42] In addition, 2 studies [20, 22] we reviewed show that patients with MTBI tend to minimize symptoms that they have before being injured. Thus, we recommend that the term postconcussion syndrome be replaced with posttraumatic symptoms because they are common to all injuries.
Even though posttraumatic symptoms are not specific to MTBI, they are a problem for patients and clinicians. The literature reviewed by the WHO Collaborating Centre Task Force suggested that most patients recover within 3 months to 1 year.  Our update supports this, but there is evidence that a significant minority continue to have subjective complaints. Hou et al  found that 22% of patients had 3 or more posttraumatic symptoms at 3 months, and there was no significant recovery by 6 months. Norrie et al  found that 30% of patients complained of fatigue at 3 months, and this remained relatively stable at 26% by 6 months.
Stulemeijer et al  found that 36% of patients with MTBI continued to have 3 or more posttraumatic symptoms at 6 months. Cassidy et al  reported that the median time to self-reported recovery was 100 days in patients with MTBI after traffic collisions, and that about 23% report not being recovered by 1 year. However, these same studies show that persistent posttraumatic symptoms are associated with psychosocial factors such as depression,  posttraumatic stress,  negative injury perceptions,  and poor expectations for recovery. 
Other psychosocial factors associated with posttraumatic symptoms at follow up include mental health status, [17, 32, 33] anxiety in women,  and litigation or lawyer involvement. [22, 32] In fact, these psychosocial factors are more strongly associated with outcomes than the traditional biomedical factors thought to determine recovery. For example, several studies [14, 26, 29, 32] found that loss of consciousness (LOC) and posttraumatic amnesia (PTA) were not associated with recovery from self-reported symptoms. The results of our review suggest that patients with persistent posttraumatic symptoms might benefit from psychosocial interventions, and this should be a focus of future intervention studies.
One purpose of prognosis is the early recognition of patients at risk of a poor or good outcome. Clinical prediction rules are prognostic tools that can help stratify patients into different risk sets at the onset of a disorder and can inform the clinician and patient of the likely course of recovery and aid in treatment decisions.  Our review found 1 clinical prediction rule. Stulemeijer  developed a clinical prediction rule in patients admitted to the ED with MTBI in The Netherlands. They defined a good outcome as a score of less than 3 on at least 13 of 16 posttraumatic symptoms measured by the RPSQ. An absence of comorbid physical problems, low levels of early posttraumatic symptoms, and low levels of early posttraumatic stress predicted a good outcome at 6 months with a probability of 90%. However, these results need to be validated in another setting before being recommended for widespread use.
Study limitations and strengths
Our study has some limitations and strengths. Although we followed a strict PRISMA-compliant protocol, our conclusions are only as good as the literature that we have accepted, and we found it to be generally weak and heterogeneous. Of the 173 studies we reviewed on adult prognosis of MTBI, only 51 (29%) were considered to have a low risk of bias, and 21 of these included self-reported outcomes relevant to this article. It is disappointing that so few good prognostic studies have been published since the WHO Task Force reviewed the same literature up to the year 2000. Also, only 1 of our accepted articles was a phase III confirmatory prognostic study. However, we may have excluded some good studies that included intentional injuries, or included both adults and children without stratifying the results. We a priori decided to do this because we think children and those with intentional injuries may have a different trajectory for recovery. In addition, most of the prognostic studies we reviewed did not take into account potential confounding effects of varying levels of treatment on prognosis. However, since there is little evidence of treatment effectiveness in MTBI, we do not think this is a major problem.  Although our search strategy was comprehensive, we may have missed some good studies that were not in the searched databases or not in languages included in our protocol.
A strength of the ICoMP is that our group includes a mix of methodological and clinical scientists with a spectrum of experience in systematic reviews and clinical care of MTBI. Our group also carefully considered the strength of the evidence on MTBI prognosis and report only on studies that have a low risk of bias. Thus, our results include only the best current evidence.
Since the prognosis review of the WHO Collaborating Centre Task Force, there has been some progress in understanding MTBI prognosis. Our results add to the growing evidence that postconcussion symptoms are not specific to MTBI and occur commonly in the general population and after other nonhead injuries. Our results also confirm the importance of psychosocial determinants of recovery. We conclude that self-reported symptoms can be persistent after MTBI, and there is a need for more intervention research targeting modifiable prognostic factors. Finally, we found only 1 study of a clinical prediction rule, and we recommend more focus on this issue because it holds the potential of identifying those at risk of a poor recovery who might benefit from more focused clinical attention.
We thank the other members of ICoMP—Jean-Luc af Giejerstam, MD, PhD; Eleanor Boyle, PhD; Victor G. Coronado, MD, MPH; Alison K. Godbolt, MBChB, MD; Ryan Hung, MD, MSc; Michelle Keightley, PhD; Alvin Li, BHSc; Connie Marras, MD, PhD; Peter Rumney, MD; and Britt-Marie Stålnacke, MD, PhD—for their contribution to this work; Panos Lambiris, MSc, Information Scientist, University Health Network, for assisting in developing, testing, and updating the search strategies; and Meijia Zhou, BSc, for assistance with retrieving and screening articles.
Styrke, J., Stålnacke, B.M., Sojka, P., and Björnstig, U.
Traumatic brain injuries in a well-defined population: epidemiological aspects and severity.
J Neurotrauma. 2007; 24: 1425–1436
Cassidy, J.D., Carroll, L.J., Peloso, P.M. et al.
Incidence, risk factors and prevention of mild traumatic brain injury: results of the WHO Collaborating Centre Task Force on Mild Traumatic Brain Injury.
J Rehabil Med. 2004; 36: 28–60
McCrory, P., Meeuwisse, W., Johnston, K. et al.
Consensus statement on concussion in sport—the 3rd International Conference on Concussion in Sport, held in Zurich, November 2008.
J Clin Neurosci. 2009; 16: 755–763
Carroll, L.J., Cassidy, J.D., Peloso, P.M. et al.
Prognosis for mild traumatic brain injury: results of the WHO Collaborating Centre Task Force on Mild Traumatic Brain Injury.
J Rehabil Med. 2004; 36: 84–105
Cancelliere, C., Cassidy, J.D., Côté, P. et al.
Protocol for a systematic review of prognosis after mild traumatic brain injury: an update of the WHO Collaborating Centre Task Force findings.
Syst Rev. 2012; 1: 17
Cancelliere, C., Cassidy, J.D., Li, A., Donovan, J., Côté, P., and Hincapié, C.A.
Systematic search and review procedures: results of the International Collaboration on Mild Traumatic Brain Injury Prognosis.
Arch Phys Med Rehabil. 2014; 95: S101–S131
Carroll, L.J., Cassidy, J.D., Holm, L., Kraus, J., Coronado, V.G., and WHO Collaborating Centre Task Force on Mild Traumatic Brain Injury.
Methodological issues and research recommendations for mild traumatic brain injury:
The WHO Collaborating Centre Task Force on Mild Traumatic Brain Injury
J Rehabil Med. 2004 (Feb); 36 (43): 113–125
Scottish Intercollegiate Guidelines Network (SIGN). Available at:
Accessed February 1, 2011.
Côté, P., Cassidy, J.D., Carroll, L., Frank, J.W., and Bombardier, C.
A systematic review of the prognosis of acute whiplash and a new conceptual framework to synthesize the literature.
Spine. 2001; 26: E445–E458
Moher D, Liberati A, Tetzlaff J, Altman DG (2009)
Preferred Reporting Items for Systematic Reviews and Meta-Analyses:
The PRISMA Statement
Int J Surg 2010; 8 (5): 336–341
Cancelliere, C., Hincapié, C.A., Keightley, M. et al.
Systematic review of prognosis and return to play after sport concussion: results of the International Collaboration on Mild Traumatic Brain Injury Prognosis.
Arch Phys Med Rehabil. 2014; 95: S210–S229
Boyle, E., Cancelliere, C., Hartvigsen, J., Carroll, L.J., Holm, L.W., and Cassidy, J.D.
Systematic review of prognosis after mild traumatic brain injury in the military: results of the International Collaboration on Mild Traumatic Brain Injury Prognosis.
Arch Phys Med Rehabil. 2014; 95: S230–S237
Carroll, L.J., Cassidy, J.D., Cancelliere, C. et al.
Systematic review of the prognosis after mild traumatic brain injury in adults: cognitive, psychiatric, and mortality outcomes: results of the International Collaboration on Mild Traumatic Brain Injury Prognosis.
Arch Phys Med Rehabil. 2014; 95: S152–S173
Cassidy, J.D., Boyle, E., and Carroll, L.J.
Population-based, Inception Cohort Study of the Incidence, Course, and Prognosis
of Mild Traumatic Brain Injury After Motor Vehicle Collisions
Arch Phys Med Rehabil. 2014 (Mar); 95 (3 Suppl): S278–285
Hartvigsen, J., Boyle, E., Cassidy, J.D., and Carroll, L.J.
Mild Traumatic Brain Injury After Motor Vehicle Collisions: What Are the Symptoms and Who Treats Them?
A Population-Based 1-Year Inception Cohort Study
Arch Phys Med Rehabil. 2014 (Mar); 95 (3 Suppl): S286–294
Brewer, T.L., Metzger, B.L., and Therrien, B.
Trajectories of cognitive recovery following a minor brain injury.
Res Nurs Health. 2002; 25: 269–281
Landre, N., Poppe, C.J., Davis, N., Schmaus, B., and Hobbs, S.E.
Cognitive functioning and postconcussive symptoms in trauma patients with and without mild TBI.
Arch Clin Neuropsychol. 2006; 21: 255–273
Rush, B.K., Malec, J.F., Moessner, A.M., and Brown, A.W.
Preinjury personality traits and the prediction of early neurobehavioral symptoms following mild traumatic brain injury.
Rehabil Psychol. 2004; 49: 275–281
American Psychiatric Association.
Diagnostic and statistical manual of mental disorders (DSM-IV-TR).
4th ed. APA, Washington (DC); 2000
Self-perception in mild traumatic brain injury.
Am J Phys Med Rehabil. 2002; 81: 609–621
Kashluba, S., Paniak, C., Blake, T., Reynolds, S., Toller-Lobe, G., and Nagy, J.
A longitudinal, controlled study of patient complaints following treated mild traumatic brain injury.
Arch Clin Neuropsychol. 2004; 19: 805–816
Lange, R.T., Iverson, G.L., and Rose, A.
Post-concussion symptom reporting and the “good-old-days” bias following mild traumatic brain injury.
Arch Clin Neuropsychol. 2010; 25: 442–450
Nygren-de Boussard, C., Lundin, A., Karlstedt, D., Edman, G., Bartfai, A., and Borg, J.
S100 and cognitive impairment after mild traumatic brain injury.
J Rehabil Med. 2005; 37: 53–57
Dischinger, P.C., Ryb, G.E., Kufera, J.A., and Auman, K.M.
Early predictors of postconcussive syndrome in a population of trauma patients with mild traumatic brain injury.
J Trauma. 2009; 66: 289–297
De Silva, M.J. and for the CRASH Trial Collaborators.
Patient outcome after traumatic brain injury in high-, middle- and low-income countries: analysis of data on 8927 patients in 46 countries.
Int J Epidemiol. 2009; 38: 452–458
Hou, R., Moss-Morris, R., Peveler, R., Mogg, K., Bradley, B.P., and Belli, A.
When a minor head injury results in enduring symptoms: a prospective investigation of risk factors for postconcussional syndrome after mild traumatic brain injury.
J Neurol Neurosurg Psychiatry. 2012; 83: 217–223
Gil, S., Caspi, Y., Ben-Ari, I.Z., Koren, D., and Klein, E.
Does memory of a traumatic event increase the risk for posttraumatic stress disorder in patients with traumatic brain injury? A prospective study.
Am J Psychiatry. 2005; 162: 963–969
Norrie, J., Heitger, M., Leathem, J., Anderson, T., Jones, R., and Flett, R.
Mild traumatic brain injury and fatigue: a prospective longitudinal study.
Brain Inj. 2010; 24: 1528–1538
Stulemeijer, M., van der Werf, S., Borm, G.F., and Vos, P.E.
Early prediction of favourable recovery 6 months after mild traumatic brain injury.
J Neurol Neurosurg Psychiatry. 2008; 79: 936–942
Tellier, A., Marshall, S.C., Wilson, K.G., Smith, A., Perugini, M., and Stiell, I.G.
The heterogeneity of mild traumatic brain injury: where do we stand?.
Brain Inj. 2009; 23: 879–887
Friedland, J.F. and Dawson, D.R.
Function after motor vehicle accidents: a prospective study of mild head injury and posttraumatic stress.
J Nerv Ment Dis. 2001; 189: 426–434
de Leon, M.B., Kirsch, N.L., Maio, R.F. et al.
Baseline predictors of fatigue 1 year after mild head injury.
Arch Phys Med Rehabil. 2009; 90: 956–965
McLean, S.A., Kirsch, N.L., Tan-Schriner, C.U. et al.
Health status, not head injury, predicts concussion symptoms after minor injury.
Am J Emerg Med. 2009; 27: 182–190
Stovner, L.J., Schrader, H., Mickeviciene, D., Surkiene, D., and Sand, T.
Headache after concussion.
Eur J Neurol. 2009; 16: 112–120
Ferrari, R., Obelieniene, D., Russell, A.S., Darlington, P., Gervais, R., and Green, P.
Symptom expectation after minor head injury. A comparative study between Canada and Lithuania.
Clin Neurol Neurosurg. 2001; 103: 184–190
De Almeida Lima, D.P., Filho, C.S., de Campos Vieira Abib, S., and Poli de Figueiredo, L.F.
Quality of life and neuropsychological changes in mild head trauma. Late analysis and correlation with S100B protein and cranial CT scan performed at hospital admission.
Injury. 2008; 39: 604–611
Sundström, A., Nilsson, L.G., Cruts, M., Adolfsson, R., Van Broeckhoven, C., and Nyberg, L.
Fatigue before and after mild traumatic brain injury: pre-post-injury comparisons in relation to apolipoprotein E.
Brain Inj. 2007; 21: 1049–1054
Ozen, L.J. and Fernandes, M.A.
Effects of “diagnosis threat” on cognitive and affective functioning long after mild head injury.
J Int Neuropsychol Soc. 2011; 17: 219–229
Iverson, G.L. and Lange, R.T.
Examination of “postconcussion-like” symptoms in a healthy sample.
Appl Neuropsychol. 2003; 10: 137–144
McCracken, L.M. and Iverson, G.L.
Predicting complaints of impaired cognitive functioning in patients with chronic pain.
J Pain Symptom Manage. 2001; 21: 392–396
Merrick, D. and Stålnacke, B.-M.
Five years post whiplash injury: symptoms and psychological factors in recovered versus non-recovered.
BMC Res Notes. 2010; 3: 190
Ferrari, R., Russell, A.S., Carroll, L.J., and Cassidy, J.D.
A re-examination of the whiplash associated disorders (WAD) as a systemic illness.
Ann Rheum Dis. 2005; 64: 1337–1342
Royston, P., Moons, K.G., Altman, D.G., and Vergouwe, Y.
Prognosis and prognostic research: developing a prognostic model.
BMJ. 2009; 338: b604
Nygren-de Boussard, C., Holm, L.W., Cancelliere, C. et al.
Nonsurgical interventions after mild traumatic brain injury: a systematic review. Results of the International Collaboration on Mild Traumatic Brain Injury Prognosis.
Arch Phys Med Rehabil. 2014; 95: S257–S264
Return to MILD TRAUMATIC BRAIN INJURY