Content uploaded by Rajiv Singh
Author content
All content in this area was uploaded by Rajiv Singh on Sep 03, 2019
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=ibjn20
British Journal of Neurosurgery
ISSN: 0268-8697 (Print) 1360-046X (Online) Journal homepage: https://www.tandfonline.com/loi/ibjn20
Description of the predictors of persistent post-
concussion symptoms and disability after mild
traumatic brain injury: the SHEFBIT cohort
James Booker, Saurabh Sinha, Kishor Choudhari, Jeremy Dawson & Rajiv
Singh
To cite this article: James Booker, Saurabh Sinha, Kishor Choudhari, Jeremy Dawson & Rajiv
Singh (2019): Description of the predictors of persistent post-concussion symptoms and disability
after mild traumatic brain injury: the SHEFBIT cohort, British Journal of Neurosurgery, DOI:
10.1080/02688697.2019.1598542
To link to this article: https://doi.org/10.1080/02688697.2019.1598542
Published online: 09 Apr 2019.
Submit your article to this journal
Article views: 68
View Crossmark data
Description of the predictors of persistent post-concussion symptoms and
disability after mild traumatic brain injury: the SHEFBIT cohort
James Booker
a
, Saurabh Sinha
b
, Kishor Choudhari
b
, Jeremy Dawson
c
and Rajiv Singh
a,d
a
Sheffield Institute for Translational Neuroscience, University of Sheffield, Sheffield, UK;
b
Department of Neurosurgery, Sheffield Teaching
Hospitals, Sheffield, UK;
c
Institute of Work Psychology, University of Sheffield Management School, Sheffield, UK;
d
Osborn Neurorehabilitation
Unit, Department of Rehabilitation Medicine, Sheffield Teaching Hospitals, Sheffield, UK
ABSTRACT
Introduction: Several patients who suffer Mild Traumatic Brain Injury (mTBI) develop Persistent Post-
Concussion Symptoms (PPCS) and long-term disability. Current prognostic models for mTBI have a large
unexplained variance, which limits their use in a clinical setting.
Aim: This study aimed to identify background demographics and mTBI details that are associated with
PPCS and long-term disability.
Methods: Patients from the SHEFfield Brain Injury after Trauma (SHEFBIT) cohort with mTBI in the
Emergency Department (ED) were analysed as part of the study. PPCS and long-term disability were meas-
ured using the Rivermead Post-Concussion Questionnaire and the Rivermead Post-Injury Follow-up
Questionnaire respectively, during follow up brain injury clinics.
Results: A representative mTBI sample of 647 patients was recruited with a follow-up rate of 89%. Non-
attenders were older (p<0.001), a greater proportion were retired (p<0.001) and had a greater burden
of comorbidity (p¼0.009). Multivariate analysis identified that female gender, previous psychiatric history,
GCS <15, aetiology of assault and alcohol intoxication, were associated with worse recovery.
Conclusion: These findings will support and add to current understanding of MBTI recovery in pursuit of
developing a validated prognostic model. This will allow for more accurate prognostication and eventual
improved treatment for sufferers of this complex disorder.
ARTICLE HISTORY
Received 12 February 2019
Accepted 18 March 2019
KEYWORDS
TBI; prognosis;
neurorehabilitation; PPCS
Introduction
Traumatic brain injury (TBI) is a major cause of disability and
death across the world. In England and Wales, 1.4 million
patients are admitted to the Emergency Department (ED) with a
head injury each year and TBI is the most common cause of
death in adults under 40.
1
Mild TBI (mTBI) comprises approximately 80–90% of TBI. It
is defined by a Glasgow Coma Scale (GCS) Score of 13–15 in the
ED and post-concussion symptoms that usually resolve spontan-
eously within a few weeks.
2
Nevertheless, 15–30% of mTBI
patients develop post-concussion symptoms that are debilitating
and have a negative impact on quality of life.
3,4
Post-concussion symptoms lasting longer than a month after
injury are often categorised under Post-Concussion Syndrome
(PCS).
5
However, this term is controversial; there is no clear con-
sensus on diagnostic criteria and it is unclear whether symptoms
are due to mTBI or a secondary process, such as post-traumatic
stress disorder.
6–8
Consequently, it is more accurate to class the
somatic, cognitive, sleep and affective sequelae of mTBI as
Persistent Post-Concussion Symptoms (PPCS), rather than cate-
gorising the symptoms under the indiscriminate term, PCS.
9
It is important to provide an accurate prognosis for patients
with mTBI. It will be valuable for clinicians, allowing them to
give more reliable information to patients and relatives. It will
also allow for the implementation of early intervention for
patients at high risk of developing PPCS. However the current
prognostic models for mTBI are unsatisfactory with a large pro-
portion of unexplained variance and limited external validation.
10
Many of these models predict recovery by dichotomising the
Glasgow Outcome Scale-Extended (GOSE), into ‘favourable’and
‘unfavourable’outcome.
11
This methodology is inappropriate for
use in mTBI prognostication because it results in a ‘ceiling
effect’, where the majority of patients achieve the best outcome.
Consequently, it is difficult to identify patients who have had a
‘favourable’recovery but where symptoms and disabil-
ity persist.
12
Accordingly more refined differentiation of outcome is
required. This can be achieved by ordinal analysis of the GOSE
(unpublished work Booker J, Singh R, et al. 2019) and the use of
additional outcome measures to more accurately map out the
symptom burden after mTBI.
10,13
Recently, prognostic models
for mTBI are beginning to be centred around validated outcome
measures other than the GOSE, and these appear promising.
14–16
This study aims provide a robust and representative descrip-
tion of the factors that predict PPCS by utilising outcome meas-
ures of PPCS and disability.
Methods
This observational cohort study is part of the SHEFfield Brain
Injury after Trauma (SHEFBIT) study, based at a large
University Hospital and investigates the recovery of patients after
CONTACT James Booker jb38g14@soton.ac.uk Sheffield Institute for Translational Neuroscience, University of Sheffield, Sheffield, UK
ß2019 The Neurosurgical Foundation
BRITISH JOURNAL OF NEUROSURGERY
https://doi.org/10.1080/02688697.2019.1598542
a TBI. Patients are routinely followed-up at Brain Injury Clinics
(BICs) at 8–10 weeks and one year after TBI. Patients provided
informed consent. As part of the BICs, patients completed four
validated questionnaires of TBI recovery and an extensive history
was taken. These self-rated questionnaires were used as outcome
measures of TBI recovery.
The inclusion criteria for this study were patients presenting
to the ED with mTBI (GCS 13–15), between August 2011 and
July 2015. Patients must be registered with a local GP to ensure
follow-up and have had a CT scan while admitted in the ED.
Patients in the SHEFBIT cohort were excluded if they were below
16 years old or had previously received in-patient care for a TBI.
The cause of mTBI was categorised using the UK Trauma
Audit and Research Network classification. An additional cat-
egory ‘sports’was added and falls greater than two metres were
categorised under ‘other’.
17
Each CT report was reviewed by the Principal Investigator
and classified using the “overall appearance”classification.
18
In
this study, CT findings were dichotomised into ‘CT positive’and
‘CT negative’findings.
Patients’socioeconomic status was categorised according to
the National Statistics Socioeconomic Classification. This has
nine categories including a student group.
19
The key outcome measures are the Rivermead Post-
Concussion Questionnaire (RPQ) and the Rivermead Head
Injury Follow-up Questionnaire (RHFUQ). The RPQ is a short
and reliable tool for measuring symptoms after mTBI.
20
It con-
sists of 16 brief questions that are self-rated on a scale between 0
(not experienced at all) and 4 (a severe problem), giving a total
score out of 64.
21
The RHFUQ is particularly sensitive in mTBI
and considers the individual’s perception of the everyday conse-
quences of loss of function as opposed to only considering the
impairment itself. It is comprised of 10 questions with subjective
rating outcomes ranging from 0 (no change) to 4 (a very marked
change), to give a total score of 0–40.
22
Univariate analysis of the RPQ and RHFUQ was done using
Mann-Whitney U test for variables with two categories and
Kruskal-Wallis test for variables with three categories. A post-hoc
Benjamini-Hochman correction was done on the univariate
results to control for type-I errors in the context of multiple
tests. Secondary multivariate analysis of the RPQ and RHFUQ
was performed using an ANCOVA model to determine whether
individual predictors significantly affected the outcome measures,
whilst controlling for covariates. The continuous variable age was
adapted to form 10 groups of equal numbers, to allow for ana-
lysis using an ANCVOA model.
Results
Between August 2011 and July 2015 there were 647 patients in
the SHEFBIT cohort that attended the BIC 8–10 weeks after
mTBI. Of these, 72 patients were excluded from the study; 51
did not attend the BIC at one-year and 21 died. The remaining
575 patients were analysed as part of the study. This represented
a high follow-up rate, with 89% of patients eligible at induction
being analysed (Figure 1).
Of the 575 patients who attended the BIC at one-year after
mTBI: 64% were male and 92% were of white ethnicity. Mean
age was 45.4 (SD 19.9) years and 70% were employed. In the
study population, 17% of patients had a psychiatric history before
the mTBI and 26% had an existing medical comorbidity
(Table 1).
A comparison was made between the demographics of attend-
ers and non-attenders using chi-squared and independent t-tests.
This showed a number of background variables that were signifi-
cantly different (p<0.05). Non-attenders were older (p<
0.001), a greater proportion were retired (p<0.001) and had a
greater burden of comorbidity (p¼0.009) (Table 1).
In terms of injury characteristics, the majority of patients had
GCS 15 (62%) and had normal CT scans (70%). The commonest
aetiology was falls (36%), followed by Road Traffic Collisions
(RTC) (26%) and assault (20%). Of the patients, 24% were intoxi-
cated at the time of injury and 9% were on long-term anticoagu-
lation treatment.
At the one-year follow-up, post-concussion symptoms and
perceived disability scores were high, although both results were
skewed to the lower scores. Mean scores were 12.34 (SD 10.39)
for RPQ and 8.67 (SD 8.66) in RHFUQ (see Figure 2).
A multivariate analysis of the independent variables using
ANCOVA was performed. Several dichotomous factors had a
statistically significant (p<0.05) effect on both outcome meas-
ures: gender (p<0.05), psychiatric history (p<0.001) and alco-
hol intoxication (p<0.05). Other independent variables with
multiple categories showed that they had a significant effect on
Figure 1. Study flow diagram. TBI: Traumatic Brain Injury; mTBI: mild Traumatic Brain Injury; GCS: Glasgow Coma Scale; GOSE: Glasgow Outcome Scale-Extended;RPQ:
Rivermead Post-Concussion Questionnaire; RHFUQ: Rivermead Post-Injury Follow-up Questionnaire.
2 J. BOOKER ET AL.
outcome in comparison to the baseline variable (0
a
): mTBI with
GCS 13 (p<0.001) and 14 (p<0.001) in comparison to 15,
aetiology of assault and “other”in comparison to mTBI from
falls as baseline.
However there were some differences between the RPQ and
RHFUQ scores. Age below 75 was a significant predictor of
higher RHFUQ scores but did not affect RPQ scores.
Additionally, being unemployed or in a higher socioeconomic-
class before the mTBI, predicted a higher RPQ score but this was
not seen in the RHFUQ. Several predictive factors had no effect
on any of the outcome measures: ethnicity, CT lesion status,
medical comorbidity and warfarin use (Tables 2 and 3).
Levene’s test and normality checks were carried out for the
ANCOVA model and assumptions were met.
The interaction of the highly significant variables on the RPQ
and RHFUQ was plotted using scatter graphs (see Figures 3
and 4). Gender showed no obvious variation between male and
female. Previous psychiatric history showed a clear variation with
patients who did not have a psychiatric history focused around
the lower scores of the RPQ and RHFUQ. Additionally, accord-
ing to the lines of best fit the presence of a psychiatric history
predicted worse RPQ but not worse RHFUQ. Exploring the
interaction of the GCS shows a clear difference in patients with
GCS 15, that are centred at lower RPQ and RHFUQ scores rela-
tive to GCS 13 and 14. The interaction of alcohol intoxication on
the RPQ and RHFUQ is seen to be variable from the crossing of
the lines of best fit. At low RPQ scores, alcohol intoxication pre-
dicts higher RHFUQ scores but at high RPQ scores, alcohol
intoxication predicts lower RHFUQ scores.
Discussion
This study analysed patients from the SHEFBIT cohort and
aimed to identify factors that were significantly associated with
worse symptoms and disability at one year after mTBI. The
results indicate that female gender previous psychiatric history,
GCS 13 or 14, aetiology of assault and alcohol intoxication were
associated with worse outcome.
There was considerable variation within the mTBI group.
There is an especially large range in the age of patients with
mTBI; this is in-line with previously observations showing that
mTBI follows a tri-modal age demographic in the UK,
23
children
(0–4 years) and the elderly (>75 years) are at risk of fall-related
mTBI, while adolescents (13–19 years) are at risk of RTC and
assault-related mTBI, often involving alcohol.
1,23–25
When com-
paring patients who were lost in follow-up with patients who
attended, several significant differences were found between the
groups. Non-attenders were significantly older, a greater propor-
tion was retired and a larger percentage had an existing comor-
bidity. Previous work has indicated that younger patients are
more aware of the PPCS after mTBI and accordingly are more
motivated to attend follow-up appointments.
26
ANCOVA analysis of the RPQ and RFUQ variables found
background variables of female gender and a psychiatric history
led to a significantly worse outcome after mTBI. Additionally,
graphical interpretation of the interactions showed that previous
psychiatric history has a discernable difference on outcome
scores, whereas this is not seen with gender (see Figure 3). There
is currently conflicting evidence as to whether there is a signifi-
cant association with female gender and PPCS.
27,28
Gender is a
complex background characteristic to dissect causality from. Of
note gender norms and role expectation differences between gen-
ders are particularly difficult to control for in human studies.
Under male gender norms it has been observed that individuals
express more ‘risk taking’behaviour and report fewer signs and
symptoms after TBI.
29
We reason this is a possible explanation
for the association between female gender and a larger burden of
reported PPCS and disability in this study. Nevertheless, further
research is required to determine if treatment following mTBI
should be stratified according to gender to reflect this difference
in recovery.
Table 1. Patient background demographics. This table compares the background demographics of patients lost in follow-up and
enrolled patients.
Demographic
Lost in follow-up
n¼72
Enrolled
n¼575 v
2-
test/t test df p-value
Gender
i. Male 47 (65%) 370 (64%) 0.001 1 0.980
ii. Female 25 (35%) 205 (36%)
Average Age in Years 57.4 (SD 21.4) 45.4 (SD 19.9) 4.783 645 <0.001
Ethnicity
i. White 69 (96%) 532 (92%) 0.620 1 0.431
ii. Non-white 3 (4%) 43 (8%)
Socioeconomic Class (NC-SEC)
i. Professional 3 (4%) 45 (8%) 11.553 8 0.172
ii. Lower Management 10 (14%) 93 (16%)
iii. Intermediate 7 (10%) 57 (10%)
iv. Small Employer 6 (8%) 42 (7%)
v. Lower Supervisory 21 (29%) 94 (16%)
vi. Semi-routine 16 (22%) 115 (20%)
vii. Routine 4 (6%) 59 (10%)
viii. Never Worked 4 (6%) 35 (6%)
ix. Students 1 (1%) 35 (6%)
Employment before mTBI
i. Employed 32 (44%) 405 (70%) 24.401 2 <0.001
ii. Unemployed 13 (18%) 1.81 (14%)
iii. Retired 27 (38%) 89 (16%)
Comorbidity
i. Yes 30 (42%) 151 (26%) 6.792 1 0.009
ii. No 42 (58%) 424 (74%)
mTBI, mild traumatic brain injury.
indicates significance at p¼0.05. indicates sigificance of p¼<0.001.
BRITISH JOURNAL OF NEUROSURGERY 3
Previous work in the SHEFBIT cohort has found a significant
association between depression and PPCS.
30
However, it is
unclear in the literature as to whether the presence of a psychi-
atric history predisposes to PPCS. A recent study found that
patients with a psychiatric history had lower probability of
returning to work.
31
We reason that this makes it more difficult
for patients to return to their normal lives, leaving a residual
symptomology and disability from the mTBI to persist.
Intriguingly, our results suggest that older age leads to a
minor disability after mTBI according to the RHFUQ, but has no
difference in PPCS. However, the RHFUQ score is a measure of
disability in reference to pre-injury functionality.
21
Therefore,
patients who are in the oldest age group are likely to have had
reduced functionality prior to the mTBI and subsequently experi-
ence a smaller relative change in function.
With respect to injury features, ANCOVA analysis of the
RPQ and RHFUQ also showed that a mTBI with an aetiology of
assault and GCS score of 13 or 14 was associated with worse out-
come. The effect of GCS <15 on the outcome measures are seen
in Figure 4 with the majority of patients with GCS achieving few
PPCS and long-term disability. In both GCS <15 and aetiology
of assault, the mechanism of injury is caused by a more substan-
tive impact to the head than in other mTBIs. As a result, there is
a more severe brain injury, which may lead to a greater burden
of symptoms which could lead to enduring disability after
mTBI.
32
Interestingly, aetiology of RTC was found to be associ-
ated with worse disability after mTBI but not PPCS. There are
multiple factors that could influence this result. For example, fol-
lowing assault or RTC, it is common for patients to be involved
in litigation, the role of which is controversial in TBI outcome.
This may impact on the psychological recovery of patients as
they are constantly reliving their injury.
33
Additionally, patients
involved in an accident perceived as ‘not their fault’, might also
be anxious about returning to a similar environment. Adoption
of a sickness role may result in a greater perceived disability
after mTBI.
The results show a worse outcome with alcohol intoxication is
present, in both the RPQ and RHFUQ measures. Graphical inter-
pretation of this interaction shows that when patients have few
PPCS, alcohol intoxication leads to worse long-term disability.
Figure 2. Rivermead Post-Concussion Questionnaire and Rivermead Post-Injury Follow-up Questionnaire scores. RPQ: Rivermead Post-Concussion Questionnaire; RHFUQ:
Rivermead Post-Injury Follow-up Questionnaire.
4 J. BOOKER ET AL.
However, when patients suffer many PPCS, alcohol intoxica-
tion appears to have a protective affect against long-term dis-
ability (see Figure 3). This reflects the conflicting evidence in
the literature with regards to the effect of alcohol on recov-
ery after mTBI.
30,34
It is known that alcohol intoxication
lowers the apparent GCS score and hence, an exaggerated
perception of TBI severity.
35
As a result, patients with
alcohol intoxication have been shown to have better recovery
trajectories compared to non-intoxicated patients.
36
Anecdotally, in this study alcohol intoxication often seemed
to reflect patients with chaotic-lifestyles and social problems.
This is likely to have negatively impacted patient’s coping
strategy and recovery. Further research is needed to elaborate
on this finding.
Table 2. Analysis of covariance of the rivermead Post-Concussion questionnaire.
Predictor b
95%
Confidence
Interval
Lower Upper p-value
Gender
Male 1.8 3.5 0.1 0.040
Female 0
a
–– –
Age at injury (groups)
15–20 0.4 3.8 4.6 0.840
20–25 1.1 5.3 3.1 0.609
25–31 2.6 1.5 6.7 0.217
31–39 1.7 2.4 5.8 0.405
39–45 2.4 1.6 6.5 0.240
45–51 3.0 1.0 6.9 0.141
51–59 2.9 6.7 0.9 0.132
59–66 1.6 5.2 2.1 0.396
66–75 1.5 5.1 2.1 0.401
75þ0
a
–– –
Ethnicity
White 0
a
–– –
Non-white 2.3 0.8 5.4 0.140
Socioeconomic Class (NC-SEC)
Professional 4.9 0.4 9.3 0.032
Lower Management 4.3 0.3 8.2 0.033
Intermediate 3.5 0.7 7.7 0.100
Small Employer 2.3 2.2 6.8 0.323
Lower Supervisor 4.0 0.1 8.0 0.043
Semi-routine 2.6 1.2 6.4 0.178
Routine 4.0 0.2 8.2 0.062
Never Worked 0
a
–– –
Students 3.0 2.0 7.9 0.238
Employment status before mTBI
Employed 0
a
–– –
Unemployed 2.8 0.3 5.3 0.029
Retired 1.1 4.0 1.8 0.449
Psychiatric history
No 6.8 9.0 4.6 <0.001
Yes 0
a
–– –
Comorbidity
No 0.7 1.6 2.9 0.557
Yes 0
a
–– –
Aetiology
Fall 0
a
–– –
RTC 2.2 0.1 4.6 0.061
Assault 2.9 0.4 5.5 0.026
Sport 1.5 1.7 4.8 0.355
Other 3.1 0.1 6.1 0.043
GCS
13 6.3 3.7 8.9 <0.001
14 4.0 2.0 6.0 <0.001
15 0
a
–– –
CT Lesion
Negative 0.1 2.0 1.8 0.912
Positive 0
a
–– –
Alcohol Intoxication
No 2.6 4.6 0.6 0.011
Yes 0
a
–– –
Warfarin Use
No 0.2 2.8 3.2 0.897
Yes 0
a
–– –
mTBI: mild Traumatic Brain Injury; GCS: Glasgow Coma Scale; RTC: Road
Rraffic Collision.
indicates p<0.05. indicates p<0.001.
0
a
, reference category.
Table 3. Analysis of covariance of the rivermead Post-Injury follow-up
questionnaire.
Predictor b
95%
Confidence
Interval
Lower Upper p-value
Gender
Male 1.9 3.3 0.5 0.012
Female 0
a
–– –
Age at injury (groups)
15–20 2.7 0.8 6.1 0.004
20–25 1.2 2.2 4.6 <0.001
25–31 3.9 0.5 7.3 0.009
31–39 3.4 <0.0 6.7 0.007
39–45 4.0 0.6 7.3 0.009
45–51 4.8 1.6 8.1 0.015
51–59 0.8 2.3 3.9 <0.001
59–66 0.9 2.1 3.9 <0.001
66–75 0.4 2.6 3.4 <0.001
75þ0
a
–– –
Ethnicity
White 0
a
–– –
Non-white 0.6 1.9 3.1 0.624
Socioeconomic Class (NC-SEC)
Professional 3.5 0.2 7.2 0.061
Lower Management 3.1 0.1 6.4 0.059
Intermediate 3.5 >0.0 6.9 0.049
Small Employer 2.6 1.1 6.3 0.171
Lower Supervisor 3.1 0.1 6.3 0.059
Semi-routine 2.4 0.7 5.5 0.126
Routine 3.4 0.1 6.8 0.055
Never Worked 0
a
–– –
Students 3.7 0.4 7.7 0.077
Employment status before mTBI
Employed 0
a
–– –
Unemployed 1.7 0.4 3.7 0.111
Retired 1.3 3.6 1.1 0.295
Psychiatric history
No 6.0 8.0 4.1 <0.001
Yes 0
a
–– –
Comorbidity
No 0.6 2.4 1.2 0.522
Yes 0
a
–– –
Aetiology
Fall 0
a
–– –
RTC 2.4 0.5 4.3 0.015
Assault 3.8 1.7 5.9 <0.001
Sport 1.1 1.6 3.8 0.425
Other 3.3 0.7 5.8 0.008
GCS
13 5.4 3.2 7.6 <0.001
14 3.6 2.0 5.2 <0.001
15 0
a
–– –
CT lesion
Negative 0.1 1.5 1.6 0.938
Positive 0
a
–– –
Alcohol intoxication
No 2.7 4.3 1.0 <0.001
Yes 0
a
–– –
Warfarin use
No 0.9 3.4 1.5 0.461
Yes 0
a
–– –
mTBI: mild Traumatic Brain Injury; GCS: Glasgow Coma Scale; RTC: Road
Rraffic C ollision.
indicates p<0.05. indicates p<0.001.
0
a
, reference category.
BRITISH JOURNAL OF NEUROSURGERY 5
Finally, an important negative finding identified is a lesion on
CT scan was not associated with worse PPCS or long-term dis-
ability. This supports previous research indicating that CT scan-
ning is insensitive at identifying the pathology within the brain
caused by mTBI.
37
This brings into question the use of CT scans
as a measure of mTBI severity and prognosis in current practice.
The evidence should inform clinicians not to rely on CT scan
findings when treating patients after mTBI. In contrast, lesions
identified using diffusion tensor imaging (DTI) has been shown
to correlate closely with PPCS.
38
However, DTI is not widely
available at trauma centres and is used more as a research tool.
This study has a lot of strengths. Foremost is the large popu-
lation of patients with mTBI that were prospectively recruited,
with very few exclusion criteria. There is an excellent follow-up
rate with few patients lost; TBI studies are notorious for their
very high attrition rate of up to 70% in 36 months.
39
This results
in a highly representative population of patients with mTBI, with
results that are applicable to the general population. Many stud-
ies have recruited from select populations e.g. soldiers or sports
players and are not easily compared to other groups of
mTBI.
40,41
Additionally, this study used sensitive measures of
recovery after mTBI that will be better able to differentiate
between subtle differences in post-injury recovery and disability
between patients.
10
While the GOSE is the most common global
outcome measure for TBI [11], it is less useful for the assessment
of outcome after mTBI, where the symptoms and disability are
Figure 3. Interaction of Gender and Previous Psychiatric History. RPQ: Rivermead Post-Concussion Questionnaire; RHFUQ: Rivermead Post-Injury Follow-up
Questionnaire; Prev Psych: Previous Psychiatric History.
6 J. BOOKER ET AL.
often less obvious. This study also had a long follow-up period
of one-year post mTBI. This ensures that the long-term recovery
of mTBI is truly considered. The majority of published studies of
mTBI recovery followed up patients for three to six months. This
is largely a result of the International Classification of Diseases-
10 and the Diagnostic and Statistical Manual of Mental
Disorders-IV criteria of PCS, which defines the minimum symp-
tom duration of one-month and three-months, post-mTBI
respectively. However, it is known that a proportion of mTBI
patients have prolonged symptoms which exist beyond these
time-points.
6
Additionally, the use of face-to-face interviews and
observations made by a single clinician (principal investigator) is
another strength, limiting inter-observer bias.
A few weaknesses must also be outlined. A number of poten-
tial confounding factors were not accounted for in this study that
have been shown to modulate outcome of mTBI in other studies.
These include litigation, personality type and intelligence.
33,42,43
Unlike some studies, we did not have a control population of
similar demographics without mTBI.
44–48
Therefore, the general-
ised symptoms may be attributable to other causes that are pre-
sent in the background symptoms of the population and are not
specific to mTBI. This includes symptoms of depression and anx-
iety that are common in the population.
New developments in research methodology are changing the
measurement of TBI outcome. This is part a result of a concerted
effort by the International Initiative for Traumatic Brain Injury
Figure 4. Interaction of Glasgow Coma Scale Score and Alcohol Intoxication in the Emergency Department. RPQ: Rivermead Post-Concussion Questionnaire; RHFUQ:
Rivermead Post-Injury Follow-up Questionnaire; GCS: Glasgow Coma Scale; Intoxicated: Alcohol Intoxication in Emergency Department.
BRITISH JOURNAL OF NEUROSURGERY 7
Research.
10
The emergence of more accurate prognostic models
that use newer, sensitive outcome measures of recovery may help
better elucidation of risk factors..
14,16
However, even in these
models there remains a substantial unexplained variance in out-
come. In the future more needs to be done to investigate the
remaining variables responsible for the variance. Of note, the
role of genetic variation remains relatively unexplored. The apoli-
poprotein gene (APOE) is the most widely researched, but the
exact effect it has on TBI remains poorly understood. One study
found that APOE E4 has a detrimental effect on recovery
49
but
this effect could not be replicated.
50
A prognostic model of mTBI
would provide valuable information to clinicians, patients and
families but it will also improve clinical judgement and decision-
making in the treatment of this complex condition.
Conclusion
The aim of this study was to identify factors that are significantly
associated with a greater burden of PPCS and disability after
mTBI. Analysis showed that the background variables: female
gender and psychiatric history, as well as a mTBI with decreased
GCS score, aetiology of assault and alcohol intoxication all lead
to worse outcomes. These findings will support and add to cur-
rent understanding of mTBI recovery in pursuit of developing a
validated prognostic model. Future research into the role of gen-
etic variation within mTBI recovery may lead to more accurate
prognostication and eventual improved treatment for sufferers of
this complex disorder.
Disclosure statement
No potential conflict of interest was reported by the authors.
References
1. Lawrence T, Helmy A, Bouamra O, et al. Traumatic brain injury in
England and Wales: prospective audit of epidemiology, complications
and standardised mortality. BMJ Open 2016;6:e012197.
2. Blennow K, Brody DL, Kochanek PM, et al. Traumatic brain injuries.
Nat Rev Dis Primers 2016;2:16084.
3. Hou R, Moss-Morris R, Peveler R, et al. When a minor head injury
results in enduring symptoms: a prospective investigation of risk fac-
tors for postconcussional syndrome after mild traumatic brain injury.
J Neurol Neurosurg Psychiatry 2012;83:217–23.
4. Zumstein MA, Moser M, Mottini M, et al. Long-term outcome in
patients with mild traumatic brain injury: a prospective observational
study. J Trauma Acute Care Surg 2011;71:120–27.
5. World Health Organization, 1992. The ICD-10 Classification of Mental
and Behavioural Disorders: Clinical descriptions and diagnostic guide-
lines. Geneva: World Health Organization.
6. Barlow KM. Postconcussion syndrome: a review. J Child Neurol 2016;
31:57–67.
7. Bramley H, Hong J, Zacko C, et al. Mild traumatic brain injury and
post-concussion syndrome: treatment and related sequela for persistent
symptomatic disease. Sports Med Arthrosc 2016;24:123–9.
8. Lagarde E, Salmi L-R, Holm LW, et al. Association of symptoms fol-
lowing mild traumatic brain injury with posttraumatic stress disorder
vs. postconcussion syndrome. JAMA Psychiatry 2014;71:1032–40.
9. Arciniegas DB. Mild traumatic brain injury: a neuropsychiatric
approach to diagnosis, evaluation, and treatment. Neuropsychiatr Dis
Treat 2005;1:311–27.
10. Maas AIR, Menon DK, Adelson PD, et al. Traumatic brain injury:
integrated approaches to improve prevention. clinical care, and
research. Lancet Neurol 2017;16:987–1048.
11. Silverberg ND, Gardner AJ, Brubacher JR, et al. Systematic review of
multivariable prognostic models for mild traumatic brain injury.
J Neurotrauma 2015;32:517–26.
12. Shukla D, Devi BI, Agrawal A. Outcome measures for traumatic brain
injury. Clin Neurol Neurosurg 2011;113:435–41.
13. Weir J, Steyerberg EW, Butcher I, et al. Does the extended Glasgow
Outcome Scale add value to the conventional Glasgow Outcome Scale?.
J Neurotrauma 2012;29:53–58.
14. Cnossen MC, et al. Development of a prediction model for post-con-
cussive symptoms following mild traumatic brain injury: a TRACK-
TBI pilot study. J Neurotrauma 2017;34:2396–409.
15. Stulemeijer M, van der Werf S, Borm GF, et al. Early prediction of
favourable recovery 6 months after mild traumatic brain injury.
J Neurol Neurosurg Psychiatry 2008;79:936–42.
16. Cnossen MC, van der Naalt J, Spikman JM, et al. Prediction of persist-
ent post-concussion symptoms after mild traumatic brain injury.
J Neurotrauma 2018;35:2691–8.
17. Lecky F, Woodford M, Yates DW. Trends in trauma care in England
and Wales 1989-97. UK Trauma Audit and Research Network. Lancet
2000;355:1771–75.
18. Wardlaw JM, Easton VJ, Statham P. Which CT features help predict
outcome after head injury? J Neurol Neurosurg Psychiatry 2002;72:
188–92–discussion 151.
19. Rose D, Pevalin D. A researcher’s guide to the national statistics socio-
economic classification, 6 bonhill street, London England EC2A 4PU
United Kingdom: SAGE Publications, Ltd.; 2003.
20. de Guise E, B
elanger S, Tinawi S, et al. Usefulness of the rivermead
postconcussion symptoms questionnaire and the trail-making test for
outcome prediction in patients with mild traumatic brain injury. Appl
Neuropsychol Adult 2016;23:213–22.
21. King NS, Crawford S, Wenden FJ, et al. The Rivermead Post
Concussion Symptoms questionnaire: a measure of symptoms com-
monly experienced after head injury and its reliability. J Neurol 1995;
242:587–92.
22. Crawford S, Wenden FJ, Wade DT. The Rivermead head injury follow
up questionnaire: a study of a new rating scale and other measures to
evaluate outcome after head injury. J Neurol Neurosurg Psychiatry
1996;60:510–4.
23. National Clinical Guideline Centre (UK), 2014. head injury: triage,
assessment, investigation and early management of head injury in chil-
dren, young people and adults. 10 Spring Gardens London SW1A
2BU: National Institute for Health and Care Excellence.
24. Faul M, Xu L, Wald MM, et al. Traumatic brain injury in the United
States: national estimates of prevalence and incidence, 2002–2006.
Injury Prevention 2011;16:A268.
25. Fu TS, Jing R, McFaull SR, et al. Health & economic burden of trau-
matic brain injury in the emergency department. Can J Neurol Sci
2016;43:238–47.
26. Patel MB, Wilson LD, Bregman JA, et al. Neurologic functional and
quality of life outcomes after tbi: clinic attendees versus non-attendees.
J Neurotrauma 2015;32:984–89.
27. Ponsford J, Willmott C, Rothwell A, et al. Factors influencing outcome
following mild traumatic brain injury in adults. J Inter Neuropsych Soc
2000;6:568–79.
28. Waljas M, Iverson GL, Lange RT, et al. A prospective biopsychosocial
study of the persistent post-concussion symptoms following mild trau-
matic brain injury. J Neurotrauma 2015;32:534–47.
29. Kroshus E, Baugh CM, Stein CJ, et al. Concussion reporting, sex, and
conformity to traditional gender norms in young adults. J Adolescence
2017;54:110–19.
30. Singh R, Mason S, Lecky F, et al. Prevalence of depression after TBI in
a prospective cohort: The SHEFBIT study. Brain Inj 2018;32:84–90.
31. Bertisch H, Satris G, Temkin N, et al. Rehabilitation trajectories and
outcomes in individuals with mild traumatic brain injury and psychi-
atric histories: A TRACK-TBI Pilot Study. J Head Trauma Rehabil
2019;34:36–44.
32. Stulemeijer M, van der Werf S, Bleijenberg G, et al. Recovery from
mild traumatic brain injury: a focus on fatigue. J Neurol 2006;253:
1041–47.
33. Bayen E, Jourdan C, Ghout I, et al. Negative impact of litigation proce-
dures on patient outcomes four years after severe traumatic brain
injury: results from the PariS-traumatic brain injury study. Disabil
Rehabil 2018;40:2040–7.
34. van der Naalt J, Timmerman ME, de Koning ME, et al. Early predic-
tors of outcome after mild traumatic brain injury (UPFRONT): an
observational cohort study. Lancet Neurol 2017;16:532–40.
35. Brickley MR, Shepherd JP. The relationship between alcohol intoxica-
tion, injury severity and Glasgow Coma Score in assault patients.
Injury 1995;26:311–4.
8 J. BOOKER ET AL.
36. Shahin H, Gopinath SP, Robertson CS. Influence of alcohol on early
Glasgow Coma Scale in head-injured patients. J Trauma 2010;69:
1176–81–discussion 1181.
37. Eierud C, Craddock RC, Fletcher S, et al. Neuroimaging after mild
traumatic brain injury: review and meta-analysis. NeuroImage Clinical
2014;4:283–94.
38. Khong E, Odenwald N, Hashim E, Cusimano MD. Diffusion tensor
imaging findings in post-concussion syndrome patients after mild trau-
matic brain injury: a systematic review. Frontiers Neurol 2016;7:156.
39. Corrigan JD, Bogner JA, Mysiw WJ, et al. Systematic bias in outcome
studies of persons with traumatic brain injury. Arch Phys Med Rehabil
1997;78:132–7.
40. Hoge CW, McGurk D, Thomas JL, et al. Mild traumatic brain injury
in U.S. Soldiers returning from Iraq. N Engl J Med 2008;358:453–63.
41. Cancelliere C, Hincapi
e CA, 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–S29.
42. Stulemeijer M, Andriessen TMJC, Brauer JMP, et al. Cognitive per-
formance after mild traumatic brain injury: the impact of poor effort
on test results and its relation to distress, personality and litigation.
Brain Injury 2007;21:309–18.
43. Fay TB, Yeates KO, Taylor HG, et al. Cognitive reserve as a moderator
of postconcussive symptoms in children with complicated and
uncomplicated mild traumatic brain injury. J Int Neuropsychol Soc
2010;16:94–105.
44. Meares S, Shores EA, Taylor AJ, et al. The prospective course of post-
concussion syndrome: the role of mild traumatic brain injury.
Neuropsychology 2011;25:454–65.
45. Losoi H, Silverberg ND, W€
aljas M, et al. Recovery from mild traumatic
brain injury in previously healthy adults. J Neurotrauma 2016;33:
766–76.
46. Dean PJA, O’Neill D, Sterr A. Post-concussion syndrome: prevalence
after mild traumatic brain injury in comparison with a sample without
head injury. Brain Inj 2012;26:14–26.
47. Caplain S, Blancho S, Marque S, et al. Early detection of poor outcome
after mild traumatic brain injury: predictive factors using a multidi-
mensional approach a pilot study. Front Neurol 2017;8:666.
48. McCauley SR, Wilde EA, Miller ER, et al. Preinjury resilience and
mood as predictors of early outcome following mild traumatic brain
injury. Journal of Neurotrauma 2013;30:642–52.
49. Ponsford J, McLaren A, Sch€
onberger M, et al. The association between
apolipoprotein E and traumatic brain injury severity and functional
outcome in a rehabilitation sample. Journal of Neurotrauma 2011;28:
1683–92.
50. Teasdale GM, Murray GD, Nicoll JAR. The association between APOE
epsilon4, age and outcome after head injury: a prospective cohort
study. Brain 2005;128:2556–61.
BRITISH JOURNAL OF NEUROSURGERY 9
