Current Pain and Headache Reports (2018) 22:64

https://doi.org/10.1007/s11916-018-0719-z

CONCUSSION AND HEAD INJURY (A FINKEL, SECTION EDITOR)

Imaging Post-Traumatic Headache

Jill C. Rau 1 & Gina M. Dumkrieger 1 & Catherine D. Chong 1 & Todd J. Schwedt 1

# Springer Science+Business Media, LLC, part of Springer Nature 2018

Abstract
Purpose of Review Headache is a frequent and debilitating symptom after mild traumatic brain injury, yet little is known about its
pathophysiology and most effective treatments. The goal of this review is to summarize findings from imaging studies used
during the clinical evaluation and research investigation of post-traumatic headache (PTH).
Recent Findings There are no published recommendations or guidelines for when to acquire imaging studies of the head or neck
in patients with PTH. Clinical acumen is required to determine if imaging is needed to assess for a secondary cause of headache
which may have been precipitated or unmasked by the trauma. Several guidelines for when to image the patient with mild
traumatic brain injury (mTBI) in the emergency setting consider headache among the deciding factors. In the research arena,
imaging techniques including proton spectroscopy magnetic resonance imaging, diffusion tensor imaging, magnetic resonance
morphometry, and functional neck x-rays have been employed with the goal of identifying diagnostic and prognostic factors for
PTH and to help understand its underlying pathophysiologic mechanisms. Results indicate that changes in regional cortical
thickness and damage to specific white matter tracts warrant further research. Future research should interrogate whether these
imaging findings contribute to the classification and prognosis of PTH.
Summary Current research provides evidence that imaging findings associated with PTH may be distinct from those attributable
to mTBI. A variety of imaging techniques have potential to further our understanding of the pathophysiologic processes
underlying PTH as well as to provide diagnostic and prognostic indicators. However, considerable work must be undertaken
for this to be realized.

Keywords Post-traumatic headache . Mild traumatic brain injury . Concussion . Magnetic resonance imaging . Diffusion tensor
imaging . Magnetic resonance spectroscopy

Introduction The term mild traumatic brain injury (mTBI) is often used
interchangeably with “concussion.” The International
There are over 2.5 million hospital visits for traumatic brain Classification of Headache Disorders III (ICHD-3) diagnostic
injury (TBI) each year in the USA and the Centers for Disease criteria for mTBI are shown in Table 1. The most frequently
Control and Prevention (CDC) has labeled concussion a seri- occurring symptom of mTBI, reported in up to 90% of pa-
ous public health issue [1]. “Mild” TBIs account for 70–90% tients, is headache [3, 4]. Individuals experiencing mTBI more
of all TBIs and worldwide are estimated to affect 42 million frequently report having headaches than patients with moder-
people yearly [2]. These TBIs result in considerable cost to ate or severe TBI [5–7]. Post-traumatic headaches are more
society in terms of healthcare dollars, lost productivity, and debilitating than those of non-traumatic origin [8]. Moreover,
disability. headache is a poor prognostic factor; patients who report
headache as a symptom of mTBI in the acute setting have a
longer recovery periods [3, 9–11] and are more likely to have
This article is part of the Topical Collection on Concussion and Head persistent post-concussion symptoms at 1 and 3 months [3,
Injury 12–15]. As many as 33% of individuals will continue to have
headaches at 1 and 5 years after mTBI [16].
* Todd J. Schwedt
Whiplash, a common accompaniment of rapid
[email protected]
acceleration-deceleration injuries, involves a sudden uncon-
1
Mayo Clinic Arizona, 5777 East Mayo Boulevard, trolled flexion-extension of the neck and can cause neurolog-
Phoenix, AZ 85054, USA ic, psychologic, cognitive, and sleep symptoms that are very
64 Page 2 of 9 Curr Pain Headache Rep (2018) 22:64

Table 1 ICHD-3 diagnostic

criteria Mild traumatic injury to the head Headache attributed to mild traumatic injury
to the head

*Injury to the head fulfilling both of the following: A. Any headache fulfilling criteria C and D
A. Associated with none of the following: B. Mild traumatic injury to the head has occurred
1. Loss of consciousness for > 30 min C. Headache is reported to have developed within
2. Glasgow Coma Scale (GCS) score < 13 7 days after one of the following:
3. Post-traumatic amnesia lasting > 24 h 1. The injury to the head
4. Altered level of awareness for > 24 h 2. Regaining of consciousness following the injury
to the head
5. Imaging evidence of a traumatic head injury such as
skull fracture, intracranial hemorrhage and/or brain 3. Discontinuation of medication(s) impairing ability
contusion to sense or report headache following the injury
to the head
B. Associated with one or more of the following
symptoms and/or signs: D. For ACUTE headache
1. Transient confusion, disorientation, or impaired Either of the following:
consciousness 1. Headache has resolved within 3 months after its
2. Loss of memory for events immediately before onset
or after the head injury 1. Headache has not yet resolved but 3 months
3. Two or more of the following symptoms suggestive have not yet passed since its onset
of mild traumatic brain injury: E. For PERSISTENT headache
Nausea 1. Headache persists for > 3 months after its onset
Vomiting F. Not better accounted for by another ICHD-3
Visual disturbances diagnosis.
Dizziness and/or vertigo
Gait and/or postural imbalance
Impaired memory and/or concentration

*Traumatic injury to the head is defined as a structural or functional injury resulting from the action of external
forces upon the head. These include impact between the head and an object, penetration of the head by a foreign
body, forces generated from blasts or explosions, and other forces yet to be defined

similar to those of mTBI. Between 50 and 80% of individuals mixed phenotypes [20••]. Migraine phenotype is most com-
with whiplash injury report headache as a symptom [17, 18]. mon and is seen in 23–49% of patients with PTH [7, 9, 21,
Like headaches attributed to mTBI, headaches that occur after 22]. Patients with PTH of a migraine phenotype (PTH-MP)
whiplash injury typically resolve within several weeks to a have a greater risk of prolonged symptom recovery than indi-
month, but persistent headaches have been shown in over viduals with other PTH phenotypes or mTBI without head-
30% of patients at 1 and 3 months and 15% at 1 year [17]. ache [3, 9]. Moreover, patients with PTH-MP respond poorly
Post-traumatic headache (PTH) is classified by the ICHD-3 to current migraine treatments [23, 24] suggesting that there
[19] as headache attributed to trauma or injury to the head may be a different pathophysiology underlying these entities.
and/or neck. This headache must be a new headache, or a
significant worsening of a pre-existing type of headache, that
is reported to have begun within 7 days of the injury. This Clinical Imaging of PTH
interval between the injury and onset of headache can be ex-
tended if there is a period during which the individual is un- Imaging of the head and brain after mTBI is not always indi-
able to sense or report a PTH (e.g., due to loss of conscious- cated, but when performed, assesses for cranial lesions that
ness or use of certain medications). The ICHD-3 identifies can be associated with “complicated mTBI,” moderate, and
headache by the type of injury (craniotomy, whiplash, or trau- severe TBI. When indicated, computed tomography (CT) of
ma to the head), duration (acute, less than 3 months or persis- the head is the imaging modality most often used in the acute
tent, greater than 3 months), and if by head injury, then by setting. Rules and clinical policies for when head CT is needed
severity of the injury (mild versus moderate or severe) (see to evaluate the patient with suspected mTBI are available
Table 1). For the purpose of this review, we will consider only [25–27]. These guidelines are designed to identify those pa-
those PTHs secondary to mTBI or whiplash. tients with gross cranial lesions such as intracranial hemor-
PTH can take on a variety of headache phenotypes includ- rhage, bone fractures, and cerebral edema, conditions that
ing but not limited to migraine, tension-type, cluster, and can require acute interventions. Of note, PTH is included as
Curr Pain Headache Rep (2018) 22:64 Page 3 of 9 64

an indication for head CT within several of these guidelines. examined the levels of N-acetylaspartate (NAA), a marker of
For example, the presence of headache in the patient with neuronal vitality, as well as choline (Cho) levels which repre-
mTBI who also has loss of consciousness or amnesia suggests sent neuronal cell membrane turnover using 1H-MRS. Twelve
the need for acute imaging within the New Orleans Criteria healthy controls were compared with 17 individuals with
[28], and headache is a symptom that contributes to the deci- PTH, 9 with acute PTH (one attributed to whiplash), and 8
sion for CT imaging in the clinical policy from the American with persistent PTH (PPTH). The headaches were of a variety
College of Emergency Physicians and Centers for Disease of different phenotypes including migraine, tension-type,
Control and Prevention [29]. The guidelines for CT imaging cervicogenic, hemicrania continua, and mixed. Nine regions
for mTBI may be re-evaluated in the near future because of the were analyzed (left and right anterolateral and posterolateral
recent FDA approval of a blood test that predicts with over frontal, lobes, anteromedial and posteromedial frontal lobe,
99% accuracy whether or not there is intracranial bleeding left and right lateral parietal lobes, and medial parietal lobe).
after a TBI [30–32]. This will likely have the desirable impact The investigators found a significant reduction in NAA in
of reducing the number of individuals who are exposed to PTH subjects compared with controls in the right and left
radiation and reduce medical costs. anterolateral frontal lobe white matter, anterior and posterior
Unfortunately, there are no established guidelines that help medial frontal lobes, and medial parietal lobes. Additionally,
to identify when to image patients with PTH in the subacute or increased Cho was found in the right posterolateral frontal
chronic settings. Thus, we must rely on clinical acumen to lobe white matter and the anterior medial frontal and medial
make these decisions, including but not limited to assessment parietal regions. These results were consistent across different
of specific headache characteristics, identification of symp- PTH phenotypes.
toms and signs that suggest an increased likelihood of the The authors concluded that NAA decreases were represen-
headache being due to an injury secondary to the TBI, and tative of axon morphologic changes occurring in the white
worsening or improving headache patterns over time. Several matter and that increased Cho levels indicated cell membrane
types of secondary headaches unmasked by or resulting from turnover and hence healing. Because the control group was
mTBI and/or whiplash injury have been reported, including healthy controls who had not experienced mTBI, the authors
cervical artery dissection [33], cerebral venous thrombosis could not relate the changes directly to headache and surmised
[34, 35], cerebrospinal fluid (CSF) leak [36, 37], and intracra- that the changes were likely from TBI as they reasoned that
nial hypertension [38]. In our clinical experience, the majority headache itself was unlikely to affect neuronal vitality.
of patients who have persistent PTH (with or without other
persistent post-TBI symptoms) undergo brain magnetic reso-
nance imaging (MRI). However, the specific imaging para- Diffusion Tensor Imaging—Interrogating
digm used is dependent on the suspected cause of the head- White Matter Integrity
ache (Table 2).
Diffusion tensor imaging (DTI) is a method of mapping white
matter tracts in the brain by observing the restriction of water
Research Imaging of PTH movement. Anisotropy is the property of being directionally
dependent; fractional anisotropy (FA) is a value between 0 and
In the research arena, several imaging techniques have been 1 which indicates restriction of movement, with 0 representing
used to further our understanding of mTBI by examining free diffusion in any direction and 1 indicating that diffusion
changes in structure, function, connectivity, and brain metab- occurs along one axis only. FA may reflect axonal diameter,
olism [39••, 40, 41, 42••]. The goals of research imaging of myelination, and/or fiber density. Decreased FA values are
PTH are many, including elucidation of pathophysiology, dif- considered to be indicative of injury [44–47].
ferentiation of PTH from primary headache phenotypes that In a retrospective analysis, Alhilali et al. [48•] compared FA
they resemble, such as migraine, and to develop prognostic in 58 subjects with PTH-MP to 17 controls with mTBI (con-
markers. Below we discuss the current research imaging liter- trol subjects could have PTH without migraine phenotype).
ature in PTH following mTBI. They determined that subjects with PTH-MP had a reduction
in FA in the corpus callosum and fornix/septohippocampal
circuit; areas which the authors reason are involved in mi-
Proton Spectroscopy Magnetic Resonance graine pathophysiology. Additionally, neurocognitive-testing
Imaging—Interrogating Brain Metabolism results showed that decreases in the FA in the fornix/
septohippocampal circuit correlated with decreased visual
Proton spectroscopy magnetic resonance imaging (1H-MRS) memory, while a reduced FA in the corpus callosum was re-
is a method of measuring metabolites of functional processes lated to a trend toward prolonged recovery from post-
in different regions of the brain. Sarmento et al. [43•] concussion symptoms.
64 Page 4 of 9 Curr Pain Headache Rep (2018) 22:64

Table 2 Etiologies for headaches

due to causes secondary to TBI Headache attributed Recommended Headache features and associated signs/symptoms
to: imaging

Carotid artery Angiography— Can be thunderclap. Often unilateral and associated with face and
dissection [33] head and neck neck pain. May have ptosis, miosis, and/or anhidrosis. Carotid
MRI—brain** bruit can frequently be detected.*
Vertebral artery Angiography— Can be thunderclap. Often unilateral posterior head and neck pain.
dissection [33] head and neck May have dizziness, vertigo, ataxia, ptosis, miosis, and/or
MRI—brain** anhidrosis. *
CSF leak [36, 37] MR with Typically orthostatic—worsens with upright position and
contrast— improves when supine. May worsen throughout the day and
brain with Valsalva.
MR—spine May have neck pain/stiffness, changes in hearing, tinnitus,
May consider disequilibrium, photophobia, and nausea/vomiting.
myelography.
Cerebral venous MRI—brain Variable presentations. Often a continuous headache, may be
thrombosis [34, MRV—head Valsalva exacerbated and associated with visual symptoms*
35]
Intracranial MRI—brain May be worst in the morning or with lying down. Visual
hypertension [38] MRV—head symptoms include transient visual obscurations, tunnel vision,
and diplopia. Papilledema likely. May have abducens nerve
palsy and/or pulsatile tinnitus.

*May develop focal or syndromic neurologic symptoms due to stroke

**Several organizations have developed guidelines for the use of imaging for screening of the cerebrovasculature
after traumatic injury [33]
CSF, cerebrospinal fluid; MRI, magnetic resonance imaging; MRV, magnetic resonance venography

In a previous study by the same group, Delic et al. [49•] collected 9–12 weeks after injury and the remaining 12% were
described a mathematical technique of interpreting DTI data collected more than 12 weeks after injury. There was no sig-
which does not rely on the observer defining or interpreting nificant difference in SE between healthy controls and sub-
anatomical regions of interest. The technique uses whole- jects with migraine without mTBI. Subjects with mTBI had
brain histograms of FA and calculates a single value, significantly lower SE than healthy controls and separately,
Shannon entropy (SE), which is a measure of data complexity than subjects with migraine without TBI. Subjects with
that was developed in the field of information theory. PTH-MP had significantly lower SE than subjects with
Water diffuses in a directed manner for areas of the brain mTBI without PTH-MP. Optimality criterion values with sen-
that are highly structured, such as white matter. When an area sitivities and specificities were calculated for these compari-
is injured, the diffusion becomes less directed and the FA sons. It was determined that an SE value of less than 0.751
value decreases. This has been reflected in studies that show could discriminate between subjects with mTBI and controls
mean FA decreases with mTBI [44–46]. When a TBI occurs, with 77% sensitivity (95% CI, 65, 86) and 95% specificity
FA values decrease for some voxels and the histogram of FA (95% CI, 74, 100) with a likelihood ratio of 16.1 (95% CI,
values changes accordingly. The higher FA bins of the histo- 2.4, 109.7). Additionally, SE of less than 0.750 predicted
gram see a decrease in voxel count and the lower FA bins see PTH-MP compared to mTBI without PTH-MP with 81%
an increase in voxel count. When the histogram changes, the (95% CI, 67, 90) sensitivity and 75% specificity (95% CI,
SE also changes. The SE measures the complexity of the 43, 93) and a likelihood ratio of 3.2 (95% CI, 1.2, 8.7).
histogram. More complex histograms have higher SE values. Similar comparisons of mean FA had much less robust pre-
To illustrate, if all the FA measurements for a patient were of dictive capacities. Results also demonstrated an inverse corre-
the same value, the SE of that histogram would be smaller lation of SE with time to recovery from post-concussion
than the SE of a histogram with FA measurements evenly symptoms.
divided between two or more values. This study presents a novel approach to DTI analysis that
In the Delic et al. study, the investigators use retrospective- provides a promise of a possible biomarker for mTBI com-
ly collected data to calculate the SE and mean FA of the pared to controls and for PTH-MP compared with mTBI with-
whole-brain histograms of FA in patients with mTBI without out PTH-MP. Unfortunately, the mTBI group without PTH-
PTH-MP (n = 17), PTH-MP (m = 57), migraine but no TBI MP may have had headache as a symptom and the frequency
(n = 20), and healthy controls (n = 22). In 83% of the subjects, of this was not reported. Therefore, it is difficult to determine
scans were collected within 2 months of injury, 5% were if SE is reflecting headache after mTBI or specifically PTH-
Curr Pain Headache Rep (2018) 22:64 Page 5 of 9 64

MP, although either result is thought-provoking. Additionally, frontal, and precentral cortex as well as in the right
comparisons were not reported between PTH-MP and control supramarginal, superior and inferior parietal, and precuneus
patients with migraine. This would have been a very appealing regions. Cortical thickness was not greater in PPTH patients
comparison, to see if perhaps a difference in the origination of compared to controls in any brain region. Further analysis
these entities is reflected in the SE. probed the relationship between cortical thickness in PPTH
and headache burden as measured by headache frequency and
duration of PPTH. Examining the cortical thickness in the
MRI—Advanced Measures of Brain Structure abovementioned regions revealed an inverse correlation be-
tween headache frequency with left and right superior frontal
The density, volume, and thickness of different regions of the thickness. No association was found between years lived with
brain can be determined from routine T1-weighted imaging PPTH and cortical thickness. The finding of decreased cortical
using advanced post-processing techniques. Generally, mea- thickness in PPTH is consistent with studies of mTBI which
surements lower than normal are thought to reflect loss of have also showed multifocal cortical thinning primarily in the
integrity, and values greater than normal are thought to reflect frontal, temporal, and parietal cortices. Additionally, post-
enhanced functionality or reorganization perhaps due to up- concussion symptoms and their severity have been correlated
regulation or increased utilization of the area. Multifocal vol- with cortical thinning in other studies [50, 51, 54]. These
ume loss and cortical thinning have been shown in patient studies were not specific to headache; however, since head-
groups after mTBI [50–54]. Several studies have investigated ache is the most common post-concussive symptom, it is like-
brain structure in patients with PTH. ly that individuals in these studies had PTH. Despite the fact
Obermann et al. [55•] acquired brain MRIs from 32 sub- that patients were specifically studied with PTH, in Chong et
jects with PTH attributed to whiplash within 14 days of their al. [56•], the control group was healthy controls, and therefore,
injury and again at 3 months post-injury. Twelve of the 32 it is not known if the structural differences found in those with
patients continued to have headache at 3 months and these PTH are due to the headache, to other accompanying symp-
patients were re-evaluated again at 1 year post-trauma. toms, or to the inciting mTBI. Future analyses are planned in
Using the initial scan, voxel-based regional brain densities which those with PTH attributed to mTBI will be compared to
were compared between patients with acute PTH (n = 20), those with mTBI and persistent post-TBI symptoms not in-
those who developed persistent post-traumatic headache cluding headache.
(PPTH) (n = 12) and sex-matched and age-matched healthy Another approach to exploring the pathophysiologic under-
controls without headache or brain or neck trauma (n = 30). pinnings of PTH is to compare patients with a specific PTH
At initial assessment, within 14 days of injury, there were phenotype to patients with the primary headache condition
no significant differences between the groups in regional gray that the PTH most resembles. This could help determine if
matter densities. However, compared to controls and subjects mTBI triggers the same physiologic process underlying pri-
with acute PTH, subjects who developed PPTH showed de- mary headaches or if there is a separate mechanism that drives
creases in gray matter density in the anterior cingulate and PTH.
dorsolateral prefrontal cortex at 3 months. At 1 year, 11 of Schwedt et al. [57•] calculated regional brain volumes, sur-
the 12 PPTH patients had resolution of their headaches and face areas, cortical thicknesses, and curvatures in individuals
the differences in the brain morphometry were no longer pres- who had PPTH (n = 28), migraine without a history of TBI
ent. Additionally, comparing the gray matter at 3 months and (n = 28), and healthy controls (n = 28). Of the patients with
1 year of patients with PPTH, there was an increased density PPTH, 89% had a phenotype consistent with migraine or
in a portion of the midbrain, bilateral thalami, and bilateral probable migraine. The PPTH and migraine groups had sim-
cerebellum. The authors interpret these areas as anti- ilar headache frequencies. Structural differences were found
nociceptive regions that are upregulated in response to pain. between the PPTH and migraine cohorts in seven measure-
The authors surmise that the changes seen in the PTH group ments: right lateral orbitofrontal surface area, curvature and
represent adaptive changes in the gray matter of pain process- thickness, left caudal middle frontal thickness, left superior
ing structures. This study helps to distinguish brain regions frontal thickness, left precuneus thickness, and right
that may be important for the development of headache (or supramarginal thickness. For each of these differences, those
persistent headache) in the setting of mTBI separate from the with PPTH showed decreased structural integrity (smaller sur-
brain changes occurring in mTBI, and identifies areas that face area, volume or thickness, or higher curvature). There
may be contributing to headache resolution. were not correlations between number of years with PTH
Also investigating changes in PPTH, Chong et al. 2018 and number of TBIs with the structural measures in these
[56•] compared cortical thickness in 33 patients with PPTH regions. When these same regions in PPTH subjects were
to 33 healthy controls. The PPTH group had less cortical compared to healthy controls, only three of the structural var-
thickness bilaterally in the superior frontal, caudal middle iations were evident. (PPTH had smaller right lateral
64 Page 6 of 9 Curr Pain Headache Rep (2018) 22:64

orbitofrontal surface area, smaller right supramarginal gyrus In the research arena, the studies using imaging to investi-
thickness, and smaller left superior frontal thickness.) There gate PTH signify progress toward our understanding of PTH.
were no differences between migraine and healthy controls in With such a small quantity of studies, each asking different
any of the seven regions found to be different between mi- questions, it is difficult to draw any meaningful conclusions
graine and PPTH patients. from this body of research as a collective. However, the results
Each of the three brain regions found to differ between indicate that multiple different imaging modalities have been
groups has been implicated in pain processing, lending cre- effective in demonstrating brain changes in those with PTH.
dence to the findings. The fact that significant differences The results are largely consistent with imaging studies investi-
were demonstrated between these groups and that only three gating mTBI, showing multifocal neuronal injury in both gray
of these same regions showed differences between PPTH and and white matter with predominance in the frontal, temporal,
healthy controls is persuasive in the argument that PPTH is and parietal regions [39••, 42••]. Given that the majority of
driven by a different mechanism than primary migraine. patients with mTBI experience new or worsening headache in
the post-traumatic period, it is difficult to recruit a large control
sample of mTBI patients without headache, and therefore dif-
Imaging the Neck in PPTH ficult to extricate findings that are due to the injury with those
that may be specific to PTH. Three of the above studies com-
Jensen et al. [58•] measured segmental and overall extension- pared PTH with a healthy control group [43•, 56•, 58•] and so,
flexion motion of the cervical spine using flexion/extension x- the differences detected may reflect brain injury rather than
rays in 19 patients with PPTH and 19 age-matched and sex- mechanisms related to headache. Even when comparisons were
matched healthy controls. The investigators found a signifi- made between patients with mTBI with and without headache,
cant 10% reduction in overall C1–C7 extension-flexion mo- it is unlikely that PTH was the only post-concussive symptom
tion in patients with PTH compared to controls. Headache being experienced. None of the studies have a large enough
severity was negatively correlated with age-corrected C1–C7 sample size to perform sub-group comparisons for each post-
motion in patients with PTH. Finally, the frequency of asso- TBI symptom. So, while studies that compare PTH to mTBI
ciated symptoms including dizziness, aural symptoms, and may be examining pathology specific to headache, alternative-
visual disturbance was negatively correlated with age- ly, results could be reflective of total symptom burden.
corrected segmental motion at C5–6. In this study, PTH re- In studies comparing PTH-MP to mTBI patients without
sulted from a variety of injuries; 15/19 were from direct blow headaches of migraine phenotype [48•], results may indicate
to the head or falls and 4/19 were from motor vehicle colli- the underlying differences in pathophysiology relating to the
sions. It is unclear how many of the patients sustained direct migrainous nature of the headache. However, while there is
head injuries, whiplash injuries, or both. Neck pain was not suspicion, it is not yet clear that there is an underlying patho-
evaluated. This makes interpreting the results challenging. physiologic difference between PTH-MP and other
Reduced overall cervical motion in PPTH patients may repre- PTH. Alhilali et al. [48•] did demonstrate reduced integrity
sent reduction of motion due to head and/or neck pain, or it of the white matter of the corpus callosum and fornix/
may have a different physiologic explanation. However, the septohippocampal circuit in PTH-MP compared to the mTBI
finding that frequency of associated symptoms was correlated group. The changes in the fornix/septohippocampal circuit
with segmental C5–6 motion serves as a reminder that the were correlated not with headache but with reduction in visual
origins and pathophysiology of PTH may include the neck memory. As this is not a feature of migraine, but rather a post-
in addition to the brain. concussive symptom, it is uncertain whether the findings are
truly reflecting the underlying process of the PTH or generally
of post-concussive symptoms. Nonetheless, this study does
Discussion show a prognostic marker; reduced FA in the corpus callosum
is predictive of prolonged recovery in patients with PTH-MP.
Imaging is a useful tool in the diagnosis and understanding of Delic et al. [49•] provide a novel mathematical analysis of
mTBI and its sequelae such as PTH. In the acute setting, DTI data that with relatively high sensitivity and specificity, it
guidelines and clinical policies exist to help clinicians deter- was able to differentiate mTBI subjects from healthy controls
mine when imaging is indicated. However, there is little in the and patients with PTH-MP from patients with mTBI without
way of formal guidance to direct clinicians as to when and PTH-MP. This is based on SE analysis of whole-brain FA, and
what clinical imaging should be used in the subacute or chron- while this does not reveal insight into different mechanisms of
ic setting in patients with PTH. As there are several secondary the conditions, it delivers potential for the application of this
headaches that can arise from trauma, research and develop- method for distinguishing mTBI from PTH and PTH from
ment of robust guidelines for imaging in these circumstances migraine (although these comparisons were not made in this
would be of benefit to patients and medical practitioners. paper).
Curr Pain Headache Rep (2018) 22:64 Page 7 of 9 64

Schwedt et al. [57•] identified areas of gray matter that may References
be involved in the pathophysiology of PPTH and not primary
migraine, providing a stepping stone for further research de- Papers of particular interest, published recently, have been
veloping tools to distinguish the entities and understanding highlighted as:
their mechanisms. • Of importance
Obermann et al. [55•] revealed changes that were associat- •• Of major importance
ed with the persistence of PTH and also with its resolution.
The regions of gray matter identified as significantly different 1. Centers for Disease Control and Prevention (CDC) NCfIPaC.
in patients with PTH at 3 months compared to within 2 weeks Report to Congress on mild traumatic brain injury in the United
States: steps to prevent a serious public health problem. Atlanta:
have all been previously associated with pain processing,
Centers for Disease Control and Prevention; 2003.
while regions of gray matter that showed increased volume 2. Cassidy JD, Carroll LJ, Peloso PM, Borg J, von Holst H, Holm L, et
concordant with PTH resolution have been previously identi- al. Incidence, risk factors and prevention of mild traumatic brain
fied in anti-nociception. injury: results of the WHO Collaborating Centre Task Force on
Mild Traumatic Brain Injury. J Rehabil Med. 2004;36(43 Suppl):
28–60.
3. Kontos AP, Elbin RJ, Lau B, Simensky S, Freund B, French J, et al.
Conclusions Posttraumatic migraine as a predictor of recovery and cognitive im-
pairment after sport-related concussion. Am J Sports Med.
2013;41(7):1497–504. https://doi.org/10.1177/0363546513488751.
This body of work provides some groundwork for future stud- 4. Holtkamp MD, Grimes J, Ling G. Concussion in the military: an
ies investigating PTH. Brain regions of potential importance evidence-base review of mTBI in US military personnel focused on
have been identified and novel mathematical techniques have posttraumatic headache. Curr Pain Headache Rep. 2016;20(6):37.
5. Hoffman JM, Lucas S, Dikmen S, Braden CA, Brown AW, Brunner
been introduced. Imaging techniques that have been applied to R, et al. Natural history of headache after traumatic brain injury. J
headache [41, 59–62] or mTBI [39••, 42••, 63] with some Neurotrauma. 2011;28(9):1719–25. https://doi.org/10.1089/neu.
success and that have not yet yielded publications in PTH 2011.1914.
include fMRI, PET, and SPECT (with phosphorous—31P 6. Yamaguchi M. Incidence of headache and severity of head injury.
Headache. 1992;32(9):427–31.
MRS) and arterial spin labeling. While it is recognizably diffi- 7. Lucas S, Hoffman JM, Bell KR, Dikmen S. A prospective study of
cult to achieve, ideal studies will compare mTBI with persis- prevalence and characterization of headache following mild trau-
tent symptoms not including headache to PTH attributed to matic brain injury. Cephalalgia. 2014;34(2):93–102. https://doi.org/
mTBI, or PTH to primary headache types, controlling for other 10.1177/0333102413499645.
8. Marcus DA. Disability and chronic posttraumatic headache.
post-concussive symptoms. Additional longitudinal studies Headache. 2003;43(2):117–21.
will be beneficial in understanding mechanisms for persistence 9. Mihalik JP, Register-Mihalik J, Kerr ZY, Marshall SW, McCrea MC,
and resolution of PTH and long-term consequences of PTH. Guskiewicz KM. Recovery of posttraumatic migraine characteristics
Ideally, imaging would assist with prognosticating PTH out- in patients after mild traumatic brain injury. Am J Sports Med.
2013;41(7):1490–6. https://doi.org/10.1177/0363546513487982.
comes, assist with identification of individuals who are in 10. Benson BW, Meeuwisse WH, Rizos J, Kang J, Burke CJ. A pro-
greatest need of PTH treatment, and help to differentiate PTH spective study of concussions among National Hockey League
from worsening of a primary headache disorder that occurred players during regular season games: the NHL-NHLPA
following TBI. Although much work needs to be completed Concussion Program. CMAJ. 2011;183(8):905–11. https://doi.org/
10.1503/cmaj.092190.
before reaching these goals, current studies have already 11. Asplund CA, McKeag DB, Olsen CH. Sport-related concussion:
helped to show objective evidence for PTH, a condition that factors associated with prolonged return to play. Clin J Sport
for many could otherwise be considered an “invisible” injury. Med. 2004;14(6):339–43.
12. Heidari K, Asadollahi S, Jamshidian M, Abrishamchi SN,
Nouroozi M. Prediction of neuropsychological outcome after mild
Compliance with Ethical Standards traumatic brain injury using clinical parameters, serum S100B pro-
tein and findings on computed tomography. Brain Inj. 2015;29(1):
Conflicts of Interest Jill C. Rau, Gina M. Dumkrieger, and Catherine D. 33–40.
Chong declare no conflicts of interest. Todd Schwedt reports personal 13. Zemek R, Barrowman N, Freedman SB, Gravel J, Gagnon I,
fees from Alder, Allergan, Amgen, ATI, Avanir, Dr. Reddy’s, Eli Lilly, McGahern C, et al. Clinical risk score for persistent postconcussion
Ipsen Bioscience, Novartis, Nocira, Aural Analytics, and Teva. He also symptoms among children with acute concussion in the ED.
reports stock options from Aural Analytics, Nocira, and Second Opinion; JAMA. 2016;315(10):1014–25. https://doi.org/10.1001/jama.
all of the aforementioned are outside of the submitted work. 2016.1203.
14. Merritt VC, Rabinowitz AR, Arnett PA. Injury-related predictors of
Human and Animal Rights and Informed Consent All reported studies/ symptom severity following sports-related concussion. J Clin Exp
experiments with human or animal subjects performed by the authors Neuropsychol. 2015;37(3):265–75. https://doi.org/10.1080/
have been previously published and complied with all applicable ethical 13803395.2015.1004303.
standards (including the Helsinki declaration and its amendments, 15. Sheedy J, Harvey E, Faux S, Geffen G, Shores EA. Emergency
institutional/national research committee standards, and international/na- department assessment of mild traumatic brain injury and the pre-
tional/institutional guidelines). diction of postconcussive symptoms: a 3-month prospective study.
64 Page 8 of 9 Curr Pain Headache Rep (2018) 22:64

J Head Trauma Rehabil. 2009;24(5):333–43. https://doi.org/10. head computed tomography findings in patients with suspected
1097/HTR.0b013e3181aea51f. mild or moderate traumatic brain injury. J Neurotrauma.
16. Stacey A, Lucas S, Dikmen S, Temkin N, Bell KR, Brown A, et al. 2016;33(2):203–14. https://doi.org/10.1089/neu.2015.4149.
Natural history of headache five years after traumatic brain injury. J 32. United States Food and Drug Administration. FDA authorizes mar-
Neurotrauma. 2017;34(8):1558–64. https://doi.org/10.1089/neu. keting of first blood test to aid in the evaluation of concussion in
2016.4721. adults. In: FDA news release. 2018. https://www.fda.gov/
17. Sjaastad O, Fredriksen T, Bakketeig L. Headache subsequent to NewsEvents/Newsroom/PressAnnouncements/ucm596531.htm.
whiplash. Curr Pain Headache Rep. 2009;13(1):52–8. 33. Rutman AM, Vranic JE, Mossa-Basha M. Imaging and manage-
18. Balla J, Karnaghan J. Whiplash headache. Clin Exp Neurol. ment of blunt cerebrovascular injury. Radiographics. 2018;38(2):
1987;23:179–82. 542–63. https://doi.org/10.1148/rg.2018170140.
19. Headache Classification Committee of the International Headache 34. Ghuman MS, Salunke P, Sahoo SK, Kaur S. Cerebral venous sinus
Society (IHS) The International Classification of Headache thrombosis in closed head trauma: a call to look beyond fractures
Disorders, 3rd edition. Cephalalgia. 2018;38(1):1–211. https://doi. and hematomas! J Emerg Trauma Shock. 2016;9(1):37–8. https://
org/10.1177/0333102417738202. doi.org/10.4103/0974-2700.173865.
20.•• Minen MT, Boubour A, Walia H, Barr W. Post-concussive syn- 35. Hsu PJ, Lee CW, Tang SC, Jeng JS. Pearls & oy-sters: delayed
drome: a focus on post-traumatic headache and related cognitive, traumatic intracerebral hemorrhage caused by cerebral venous sinus
psychiatric, and sleep issues. Curr Neurol Neurosci Rep. thrombosis. Neurology. 2014;83(14):e135–7. https://doi.org/10.
2016;16(11):100. https://doi.org/10.1007/s11910-016-0697-7. 1212/WNL.0000000000000836.
This paper provides a recent review of post-traumatic head- 36. Huntoon MA, Watson JC. Intracranial hypotension following mo-
ache, presenting phenotypes and associated symtoms. tor vehicle accident: an overlooked cause of post-traumatic head
21. Couch J, Stewart K. Migraine with aura (MA) is the most common and neck pain? Pain Practice. 2007;7(1):47–52.
phenotype for post-traumatic headache (PTH) (P1.183). Neurology. 37. Sarrafzadeh AS, Hopf SA, Gautschi OP, Narata AP, Schaller K.
2017;88(16S):183. Intracranial hypotension after trauma. Springerplus. 2014;3:153.
22. Lew HL, Lin PH, Fuh JL, Wang SJ, Clark DJ, Walker WC. https://doi.org/10.1186/2193-1801-3-153.
Characteristics and treatment of headache after traumatic brain in-
38. Sabo T, Supnet C, Purkayastha S. Secondary intracranial hyperten-
jury: a focused review. Am J Phys Med Rehabil. 2006;85(7):619–
sion (pseudotumor cerebri) presenting as post-traumatic headache
27. https://doi.org/10.1097/01.phm.0000223235.09931.c0.
in mild traumatic brain injury: a case series. Childs Nerv Syst.
23. Erickson JC. Treatment outcomes of chronic post-traumatic head-
2018;34(4):681–90. https://doi.org/10.1007/s00381-017-3681-x.
aches after mild head trauma in US soldiers: an observational study.
39.•• Dean PJ, Sato JR, Vieira G, McNamara A, Sterr A. Multimodal
Headache. 2011;51(6):932–44. https://doi.org/10.1111/j.1526-
imaging of mild traumatic brain injury and persistent
4610.2011.01909.x.
postconcussion syndrome. Brain Behav. 2015;5(1):45–61. https://
24. DiTommaso C, Hoffman JM, Lucas S, Dikmen S, Temkin N, Bell
doi.org/10.1002/brb3.292. This paper provides an overview of
KR. Medication usage patterns for headache treatment after mild
techniques that have been used to image mTBI.
traumatic brain injury. Headache. 2014;54(3):511–9.
25. Stiell IG, Wells GA, Vandemheen K, Clement C, Lesiuk H, 40. Wu X, Kirov II, Gonen O, Ge Y, Grossman RI, Lui YW. MR
Laupacis A, et al. The Canadian CT Head Rule for patients with imaging applications in mild traumatic brain injury: an imaging
minor head injury. Lancet. 2001;357(9266):1391–6. update. Radiology. 2016;279(3):693–707. https://doi.org/10.1148/
26. Papa L, Stiell IG, Clement CM, Pawlowicz A, Wolfram A, Braga C, radiol.16142535.
et al. Performance of the Canadian CT Head Rule and the New 41. Younis S, Hougaard A, Vestergaard MB, Larsson HBW, Ashina M.
Orleans Criteria for predicting any traumatic intracranial injury on Migraine and magnetic resonance spectroscopy: a systematic re-
computed tomography in a United States level I trauma center. view. Curr Opin Neurol. 2017;30(3):246–62. https://doi.org/10.
Acad Emerg Med. 2012;19(1):2–10. https://doi.org/10.1111/j. 1097/WCO.0000000000000436.
1553-2712.2011.01247.x. 42.•• Chong CD, Schwedt TJ. Research imaging of brain structure
27. Unden J, Ingebrigtsen T, Romner B, Scandinavian Neurotrauma C. and function after concussion. Headache. 2018;58:827–35.
Scandinavian guidelines for initial management of minimal, mild https://doi.org/10.1111/head.13269. This paper provides
and moderate head injuries in adults: an evidence and consensus- information on the newest imaging techniques being used
based update. BMC Med. 2013;11:50. https://doi.org/10.1186/ to investigate mTBI.
1741-7015-11-50. 43.• Sarmento E, Moreira P, Brito C, Souza J, Jevoux C, Bigal M.
28. Haydel MJ, Preston CA, Mills TJ, Luber S, Blaudeau E, DeBlieux Proton spectroscopy in patients with post-traumatic headache at-
PM. Indications for computed tomography in patients with minor tributed to mild head injury. Headache. 2009;49(9):1345–52.
head injury. N Engl J Med. 2000;343(2):100–5. https://doi.org/10. https://doi.org/10.1111/j.1526-4610.2009.01494.x. The authors
1056/NEJM200007133430204. investigate regional metaboloic changes in the brain of PTH
29. Jagoda AS, Bazarian JJ, Bruns JJ Jr, Cantrill SV, Gean AD, Howard subjects compared to healthy controls using magnetic
PK, et al. Clinical policy: neuroimaging and decisionmaking in resonance proton spectroscopy.
adult mild traumatic brain injury in the acute setting. Ann Emerg 44. Lipton ML, Gellella E, Lo C, Gold T, Ardekani BA, Shifteh K, et al.
Med. 2008;52(6):714–48. https://doi.org/10.1016/j.annemergmed. Multifocal white matter ultrastructural abnormalities in mild trau-
2008.08.021. matic brain injury with cognitive disability: a voxel-wise analysis of
30. Papa L, Brophy GM, Welch RD, Lewis LM, Braga CF, Tan CN, et diffusion tensor imaging. J Neurotrauma. 2008;25(11):1335–42.
al. Time course and diagnostic accuracy of glial and neuronal blood https://doi.org/10.1089/neu.2008.0547.
biomarkers GFAP and UCH-L1 in a large cohort of trauma patients 45. D'Souza MM, Trivedi R, Singh K, Grover H, Choudhury A, Kaur P,
with and without mild traumatic brain injury. JAMA Neurol. et al. Traumatic brain injury and the post-concussion syndrome: a
2016;73(5):551–60. https://doi.org/10.1001/jamaneurol.2016. diffusion tensor tractography study. Indian J Radiol Imaging.
0039. 2015;25(4):404–14. https://doi.org/10.4103/0971-3026.169445.
31. Welch RD, Ayaz SI, Lewis LM, Unden J, Chen JY, Mika VH, et al. 46. Benson RR, Meda SA, Vasudevan S, Kou Z, Govindarajan KA,
Ability of serum glial fibrillary acidic protein, ubiquitin C-terminal Hanks RA, et al. Global white matter analysis of diffusion tensor
hydrolase-L1, and S100B to differentiate normal and abnormal images is predictive of injury severity in traumatic brain injury. J
Curr Pain Headache Rep (2018) 22:64 Page 9 of 9 64

Neurotrauma. 2007;24(3):446–59. https://doi.org/10.1089/neu. 55.• Obermann M, Nebel K, Schumann C, Holle D, Gizewski ER,
2006.0153. Maschke M, et al. Gray matter changes related to chronic posttrau-
47. Chong CD, Schwedt TJ. White matter damage and brain network matic headache. Neurology. 2009;73(12):978–83. https://doi.org/
alterations in concussed patients: a review of recent diffusion tensor 10.1212/WNL.0b013e3181b8791a. The authors use MRI
imaging and resting-state functional connectivity data. Curr Pain morphometry to examine grey matter changes correlate with
Headache Rep. 2015;19(5):485. https://doi.org/10.1007/s11916- the persistence of PTH at 3 months and the changes that occur
015-0485-0. with headache resolution within 1 year.
48.• Alhilali LM, Delic J, Fakhran S. Differences in callosal and 56.• Chong CD, Berisha V, Chiang CC, Ross K, Schwedt TJ. Less
forniceal diffusion between patients with and without cortical thickness in patients with persistent post-traumatic head-
postconcussive migraine. AJNR Am J Neuroradiol. 2017;38(4): ache compared with healthy controls: an MRI study. Headache.
691–5. https://doi.org/10.3174/ajnr.A5073. The investigators 2018;58(1):53–61. https://doi.org/10.1111/head.13223. The study
examine regional changes that occur in PTH-MP with DTI. employs MRI morphometry to explore difference in cortical
49.• Delic J, Alhilali LM, Hughes MA, Gumus S, Fakhran S. White thickness between patients with PPTH and healthy controls.
matter injuries in mild traumatic brain injury and posttraumatic 57.• Schwedt TJ, Chong CD, Peplinski J, Ross K, Berisha V. Persistent
migraines: diffusion entropy analysis. Radiology. 2016;279(3): post-traumatic headache vs. migraine: an MRI study demonstrating
859–66. https://doi.org/10.1148/radiol.2015151388. This paper differences in brain structure. J Headache Pain. 2017;18(1):87. In
describes a mathematical method when applied to DTI data this paper, MRI morphometry is used to compare differences
provided a novel biomarker that with reasonable sensitivity between subjects with PPTH and those with migraine.
and specificity was able discriminate between patient with 58.• Jensen OK, Justesen T, Nielsen FF, Brixen K. Functional radio-
mTBI and healthy controls and also patients with PTH-MP graphic examination of the cervical spine in patients with post-
from patients with mTBI without PTH-MP. traumatic headache. Cephalalgia. 2007;10(6):295–303. Overall
50. Dean PJ, Sato JR, Vieira G, McNamara A, Sterr A. Long-term and segmental motion in the cervical spine using flexion/
structural changes after mTBI and their relation to post- extension x-rays in patients with PTH are compared to healthy
concussion symptoms. Brain Inj. 2015;29:1–8. https://doi.org/10. controls and correlated with associated symptoms.
3109/02699052.2015.1035334. 59. Schwedt TJ, Chong CD. Medication overuse headache: pathophys-
iological insights from structural and functional brain MRI re-
51. Wang X, Xie H, Cotton AS, Tamburrino MB, Brickman KR, Lewis
search. Headache. 2017;57(7):1173–8. https://doi.org/10.1111/
TJ, et al. Early cortical thickness change after mild traumatic brain
head.13037.
injury following motor vehicle collision. J Neurotrauma.
60. Russo A, Silvestro M, Tedeschi G, Tessitore A. Physiopathology of
2015;32(7):455–63. https://doi.org/10.1089/neu.2014.3492.
migraine: what have we learned from functional imaging? Curr
52. Tremblay S, De Beaumont L, Henry LC, Boulanger Y, Evans AC, Neurol Neurosci Rep. 2017;17(12):95. https://doi.org/10.1007/
Bourgouin P, et al. Sports concussions and aging: a neuroimaging s11910-017-0803-5.
investigation. Cereb Cortex. 2013;23(5):1159–66. https://doi.org/ 61. Chong CD, Schwedt TJ, Dodick DW. Migraine: what imaging re-
10.1093/cercor/bhs102. veals. Curr Neurol Neurosci Rep. 2016;16(7):64. https://doi.org/10.
53. Tate DF, York GE, Reid MW, Cooper DB, Jones L, Robin DA, et al. 1007/s11910-016-0662-5.
Preliminary findings of cortical thickness abnormalities in blast 62. Aurora SK, Brin MF. Chronic migraine: an update on physiology,
injured service members and their relationship to clinical findings. imaging, and the mechanism of action of two available pharmaco-
Brain Imaging Behav. 2014;8(1):102–9. https://doi.org/10.1007/ logic therapies. Headache. 2017;57(1):109–25. https://doi.org/10.
s11682-013-9257-9. 1111/head.12999.
54. Albaugh MD, Orr C, Nickerson JP, Zweber C, Slauterbeck JR, 63. Ellis MJ, Ryner LN, Sobczyk O, Fierstra J, Mikulis DJ, Fisher JA,
Hipko S, et al. Postconcussion symptoms are associated with cere- et al. Neuroimaging assessment of cerebrovascular reactivity in
bral cortical thickness in healthy collegiate and preparatory school concussion: current concepts, methodological considerations, and
ice hockey players. J Pediatr. 2015;166(2):394–400 e1. https://doi. review of the literature. Front Neurol. 2016;7:61. https://doi.org/10.
org/10.1016/j.jpeds.2014.10.016. 3389/fneur.2016.00061.