Abstract
Objectives
Aim of the review is to summarize the knowledge about the sensory function and pain modulatory systems in posttraumatic headache and discuss its possible role in patients with posttraumatic headache.
Background
Posttraumatic headache is the most common complication after traumatic brain injury, and significantly impacts patients’ quality of life. Even though it has a high prevalence, its origin and pathophysiology are poorly understood. Thereby, the existing treatment options are insufficient. Identifying its mechanisms can be an important step forward to develop target-based personalized treatment.
Methods
We searched the PubMed database for studies examining pain modulation and/or quantitative sensory testing in individuals with headache after brain injury.
Results
The studies showed heterogenous alterations in sensory profiles (especially in heat and pressure pain perception) compared to healthy controls and headache-free traumatic brain injury-patients. Furthermore, pain inhibition capacity was found to be diminished in subjects with posttraumatic headache.
Conclusions
Due to the small number of heterogenous studies a distinct sensory pattern for patients with posttraumatic headache could not be identified. Further research is needed to clarify the underlying mechanisms and biomarkers for prediction of development and persistence of posttraumatic headache.
Keywords
Introduction
Posttraumatic headache (PTH) is a common complication of traumatic brain injury (TBI). It is defined by the International Classification of Headache Disorders (ICHD-3, 2018) as a secondary headache, which develops within seven days after trauma, regaining consciousness or recovering the ability to sense and report pain. If the headache lasts for more than 3 months, it is defined as persistent posttraumatic headache (1).
The prevalence rates of TBI-patients developing headache within 1 year after trauma vary from 10 to 95% (2–4). Taking only the largest prospective study into consideration it is estimated that 54 to 69% of patients with TBI suffer from a headache within 1 year after brain injury (2). The prevalence rates vary depending on the severity of the trauma and the pain type, but PTH is still an underreported entity (2–4). Notably, PTH develops most frequently after mild TBI (5,6). However, the defining criteria for mild TBI vary between the studies widely, which could be one of the reasons for the heterogeneous prevalence rates (7,8). Further, female sex has claimed to be a risk factor for developing PTH (4).
The most prevalent phenotypes of PTH are migraine-like headache and tension-type like headache (1). Besides the headache characteristics like localization, quality and intensity, as well as accompanying symptoms like photo- and phonophobia, in case of migraine-like headache, PTH often shares other additional features with migraine or tension-type headache (9), e.g. common reported comorbidities are anxiety, depression, cognitive dysfunction. Thereby, posttraumatic headache has a further significant impact on quality of life (10,11). The similarities to these primary headaches lead to the presumption that they may share similar mechanisms. However, this seems to be only partly the case and the origin and pathomechanisms of PTH need further investigations (12).
Even though persistent PTH accounts for 4% of all secondary headache disorders (13), the underlying mechanisms and etiology are poorly understood, and are probably of multifactorial origin. It is assumed that genetics, impaired descending modulation with remodelling, neurometabolic changes, trigeminal system activation, cortical spreading depression and neuroinflammation are, amongst others, possible mechanisms for developing chronic pain (14). Little is known about risk factors for developing PTH after a brain injury. The identified influencing factors diverged in different studies. History of headache before trauma, female sex, age, the type of accident and CT abnormalities in mild TBI have been identified as possible risk factors (15,16).
Despite the high prevalence and the socioeconomic burden of PTH, there is a lack of specific evidence-based treatments so far. Current treatment options are mostly based on acute or preventive medications for primary headache disorders, but are not very effective (17,18).
In animal studies after a mild trauma PTH was associated with tactile allodynia in both cephalic and extracephalic body regions (19). The injury pattern has been reported to affect the risk of developing persistent PTH. Moreover, mice reacted hypersensitive to known migraine triggers after TBI, even after recovering from tactile hypersensitivity (19). Male and female rodents did not differ in sensitivity, but different specific transmitter and hormonal cycle changes were reported in females (20). Mild TBI led to a loss of diffuse noxious inhibitory control responses at an early time point after a TBI (21). Further, animal models indicated of altered sensory processing after TBI as a risk factor for PTH, and attendant symptoms like auditory hypersensitivity and increased fear learning (22). However, animal models of TBI are not able to capture the whole complexity of changes in patients with PTH.
Understanding of the pathophysiology of chronic pain after TBI is an important step for identifying treatment targets. The aim of this review is to summarize the knowledge from the existing literature about the kind of alterations in the somatosensory function and pain modulation associated with PTH. Further, based on that, we discuss their possible contribution to the persistent headache development.
Methods
For this systematic review we searched PubMed database for primary literature published in English. We looked for any studies that explored the somatosensory function in patients suffering from headache after a traumatic brain injury based on quantitative sensory testing or conditioned pain modulation disregarding the used protocols. The keywords (“sensory” AND/OR “sensory system” AND/OR “sensory function” AND/OR “sensory testing” AND/OR “quantitative sensory testing” AND/OR “somatosensory” AND/OR “pain modulation” AND/OR “conditioned pain modulation”) AND (“posttraumatic headache” AND/OR “post concussion headache”) were used. The search was conducted between 26th January and 4th February 2021. Only studies on humans were included. There was no restriction on the publication date. The search resulted in 1,227 hits. After removing duplicates and examining the titles and abstracts, 6 articles could be included in this review.
Results
The characteristics of the included studies are summarized in Table 1.
Somatosensory function studies.
CDT: Cold detection threshold, CPM: Conditioned pain modulation, CPT: Cold pain threshold, DMA: Dynamic mechanical allodynia, HPT: Heat pain threshold, MDT: mechanical detection threshold, MPS: Mechanical pain summation, PPT: Pressure pain threshold, PTH: posttraumatic headache, QST: Quantitative sensory testing, SMA: Static mechanical allodynia, TBI: traumatic brain injury, VDT: Vibration detection threshold, WDT: Warmth detection threshold, WUR: Wind-up ratio.
Sensory testing
Quantitative sensory testing (QST) provides a comprehensive analysis of the function of the somatosensory system (23). Its aim is to detect the individual sensory phenotype which can contribute to the understanding of the underlying pain mechanisms as well as mechanism-based diagnosis and treatment (24). The assessed parameters include thermal detection (cold, warmth) and pain thresholds (cold, heat), mechanical detection (touch, vibration) and pain thresholds (pinprick, pressure) as well as assessment of sensitivity to suprathreshold stimuli (23). It has been shown that the somatosensory profile can predict the tendency to develop chronic pain, e.g. in patients with neuropathic pain (25).
All studies on QST included in this review found alterations in sensory processing in patients with PTH compared to the control groups. However, not all studies included a control group of both healthy subjects and pain-free subjects after TBI.
Levy et al. (2020) (9) reported differences in the sensory profiles depending on the clinical phenotype of patients with PTH. Patients with tension-type-like PTH had a cephalic and extracephalic hyposensitivity for thermal stimuli (both nociceptive and non-nociceptive warmth/heat stimuli) that was accompanied by cephalic pressure hyperalgesia. Migraine-like PTH patients, on the other hand, had warmth hyposensitivity only in extracephalic body regions. Both headache phenotypes had cold hyposensitivity in common, compared to headache-free subjects and healthy controls. The distinct sensory changes in cephalic areas in both phenotypes of persistent PTH were partly reflected by the extracephalic sensory abnormalities. In headache-free TBI subjects the cold sensation thresholds were similar to those observed in healthy controls, but the warmth sensitivity in extracephalic regions was reduced. Headache intensity did not correlate with the extent of sensory abnormalities, with one exception: the stronger the headache severity was, the higher the cephalic warmth sensation thresholds (thermal hyposensitivity) were in subjects with tension-type-like symptoms (9).
Bouferguene et al. (2020) (26) focused on differences between young and elderly (>60 years) PTH patients. They found more pronounced sensory loss in heat and pressure pain (thermal and mechanical hypoalgesia) in the elderly group, in contrast to increased warmth sensitivity (warmth hyperesthesia) in the young group. Elderly PTH patients had an increased sensitivity to warm temperature only in the painful body side. In this study chronic pain was associated with a reduced heat pain and vibration sensitivity in both young and elderly TBI survivors. There were differences between the affected and the contralateral side, especially in mechanical detection. Interestingly, alterations were not always associated with chronic pain, as cold hypoesthesia was found in elderly patients without pain (26).
Naugle et al. (2020) (27) compared a group with persistent PTH after TBI with a group that did not suffer from PTH after TBI. The pressure pain sensitivity and temporal pain summation 1 week after the injury were not associated with the development of persistent headache; thus the authors stated that these sensory alterations cannot be used as a predictor for pain chronification (27).
The study from Carey et al. (2019) (28) reported significantly decreased pressure pain thresholds (mechanical hyperalgesia) of the head in PTH participants. The measures were significantly associated with headache pain intensity. Pressure pain thresholds of the forearm, as a non-painful site, did not differ between the study groups (28).
Defrin et al. (2015) (29) reported generalized heat hypoalgesia (in the forehead and forearm) and cranial mechanical hyperalgesia in patients with persistent PTH. Besides this, the authors found that the larger the headache intensity, the lower was the pressure pain threshold (29).
In the study by Defrin et al. (2010) (30) subjects with persistent PTH had significantly higher thermal thresholds in both the head and hand and significantly lower pressure-pain threshold in the head than the pain-free TBI subjects and healthy controls (30).
In conclusion, the most common sensory alteration in patients with headache after TBI was a higher detection threshold for heat pain (heat hypoalgesia) (9,26,29,30) and lower pressure pain threshold (mechanical hyperalgesia) (9,28–30). Two studies detected an increased cold perception threshold compared to pain-free subjects (9,30), whereas two others found a decreased warmth perception threshold (warmth hyperesthesia) (9,26). Notably, the sensory profiles of patients without headache after TBI were similar to healthy controls. Whether the somatosensory changes in PTH patients are restricted to the painful (cephalic) sites or occur generalized, was inconsistent between the studies. Overall, there was no definite accordance between the changes of sensory profiles in PTH patients and their predictive function between the studies.
Pain modulation
Conditioned pain modulation (CPM) describes a psychophysical measure of pain modulatory pathways in humans, also known as the ‘pain inhibits pain’ phenomena (31). Common testing paradigms apply a painful stimulus, called ‘test stimulus’, followed by a second painful stimulus (‘conditioning stimulus’) on a different body site, that modulates the perception of the first one (32), normally resulting in a pain inhibition quantified as a CPM effect (33). Until now, there is no consensus on the most suitable testing algorithm, e.g. regarding the type or the intensity of the used test and conditioning stimulus (32). Different studies on CPM in chronic pain states (e.g. migraine, tension-type headache, fibromyalgia, postsurgical pain) indicate that the CPM capacity is less efficient in patients with pain syndromes, compared to healthy controls, and may predict the development of chronic pain (33,34).
Three of the included articles assessed CPM in PTH patients. They all used pressure pain threshold for the test stimulus, but different conditioned stimuli (Table 2).
Characteristics of the conditioned pain modulation test paradigms.
CPM: Conditioned pain modulation, HPT: Heat pain threshold, PPT: Pressure pain threshold.
In the study by Naugle et al. (2020) (27) patients with persistent PTH after mild TBI exhibited significantly reduced endogenous pain inhibition, compared to those without PTH. That inefficient endogenous pain inhibitory function at 1–2 weeks after injury predicted the presence of persistent PTH at 4 months after injury, indicating that persistent PTH is characterized by dysfunctional alterations in endogenous pain modulatory function in the early stages following mild TBI (27).
In the study by Carey et al. (2019) (28) using the same examination protocol like Naugle et al. the deficient descending pain inhibition and altered pain modulatory profiles were described to be notable only a few days after mild TBI. Though, the pain inhibitory capacity and headache intensity did not correlate (28).
Similar results were reported in an earlier study of Defrin et al. (2015) (29). Adaptation to pain and CPM were diminished in the persistent PTH group compared with the two control groups. Moreover, there was a negative correlation with the headache intensity, i.e. the stronger the headache was, the lower the CPM effect. The dysfunctional pain modulation capabilities are related specifically to the persistent PTH, rather than to the TBI, because subjects without headache after TBI and healthy controls had the same degree of CPM function (29).
All studies assessing pain modulatory function found a decreased CPM effect in patients suffering from persistent headache after TBI (Table 3).
Study results.
CPM: conditioned pain modulation, CPT: Cold pain threshold, HPT: Heat pain threshold, PPT: Pressure pain threshold, QST: Quantitative sensory testing, PTH: Posttraumatic headache, WPT: Warmth perception threshold.
Discussion
The aim of the review was to summarize data on sensory alterations in PTH and to analyse whether findings indicating central sensitization and dysfunctional pain modulation are indicative of development of chronic headache after brain trauma. Most studies included in this review were cross-sectional and all showed changes in the sensory profiles indicating a dysfunctional pain modulatory system.
Sensory testing
The sensory system alterations were inconsistent between the studies. Four articles described a heat hypoalgesia (9,26,29,30). Notably, in one study (9) this was only found in patients with tension-type-like headache, and in another (26) it was only found in the elderly (>60 years) participants. The second common alteration was a pressure hyperalgesia reported in 4 studies (9,28–30). Again, in one study (9) this was only present in tension-type-like headaches. In contrast, pressure hypoalgesia was found in the elderly subjects in one study (26). The warmth perception threshold was reported to be decreased in PTH patients in one study (26), though in another it was the case only in migraine-like headache patients (9). Moreover, a cold pain hypoalgesia was found in two studies (9,30).
Two of the studies examining possible influencing factors found different sensory profiles depending on age and PTH phenotype (9,26), indicating possibly different underlying pain mechanisms. Thus, different age groups and headache phenotypes may also require different therapeutic approaches.
One longitudinal study (27) aimed to assess the sensory function over time after TBI and find possible predictors or biomarkers. Against their hypothesis, the sensory alterations at first week after trauma could not predict the occurrence or persistence of headache at later dates.
Overall, abnormal thermal sensitivity was found in most studies and was interpreted as a feature of central neuropathic pain (9). Damages specially in the temperature system can be at the spinothalamic or the thalamocortical level or both. They are leading to increased neuronal hyperexcitability, especially in the trigeminal system, and can occur even after mild TBI, which by definition does not exclude structural damage (29). This mechanism is suggested to play an important role in central pain after spinal cord injury, severe TBI or stroke (29). On the other hand, Defrin et al. (2010) (30) reported less pronounced changes in sensory function also in headache-free patients after TBI. This allows the assumption that damage to the somatosensory nervous system is a necessary but not a sufficient condition for the development of persistent PTH. Additional factors have to contribute to its development (30).
Changes in the trigeminal sensory system are known in other headache types. It was found that primary headache entities also show a specific pattern of mechanical sensory disturbance compared to healthy controls. Recently, decreased pressure pain threshold in cranial regions were reported in subjects with tension-type headache and migraine (mechanical hyperalgesia) (35), similarly as most of the included articles in this review showed in PTH patients. It is suspected that damage to the peripheral sensory nerve endings, as well as persistent inflammation, in cephalic regions could lead to peripheral and central sensitization, expressed by alterations in tactile sensation (9). Thus, central sensitization is considered as a relevant underlying mechanism in different chronic headache types (35). In line with that, the finding of cephalic mechanical hyperalgesia in subjects with primary tension-type headache indicates a shared mechanism with tension-type-like persistent PTH. Nonetheless, thermal hypoalgesia was reported in tension-type-like PTH, whereas in primary tension-type headache it was thermal hyperalgesia. Similarly, when comparing migraine and migraine-like PTH, cold hypoesthesia was found in migraine-like PTH subjects, whereas migraine patients were characterized by a heat hyperesthesia and normal cold detection threshold. Taken together, persistent PTH seems to underlie different or additional mechanisms compared to the matching primary headaches (9,35), which differ between the clinical PTH phenotypes (9). The distinct sensory features could contribute to the differentiation of the symptom profiles, i.e. classifying as migraine-like or tension-type-like PTH.
The diverging results of the examined studies can be explained by different factors. First of all, not all studies assessed all QST components, and they did not use the same standardized protocol. Second, the studies included patients after different severities of trauma (mild, moderate, severe TBI) and the inclusion criteria differed among the studies. Further, only one study assessed the sensory function depending on the clinical phenotype (9), and the examination time after trauma was heterogenous. Only one study (27) assessed the somatosensory function at different time points in a longitudinal manner. Bouferguene et al. (26) found differences in different age groups, so that could be another possible influencing factor. Moreover, the participants were mostly under influence of pain medication or other modulatory drugs, like antidepressants, but the effect of pain medication has not been reported in the studies.
We also like to point out that the results do not suggest causality. If central sensitization is the cause for developing chronic pain, or chronic pain leads to central sensitization remains unclear, especially due to the limited number of longitudinal studies. It is also unknown if the sensitization is pre-existing before the TBI and could, by that, be a predisposing factor for PTH. It could likewise be a reciprocal influence between cranial hyperalgesia and persistent PTH, maintaining each other like a vicious cycle.
Pain modulation
The CPM capacity in PTH patients was consistently diminished among all three included studies (27–29). Naugle et al. (27) was the first human study showing that decreased CPM function at 1–2 weeks after injury has the potential to predict the development of persistent headache. Altered central pain modulatory function seems to be an important mechanism contributing to the development of PTH. However, its origin remains still unknown. Just like the trigeminal sensitization, TBI-induced damage to the ascending spinothalamic and thalamocortical system, as well as diminished descending pain inhibitory control system could result in a dysfunctional pain inhibitory capacity and thereby contribute to the development of chronic pain (36,37). In contrast, there were no significant differences between headache-free subjects after TBI and healthy controls (29). This shows that dysfunctional pain modulation is associated with the development of chronic pain after trauma, but not with TBI itself.
Literature on the magnitude of CPM in primary headache types, e.g. in migraine patients, is inconsistent. Some studies demonstrated less efficient pain modulation in migraine patients compared with healthy controls, while others found no or only subtle difference (38). However, the pain perception and the endogenous pain inhibition are affected by pain medication, especially antidepressants (39,40). Furthermore, psychological factors, like pain catastrophizing, can modulate the response (40). This could be a reason why some studies found a deficient pain modulation, while others have not. Also, a missing CPM effect in chronic pain states can represent either an insufficient endogenous pain inhibition a priori or a maximally activated inhibitory pathways. Overall, CPM cannot be described as a method free from interference (40). Nevertheless, in previous research CPM has been reported to be able to predict the occurrence of chronic post-operative pain when assessed before surgery in pain-free subjects (34). This allows the presumption that CPM effect could be a biomarker for chronic pain development also after a trigger like TBI. However, this has to be further investigated.
Psychological factors
An important factor, that cannot be ignored, is associated psychological symptoms, mostly assessed with questionnaires in the studies included in the review. All studies reported an association of psychological disorders with PTH in patients. Higher levels of depressive mood (26–28), anxiety symptoms (28), pain catastrophizing (27,28) or PTSD symptomatology (9,29,30) were found in patients with chronic headache after TBI compared to controls. These disorders are known to contribute to pain chronification and some studies reported a positive correlation between psychological symptom level and headache severity (9,28,29). Pain intensity and the psychological state are strongly interacting and constantly influencing each other (10,41).
Future implications
Chronic pain can be understood as an imbalance between excitatory and inhibitory input to the brain due to morphological and molecular changes. These changes can be caused by damage to the brain as happens in patients after TBI. The excitation/inhibition-balance plays an important role in neuronal plasticity and can shift the sensory patterns and pain perception (42). This leads to the conclusion that, if the imbalance could be compensated, pain chronification could be stopped or prevented. Studies in animal models have shown that a stimulating environment can improve the neuronal function recovery after brain damage (42). Early (during first week) stimulation therapy led to lower levels of sensory hypersensitivity in rats, whereas deprivation therapy elicited greater sensory hypersensitivity (43). Transferred to humans the results of these animal studies could assume that an early rehabilitation with sensory stimulation can improve the outcome of patients after TBI, possibly also concerning the development of PTH (42,43).
Based on the existing data, the persistence of pain after TBI is not reliably predictable and the sensory profile changes not specific. Further resolving the mechanisms for pain development after brain trauma may help identify biomarkers, influencing the outcome, such as phenotype, age, concomitant symptoms, or type of accident. In the last few years progress has been made on this topic, but future research is necessary, especially longitudinal studies examining different time points after trauma and categorizing the subjects according to already known factors.
Limitations
This review has several limitations, most importantly the small number of articles we found on the topic. The studies differed in their examination methods, inclusion criteria of the patients and TBI severity, so that the comparability is limited. Most studies were lacking information on prior primary headaches, and medication-overuse headache, which is another critical issue for the data interpretation. The results were also incongruent and partially conflicting, therefore, the conclusions for clinical practice are limited.
Conclusions
PTH is a frequent complication of TBI. The underlying mechanisms are still not fully understood. Studies on sensory function in PTH indicate a disrupted endogenous pain modulation and central trigeminal sensitization, among other factors. The alterations found in sensory profiles vary within the studies, but conditioned pain modulation was shown consistently diminished. Further research using longitudinal designs is needed to determine the changes more precisely and put them in a context with other pathophysiological changes to be able to draw conclusions for optimized target-based treatment options.
Clinical Implications
Damage to pain modulatory systems along with cranial sensitization is probably involved in the development of persistent posttraumatic headache (PTH). Diminished conditioned pain modulation seems to be an important and predictive change in headache patients. It is important to further uncover the mechanisms for pain development after brain trauma and find biomarkers for patients at risk to develop persistent PTH.
Footnotes
Data Availability Statement
Conflicts of Interest
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: EEK has received personal fees from Novartis GmbH, Casquar GmbH and painCert GmbH.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Deutsche Forschungsgemeinschaft, SFB 874/A1 and A5, Project No.: 122679504. JJ has received a research grant from the International Graduate School of Neuroscience at the Ruhr University Bochum. EKEK holds an endowed professorship funded by the German Social Accident Insurance (DGUV) for the time of 6 years (2020-2026). Ö.S.Ö. received intramural funding from the Ruhr University Bochum, Germany (FoRUM grant nr. K120-18).