Mechanisms and treatments of chronic pain after traumatic brain injury

Abstract

While pain after trauma generally resolves, some trauma patients experience pain for months to years after injury. An example, relevant to both combat and civilian settings, is chronic pain after traumatic brain injury (TBI). Headache as well as pain in the back and extremities are common locations for TBI-related chronic pain to be experienced. TBI-related pain can exist alone or can exacerbate pain from other injuries long after healing has occurred. Consequences of chronic pain in these settings include increased suffering, higher levels of disability, serious emotional problems, and worsened cognitive deficits. The current review will examine recent evidence regarding dysfunction of endogenous pain modulatory mechanisms, neuroplastic changes in the trigeminal circuitry and alterations in spinal nociceptive processing as contributors to TBI-related chronic pain. Key pain modulatory centers including the locus coeruleus, periaqueductal grey matter, and rostroventromedial medulla are vulnerable to TBI. Both the rationales and existing evidence for the use of monoamine reuptake inhibitors, CGRP antagonists, CXCR2 chemokine receptor antagonists, and interventional therapies will be presented. While consensus guidelines for the management of chronic post-traumatic TBI-related pain are lacking, several approaches to this clinically challenging situation deserve focused evaluation and may prove to be viable therapeutic options.

1. Introduction

Traumatic brain injury (TBI) is a significant health issue that affects members of the military, Veterans, and the general public across the lifespan. Defined as an injury to the brain by an outside force, most cases of TBI are caused by motor vehicle accidents, falls, sports-related injuries, and explosions or impacts from military service. Many of these injuries are mild and involve only brief loss of consciousness (Lefevre--Dognin et al., 2021), although some are life-threatening. Disabilities frequently experienced by TBI patients include issues with cognition, sensory processing, motor control, emotion, and chronic pain (Bales et al., 2009; Blennow et al., 2016; Irvine and Clark, 2018; Thomas et al., 2015). Chronic pain in the setting of TBI contributes to disability, causes suffering, complicates rehabilitative efforts and poses a significant overall challenge to management teams (Lippa et al., 2015).

Unfortunately, very little information is available concerning why patients with TBI develop various sequelae, including chronic pain. Disparities in pain-related TBI outcomes exist; recent analyses indicate that pain severity and resulting disability may be greater for non-Hispanic Blacks, particularly in elderly populations (Sander et al., 2023). Regrettably, there is limited information regarding treatment of TBI-related pain and other chronic symptoms (Heslot et al., 2022). Opioids are commonly prescribed, but they are particularly prone to causing side effects, dependence, and addiction (Starosta et al., 2021). Furthermore, history of TBI confers an elevated risk of opioid use and abuse (Bertenthal et al., 2018; Seal et al., 2018; Starosta et al., 2021), which is especially problematic for young Veterans with TBI (Golub and Bennett, 2013). Likewise, laboratory studies suggest a window for enhanced vulnerability to opioid abuse after TBI (Chiariello et al., 2023). An often-used alternative class of analgesics, non-steroidal anti-inflammatory drugs (NSAIDs), is substantially less potent and has been associated with poorer outcomes from TBI in laboratory models (Browne et al., 2006). Innovative approaches to pain treatment in the setting of TBI would be of tremendous benefit to military personnel, Veterans, and general community members.

1.1. The characteristics of pain after traumatic brain injury

A substantial body of information has accrued concerning the occurrence of chronic pain after TBI involving civilian and military cohorts, and this information provides clues about likely etiologies as well as possible treatments. The most common and highly disabling type of pain after TBI, especially mild TBI (mTBI), is post-traumatic headache (PTH) (Lefevre-Dognin et al., 2021). About 90% of patients with a history of mTBI, also known as concussion, show features of PTH. Although most PTH resolves in the first three months after an injury, headaches can persist for months in 40% of patients with acute PTH (Bales et al., 2009). Other commonly reported painful areas include the spine, and limbs, with the overall pain prevalence ranging from 22 to 95% depending on the pain site, and an overall 75.3% for those with mild TBI (mTBI) as reported in one systematic review (Nampiaparampil, 2008). Additionally, disability and movement-related pain are highly prevalent among those with TBI (Ofek and Defrin, 2007). Results of a very recent study involving soldiers demonstrated that a history of TBI was associated with a greater likelihood of reporting pain in the extremities (Odds ratio, OR 5.8), back pain (OR 6.1) or headaches (OR 8.5) compared with soldiers without TBI (Englert et al., 2023). In another review, evidence of chronic pain was identified in 81.5% of those Veteran patients seen at a polytrauma (e.g., TBI with additional sites of injury) network site (Lew et al., 2009). Interactions between TBI and the prevalence and severity of chronic pain have been demonstrated in several studies (Higgins et al., 2014; Nampiaparampil, 2008), and TBI worsens the functional outcomes of injuries to the extremities (Andruszkow et al., 2013).

Interestingly, mild rather than more severe TBI appears more likely to cause chronic pain (Nampiaparampil, 2008), indicating that the global severity of brain injury is not the only determinative factor. Unfortunately, mTBI is often missed during emergency visits related to other traumas, such as motor vehicle collisions, but many of these patients will go on to develop chronic pain (Peixoto et al., 2023). In fact, laboratory studies in which mTBI is combined with additional limb bone and muscle trauma have been shown to worsen pain outcomes measured months later (Sahbaie et al., 2018; Wong et al., 2023). Likewise, pain-related sensitization of surgical wounds and fractured limbs of animals previously given mTBI injuries outlasts the sensitization observed in control animals (Chen et al., 2023; Sahbaie et al., 2018).

2. Central mechanisms contributing to chronic pain after TBI

2.1. Endogenous pain modulation

Pain is a universal experience in animals that can be both protective e.g., avoiding potentially damaging sources of heat, guarding an injured limb and maladaptive e.g., disabling pain experienced after the resolution of tissue damage, pain experienced in response to non-noxious stimulation. A key factor for determining the animal’s pain state is the bidirectional action of its endogenous pain control systems. These systems are capable of inhibiting pain or, under other circumstances, augmenting the intensity of pain in response to a noxious stimulus. Understanding these systems is of critical importance to evaluating, predicting, and treating chronic pain as links between endogenous pain control dysfunction and pain have been shown for chronic postoperative pain, fibromyalgia, arthritic pain, chronic headache and chronic pain after TBI (Carey et al., 2019; Naugle et al., 2020; Staud, 2012; Yarnitsky, 2015). The circuitry and mechanisms of endogenous pain control systems are complex and not yet fully delineated. However, we understand that mechanisms resident to the spinal cord, including interneuron modulatory functions work in concert with top-down modulation from the brainstem (Seifert et al., 2009). The top-down circuitry is of particular interest given the vulnerability of brainstem structures and adjacent fiber tracts to the forces of TBI (Kim et al., 2021).

The principal components of the top-down or descending pain modulatory circuitry are important to consider in order to rationally design interventions to limit pain. A structure long established to regulate pain is the periaqueductal grey matter (PAG), a structure surrounding the cerebral aqueduct near the top of the brainstem. This structure has roles in defensive responses seen in animals, in addition to behaviors related to reproduction and autonomic regulation (Behbehani, 1995). Providing an endogenous source of analgesia could be critical in pain after TBI. This center receives input from the cortex, amygdala, and ascending sensory tracts via intervening nuclei in addition to other structures and is activated by a host of neurotransmitters, including opioids. Direct stimulation of the PAG with opioids or via electrodes can provide profound analgesia occasionally exploited in patients through deep brain stimulation (Sims-Williams et al., 2017). Dysfunction of the PAG has been linked to migraine, fibromyalgia and other painful diseases (Coulombe et al., 2017; Welch et al., 2001).

The target for fibers from PAG as well as fibers from sensory relay centers in the brain, such as the thalamus and parabrachial nuclei, is the rostral ventromedial medulla (RVM). This portion of the medulla, particularly the nucleus raphe magnus, forms a critical component of the endogenous pain control system capable of both enhancing and suppressing pain (Chen and Heinricher, 2022). Functionally, the RVM is often discussed as being composed of three classes of neurons, ON-, OFF- and NEUTRAL cells, that describe the changes in the firing patterns of the cells in association with nociceptive responses in animals. Opioids tend to silence ON-cells (known to be pro-nociceptive) and excite OFF-cells (known to be anti-nociceptive), both of which project to the dorsal horns of the spinal cord via the dorsolateral funiculus. Although none of these cell classes can be marked by a particular neurotransmitter, serotonin (5-HT) seems to be a key neurotransmitter in this descending regulatory pathway (Chen and Heinricher, 2022). Furthermore, serotonin release at the spinal level can also be pro- or anti-nociceptive depending on which spinal serotonin receptors are activated and which cell type is targeted. In this regard, activation of serotonin 5-HT_1a, 5-HT_1b, 5-HT_2c, 5-HT₃, 5-HT₄, and 5-HT₇ receptors have been observed to inhibit nociceptive signal transmission at the trigeminal and spinal level. On the other hand, 5-HT_2A, 5-HT_2B and 5-HT₃ receptors tend to promote nociception (Heijmans et al., 2021). As a major source of pain modulatory fibers, dysfunction of the RVM has been linked to many types of pain, including neuropathic pain, opioid-induced hyperalgesia, headache and arthritis (Mills et al., 2018; Schulte et al., 2016; Soni et al., 2019; Vera-Portocarrero et al., 2011).

Another important group of related nuclei to consider are the noradrenergic pontine nuclei A5, A6 (the locus coeruleus, LC), and A7. These nuclei project to the PAG, RVM, and spinal cord dorsal horn where they directly release noradrenaline. The LC has many functions, including sleep and pain modulation. At the level of the spinal cord, the noradrenaline released normally interacts with alpha-2 adrenergic receptors to provide analgesia. It is through these receptors that drugs such as clonidine, dexmedetomidine, serotonin-norepinephrine reuptake inhibitors (SNRIs), and tricyclic antidepressants (TCAs) exert much of their analgesic activities. In addition, spinal alpha-2 adrenergic receptors are critical for the diffuse analgesia produced in response to noxious stimulation via descending fibers termed diffuse noxious inhibitory controls (DNIC) in animals and conditioned pain modulation (CPM) in humans (Sirucek et al., 2023). While activation of the LC may be involved in the placebo response and in recovery from nerve damage, LC dysfunction may worsen pain in complex regional pain syndrome and migraine (Alba-Delgado et al., 2021; Crawford et al., 2021; Drummond and Finch, 2022; Vila-Pueyo et al., 2019).

2.2. TBI disrupts endogenous pain control mechanisms

The complex brainstem endogenous pain control mechanisms just discussed are powerful, although dysfunction has been linked to several forms of chronic pain as discussed. Recently, TBI has been shown to generate dysfunction in these systems and predispose if not cause pain in TBI patients. Some of the earliest observations using quantitative sensory and psychophysical testing showed in case-control and cross-sectional studies that TBI patients with greater levels of headache had weaker CPM capability (Carey et al., 2019; Defrin, 2014; Defrin et al., 2015). The same patients sometimes had evidence of mechanical allodynia. More recent studies have established prospectively that after mTBI, patients showing more limited CPM have a higher likelihood of headache pain persisting at four months from the time of injury (Naugle et al., 2020).

Other investigators have made progress in delineating types and locations of damage to the brain using imaging techniques, most often MRI. Some of this work has identified alterations in endogenous pain control structures. For example, “central pain” after TBI, defined according to standardized criteria, was correlated with changes in fractional anisotropy (FA) in the PAG (Jang et al., 2016). Interestingly, PAG-related changes in the default mode network (DMN) predict the persistence of post-traumatic headache (Niu et al., 2019). In patients with pain persisting 12 months from the time of TBI, the functional connectivity (FC) of the PAG is diminished with the somatosensory cortex, while FC is enhanced with the rostral anterior cingulate cortex (ACC) (Bosak et al., 2022).

Animal studies to date also support an exquisite sensitivity of the endogenous pain control system to TBI. Experimental evidence suggests that TBI causes damage to endogenous pain modulation systems in the brainstem disrupting the balance of inhibition versus facilitation of descending pain modulation (Chen et al., 2023; Irvine and Clark, 2017; Irvine and Clark, 2018; Sahbaie et al., 2018). Studies of mTBI in rats using the lateral fluid percussion model and in mice using a closed head injury approach have shown that changes in nociceptive signaling after TBI manifest in two distinct phases. The initial serotonin-dependent phase occurs within 24 h of TBI and involves mechanical allodynia of the hindlimbs that resolves within 28 days post-injury (Irvine et al., 2019). During the initial phase, there is a significant increase in neuroinflammation within the PAG, LC, RVM and superficial dorsal horn which are all key areas involved in the descending modulation of pain (Irvine et al., 2018). It has been shown after mTBI, mice show a substantially longer period of hindpaw sensitization in a surgical incision model (weeks versus a few days) (Chen et al., 2023). In the same animals, deletion of descending serotonergic signaling using a neuro-toxin restored the normal pattern of recovery, suggesting that functional endogenous pain control is required for normal recovery from pain (Chen et al., 2023). This is followed by a more slowly developing second phase characterized by the failure of DNIC that can persist to 180 days post-injury (Irvine et al., 2020; Sahbaie et al., 2022). In both rodent species, DNIC is dependent on descending noradrenalin/alpha-2 adrenergic signaling from the LC in uninjured animals and serotonin signaling after TBI (Irvine et al., 2018, 2020; Sahbaie et al., 2022).

2.3. The use of animal models in TBI research

In interpreting the findings from animal models with respect to TBI and other medical conditions, caution must be used when interpreting the findings. While indispensable in advancing biomedical science, much has been written about the injudicious application of poorly selected animal models (Mukherjee et al., 2022). Specific problems in addressing the construct, face and other types of validity, the representation of both sexes in experimental plans, experimenter blinding and the time course of study are amongst the concerns (Clark, 2016). Guidelines to enhance the reliability and impact of animal research have been written and widely promulgated (Percie du Sert et al., 2020). Although painstakingly collected, the data available from animal models need to be interpreted in the broad context of research on brain injury. Ultimately the translation of animal and other laboratory findings to human experimental and clinical studies is an important goal.

Taken together, these findings indicated that an imbalance of pain modulatory pathways after TBI can diminish descending inhibition and/or augment descending facilitation, thereby supporting the development of central sensitization in the spinal cord - a key feature in chronic pain (Arendt-Nielsen et al., 2010; Lautenbacher and Rollman, 1997; Yarnitsky et al., 2012).

3. Therapeutic opportunities for the treatment of pain after TBI

While pain of many types is common after TBI, no widely promulgated guidelines exist for the management of the array of pain syndromes encountered in these patients. The current pain management practice tends to focus on specific painful sequalae of TBI, and seldom involves a focus on the underlying mechanisms that might be relevant to multiple types of pain. The following section highlights several targeted therapeutic opportunities based on the recent knowledge of the mechanisms underlying TBI-related pain.

3.1. Therapies targeting serotoninergic and noradrenergic signaling

As mentioned in the previous section, the knowledge of descending serotonergic and adrenergic interactions gives rise to several potential therapeutic targets. Some existing pharmacological agents, such as selective serotonin reuptake inhibitors (SSRIs), SNRIs, or TCAs, could restore the balance of the pain modulatory pathways and have translational potentials for treating TBI-induced chronic pain.

3.2. Selective serotonin reuptake inhibitors (SSRIs)

The triad of chronic pain, post-traumatic stress disorder (PTSD), and TBI have classically been recognized as the “polytrauma clinic triad” (Cifu et al., 2013). SSRIs, such as sertraline and fluoxetine, are recommended as one of the first-line treatments for depression and PTSD symptoms following TBI (Fann et al., 2009; Yue et al., 2017). Although the analgesic effect of SSRIs for treating post-TBI chronic pain has not been directly studied, recent clinical data recognized that TBI patients on SSRIs reported significantly lower pain scores than those who did not (Mollayeva et al., 2017). Laboratory data using animal models of TBI also showed anti-nociceptive effective through restoration of descending inhibitory drive that was disrupted by the head injury (Irvine et al., 2019; Sahbaie et al., 2022). Intriguingly, blockade of spinal 5-HT₃ receptors also reversed the behavioral allodynia in animals following TBI, suggesting an upregulation of descending pain facilitation (Chen et al., 2023; Irvine et al., 2019; Sahbaie et al., 2019). Collectively, these data imply that TBI dysregulates the complex interplay between descending pronociceptive and antinociceptive drives, and SSRIs may help restore the balance within the endogenous serotoninergic pain modulatory system.

3.3. Serotonin-norepinephrine reuptake inhibitors (SNRIs)

SNRIs (e.g., duloxetine, venlafaxine) have long been used for treating neuropathic pain, such as diabetic neuropathy and posthepatic neuralgia, or central pain syndromes, such as poststroke pain (Finnerup et al., 2015; Kremer et al., 2016). Although much less evidence directly examines the effect of SNRIs on post-TBI pain, they are still widely used clinically for pain conditions related to TBI. The mechanism underlying the analgesic effect of SNRIs is not well delineated but is thought to involve augmentation of both descending serotonergic and aminergic pathways (Kremer et al., 2016). Diminished descending adrenergic signaling through spinal α2 adrenoceptor signaling was seen in animal models of peripheral nerve injury, which could be restored by administration of SNRIs (Arora et al., 2016; Burnham and Dickenson, 2013). Animals with TBI were shown to have diminished brain and spinal cord noradrenalin content and loss of cell bodies in the LC and other noradrenergic nuclei associated with descending motor and nociceptive inhibition (Bose et al., 2013), and restoring spinal NA content is associated with reductions in spasticity and gait abnormalities and restoration of DNIC (Bannister et al., 2015; Bose et al., 2013). A recent study demonstrated that administration of the noradrenergic reuptake inhibitor (NRI), reboxetine, reverses peripheral allodynia observed during the first pain phase in animals following TBI, although reboxetine failed to restore DNIC in this particular study (Irvine et al., 2020). This suggests that the antinociceptive mechanism of SNRI in post-TBI pain may be distinct from pain arising from peripheral sources.

3.4. Tricyclic antidepressants (TCAs)

Similar to SNRIs, the principal mechanism of anti-neuropathic action for TCAs is by inhibiting the presynaptic reuptake of serotonin and noradrenaline (Kremer et al., 2016). TCAs, most notably nortriptyline and amitriptyline, are often used for treating neuropathic pain (Kremer et al., 2016), and have been effective in treating both chronic tension headache and migraine as preventive therapies (Bettucci et al., 2006; Silberstein et al., 2012). In the context of PTH, recent clinical data also demonstrated their efficacious for PTH as an abortive therapy, though their benefits appear to diminish when used long-term. (Cushman et al., 2019).

3.5. Therapies targeting calcitonin-gene related peptide (CGPR) signaling

Headache, the most common type of pain experienced after TBI, is named as a specific category of headache by The International Classification of Headache Disorders 3rd edition but is typically described as migraine or tension-like. Consequently, treatment recommendations generally follow the principles used to manage headache complaints in the non-TBI patients, such as those targeting the CGRP signaling pathways. However, it is important to note the current lack of adequate trials for medications (Larsen et al., 2019) or alternative non-pharmacological approaches (Lee et al., 2022) specific to post-traumatic headache (PTH) in TBI patients.

3.6. Direct calcitonin-gene related peptide (CGPR) antibodies and antagonists

CGRP-positive neurons and nerve fibers are prominent in multiple important brain pain-modulating regions, including RVM and LC, and are thought to play an important role in headache and migraine (Holland et al., 2019). Two classes of therapeutic agents have been developed to interrupt the function of CGRP: CGRP-targeted monoclonal antibodies (mAbs) and small-molecule antagonists (gepants). Preclinical studies show that mTBI leads to the development of headache-related pain behaviors that are mediated through CGRP-dependent and CGRP-independent mechanisms (Bree and Levy, 2018; Navratilova et al., 2019). Four CGRP-targeted mAbs and three gepants have been US Food and Drug Administration (FDA) approved for the treatment of migraine (Cohen et al., 2022). While the use of these mAbs and gepants has been FDA-approved for clinical use in migraine, their specific efficacy and safety for PTH are still being explored.

CGRP mAbs:

Monoclonal antibodies, such as erenumab, fremanezumab, and galcanezumab, have been developed to target CGRP or its receptor and are approved for the prevention of migraines. These antibodies work by blocking the activity of CGRP, which is involved in pain transmission and inflammation. Studies investigating the use of CGRP monoclonal antibodies for PTH are limited but have shown some positive results. For example, a sentinel study published in 2020 evaluated the use of erenumab in patients with persistent PTH. The study found that erenumab treatment among patients with persistent PTH showed a lower frequency of moderate to severe headache days was associated with a significant reduction in headache days and improved quality of life (Ashina et al., 2020).

Preclinical studies by Navratilova et al. showed that continuous sequestration of CGRP peptide prevented both acute and persistent PTH (Navratilova et al., 2019). On the other hand, delayed CGRP mAb treatment (after central sensitization was established) did not prevent persistent PTH. This study suggests that early and continuous CGRP peptide blocking after mTBI may be a viable treatment option. A recent study from the same group further demonstrated that intraperitoneal fremanezumab, a CGRP mAb, blocked not only the acute PTH-related pain but also prevented the loss of net descending inhibition within central pain modulation pathways (Kopruszinski et al., 2021).

Gepants:

Although gepants are approved for migraine treatment, there are no preclinical studies available to help understand possible effects on PTH. On the other hand, recent data do show that the CGRP receptor antagonist olcegepant significantly reduced allodynia in rodent models of migraine-like pain (Christensen et al., 2019). In clinical settings, PTH patients commonly exhibit migraine-like or probable migraine-like headaches suggesting possible utility in this population. Analgesics, such as gepants, are prescribed for the acute treatment of migraine and have a favorable safety profile (van Hoogstraten and MaassenVanDenBrink, 2019). Further research is needed to establish the effectiveness and safety of CGRP monoclonal antibodies specifically for PTH. The underlying mechanisms and pathophysiology of PTH may differ from primary migraines (Guglielmetti et al., 2020), and the response to treatment could vary as well.

3.7. Botulinum toxin

Onabotulinumtoxin-A, commonly known as Botox, has been FDA approved for the prevention of chronic migraine and has been studied as a potential treatment for various types of chronic headaches. New evidence and specific research on the use of onabotulinumtoxin-A for PTH is emerging and it has been considered as a potential treatment option. The mechanism of action of onabotulinumtoxin-A involves blocking the release of neurotransmitters involved in pain signaling. It is thought to inhibit the release of CGRP, glutamate, and other pain mediators. These properties make it a potentially useful treatment for chronic headaches, including PTH.

A recent study demonstrated that administration of onabotulinumtoxin-A subcutaneously over the cranial sutures in animals with mTBI reduced their pain behaviors (Navratilova et al., 2022). In these studies, administration of onabotulinumtoxin-A 72 h after mTBI, resolved acute PTH but only transiently diminished persistent PTH, suggesting that the underlying mTBI-induced pathophysiological deficits were not fully restored. On the other hand, administration of onabotulinumtoxin-A 2 h after mTBI, fully blocked both acute and persistent PTH-related pain, suggesting blocking of the mTBI-induced pathophysiological deficits leading to headache-like pain.

Several studies have explored the use of onabotulinumtoxin-A PTH in clinical settings. Overall, these studies showed treatment with onabotulinumtoxin-A significantly reduced the headache frequency and severity in patients with PTH (Zirovich et al., 2021).

3.8. Oxytocin receptor agonists

Oxytocin, a neuropeptide primarily associated with social bonding and childbirth, has been found to have potential effects on pain modulation. Specifically, application of nasal oxytocin has been shown in small studies to reduce pain in painful conditions such as migraine headache as well and in other chronic pain conditions (Tzabazis et al., 2017). Oxytocin was effective in reducing pain sensitivity in a preclinical model of PTH as well (Meidahl et al., 2018).

The mechanisms through which oxytocin reduces pain are yet to be fully understood, but oxytocin has been reported to block capsaicin-induced CGRP release, in vitro, from dura (Tzabazis et al., 2016). In addition, studies showed that activation of oxytocin receptors on trigeminal ganglia neurons in vitro decreased their excitability (Bharadwaj et al., 2022). Modulation of neuronal activity of limbic and cortical brain regions, as well as ascending and descending pain pathways in the spinal cord have been suggested as potential mechanisms for oxytocin’s pain-modulating effects (Boll et al., 2018). Oxytocin has been reported to reduce trigeminal neuronal firing evoked by meningeal electrical stimulation, a well-established model of migraine (Garcia-Boll et al., 2020). Nasally applied oxytocin has been reported to concentrate in the trigeminal nerve, ganglia, and nucleus, inhibits firing of peripheral and central trigeminal nociceptive neurons (Tzabazis et al., 2017). Thus, oxytocin may act via several convergent mechanisms to limit PTH pain.

3.9. Therapies targeting chemokine receptors

Chemokines (chemotactic cytokines) are classified based on structural features that in turn give rise to their functional specificity. There are two main categories of chemokines; CXC (alpha-chemokines) and CC (beta-chemokines). CXC are primarily chemotactic for neutrophils yet CC are attractants for monocytic cells and lymphocytes (Zlotnik and Yoshie, 2000). The membrane bound CXC receptor 2 (CXCR2) has been demonstrated to be the principal receptor mediating neutrophil migration in rodents. Furthermore, studies using CXCR2 mutant mice have shown that it is responsible for the removal and regulation of circulating chemokines, identifying a unique scavenging role for CXCR2 (Semple et al., 2010b).

3.10. CXCR2 receptor antagonists

CXCR2 and its ligands (CXCL1-3, 5–8) have been shown to have a critical role in the early infiltration of neutrophils across the blood brain barriers after TBI (Semple et al., 2010a; Wang et al., 2022). CXCL1 and CXCL2 expression peaks within 4 h and remains so up to 24 h post-injury. Closed head injury of CXCR2 deficient mice revealed a reduction in the extent of neutrophil infiltration into the brain after injury compared to wildtype mice (Semple et al., 2010a). This resulted in significant decreases in tissue damage, neuronal loss and overall cell death (Semple et al., 2010a) compared to wildtype mice. CXCR2 has also been shown to be expressed on neurons in several areas of the brain and spinal cord, which extends to both astrocytes and microglia following activation (Liang et al., 2017; Semple et al., 2010a; Xia et al., 2022).

CXCR2 and its ligands have also been shown to be involved in regulating pain after several conditions including cancer (Xu et al., 2022; Zhou et al., 2015), inflammation (Cao et al., 2016; Sahbaie et al., 2019; Zhang et al., 2013), surgery (Kiguchi et al., 2012; Sun et al., 2013) and chronic opioid administration (Sun et al., 2014). Nociceptive sensitization after TBI has been shown to occur, in part, due to epigenetic upregulation of spinal CXCR2 expression (Liang et al., 2017). Spinal CXCR2 and 5-HT signaling appears to be closely intertwined. Blocking the TBI-mediated increase in spinal CXCR2, with an antagonist (SCH5 27123) or spinal 5-HT depletion, significantly reduced nociceptive sensitization (Sahbaie et al., 2019). Although the mechanism underlying antinociceptive effects of CXCR2 antagonism in post-TBI animals was not well delineated, it could be related to its ability to decrease spinal microglial upregulation, attenuate acute spinal neuroinflammation and reduce TBI-induced spinal CXCR2 expression (Sahbaie et al., 2019).

3.11. Exercise therapies for Post-TBI pain

Exercise has a host of benefits for TBI and other trauma patients and is often incorporated into rehabilitative recovery plans. Aside from benefits to the musculoskeletal system, exercise has received particular attention as a way to enhance cognitive recovery from brain injury and improve mood. Reviews are available describing literature support for exercise in TBI for both adult and pediatric patients (Sharma et al., 2020; Sharma and Timmons, 2019; Vanderbeken and Kerckhofs, 2017). A separate meta-analysis concluded that exercise after TBI had a moderate positive impact on overall quality of life (Chang et al., 2023). Exercise is not without controversy, however, even when used for overall rehabilitative effects; some have expressed concern that physical exercise started too early after injury might exacerbate damage to the brain or delay recovery. Consensus guidance indicates that symptom-limited aerobic exercise can begin 24–48 h after TBI, but the same guidance cites the lack of clear data on the optimal timing, type, and progression of intensity of exercise after mTBI (McCrary et al., 2017). Lastly, clinical studies directed at understanding the benefits or possible adverse consequences of aerobic exercise on pain after TBI are also lacking.

Laboratory studies for many years have focused on the ability of aerobic exercise in mice and rats to preserve cognitive ability and reduce signs of depression and anxiety. It has been demonstrated that exercise can reduce the production of inflammatory mediators, moderate glial activation, limit neuronal loss, augment the production of neurotrophic molecules such as BDNF and enhance neurogenesis (Battistini et al., 2023; Hu et al., 2023; Zhao et al., 2015). Recent preclinical evidence suggested exercise may have a substantial benefit on pain after TBI. A recent study by Bharadwaj and colleagues (Bharadwaj et al., 2023) demonstrated that voluntary exercise (e.g., running wheels) reduces hindlimb sensitization as well as periorbital sensitization (e.g., a model for headache) in animals with TBI. Furthermore, DNIC response, indicative of endogenous pain modulation, was partially preserved in the exercised animals, but was lost in the non-exercised groups. These effects were accompanied by the reduced spinal cord production of nociceptive mediators, including prodynorphin and the chemokine pathway molecules CXCL1, CXCL2, and CXCR2 (Bharadwaj et al., 2023). However, similar to the situation with exercise in TBI patients, the optimal timing and intensity of exercise in the laboratory setting is still unclear, although one recent study suggests a 48-h delay may reduce the exacerbation of inflammatory mediator production (Taguchi et al., 2019). Future studies are also required to further delineate the timing and duration of exercise and determine the durability of effects on post-TBI pain once exercise stops.

3.12. Neuromodulation therapies

Beyond medications and behavioral modifications, it has been observed that non-pharmacological treatments, such as spinal cord or dorsal root ganglion stimulation and transcranial magnetic stimulation, exert some analgesic effects through the augmentation of descending pain modulation, and therefore may be of use to TBI patients with pain (Dall’Agnol et al., 2014; Schuh-Hofer et al., 2018).

The use of spinal cord and dorsal root ganglion stimulation for the treatment of chronic pain has gained popularity in recent years. Clinical data suggests that these are safe and efficacious treatment options for treating several central neuropathic pain conditions, such as chronic back pain and complex regional pain syndrome (Harrison et al., 2018; Liem et al., 2015; Simpson et al., 2009). It was thought that electrical stimulation of the spinal cord or dorsal root ganglion inhibits the ascending nociceptive pathway as well as recruits the descending pain-inhibitory pathways (de Geus et al., 2023; Heijmans and Joosten, 2020; Schuh-Hofer et al., 2018). Similarly, transcranial magnetic stimulation aims to activate the top-down pain-controlled mechanisms and has been shown to have some positive analgesic effects in some chronic pain syndromes (Dall’Agnol et al., 2014; Pink et al., 2021). However, it is important to note that there are considerable variations in stimulation protocols, and the efficacy and safety of these treatment modalities have not been directly studied in the post-TBI population. Therefore, one should proceed with these neuromodulation techniques with caution until further clinical and basic science studies are able to delineate the underlying mechanisms and optimize treatment protocols for TBI-related pain syndromes.

4. Conclusions

With the high rate of chronic pain in TBI suffers and the lack of effective treatments, the need to understand the pathophysiology underlying the TBI-chronic pain relationship becomes increasingly important. There is growing evidence to suggest that disrupted endogenous pain control mechanisms are a crucial part of the pathogenesis of chronic pain after TBI. Several pharmacological agents targeting the descending serotonergic and noradrenergic pathways, CGRP signaling, and chemokine signaling have shown promise in treating TBI-related pain. Additionally, non-pharmacological interventions, such as exercise programs and neuromodulation techniques, could offer analgesic effects through augmentation of descending pain modulation. The challenge now lies in further understanding endogenous pain modulatory mechanisms, refining these strategies, and evaluating their effectiveness in rigorous clinical trials, which will ultimately better the quality of life for individuals with TBI.

Abbreviations

ACC

anterior cingulate cortex

CGPR

calcitonin-gene related peptide

CC

beta-chemokines

CPM

conditioned pain modulation

CXC

alpha-chemokines

CXCR2

CXC receptor 2

DMN

default mode network

DNIC

diffuse noxious inhibitory controls

FA

fractional anisotropy

FDA

Food and Drug Administration

LC

locus coeruleus

NSAID

non-steroidal anti-inflammatory drugs

NRI

Noradrenergic reuptake inhibitors

OR

odds ratio

PAG

periaqueductal grey matter

PTH

post-traumatic headache

PTSD

post-traumatic stress disorder

RVM

rostral ventromedial medulla

SSRI

selective serotonin reuptake inhibitor

5-HT

serotonin

SNRI

serotonin-norepinephrine reuptake inhibitor

TBI

traumatic brain injury

mTBI

mild traumatic brain injury

TCA

tricyclic antidepressant

Data availability

No data was used for the research described in the article.