Nonresponsive Cervicogenic Headache and Mechanisms for Action of Low Level Laser Treatment

Abstract

Cervicogenic headache (CH) is a commonly presenting pain condition that affects 2.5% to 4.1% of the population, 13% to 17% of the headache population and up to 33% of patients presenting at complex multidisciplinary pain clinics. Patients with CH experience a large burden of suffering in work and activities of daily living. It is classified as a secondary headache and is defined as a headache originating from the neck with head and neck symptoms that include dizziness, photophobia and autonomic symptoms. It can be traumatic or atraumatic in origin and generally responds well to physiotherapy intervention. There however remains a subset of patients who do not respond to treatment (25%). There are no currently identified predictors of responsiveness to physiotherapy treatment. CH also responds to other treatments, including facet joint injections, radiofrequency neurotomy and surgery to the cervical spine. These treatments similarly have a nonresponsiveness rate of 25%, also with no identified predictors for unresponsiveness. Unresponsiveness to treatment for musculoskeletal pain conditions, apart from CH, including cervical and lumbar pain, is also apparent in the clinical setting and likewise presents a burden to the health system, in terms of time off work and suffering by the patient. Evidence for the effective use of low level laser treatment (LLLT) also known as photobiomodulation (PBM) has been increasing in the literature, particularly in the area of chronic neck pain (including headache) especially in the subset of patients who are unresponsive to physiotherapy, surgery, joint injections and other therapeutic interventions. The aim, therefore, of this thesis is to investigate current features of CH that may determine unresponsiveness to treatment and, from an examination of the relevant literature, to determine which of these features would respond to intervention with LLLT and to build an enhanced understanding of LLLT mechanisms. Since there have been no predictors of unresponsiveness to treatment of CH in previous investigations, a model for investigation was used that involves a hierarchy of scientific knowledge, going from consensus amongst groups of therapists up to supported theory. Accordingly, musculoskeletal physiotherapists who regularly treated adults and children with CH were surveyed as to their views of the reasons for unresponsiveness to treatment. They were also surveyed for their opinions regarding the discriminatory feature of CH in children with the condition. The survey of 90 physiotherapists (response rate 75%) is presented in Chapter 1 of the thesis. The group of clinicians had good consensus regarding several predisposing factors of unresponsiveness to treatment. These included: a history of severe trauma, neural sensitivity, genetic history of CH or migraine, absence of strong impairment in physical signs, latency of response to treatment and immunological comorbidities. Features that some physiotherapists felt were strongly associated with unresponsiveness to treatment included: the presence comorbid temporomandibular dysfunction, the presence of leg length discrepancy and pelvic asymmetry (related to global asymmetry), the feature of dizziness (especially in children) and the presence of ‘stomach pain’. Children with CH were considered to have similar features to adult unresponsive CH but responded to treatment. One of the main consensus features of unresponsive CH suggested by the expert physiotherapists was a family history of CH. The knowledge gained from this first study was therefore utilized in an examination of sufferers and non-sufferers of CH in family members. All features suggested by the expert physiotherapists as being related to unresponsiveness to treatment were tested in study 2, whether there was consensus or not, so long as plausible linking mechanisms to CH were present. The sufferers had all undergone a course of physiotherapy (including LLLT) at some point previously. The results of this study are presented in Chapter 2. There was an exclusion criterion for severe trauma and, during the study one of the subjects sustained a severe rugby injury with resultant cranio-cervical instability. The case study is presented as a concomitant published paper in Appendix 1. The results of the examination of CH sufferers and non-sufferers demonstrated a significant difference between the groups in the areas of tandem balance, both with eyes open and shut (p<0.03). The tandem stance test is a measure of autonomic balance mechanism, which may reflect a variety of pathophysiological conditions. Other features of CH, encompassed in its definition but not always present, are photophobia, asymmetrical neural sensitivity and autonomic features such as peri-ocular swelling. These features suggest that the observed balance impairment may relate to an ion channel dysfunction (a channelopathy) of calcium or potassium, which are present in other benign balance conditions as well as the vertigo associated with migraine with aura. There was also an observed difference in the performance of cervical muscle endurance, as measured by the cranio cervical flexion test (CCFT) (p<.05), which is a discriminatory test for the presence of CH. Both muscle balance and muscle endurance are considered modifiable by physiotherapy after treatment intervention. These results raise the question as to whether children should be screened for CH where family members have unresponsive CH. It is noteworthy that the expert physiotherapists found children to be more responsive to treatment for CH than adults. The importance of balance as an impairment in the CH sufferers in the study reported in Chapter2, taken together with neural sensitivity in CH, reported in the literature, pointed to the potential role of channelopathies in the aetiology of CH, and necessitated an examination of ion channel modulation by LLLT. The presence of ion channelopathies in other unresponsive pain conditions and the potential for modulation of channelopathies by LLLT were reviewed and are presented in Chapter 3. The main findings were that LLLT acts on small potassium channels, calcium channels and TRPV channels at the cellular membrane (including neuronal and neuro-muscular membranes) and mitochondrial membranes and therefore influences the flux of calcium ions. This is in addition to its generally accepted action on chromophores within the mitochondria and its action on neurotransmission at the synapse. Since ion channels respond to perturbations of mechanical force, it was logical to then review mechanotransduction effects at the membrane and the signal transduction pathways that were subsequently invoked. Chapter 3 also includes (in draft form) an integrated model of ion channel modulation by the mechanotransduction effects of LLLT on tendon and muscle repair. The conclusion of this work was that conformational modulation of proteins (including ion channels) by LLLT, acts as a photomechano- chemical switch, which affects a cell response to up or down regulate genes, including antioxidant genes and collagen type 1 and 3 synthesis genes and also acts to repress C3 complement genes. This mechanism may mimic the endogenous molecular switching mechanisms involved in neurotrophic signalling. Because LLLT has a biphasic effect, where a small increasing dose causes a therapeutic effect, but a larger dose is ineffective and because LLLT dosage is highly individual depending on skin pigmentation, an experiment was undertaken to measure the transmission of laser light through different skin types (Chapter 3). The result of this experiment was a dose equivalence equation for patients with differing Fitzpatrick scale skin colouration, to enable achieving the desired delivered dose at the commonly used LLLT wavelength of 904nm. The equation can be summarized as a 6% increase in dose requirement for each Fitzpatrick scale unit, a finding which corresponds closely with a similar study that measured absorption by the skin, as opposed to transmission. The implication of skin pigmentation underlying the variation of delivered LLLT dose led to an investigation of the physiology of neuromelanin in the brain, spinal cord and sympathetic nervous system. Neuromelanin is synthesised in different pathways to skin melanin, involving pathways of melanocortin signalling and the POMC precursor molecule and catecholamine breakdown. Variations in these pathways lead to variations in neuromelanin. For example, red-headed women (MCR1 receptor gene variation) have a variation in catecholamine uptake and in their response to β- opioids, which may result in these individuals having a varied pain tolerance and a different anaesthetic response. This may result in a different individual LLLT dosage requirement. Other pigment variations may also display variations in LLLT dosage requirement. Taken together, this information resulted in the proposal of a novel mode of signal transduction pathways including phototransduction in the nervous system, which is included in Chapter 4. Further elucidation of the model exploring the importance of prion protein in signal transduction followed, and this work is also in Chapter 4. These concepts, as well as an exploration of research into cytoskeleton assembly/disassembly (involving microtubules and β- tubulin) was used as the basis for the proposal of an extension of current preconditioning regimes, to use PBM to intervene in the increasingly prevalent condition of post-operative cognitive dysfunction (POCD). Summarizing Chapter 4 findings gives the following logic path: the model of LLLT involves a protein to protein transfer of information from the conformational modulation of these proteins, including receptor tyrosine kinases (RTK) and ion channels with resultant signal transduction cascades of signalling pathways. This photoactivation is proposed to act as a switch and would also involve redox sensing molecules such as TRPV1 potassium and calcium ion channels and small molecules such as SUMO. This process would then trigger endogenous pathways that act to modulate the cytoskeleton, which is a signal transducer of neurotrophic signalling. The varicosities seen in LLLT neural blockade and analgesic effect of laser are also present as an endogenous mechanism for neuroprotection against NMDA poisoning, hibernation and the protective and adaptive response of cortical spreading depression, seen in the aura of migraine. The proteins involved in neuroprotective varicosities are PrPC and PSD95. It is of note that these two proteins, as well as being involved in neuroprotective varicosities, are both molecular determinants of balance. In conclusion, the strongest significant factors present in sufferers of CH compared with nonsufferers are impairments in the tandem test for balance. Despite these patients having had a course of treatment at some stage in the clinic (including LLLT) and demonstrating minimal other signs and symptoms of CH compared to non-sufferers, there were still discernible differences in this feature. The two proteins that appear to be involved with endogenous modulation of the cytoskeleton, resulting in varicosities are also molecular determinants of balance. It could be hypothesised, therefore, that although genetic features are not changeable, the ability to invoke epigenetic changes and proteomic modulation at the membrane may be a reason for the efficacy of LLLT in the immediate and medium term. The relationship between the proposed mechanism of LLLT action and SUMOylation, prionization of functional amyloid, IUP interactions and the theory of the proposed Resonance Recognition Model are also explored in Chapter 5. This research could be expanded to assess the efficacy of LLLT in preconditioning against POCD, with a view to the examination of modulating proteins at the neural membrane; a possibility that is explored in Chapter 5, with reference to further examination of proteomic mechanisms that would be involved in neural homeostasis and atom-to-organism hormesis.