Whiplash-associated disorders (WAD) are the clinical manifestations of whiplash injury that may include motor dysfunction, altered pain processing and psychological distress (Sterling, Kenardy, Jull & Vincenzino, 2003; Seroussi, Sing & Fry, 2015). These symptoms may resolve from within a few days to up to 3 months, however over one third of patients will develop chronic WAD, leading to long-term disability and reduction of health status, physical function and quality of life (Seroussi et al., 2015). The complexity of chronic WAD can pose a challenge for clinicians in the assessment, diagnosis and treatment of the condition often leading to poor prognosis and treatment outcomes for the patient. The mechanisms behind the development of chronic WAD are still poorly understood, however there is evidence that the biomechanical forces associated with initial trauma have weak correlation with long term-WAD symptoms, suggesting factors other than the biological must be explored (Myrtveit et al., 2014). Psychological comorbidities such as depression, anxiety and post-traumatic stress disorder (PTSD) along with maldapative behavioural patterns such as fear avoidance and catastrophizing have been associated with chronic WAD, warranting an investigation into the relationship between treatment interventions targeting psychological factors and their effects on WAD symptoms (Sterling et al., 2003; Sterling & Chadwick, 2010; Dunne, Kenardy & Sterling, 2013; Myrtveit et al., 2014). This paper will examine psychosocial components of chronic WAD and how these may benefit from psychological models of treatment by analysing and contrasting both Cognitive behavioural therapy (CBT) and Acceptance and Commitment Therapy (ACT). Furthermore, there will be an analysis of current research applying these models to WAD populations in order to determine their clinical significance in the management of chronic WAD.

Our understanding of pain has developed exponentially in the last 50 years, facilitating a movement away from a traditionally linear, biomedical approach. The conceptualisation of the gate control theory by Melzak and Wall (1965) was the first model of pain that recognised the integration of both physiological and psychological aspects of pain and has led to subsequent theories surrounding pain including the current neuromatrix model proposed by Melzack (1990, 1999, 2001) (Hadjistavropoulos & Craig, 2004). These concepts and those before them provide a history that reflects a similar progression in terms of pain management strategies. Current best practice protocols encourage a focus on the biopsychosocial approach by which pain is viewed as a complex, multifaceted experience emerging from the dynamic interplay of physiological, psychological, social and environmental influence (Roditi & Robinson, 2011). Chronic WAD is an example of a persistent pain condition in which management has benefited from this particular approach, with a large body of research identifying psychosocial factors as leading precursors and precipitating factors to long-term suffering and disability in patients with WAD (Sterling et al., 2003; Nieto, Miro & Huguet, 2008; Williamson, Williams, Gates and Lamb, 2008; Bostick, Carroll, Brown, Harley and Gross, 2013; Myryviet, Wilhelmsen, Petrie, Skogen & Sivertsen, 2013; Myrtveit et al., 2014; Seroussi at al., 2015).

WADs differ from many other musculoskeletal disorders as they more than likely to be preceded by a traumatic event such as a motor vehicle crash (MVC) (Sterling et al., 2003). This has triggered further research into the role post-traumatic stress symptoms (PTSS) may play in the development of persistent symptoms (Sterling et al., 2003). The shared vulnerability model proposes that PTSD and chronic pain may occur due to a common predisposing factor, with correlations suggesting anxiety sensitivity (Asmundson, Coons, Taylor & Katz, 2002). Another construct known as the maintenance model proposes that the cognitive, affective and behavioural components of chronic pain and the physiological, affective and behavioural components of PTSD have led to a mutual maintenance and exacerbation of both conditions (Asmundson, Coons, Taylor & Katz, 2002; Morasco et al., 2013). More recently, Morasco and colleagues (2013) identified the role of pain coping strategies and depressive symptoms as a partial mediator of the relationship between PTSD and both pain interference and severity. In clinical settings, failure to consider complex relationships shared between PTSD and chronic pain is likely to lead to reduced overall treatment effects and poor patient outcomes.

Poor mental health status defined by conditions such as depression and anxiety have been well documented as symptoms associated with chronic WAD (Sterling et al., 2003; Papadakaki et al., 2017). Mykletun and colleagues (2011) have suggested a reverse causality in the association between whiplash and symptoms of anxiety and depression, where anxiety and depression are the trigger of WAD rather than a product of the condition. This highlights anxiety and depression as potential key predicative risk factors chronic WAD development. Other studies have highlighted the role of depression as a mediator of pathways proposed in the fear-avoidance (FA) model.

The fear-avoidance (FA) model is a psychological construct which provides a rationale as to how individuals develop chronic pain conditions based on avoidant behaviour stemming from fear (Crombez, Eccleston, Van Damme, Vlaeyen & Karoly, 212). It suggests that cognitive and behavioural mechanisms present in WAD such as fear based pain, catastrophizing and activity avoidance, often as a result of trauma in WAD, lead to physical deconditioning and sensitised responses to pain which repeat in a vicious cycle that maintain pain and suffering (Nieto, Miro & Huguet, 2008; Sterling & Chadwick, 2013).

It has been hypothesised that psychological models of treatment “ultimately impact only a small subset of primary psychological factors: (a) what patients think (cognitive content), (b) how patients think (cognitive processes or cognitive coping), and (c) what patients do (behavioural coping)”, suggesting models that focus on these factors are likely to have the greatest impact (Jensen and Turk, 2014). CBT is a psychological treatment model based on the theory that an individual’s cognitions about a particular event or situation can influence their emotional, behavioural and physiological reactions (Beck Institute for Cognitive Behaviour Therapy, 2016). In the context of chronic pain, CBT is used help reduce pain and psychological distress and improve function by helping patients identify thoughts and beliefs connected to maladaptive pain behaviours in order to decrease these behaviours and replace them with adaptive ones (Edhe, Dillworth & Turner, 2014).  The last 30 years has seen CBT become a widely acceptable psychological treatment in patients with chronic pain conditions, however its effectiveness is still debatable (Edhe et al., 2014; Broderick et al., 2016; Wetherell et al., 2016). CBT offers a level of diversity where a variety both independent and dependant variables can be applied within the model, however this can lead to a lack of specificity with regards to core therapeutic techniques and therefore viewed as much of a weakness as it is a strength (McCracken & Morley, 2014). Additionally, this level of freedom means that fundamental application of the model will differ significantly between practitioners, making it challenging to define and operationalise CBT in clinical settings (Jensen & Turk, 2014). Despite this critique, there is evidence to support the use of CBT in both chronic pain and PTSD populations, suggesting it is a viable treatment option for chronic WAD sufferers (Morasco et al., 2013). Research surrounding the use of psychological interventions in the prevention of psychological distress post MVC have also found promising results regarding the use of CBT as a preventative treatment (Guest, Tran, Gopinath, Cameron & Craig, 2016).

ACT is a relatively new treatment model, with early research showing promising results in the management of chronic pain (Hayes, Luoma, Bond, Masuda & Lillis, 2006; Scott & McCracken, 2015). It is considered a derivative of CBT, however differences in both theoretical and philosophical foundations, scientific approach and clinical applications differentiates the two models (Scott & McCracken, 2015). ACT is based on the psychological flexibility (PF) model which considers human behaviour from a functional contextual where behaviour is determined by the function of experiences such as emotions, cognitions and pain in a particular context (Hayes et al., 2006; ;Scott & McCracken, 2015). The PF model encompasses six interrelated processes: acceptance, defusion, moment-to-moment awareness, self-as-context, values orientation and committed action (Wicksell, Olsson & Hayes, 2010; McCracken, 2013; Vowles, Snowden & Ashworth, 2014; Scott & McCracken, 2015; Wetherall et al., 2016). More recently these terms have been grouped and summarised further as behaviour that is open, centered and engaged in order to ease the application of this model for ACT in a clinical setting (Vowles, Snowden & Ashworth, 2014). This model is pragmatic in its approach to pain management as it does not aim to necessarily eliminate pain-related experiences but increase an individual’s capacity to be open and aware of their cognitions and emotions and enhance their ability to interact with these experiences so they can increase effectiveness of actions and achieve personal goals in the presence of pain (McCracken, 2013; Wetherall et al., 2016). Clinically, ACT integrates many familiar treatment methods seen in others styles of psychological therapy such as mindfulness, exposure and behavioural activation however the particular goals and focus within treatment help delineate the approach from other psychological therapies (McCracken & Vowles, 2014). Brown, Gaudino and Miller (2011) reported that in a survey of cognitive behavioural therapists which included those who identified as traditional CBT practitioners and ACT practitioners, therapists who practice with an acceptance based approach gravitated towards the use of exposure, family system techniques and mindfulness as well as a larger set of techniques in total whilst those with a traditional CBT approach reported greater use of cognitive restructuring and relaxation.

Both CBT and ACT provide frameworks which appear suitable for the treatment of WAD, and therefore these models will be further examined and compared in the context of research findings associated with WAD populations. In 2012 Dunne and colleagues conducted a RCT which examined the use of CBT for the treatment of PTSD in the context of chronic whiplash. Treatment consisted of 10 weekly 1 hour sessions of trauma-focused CBT (TF-CBT) which was comprised of psychoeducation, anxiety management, cognitive restructuring, imaginal exposure, graded in vivo exposure and relapse prevention, whilst the control group were put on a wait-list. Results showed clinically significant findings in reduction of PTSD symptoms at both 10 weeks and at 6-month follow up, replicating findings see in previous studies focused on MVC survivors (Fecteau & Nicki, 1999; Blanchard et al., 2003; Maercker et al., 2006). Reduction of PTSD symptoms also correlated with improved neck disability, psychosocial distress and function and physiological reactivity to trauma cue, however there was little to no effect on sensory pain thresholds (Dunne et al., 2012). This study is valuable as unlike preceding studies which have focused on treatment of PTSD in MVC alone, this is the first RCT concerning the effects of CBT for the treatment of PTSD in treatment outcomes related to other comorbid conditions present in WAD (Dunne et al., 2012). The fact the study does not examine the benefits of CBT in WAD suffers who do not exhibit PTSD may be viewed as a negative, however this approach has demonstrated that simplifying treatment for a particular subgroup of sufferers by isolating one element of WAD within a specific CBT framework (ie. TF-CBT) can have a wider effect on other symptoms. This approach also aims to eliminate issues associated with flexibility in the CBT model as mentioned previously, aiding the clinical application of findings. The study used one researcher to deliver CBT for the purposes of consistency within the study, however this makes it difficult to predict whether another practitioner could replicate the results. The use of several clinicians with equal skill base and training would be preferred for future studies to eliminate this issue.

The second paper is a randomised controlled trial by Wicksell, Ahlqvist, Bring, Melin & Olsson (2008) which investigated whether acceptance and exposure strategies could improve function and life satisfaction in individuals with chronic WAD. The treatment timeline was fairly similar to Dunne et al. (2012), however treatment delivery was much more structured with patients receiving treatments which were delivered in a subsequent order of pain education, shifting perspective, exposure and acceptance and defusion. This highlights an advantage of using ACT model as it can be clearly linked to the theoretical framework it is based on (Scott & McCracken, 2015). Acceptance and defusion are recognised as elements of psychological flexibility that relate to open behaviours, which are strongly linked in improving fear avoidance, anxiety and catastrophizing (Hayes et al., 2006). It is disappointing that researchers did not focus on more elements of psychological flexibility as this has been a critique of the application of the model in the past, however in the context of WAD the selected approaches are appropriate to treatment outcomes for the condition (Scott & McCracken, 2015). Results demonstrated significant changes in treatment outcomes which included an increase in life satisfaction and psychological flexibility and a decrease in pain disability, fear of movement and depression at the end of treatment (8-weeks) and at 7-month follow-up. There were no changes in pain intensity in either groups. The lack change to pain intensity in both studies echo previous findings by Söderlund, & Lindberg (2001) which examined the use of CBT in WAD, suggesting that psychological interventions for WAD may be more suitable in a multimodal approach which includes complementary interventions that better target pain intensity. It is important to remember that unlike CBT, ACT does not aim to remove painful experiences but assist patients in their presence, suggesting Wicksell et al. (2008) produced results that were more congruent with the psychological model applied.

The largest issue with both articles is small sample sizes. The CBT study had a sample of size of 23 participants whilst the ACT study only had 20, making it impossible to generalise findings to WAD populations. Both studies recommend that future research endeavours to recruit larger samples in order to make generalisations and perform more sophisticated levels of analysis. Dunne et al. (2012) specifically mentioned aims to compare CBT to other treatment modalities to see whether findings would still be statistically significant whilst Wicksell et al. (2008) commented on focusing on determining particular subgroups of patients within WAD populations as a predictor for favourable treatment outcomes. There are several papers which compare ACT and CBT to each other from a theoretical standpoint and with consideration to a wide range of conditions, however it is suggested that further research examine how each therapy performs against each other in the treatment of WAD to determine which may be more suitable for this particular presentation. It is also suggested that further research implement ACT and CBT in the acute stages of whiplash in order to ascertain whether they play a role in minimising the chances of becoming chronic suffers.

Chronic WAD is a condition which is undoubtedly affected by psychosocial factors. Management of WAD should consider a multimodal treatment approach which includes psychological treatments aimed at depression, anxiety and post-traumatic stress disorder (PTSD), as well as maldapative behavioural patterns such as fear avoidance and catastrophising. CBT and ACT are viable psychological models which have demonstrated evidence of improvement in treatment outcomes for chronic WAD symptoms, however these findings are still limited. Further research should endeavour to further investigate these models of treatment to increase our understanding surrounding their use in WAD populations and improve prognosis for sufferers.

1.1 Justification

The 2016 Global Burden of Disease (GBD) study, concluded that painful chronic musculoskeletal conditions account for 20-33% of global disability (Vos et al., 2017). The GBD further identifies LBP as the single leading cause of global pain and disability, with persistent pain, limited mobility, and restricted dexterity and or functional ability commonly presenting (World Health Organisation [WHO], 2018). Consequently, families, workplaces, health care settings and communities are significantly impacted with a reduction in people’s ability to work, frequent withdrawal from social activities fostering mental health vulnerabilities, and high health care resource demand. CSP is complex populous, where the difficulty of linking a diagnosis to symptoms and a lack of accord amongst clinicians is rife. Consensus regarding clinical assessment, definitive pathoanatomical diagnosis, relevant treatment prescriptions and measurable patient outcomes across care disciplines is lacking and particularly evident in CSP where 90% of presenting LBP signs and symptoms can only be identified as ‘non-specific’ (Koes, Tulder, Lin, Macedo, Mcauley & Maher, 2010). Increasingly, the role of central sensitisation (CS) a condition characterised by brain plasticity that leads to hyper-excitability of the central nervous system, is deemed key (Hartvigsen et al., 2018). A potential combatant to CS may be PNE, a rapidly growing but relatively immature adjunct to usual care and that engages strategies to decrease the sensitivity of the central nervous system (Moseley, 2002; Moseley & Butler, 2015). Growing evidence suggests PNE may be efficacious in decreasing pain, disability, fear-avoidance (kinesiophobia), pain catastrophisation, limited movement, and health care utilisation in the CSP populous (Moseley & Butler, 2015). A limited number of RCT’s regarding the efficacy of PNE as a standalone and multi-modality treatment strategy have been conducted in CSP, lumbar radiculopathy, fibromyalgia, chronic fatigue syndrome, whiplash and general chronic pain (Gallagher, Mcauley & Moseley, 2013). This review aims to evaluate eight RCT’s examining the efficacy of PNE, in addition to usual care in improving pain levels, function and QOL in people with CSP.

1.2 Limitations to the Research

With over 28 self-report outcome measures used, with half being used only once, drawing comparisons was difficult. Additionally, methods including PNE alone, PNE+MT and PNE+MT+TE were trialed, thus comparing improvements on pain, QOL and function was challenging. Delivery of PNE varied in time length, format and content with two articles (Beltran-Alacreu et al., 2015; Louw et al., 2016), seemingly addressing cognitive or neurobiological explanations that were divergent to the research used to deliver PNE in the remaining articles. Furthermore, Louw et al., (2016) preferred to examine the effects of a neurobiological explanation to a manual therapy technique, rather than pain itself. Moreover, all research was conducted in higher income countries thus, the global relevance may be questioned in low socio-economic countries where healthcare and education is inadequate.

1.3 Resources Accessed / Inclusions and Exclusions

Throughout September and October 2018, The Think Education Library platform enabled a thorough search of several databases including PubMed, MeSh, MEDLINE and EBSCOhost. An additional search was performed using Google Scholar. Search methodology included a PICO chart (see Appendix A), with relevant keywords (see Appendix B). Assessed were RCT’s published between 2002-2018 with efficacious measures for pain, QOL and/or function that met the criteria of: adults aged 18-75 with persisting CSP longer than 2 months, LBP with or without leg pain, failed back surgery syndrome beyond 3 years, whiplash and associated disorders and CNP. Exclusions were widespread musculoskeletal pain syndromes (fibromyalgia and chronic fatigue syndrome) and other contraindications or psychiatric disorders, pending surgery of legal actions, unstable work conditions or pregnancy.

2. Overview of Chronic Spinal Pain

2.1 Chronic Spinal Pain

CSP is synonymous with high health care utilisation, pain and disability worldwide. Despite a plethora of research, a global consensus on the diagnosis and management of CSP is yet to be reached. CNP has a high prevalence and is common in clinical practice (Fejer, Kyvik & Hartvigsen, 2005), with LBP surpassing all non-communicable diseases as the greatest cause of disability worldwide (WHO, 2018). Consequently, favourable patient relations and outcomes are compromised. Whilst clinical practice guidelines for CSP finds consistent recommendations for TE, cognitive behavioural therapy (CBT) and multidisciplinary treatment with some discrepancies regarding MT and pharmacology (Koes et al., 2010; Puentedura & Flynn, 2016), there appears to be a lack of acknowledgement to recent advances chronic pain research where RCT’s demonstrate reduced pain levels, pain catastrophisation, decreased disability and improved physical performance (Clarke, Ryan and Martin, 2011; Louw, Diener, Butler & Puentedura, 2011).

The financial burden across communities is far reaching, particularly in peak income earning age groups where the incidence of work absenteeism is significant (Briggs et al., 2018). Comorbidities including anxiety and depression are frequently encountered (Harrison, Henderson, Miller & Britt, 2016; Dominick, Blyth & Nicholas, 2012) and whilst the impact on physical and mental components of quality of life are significant in multi-morbidity presentations, the effect is magnified when the presenting person suffers from musculoskeletal chronic pain (Briggs et al., 2015). Recent investigations have identified the role of CS as key to unravelling the complexities of chronic spinal pain with various studies agreeing that the clinical importance of CS in chronic spinal pain is significant (Coombes, Bisset & Vicenzino, 2012; Smart, Blake, Staines, Thacker & Doody, 2012; Nijs et al., 2015). To lessen the burden a clear need for an approach that extends beyond the biomedical approach of ‘what’s broken, let’s fix it” exists (Puentedura & Flynn 2016; Gore, Tai, Sadosky, Leslie, & Stacey, 2012; Nijs et al., 2015).

2.2 Pathophysiology

CSP is a symptom defined by location of pain. LBP commonly presents between lower rib margins and the buttock creases with or without leg pain, and where associated neurological symptoms may arise. CNP encompasses whiplash and associated disorders and is accompanied by significant levels of pain and disability with a high correlation to fear of movement (Kregel et al., 2015). Concurrent pain in other body sites and more general physical and mental health problems are frequent (Hartvigsen et al., 2018) Combined co-morbidities of CSP and diverse problems encompassing psychological, social and biophysical factors as well as pain-processing mechanisms impact on both pain experience and the associated disability. Hyper-excitability of the central nervous system, including descending nociceptive inhibition is implicated in the ongoing maintenance of CSP (Roussel, Nijs, Meeus, Mylius, Fayt & Oostendorp, 2013), and demonstrates notable changes in gray matter volume in regions of the brain involving modulatory, emotional-affective, and sensory-discriminative pain processing (Malfliet et al., 2018; Kregel et al., 2015). CSP sufferers often present with maladaptive pain cognitions such as kinesiophobia, pain catastrophising and hypervigilance (Crombez, Vlaeyen, Heuts & Lysens, 1999; Picavet, Vlaeyen & Shouten, 2002). Hence, the pathophysiology of CSP is multifaceted with consideration to modern pain research a necessity, in the pursuit of improved patient outcomes.

2.3 Causes

Importantly a small percentage of CSP can be attributed to serious red flag pathologies such as malignancy and cauda equine syndrome, however, these cases are rare. More commonly, awkward postures, manual tasks, fatigue, lifting, bending and other physically demanding task have been associated with increased risk and/or new episodes of low back pain, particularly in low-to-middle income countries. However, there is little evidence to suggest a causal link can be drawn. CSP is considered to have multifactorial contributors (Hartvigsen et al., 2018) and whilst many clinicians address CSP with a mechanical, non-mechanical or non-specific diagnosis, this does not address the biopsychosocial model commonly preferred to the biomedical approach. Biophysical, psychological, social and genetic influences, and additional comorbidities contribute interactively. Consequently, persistent CSP does not result simply from nociceptive input (Hartvigsen et al., 2018). However, there are demonstrable biophysical factors in this populous including discrepancies in muscle composition and size (Goubert et al., 2014), and aspects of coordination (Dubois, Abboud, St-Pierre, Piche & Descarreaux, 2014), when compared with those who do not suffer from CSP. Additionally, the overlapping psychological influences of catastrophising, depression, anxiety and a lack of self-efficacy though their mechanism in CSP is not fully understood (Campbell et al., 2013). Potentially, the fear-avoidance model of chronic pain suitably explains maladaptive learning processes and disabling beliefs (avoidance thus establishing disability) which implies that pain cognitions play a major role in disability maintenance and development, more so than the pain itself (Crombez, Eccleston, Van Damme, Vlaeyen & Karoly, 2012). Jackson, Wang, Wang & Fan (2014), determined that self-efficacy across chronic pain conditions, is consistently associated with affective distress, pain severity, impairment and disability. Consequently, the global socio-economic burden of CSP is at great risk of increasing as most CSP cases are categorised as having no specific cause.

2.4 Treatment

The management of CSP is well evidenced and supported amongst clinicians with a combination of MT, TE and pharmacological interventions. Additionally, psychological interventions such as CBT, acceptance and commitment therapy, mindfulness based stress reduction and pacing are common in multimodal care. PNE challenges previous biomedical approaches that engender anatomical education, biomechanics and pathoanatomical causes, by presenting a neurobiological model that enables patient understanding of the components of pain experiences including central and peripheral sensitisation, and neuroplasticity (Moseley & Butler, 2015). The case for PNE is potentially strengthened by Louw et al.,’s (2017) view that biomedical models demonstrate “limited efficacy in helping people understand chronic pain, and may in fact even increase a person’s pain experience by increasing fear-avoidance and pain catastrophisation”. Divergence exists between PNE and manual therapy, where potential lack of practitioner understanding and education has facilitated clinicians attempting to engage PNE as a ‘hands-off’ approach only where patients may be categorised to receive PNE (with no hands-on), or hands-on with no PNE (Louw et al., 2017). Perhaps the notion of PNE and manual therapy co-existing requires more robust exploration. Well used, PNE as an intervention has the potential to place clinicians as leading practitioners in the musculoskeletal field.

3. Research Findings

3.1 Interventions

Each of the trials used some form of PNE as their primary intervention, with the addition of up to two other interventions or comparators of usual care that included MT, TE, aquatic exercise, general practitioner based care or biomechanical education. The format, delivery and length of PNE intervention varied between studies however, PNE interventions were all physiotherapist delivered, and varied from a 5-minute neuroplasticity explanation using a picture of the homunculus (somatosensory cortex) (Louw et al., 2016), to four, 60-minute PNE sessions delivered weekly over 4 weeks (Moseley, 2002). Formats included groups of 4-6, 1 on 1, online and booklets form with an average of 2 sessions. Heterogeneous PNE content included covering cognitive, operant and respondent aspects (Beltran et al., 2015), pain pathway, fear-avoidance beliefs and coping (Chaleat-Valayer et al., 2016), to reconceptualising pain (Malfliet et al., 2017; Malfliet et al., 2018) and targeted PNE (Moseley, 2002; Malfliet et al., 2017). Pires et al., (2014), contrasted by covering the transitions of acute to chronic pain, central sensitisation, the brains role in pain perception, psychological factors, cognitive and behavioural responses to pain, flare-up management and pacing.

Interestingly, six of the articles indicated PNE content was based on Butler, Moseley and Sunyata’s Explain Pain (2013), or content from the international research group, Pain in Motion. The exception was Moseley (2002) who, as co-author of Explain Pain (Butler, Moseley & Sunyata, 2013), was effectively challenging his own unpublished trials. In the authors view, this is suggestive of growing consensus on both the efficacy and understanding of PNE and its benefits. Louw et al., (2016) contrasted as their intervention was a neurobiological explanation of a manual therapy technique, rather than of pain itself. However, critically, there was no consistency indicating level of knowledge of the physiotherapist delivering PNE to CSP sufferers, or how their level of knowledge on the subject, thus outcomes may have been implicated.

PNE was combined with adjunct therapies including TE (Moseley, 2002; Chaleat-Valayer et al., 2016; Malfliet et al., 2017; Beltran-Alacreu et al., 2015; Pardo et al., 2017; Pires et al., 2014), an accepted intervention in many muscuoskeletal conditions. Not assessed was the content of each TE program as this was deemed by the author as being less relevant. This study differed from the other 7 in that the content of the dual interventions both changed. A single study selected used aquatic exercise (Pires et al., 2014). Additionally, Betran-Alacreu et al., (2015), Chaleat-Valayer et al., (2016) and Moseley (2002) supplemented PNE+TE with a course of MT delivered by a physiotherapist. Louw et al., (2016) contrasted MT+PNE only with only one study, Malfliet et al., (2017) not engaging any form of MT or TE preferring to solely to compare 3 sessions of PNE against bio-medically focused back/neck school addressing spinal anatomy and physiology.

3.2. Efficacy of Outcome Measures

Due to the heterogeneous nature of outcome measures recruited, determining the influence of PNE on CSP was difficult, with 28 measures spanning the articles (See Appendix D). All but one article (Malfliet et al., 2017), used either the numeric rating scale (NRS) (Pardo et al., 2017; Malfliet et al., 2018; Louw et al., 2016; Moseley, 2002) or visual analogue (VAS) (Pires et al., 2014; Chaleat-Valayer et al., 2016) as primary outcome measure for pain. Both tools are well established self-report tools for measuring pain intensity with the both the NRS and VAS showing strongly evidenced reliability, validity and retest reliability when compared to other tools (Cruz, Fernandes & Carnide, 2013). Additionally, Beltran-Alacreau et al., (2015) used the well supported, and most commonly reported Neck Disability Index (NDI) (Vernon, 2008), as the primary outcome determinant of neck pain creating a variance in any possible comparisons to be drawn. Malfliet et al., (2017) was somewhat disabling to pain measure comparisons as their study examined the efficacy of blended-learning PNE on improving disability, catastrophizing, kinesiophobia and illness perceptions, rather than pain intensity thus, function, quality of life and psychometric measures were appropriately engaged. They did however assess the impact of pain on daily life activities using the Pain Disability Index (PDI), a valid tool with good internal consistency and good re-test reliability (Soer, Reneman, Vroomen, Stegeman & Coppes, 1976).

Function and quality of life (QOL) suffered greater variance and often overlapped in interpretation in that some outcome measures were used to measure both outcomes in different studies making a direct comparison somewhat problematic. However, their efficacy could all be well validated. The Roland Morris Disability Questionnaire addressing pain related disability and QOL was used in two trials (Pardo et al., 2017; Moseley, 2002) and is well researched previously (Roland & Fairbank, 2000) and additionally demonstrates good reliability when administered over the phone (Cherkin, Deyo, Battie, Street & Barlow, 1998). The Quebec Back Pain Disability Scale (QBPDS), engaged by both Pires et al., (2014) and Chaleat-Valayer et al., (2016) and is well researched for functional disability (Cruz, Fernandes, Carnide Vieira & Nunes, 2013). Additionally, the Pain Disability Index (PDI) used in two trials (Malfliet et al., 2018; Malfliet et al., 2017) documents both validity and re-test reliability (Soer et al., 1976) for function, as does the Short Form Health Survey (SF36) engaged by Malfliet et al., 2018, and confirmed as a relevant QOL measure previously (Gandek, Ware, Aaronson, Apolone, Bjorner & Brazier et al., 1998), Chaleat-Valayer et al., implemented the SF-12 Health Survey (SF-12).

Significantly in the field of PNE, across the articles numerous self-report psychometric measures assessing fear avoidance, catastrophizing, pain related anxiety and CS were repeatedly used indicating the inter-related issue of thoughts, feelings and behaviours in the pathophysiology of LBP was deemed relevant. Most commonly, the Tampa Scale of Kinesiophobia (TSK) a tool demonstrating good internal consistency in LBP populations (Cordeiro, Pezarat-Correia, Gil & Cabri, 2013), to assess of Fear of Movement (Pardo et al., 2017; Beltran-Alacreu et al., 2015; Pires et al., 2014; Malfliet et al., 2017) was used. However, given the growing body of evidence implicating the role of CS in CSP, and PNE’s validity as a combatant, it is perhaps relevant to question why the CSI was not used as a relevant outcome measure across the articles given its availability from 2012. Similarly lack of inclusion of psychometric measures should be noted as a possible failing of those articles not addressing this component of CSP, a perplexing circumstance given the notoriety in the field of chronic pain of the authors of the studies.

3.3. Treatment Effects / Outcomes

3.3.1 Pain

Across the articles the greatest effects were found in improvement in pain intensity, however debate exists, perhaps due to the variance of interventions in this review. Whilst previous findings that used PNE alone (Moseley, Nicholas & Hodges, 2004), propose PNE improves pain intensity, Malfliet et al., (2017), argue that PNE compared to a biomedical educational approach demonstrates some improvement in pain intensity, effect sizes are not large enough to warrant PNE as a stand-a-lone intervention or treatment. Significantly, effect sizes were greater when PNE was delivered in addition to TE.  By delivering 2 group, face to face PNE sessions prior to administering a daily home exercise program, Pardo et al., (2017) determined that whilst lumbar pain intensity decreased in both the experimental and control groups, a greater improvement was found in the PNE+TE group at all time points to a large effect size. Malfliet et al., 2018, concurred that PNE+TE improved pain intensity though their methodology differed in that PNE was delivered in group, online and face to face formats and, in-line with previous evidence (Jack, McClean, Moffett & Gardiner, 2010) confirming face to face TE is more efficacious in improving pain in LBP patients than home exercise programs.

Curiously, early in the PNE’s development, Moseley (2002), pioneered that by adding PNE+TE+MT improved pain intensity to good effect at the longest follow-up period of 12 months. Given Malfliet et al., (2017) conclusion that a stand-a-lone intervention of 5 minutes of PNE only may not improve outcomes, studies comparing stand-a-lone PNE over longer, and more varied delivery formats may be relevant. Significantly, Pardo et al., engaged in fewer and or shorter PNE education sessions than Malfliet et al., 2018, Louw et al., 2014, Malfliet et al., 2017 and Moseley, 2004 with similar effect on pain intensity. Beltran-Alacreu et al., 2015 also found shorter, less frequent sessions agreeable, however, excluding Malfliet et al., 2017, each of the other 7 studies engaged more than one intervention.

What also appears relevant is the order in which PNE is delivered in multi-modal interventions with all eight articles preferring to deliver at least one session of PNE prior to any other intervention. Significant changes in CSI scores in two studies (Malfliet et al., 2018; Malfliet et al., 2017) could argue that clinically interpreted alongside clinically significant changes in PPTs (Pardo et al., 2017; Malfliet et al., 2018) and NRS (Pardo et al., 2017; Malfliet et al., 2018; Louw et al., 2016; Moseley, 2002), PNE demonstrates superiority, particularly when combined with MT and/or TE when compared to current best-evidence primary care models in reducing pain intensity in CSP.

These positive results however, need to be considered in context of the small sample sizes included across the studies, and the apparent lack of consideration in measuring signs and symptoms of CS in the context of CSP. Whilst 1 year positive outcomes with regards to pain are reported in CSP and various widespread chronic conditions (Malfliet et al., 2018; Moseley, 2002, Louw et al., 2014, Vibe Fersum, O’Sullivan, Skouen, Smith & Kvale, 2013), and allowing consideration of modern chronic pain research inclusive of but not limited to neuroplasticity and CS, it seems illogical that using pain rating scales alone in the measurement of pain is efficacious. It could be argued that trials that used psychometric measures alongside pain rating scales are of more relevance in chronic pain where behavior change is key to changes in health status (Louw, Zimney, Puentedura & Diener, 2016).

3.3.2. Function and Quality of Life

A consequential link from reduction of pain intensity may be drawn to function and/or QOL in that a person suffering from CSP has a greater understanding of pain and is subsequently less fearful, enabling them to feel empowered and move regularly thus improving both function and/or QOL. However, this is a relatively weakly supported conclusion as outcome measures showed only small effect sizes across the studies. Significantly, Pires et al., (2014) determined participants in their education group were more likely to report ‘perceived’ functional improvements at 3-months follow-up, no statistically significant changes were recorded. This finding was consistent across numerous studies (Malfliet et al., 2018; Beltran-Alacreu et al., 2015; Pardo et al., 2017; Moseley, 2002) however, due to the variety of intervention combinations applied, it’s difficult to say whether PNE, MT, TE or a combination of all three determines the level of reported change. Louw et al., (2014) argue there were no significant interactions for LBP, leg pain or trunk flexion but found significant improvement in straight leg raise (SLR) testing with the PNE group being 7.2 times more likely to improve beyond minimal detectable change of 4.5cm than the mechanical education group. Chaleat-Valayer et al., (2016), did not prove PNE+TE decreased LBP recurrence however, the intervention was effective in altering fear-avoidance measures and health care utilisation which again is suggestive of QOL improvement.

Two trials (Moseley, 2002; Malfliet et al., 2018) concluded improvement in function and/or QOL measures (RMDQ, PVAQ, IP-QR) at all timeframes including the longest of 12 months. Of notable significance is those trials that reported improvements in function and/or QOL, additionally used at least one psychometric measure (FABQ, TSK, PCS) and demonstrated moderate to large effect sizes at all time points. Whilst the exact reason for improved outcomes in PNE education groups is unclear, previous studies (Nijs & Houdenhove, 2009) have found correlations between a better understanding of pain not only contributes to positive changes in maladaptive pain cognitions, but also improves patient perceptions of their ability to control and manage pain. Targeting the fear-avoidance cycle and empowering patients to re-enter functional, social and daily life activities subsequently improves QOL. Whilst this notion requires more research, it is gathering momentum and has been the subject of numerous RCT’s across other chronic pain conditions (Moseley & Butler, 2015), with promising results. Arguably however, the success of PNE as an intervention remains in the quality of understanding of the clinician delivering the message which further studies should examine.

Another obvious flaw regarding the efficacy of PNE on function and/or quality of life are the lack of more long-term studies. To date, only one study (Louw et al., 2016) measured outcomes beyond 3 years however the focus of that study involved pre-operative PNE on lumbar radiculopathy surgery patients. None of the studies included in this review had follow-up periods beyond 1 year, which given behavioural change appears to be the cornerstone of PNE’s potential success, would be a limitation with regards to the validity of PNE on QOL as behavioural change requires significant time investment on the part of the sufferer. Importantly, conclusions with regards to function and/or QOL could only be loosely drawn as clinically relevant statistics were heterogeneous in outcome measures and conclusions drawn. Some bias also needs to be considered, as those studies that did find relevance were authored by prominent authors in the field of PNE.

4. Conclusion

In questioning the efficacy of PNE, in addition to usual care, in improving pain, function and quality of life in people with CSP numerous conclusions can be drawn. Whilst all articles reviewed determined that PNE has a positive effect on pain intensity in CSP in short-medium time-frames, none of the trials could be compared as like for like due to the heterogeneous nature of interventions, quality and variety of PNE content and, a very broad range of outcome measures engaged. Additionally, only small effects were noted overall with regards to function and/or QOL which, it could be suggested may have measured differently if a more logical inclusion of psychometric measures, in line with modern chronic pain research, were utilised. Whilst this is debatable, it is seemingly neglectful not to consider the psychosocial element more completely by utilising concurrent types of outcome measures.

Clear consensus proposing multi-modal delivery of PNE in addition to usual care as being superior, was noted amongst the articles and across the broader research community. This included the notion that PNE as a stand-a-lone intervention may not be efficacious in improving any of the outcomes questioned. For PNE to be accepted, definitive criteria regarding content and subsequent clinician education regarding PNE vs biomedical explanations requires attention. Problematically, a lack of clinician understanding may be hampering the advancement of PNE which ultimately impacts detrimentally on the global burden of CPS and it’s far reaching socioeconomic harms.

Future trials with larger sample sizes, in broader socioeconomic environments, assessing for changes in pain behaviours resulting from PNE, in addition to this trials outcomes, are strongly recommended as the slow turning ship of biomedical explanations vs the biopsychosocial model is dependent on progressively improving patient outcomes. Additionally, a standardised format of outcome measures that address noted CSP contributors, including CS may enable greater acceptance of PNE as a useful tool in reducing the incidence of CSP globally.

Neck and Arm Pain - Part 1

A 44 year old single mother of 3 teenage boys presents with acute neck and right arm pain for the past 2 days. She states she has had neck and upper back pain on and off for past 5 years, progressively worsening but no previous history of arm symptoms.

Currently she is experiencing acute neck pain radiating down the right arm and into the right wrist and thumb with some shoulder and mid scapula pain.

On examination she exhibits limitation and pain in cervical rotation and right lateral flexion with guarding and muscle spasm, she also experiences acute pain on right shoulder elevation.

There also appears to be slightly diminished reflexes in the right triceps and brachioradialis compared to the left.

Description & Differential Diagnosis

Neck pain is a leading cause of disability with significant physical, psychological and socioeconomic impacts.1,2 Epidemiological studies have reported an average prevalence rate of 37.2% which is known to increase with age.1,3 Pain experienced in the cervical region and upper limb can be the result of multi-factorial aetiology.4,5 Appropriate identification of the source of neck pain is assisted by identifying the underlying mechanisms behind the pain.1 Pain may be of a mechanical or neuropathic origin, or secondary to more serious conditions such as visceral disorders and illness.1 Accurate diagnosis of the source of neck pain is essential to forming decisions about how to effectively treat patients.1,5

The patient's chief complaint is acute neck pain with unilateral radiation into the right arm, wrist and thumb. Associated symptoms include pins and needles in the right thumb and forefinger and mild upper back pain localised to the medial border of the right scapula. The presence of unilateral radiating symptoms is characteristic of cervical radiculopathy (CR), a form of peripheral neuropathic pain often a result of disc pathology, osteophyte formation and or spinal stenosis.1,6 Loss of sensation to touch and cold are common findings in CR and often experienced in a dermatomal distribution.7 The patient's symptoms in the medial scapula region and first two digits correlate to the C6 and C7 dermatomes, known to be the most commonly affected nerve roots in cervical radiculopathy.1,6

The patient's history reveals considerable overlap with the known epidemiology of CR. 7 Epidemiological studies have found greater incidences of CR in the fourth and fifth decade of life with a low incidence rate of traumatic mechanisms of injury.7 Furthermore, previous episodes of progressively worsening neck and upper back pain is suggestive of an insidious onset which is common amongst older cervical radiculopathy patients.8

The patient’s postural presentation correlates closely with that of a classic CR patient. Tilting of the head away from the affected side and cradling the affected arm are typical protective postural behaviours as an attempt to reduce tension on the affected nerve root.9 An increased thoracic kyphosis is a common postural observation to be expected along with cervical muscle spasms.8

On physical examination, the observed diminished reflexes in the right triceps and brachioradialis lend further support to the involvement of the C6 and C7 nerve roots.8 Asymmetrical diminished reflexes show an 82% correlation to cervical radiculopathy.7 Assessment of the patients cervical range of movement reproduces symptoms and is limited in cervical flexion, right lateral flexion and rotation. Whilst cervical range of movement limitations are common in all types of neck pain,1reproducible arm pain in ipsilateral flexion is a common feature of cervical radiculopathy due to the narrowing of the neural foramen.10

Whilst a provisional diagnosis of cervical radiculopathy looks likely, it is necessary first to investigate any visceral disorders and overlapping conditions.11 Whilst there are clear peripheral neuropathic symptoms, it is important to also identify symptoms of central neuropathic pain stemming from spinal cord pathology.1Cervical myelopathy is a form of central neuropathic pain resulting from compression of the spinal cord.12 In this case, the patient is lacking the classic myelopathy upper motor neuron signs such as ataxic gait, hyperreflexia and bowel problems.12 Another more serious condition that can be potentially missed is Parsonage-Turner Syndrome (PTS).13 The patient presentation of atraumatic acute pain in the shoulder, neck and upper arm and decreased reflexes are characteristic of PTS.13 The onset of symptoms often follows viral infection, which is consistent with the current patient's medical history.14 Monitoring for this condition over time is necessary as decreasing pain and worsening muscle weakness make the PTS clinical presentation clearer and easier to differentiate.13

Given that thoracic outlet syndrome (TOS) typically presents as unilateral pain and is most prevalent in women aged in their 40's, it is necessary to differentiate this condition.1 Compression of the brachial plexus due to chronic muscle overuse can affect the C6 and C7 nerve roots and bring about symptoms similar to those experienced in this case.15 However, it is more common for TOS symptoms to have a wider anatomic distribution rather than restriction to dermatome patterns.16

Myofascial pain originating from the scalene muscles is a common cause of parasthesia and pain seen in TOS.17 Indeed, myofascial pain referral often mimic dermatomal patterns of spinal nerve roots.18 Specifically, myofascial pain originating from the scalenes simulates the C6 dermatomal radicular pain as seen in the current patient.17 It is possible that this patient's symptoms have been brought on by chronic over use of postural muscles.17 Whilst there are no palpatory findings in this case, the presence of tenderness or reproduction of symptoms upon palpation of the scalenes would be suggestive of concomitant myofascial pain.18

Cervical sprain and strain is another musculoskeletal disorder with overlapping presentations to the current patient.19 Restricted and painful cervical ROM and pain radiating into the interscapular region are common findings.19 However the absence of any notable trauma and the sensory deficits found on neurological examination are more suggestive of central pathologies.19

It is common for patient's to experience a "double-crush" phenomena whereby neural compression exists at both the cervical spine and peripheral nerve.6 Indeed, high concomitant occurrences of carpal tunnel syndrome and cervical radiculopathy have been observed amongst patients with cervical radiculopathy symptoms.20It is therefore necessary to consider the presence of peripheral nerve entrapments in the current patient. The patient's occupation, age and gender accompanied by the presence of paraesthesia in the median nerve distribution are indicative of carpal tunnel syndrome.21 Further diagnostic investigations are required to delineate these two conditions and/or confirm the presence of both.6