Nociceptive, peripheral neuropathic, central sensitivity – is it all Greek to us?

Coffee and biscuitFor those of you who have done one, you will know that finishing your PhD can be a bit like sailing in front of Wild Oats in the Sydney to Hobart – the flapping spinnaker, full of midnight oil and editorial mutterings, left limp as a soggy biscuit. For those of you who have not done a PhD, you might not get it. Well, here at BiM we are keen to keep the doctoral ginger nut from dropping into the tea cup of discontent by giving new graduates the chance to tell us about their work. This one is from the aptly named Keith Smart, from University College Dublin.  It is an important series of studies because it takes a proper stab at integrating modern pain science with manual therapy practice.  That is, Keith and his supervisors have said something along these lines – ‘if the physiologists argue that tissue damage, peripheral neuropathic changes and central sensitisation can underpin pain, then let’s see if clinicians can tell when they do’.  However, they haven’t just gone there, they have done the background studies too. I won’t do it justice, so have a quick read.

The development and preliminary validation of mechanisms-based classifications of musculoskeletal pain


‘Nociceptive pain’ (NP), ‘peripheral neuropathic pain’ (PNP) and ‘central (sensitisation) pain’ (CP) have been suggested as mechanisms-based classifications of pain. Some think that classifying pain in this way might i) help clinicians to better understand clinical presentations of pain and ii) improve outcomes by encouraging clinicians to direct treatment towards the underlying neurophysiological mechanisms responsible for the generation and/or maintenance of patients’ pain. But before their use in clinical practice can be recommended classification systems should be developed and validated using appropriate methodologies. The purpose of this research project was to develop mechanisms-based classifications of musculoskeletal pain and then evaluate their construct validity.

Studies and results

In the first study we used a Delphi survey method in order to generate lists of clinical criteria thought to reflect a dominance of NP, PNP and CP mechanisms. In the second study we investigated the reliability of clinical judgements and criteria associated with these mechanisms-based classifications of pain. And in the third study we investigated the discriminative validity of these categorisations by looking at whether or not patients with low back (± leg) pain classified in this way could be differentiated from one another on the basis of discriminatory clusters of symptoms and signs. The findings from the Delphi survey generated lists of clinical criteria associated with each category of pain, the findings from the reliability study suggested that such classifications might be performed reliably, and findings from the validity study suggested that NP, PNP and CP could be accurately predicted from a cluster of 7, 3 and 4 symptoms and signs respectively. We also found that patients classified with a dominance of CP reported more severe pain, poorer health-related quality of life, and greater levels of functional disability, depression and anxiety compared to those with PNP and NP. A similar pattern of findings was found for those with PNP compared to NP.

Implications for practice

Together these findings provide some evidence that clinicians might be able to distinguish between patients with an assumed dominance of NP, PNP and CP by identifying a small number of symptoms and signs. Also, the differences in health parameters between categories might reflect clinically meaningful differences associated with the multidimensionality of pain.

About Keith Smart

Keith Smart Dublin

After escaping from his original home town of Bromley in the UK Keith went to University College Dublin (and bought the t-shirt) in his adopted hometown of err…. Dublin, to do his PhD thesis. He fitted in to life in Ireland well and some have noted a tendency towards Irish-sounding colloquialisms creeping in to his day to day discourse. Since finishing up earlier on this year, he has now returned to full-time clinical practice at St Vincent’s University Hospital, Dublin. Keith has attacked this research question with the same rigour and energy that he applies to his need to catch every single Manchester United European Cup match (wherever he is in the world), his love of dark chocolate digestive biscuits and his unequivocal (yet non-violent) approach to managing Kuala Lumparan pickpockets. So it was always going to be a quality job. Clearly he didn’t write this bio…


Smart KM, Blake C, Staines A, Doody C (2010). The reliability of clinical judgments and criteria associated with mechanisms-based classifications of pain in patients with low back pain disorders: a preliminary reliability study Journal of Manual and Manipulative Therapy, 18, 102-10 : 10.1179/106698110X12640740712897

Smart KM, Blake C, Staines A, & Doody C (2010). Clinical indicators of ‘nociceptive’, ‘peripheral neuropathic’ and ‘central’ mechanisms of musculoskeletal pain. A Delphi survey of expert clinicians. Manual therapy, 15 (1), 80-7 PMID: 19679504

Smart, K., O’Connell, N., & Doody, C. (2008). Towards a mechanisms-based classification of pain in musculoskeletal physiotherapy? Physical Therapy Reviews, 13 (1), 1-10 DOI: 10.1179/174328808X251984

All blog posts should be attributed to their author, not to BodyInMind. That is, BodyInMind wants authors to say what they really think, not what they think BodyInMind thinks they should think. Think about that!


  1. Hi guys I’ve putt you both ‘in touch by email so Keith can send thu the paper

  2. Melissa says:

    I’m just trying to locate this article to read & it doesn’t seem to be available in 2010 electronic version, even though we have access through the present day. Any thoughts on how I could locate it?

    Smart KM, Blake C, Staines A, Doody C (2010). The reliability of clinical judgments and criteria associated with mechanisms-based classifications of pain in patients with low back pain disorders: a preliminary reliability study Journal of Manual and Manipulative Therapy, 18, 102-10 : 10.1179/106698110X12640740712897

    Anonymous Reply:

    Hi Melissa.
    A copy of that paper should be on its way to your inbox anytime now!
    All the best

    Melissa Reply:

    Keith, nothing came through can you please resend?

    Anonymous Reply:

    Hi Melissa, what’s your email add and i’ll send it on directly.

  3. Donna Carrillo Lopez says:

    I am at Tufts in a Pain Policy Program here in the US and am doing a paper on non-specific low back pain and the issue of change that includes the gradual adoption by physicians of evidence based medicine and even the acceptance of the biopsychosocial model that one of our academic physicians called another form of reductionism. Anyway, I look with great interest at the cortical issues involved with NSLBP and I believe there is literature from neurology detailing changes in the cortex.

    However, as you probably note, we are still early in the development of EBM with RCTs and meta-analysis (that need to evolve in order to create valid, consistent meaning for physicians).

    I still maintain that the advance of genomics in medicine and the identification of gene subtypes may be a focus for future therapies and healing interventions.

    Does anyone focus here on CRPS and I wonder if gene line therapy might be a future for this dreadful condition.

    Keith Smart Reply:

    Hi Donna.
    Thanks for your post.
    I agree with you that research into the genetics underlying could help us a lot in our understanding of the prediposition to/development of chronic pain conditions. Environmental factors, which i think are just as important, but that much harder to study, also influence pain processing and outputs hugely.
    I haven’t focused on CRPS in any of my own research so i can’t help on that score i’m afraid, sorry. But i look forward to details emerging.

  4. Colette Ridehalgh says:

    Aplogies Neil and Keith, I seem to have somehow sent the message without finishing it! I was going to say I know that Herta Flor is fairly prolific in this area. Thanks for your time and thoughts, Colette

  5. Colette Ridehalgh says:

    Dear Neil and Keith, thank you for your responses. I will look at those papers that you recommended Keith. I am not a keen advocate of clinical prediction rules at this point, maybe it is the anarchist in me, but the word rules makes my blood run cold! The accumulation of the literature together with sound clinical reasoning seems to make more sense to me, otherwise we become mere technicians. Anyway we have moved off the topic area somewhat and I have a lot of reading to do. From a novice’s perspective in the literature around fMRI and low back pain are there any particularly useful ones that you would recommend Neil? I

    Neil O'Connell Reply:

    We have recently summarised the literature here:

    Wand BM, Parkitny L, O’Connell NE, Luomajoki H, Thacker M, Moseley GL. Cortical changes in chronic low back pain: Current state of the art and implications for clinical practice. Manual Therapy Article in press

    It is available online…

  6. Colette Ridehalgh says:

    Yes very helpful Lorimer, I might need to pick your brain (so to speak) too as I am due to have a meeting in a couple of weeks with a Prof of medical imaging with the plan to incorporate fMRI into my study- granted a small pilot study. I have no idea where to start with this, but I am conscious of the fact that some are suggesting that all patients with cLBP are showing changes within the brain, but not much suggesting that there may be differences between cLBP with different pathology/presentation etc etc. I may be missing a crucial bit of literature on this, but certainly Wand et al’s suggestion that subclassification is a waste of time because of the changes in the brain made me feel uncomfortable to say the least. Are we at the stage where our knowledge of these changes in the brain are so advanced that we can rule out physical subclassification systems? Sorry not feeling too eloquent this morning- woken 3 times in the night by screaming child! Any thoughts (not about screaming child, although they would be appreciated too!), Colette

    Neil O'Connell Reply:

    Hi Collette,

    In our paper on subgrouping we didn’t suggest that the evidence of brain changes is enough to rule out physical classification systems. It wasn’t when we wrote that and it still isn’t now. The findings are interesting since they correlate with clinical signs but they could be epihenomena. They do represent a new model to explore in back pain.

    We suggested subgrouping might be a false goal based on the evidence from RCTs and Systematic reviews which, in our view does not point to the suggestion that the reasons that our treatments for non-specific chronic back pain perform so indifferently is due to a lack of subgrouping in clinical trials.

    Addtionally it is interesting that so many of the logical precitions of a mechanical model of back pain are just not reflected in the quality literature. But going back to your main point – we certainly don’t have nearly we enough data regarding brain changes to draw strong conclusions.

    Keith Smart Reply:

    Hi Colette, Neil.

    The classification of low back pain or musculoskeletal disorders in general can be a tricky issue. As Neil rightly points out, studies to date do not provide enough empirical proof that many of the proposed sub-grouping systems for low back pain actually ‘exist’.

    I think it’s likely that sub-groups based on assumed biomechanical/structural dysfunction are attractive to physios because they ‘fit’ with the biomechanical paradigms that so often underpin our teaching, often appear to make sense with what we (think we!) see in the clinic and provide a ready justification for our intervention.

    Methodologically there are a number of ways to approach the issue of classification development and validation. Ford et al (2007) summarises these nicely. My interpretation of this paper is that if it can be demonstrated that the categories within a classification system:
    1) can be distinguished from one another on the basis of symptoms and signs (or other clinical investigations) (discriminative validity) and represent clinically meaningful differences across related health parameters (discriminant validity).
    2) can predict some future outcome (i.e. in response to natural history or treatment) (predictive validity) then they may have some merit.

    Testing these classifications or ‘clinical prediction rules’ in a RCT however is probabbly the ultimate test of a classification system designed to direct treatment (Stanton et al 2010).

    Another approach, where ‘classification’ is considered synomomous with ‘diagnosis’, is to develop classification systems and categories as if they were medical diagnoses. Studies like this could/should be designed using suitable methodologies, as outlined in the STARD guidelines.

    All in all, there’s more than one way to skin the classification cat. And often a combination of methodologies are required to provide the empirical evidence necessary to justify the use of a classification system in clinical practice.

    Bossuyt et al (2003). The STARD statement for reporting studies of diagnostic accuracy: explantion and elaboration. Clinical Chemistry 49: 7-18

    Ford J et al. (2007). Classification systems for low back pain: a review of the methodology for development and validation. Physical Therapy Reviews 12: 33-42.

    Stanton TR et al. (2010). Critical appraisal of clinical prediction rules that aim to optimize treatment selection for musculoskeletal conditions. Physical Therapy (in press).

  7. Now THAT is what we are after. Thanks Keith/Colette.

  8. Colette Ridehalgh says:

    Hi Keith, can’t believe I have just stumbled on this. I am doing my PhD at the moment looking at the responses to neurodynamic mobilisations in patients with referred leg pain. I have been struggling with the definitions, and how to attempt to sub-group. I am wondering if you have had any experience in looking at more objective signs of central sensitisation? A recent paper by Jensen et al. suggested that if patients with CLBP had more than 8 tender points, they are likely to have CS, asnd I was going to use this in my present study. Looking at your paper might also help me to identify patients with CS from their subjective findings.

    Keith Smart Reply:

    Hi Colette.
    In the discriminative validity study that we did, three ‘symptoms’ and one ‘sign’ emerged as predictive of a clinical classification of ‘central pain’ (sensitivity 91.8%, specificity 97.7%), these were:
    •Disproportionate, non-mechanical, unpredictable pattern of pain provocation in response to multiple/non-specific aggravating/easing factors.
    •Pain disproportionate to the nature and extent of injury or pathology.
    •Strong association with maladaptive psychosocial factors (e.g. negative emotions, poor self-efficacy, maladaptive beliefs and pain behaviours, altered family/work/social life, medical conflict).

    •Diffuse/non-anatomic areas of pain/tenderness on palpation.

    So the ‘sign’ listed above would seem to match that in the Jensen et al. paper you mention. (Which paper is that by the way?)

    It may be that diagnostic/classification criteria associated with ‘central (sensitisation) pain’ become more clearly identified over the next few years. Our discriminative validity paper is currently under review so we hope to publish more details on these criteria in due course. Hope that helps Colette? But please stay in touch if you’d like any more information.

    Colette Ridehalgh Reply:

    Hi Keith, the Jensen et al. paper was published this year in European Journal of Pain 14 :514–522. Thanks for response, really very helpful! Will look out for your latest paper when it is published.

  9. Thanks, Keith,

    Your explanation does help a lot to really define what you studied specifically. Especially your information about how the definitions of different types of pain have changed recently. I appreciate your citation for the proposed new definitions as well. It helps others who may be conducting similar research and those of us that are consumers of research.

    Keep up the good work!!

  10. Keith Smart says:

    Hi Sue.
    Thanks for your comments/definitions.
    The definitions for nociceptive, peripheral neuropathic and central (sensitisation) pain that we used in our studies were as follows:

    Nociceptive pain: Pain derived from nociceptive activity in the target tissues of the nervous system such as skin, muscle, joint and viscera, as transmitted by afferent nerve fibres which synapse in the spinal cord.

    Peripheral neuropathic pain: Pain initiated or caused by a primary lesion or dysfunction in the peripheral nervous system’.

    Central pain: Pain initiated or caused by a primary lesion or dysfunction in the central nervous system. The term ‘central’ pain was used in this study to refer to musculoskeletal pain states characterised primarily by ‘dysfunction’ within the central nervous system (CNS) i.e. central sensitisation/hyper-excitability rather than by any distinct lesion within the CNS as might occur with CNS disorders such as multiple sclerosis.

    Definitions of terms can be a sticky issue. You might notice that our ‘peripheral neuropathic’ and ‘central’ pain definitions were based on those published by the International Association for the Study of Pain (IASP) in 1994. My study started in 2007, then sure enough in 2008 a new set of definitions were proposed (Loeser and Treede, 2008). But because my study had started i really had to keep my original definitions of terms.

    Interpretation of definitions can also be a problem; and there’s always plenty of scope for disagreement, particularly around the issue of central sensitisation/hyper-excitability pain as something very distinct from central neuropathic pain. To paraphrase something i read recently in a novel (Wolf Hall, by Hiliary Mantel – and very good it was too!) “What has been defined can be re-defined”. So our pain-related definitions are likely to change again.

    I hope that goes some way to clarifying the terms we used in our studies?

    Kind regards

    Loeser JD, Treede R-D (2008). The Kyoto protocol of IASP basic terminology. Pain 137: 473-7.

  11. Thanks so much for reporting on this study! It’s great to see research being conducted on different pain mechanisms. However, for many readers (outside the world of Ph’ds), it would be great if you provided a definition of the three different pain types. Including a basic definition would make this research information much more accessible to a wider group of people. In trying to understand your study, I found this link helpful: and these definitions helpful:
    Nociceptors are the nerves which sense and respond to parts of the body which suffer from damage. They signal tissue irritation, impending injury, or actual injury. When activated, they transmit pain signals (via the peripheral nerves as well as the spinal cord) to the brain. The pain is typically well localized, constant, and often with an aching or throbbing quality. Visceral pain is the subtype of nociceptive pain that involves the internal organs. It tends to be episodic and poorly localized.

    Nociceptive pain is usually time limited, meaning when the tissue damage heals, the pain typically resolves. (Arthritis is a notable exception in that it is not time limited.) Another characteristic of nociceptive pain is that it tends to respond well to treatment with opioids.

    Neuropathic pain is the result of an injury or malfunction in the peripheral or central nervous system. The pain is often triggered by an injury, but this injury may or may not involve actual damage to the nervous system. Nerves can be infiltrated or compressed by tumors, strangulated by scar tissue, or inflamed by infection. The pain frequently has burning, lancinating, or electric shock qualities. Persistent allodynia, pain resulting from a nonpainful stimulus such as a light touch, is also a common characteristic of neuropathic pain. The pain may persist for months or years beyond the apparent healing of any damaged tissues. In this setting, pain signals no longer represent an alarm about ongoing or impending injury, instead the alarm system itself is malfunctioning.

    Neuropathic pain is frequently chronic, and tends to have a less robust response to treatment with opioids, but may respond well to other drugs such as anti-seizure and antidepressant medications. Usually, neuropathic problems are not fully reversible, but partial improvement is often possible with proper treatment.

    CP Definition?

    Thanks so much!