Physiology of pain and nociceptors

This chapter tries to address Section K3(i) of the 2017 CICM Primary Syllabus, which expects the exam candidate to "describe the physiology of pain, including peripheral nociception, conduction, mediators and pathways, spinal cord modulation, central processing of pain, changes in the older patient". One CICM past paper question did specifically ask for this (Question 12 from the second paper of 2019). "Define pain" and "outline the processes by which pain is detected" were the specific demands. 

In summary:

This topic is clearly a well-trodden path, and not one of those areas where one would need to scrape shreds of information together from forgotten journals published in the 1950s. A Google Scholar search for "physiology of pain" yields literally a whole page of results each of which is titled "Physiology of pain".  Hudspith et al (2006), Steeds (2009), Renn & Dorsey (2005), Dzhambazova (2021) and Kendraud et al (2020) are all excellent free resources, covering essentially the same ground in five different ways, and so it might seem pointless therefore to add a sixths way, but then here we are. 

Definition of pain

From the examiner's comments to Question 12 from the second paper of 2019, it is clear that the WHO definition of pain was what they were after. That would probably be this:

"an unpleasant sensory or emotional experience associated with actual or potential tissue damage, or described in terms of such damage".

It is hard to trace this definition to back to the WHO, as they don't seem to directly own up to creating it. The definition as stated above actually comes from the International Association for the Study of Pain (IASP). Specifically, it was the IASP Subcommittee for Taxonomy, who brought this out in 1979 along with a lot of other terminology and criteria. There's nothing on the IASP website to suggest that they are a part of WHO, but one often comes across a piece of internet flotsam which conflates the two organisations and misattributes the definition. To be fair, the WHO do produce the International Classification of Diseases, and this document does include pain in it, and the definition in there is the IASP definition, so...

 Anyway. That we have a definition at all is quite remarkable. The difficulty in defining pain is very well expressed in an excellent article by Cohen et al (2018). To offer a glimpse of the controversy, the concerns included a debate over whether the word "unpleasant" is a satisfactory way to describe a sensation, or whether tissue damage is necessary, and whether self-reported experience is a vital element. In short, the definition has gone through several evolutions, and the most recent revised version (2020, at the time of writing) trails several notes and qualifying statements which have been added over the years to satisfy the critics:

"Pain is an unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage. 

• Pain is always a personal experience that is influenced to varying degrees by biological, psychological, and social factors.
• Pain and nociception are different phenomena. Pain cannot be inferred solely from activity in sensory neurons.
• Through their life experiences, individuals learn the concept of pain.
• A person’s report of an experience as pain should be respected.
• Although pain usually serves an adaptive role, it may have adverse effects on function and social and psychological well-being.
• Verbal description is only one of several behaviors to express pain; inability to communicate does not negate the possibility that a human or a nonhuman animal experiences pain."

In case one has a special interest in how we arrived at this point, Raja et al (2020) explain the tortuous path of "challenges and compromises" taken by the fourteen-member tasks force who produced it over the period of two years, by some kind of nightmarish Delphi method involving what must have seemed like endless consultation with legions of ethicists and philosophers. 

Peripheral nociceptors and the noxious stimulus

What noxious stimulus you detect clearly depends on the sort of sensory organ you have, and it would be impossible to discuss one without the other. We detect mechanical chemical and thermal stimuli because we have nociceptors specifically intended to detect these. We don't have specific nociceptive mechanisms for sensing electrical shock, but it tends to be perceived as pain anyway (even below the threshold of causing any tissue damage), as the electricity depolarises all forms of nociceptors directly. In contrast, there are some potentially quite damaging threats that we have no capacity to detect, because we do not have specific sensors for it - for example, one would not be able to feel one's DNA being dismantled by a mega-dose of gamma radiation. 

The definition of "nociceptor"

So: nociceptors. What even are they? Patel (2010) defines them as "unspecialized, free, unmyelinated nerve endings that convert (transduce) a variety of stimuli into nerve impulses", and this definition resembles others in published works, but overall pain literature seems to indicate that the term "nociceptor" has probably outlived its usefulness except as a rhetorical device. "We will use it here as a means of distinguishing afferents capable of responding to tissue damage from those that normally only encode innocuous stimuli", scoff Gold & Caterina (2008). 

A nociceptor nerve ending is said to be "free" in the sense that it does not synapse with anything else, i.e it is just hanging out in the tissues. These typically take the shape of a finely branched meshwork of fronds extending from the distal end of a sensory axon, referred to as a"terminal arbor". One sensory neuron would live with its cell body in the dorsal root ganglion of the spinal cord, sending an axon out to the periphery which would terminate in this sort of arbor. Here's some examples of terminal arbors from rat and mouse skin (Tracey, 2017), with scale bars of around 100μm:

The area innervated by a single such neuron would obviously be different depending on where the information is being collected from. For example a neuron innervating the fingertips would only branch over 1mm² of skin surface, as it would be important to have detailed spatial discrimination there. Conversely, nociceptive neurons innervating some uninformative area (eg. the bowel) would each cover many square centimetres of tissue. 

Classification of nociceptors

Nociceptors are generally classified by the type of nerve fibre that supplies them, as this feature seems to be associated with their specific functional role. 

Aδ mechanosensitive nociceptors	Large myelinated Aδ fibres
Aδ mechanothermal nociceptors	Large myelinated Aδ fibres
Polymodal nociceptors	Thin unmyelinated C fibres

This is the usual approach taken by textbooks, including Stoelting. It is therefore what CICM examiners will probably expect in an exam answer. Other classification systems which focus on neurochemical and functional subgrouping are developing, and may eventually mature to supercede the older systems. Moreover, some are beginning to question even the need for a classification system (Handwerker, 2010)

Molecular mechanisms of nociception

The process of nociception begins with transduction, which can be defined as

"A process in which environmental stimuli evoke conformational changes in the structure of specific proteins located on nociceptor terminals that directly or indirectly (i.e., via cellular signaling cascades) trigger the opening or closing of ion channels"

- Gold & Caterina (2008) 

There are a large number of proteins implicated in this process, all of which can be broadly described as nonselective cation channels. When stimulated, they open to allow an influx of sodium into the neuron, depolarising the membrane and creating an action potential which then propagates towards the spinal cord. The list is long and probably irrelevant for exam purposes. Highlights include:

• ATP-gated ion channels that detect ATP (which is released from damaged cells)
• Proton-gated ion channels that detect a change in tissue pH (usually they open for a pH < 5.0)
• TREK-1 potassium channels that detect membrane stretch, i.e. these are responsible for sensing tissue deformation
• G-protein coupled receptors that detect a variety of inflammatory ligands, like for example prostaglandins and bradykinin
• Transient Receptor Potential ion channels (TRPs), which are polymodal, i.e. triggered by a whole host of different stimuli, such as:
  • Heat (apparently the threshold for TRPV1 is 43º C)
  • pH (less than 5.0)
  • Hypo-osmolarity
  • Metabolites of arachidonic acid (released in the course of inflammation)
  • Capsaicin and related vanilloid compounds

From this list, it should become apparent that only a minority of nociceptor actually transduce mechanical deformation into action potentials, and therefore nociceptors mainly react to different chemical signals which lead to nociceptor activation or modulate the behaviour of pain receptors (and only a minority of them actually transduce mechanical deformation into action potentials). These chemical mediators are probably numerous beyond counting, but a handful need to be discussed in a bit more detail for exam purposes:

Chemical mediators of pain sensation

These can be broadly grouped into two main groups: chemicals released in the course of an inflammatory response, and substances which happen to only be found inside cells (which fall out when those cells are damaged).

Contents spilled from damaged cells

ATP and potassium are found inside cells in high concentrations and are virtually absent from the extracellular fluid. Ergo, they make convenient markers for tissue damage- if your molecular neighbourhood is suddenly inundated with potassium and ATP, that probably means that a cell has just burst somewhere nearby. This sounds good in theory, as there are definitely ATP-gated channels on nociceptive nerve endings and subcutaneous ATP injections definitely cause pain, but as Hamilton & McMahon (2000) point out, the physiological role of endogenous ATP remains to be established. 

Inflammatory mediators

Arachidonic acid metabolites and other pro-inflammatory molecules released by vascular endothelium and platelets (prostaglandins, bradykinin, histamine, serotonin) are definitely involved in the activation of nociceptors, as well as in the modulation of their function.  To paraphrase some extremely dense material from Smith (2006) and McMahon et al (2006), 

• These are usually polypeptides, eg. bradykinin is a nonapeptide
• They are usually very short-lived and broken down rapidly by plasma or local tissue enzymes, limiting their effect to the specific region of inflammation
• Their role is primarily inflammatory, i.e. they perform some sort of important immunological role (for example, slowing the flow of local capillary blood, and making the endothelium more permeable), and they just happen to also have a pain-related function
• The list of such mediators is massive, and in any case nobody could possibly ever be asked any sort of exam question where the number of inflammatory nociceptive mediators would somehow influence the mark, because that would be insane. However, it is worth knowing a few of the names, or at least the dominant chemical groups. Thus:
  • Eicosanoids (eg. arachidonic acid metabolites): prostaglandins and leukotrienes
  • Monoamines: serotonin and histamine 
  • Cytokines, such as IL-6, IL-8, IL-10, IFN-γ, etc
• The effect of some of these is mediated directly (i.e. they open a cation channel), and for others there is some indirect G-protein related activity involved, but both ultimately lead to the depolarisation of the membrane and therefore the propagation of an action potential up the axon
• Another effect is the peripheral sensitization of the nociceptors, an effect of local inflammation which increases the number of nociceptive receptor proteins expressed on the surface of the terminal arbor branches. Carlton & Coggeshall (2001) were able to demonstrate an inflammation-induced doubling in the number of capsaicin receptors. The result is a lowering of the threshold required to produce an action potential. Apart from this  there is a whole host of other different sensitization mechanisms, enough for an entire paper (Wei et al, 2019)