Importance of anti- and pro-nociceptive mechanisms in human disease

The burden of chronic pain to society is enormous. This is both in terms of physical and emotional impact to individuals and carers, in addition to the large financial burden. Current estimates suggest that 11.5–55.2% of individuals worldwide are defined as suffering from chronic widespread pain.¹ A major characteristic of functional disorders such as irritable bowel syndrome (IBS) and inflammatory/neuropathic disorders such as gastro-oesophageal reflux and chronic pancreatitis is abdominal discomfort or pain. There is an increasing awareness that many similarities exist mechanistically between somatic chronic pain conditions and the pain witnessed as chronic in IBS and chronic pancreatitis patients. With this realisation there has been a change of focus for researchers of both somatic and visceral pain conditions from peripheral structures as the preferred target of research to the central nervous system (CNS). It has long been recognised that the CNS has a major modulating nociceptive influence that alters resultant pain perception.^2–4 Recent developments in neuroimaging have enabled CNS investigations of visceral pain processing in patients and controls and such studies have highlighted the additional relevance of cognitive and emotional factors in modulating pain perception from physical changes such as plasticity and sensitisation.^5–7 Imaging studies have provided valuable objective information on what is inherently a subjective phenomenon, that for too long has relied upon patients giving a self report of their pain using coarse pain rating scales.⁸ Currently, there is a wider imaging literature on pain processing from somatic structures compared with visceral organs. This is probably because it is experimentally (and ethically) easier to perform somatic acute pain paradigms in healthy controls (for instance, using noxious thermal events) compared with more challenging oesophageal or rectal balloon distensions. However, this situation is rapidly changing and in terms of investigating relevant patient groups with pain conditions, there is a rapidly growing literature investigating visceral pain syndromes that competes with imaging studies investigating neuropathic or inflammatory pain (the interested reader is referred to these excellent reviews on imaging pain in the literature^9–11).

Since 1906 we have known that the brain can modulate in a “top-down” manner spinal cord excitability via a tonically active influence that is largely inhibitory in function. Evidence to support this came from work by Sherrington who showed that nociceptive reflexes were enhanced after the spinal cord was transected.¹² Reynolds in 1969 again emphasised the relevance of this phenomenon by showing that focal electrical stimulation in the rat midbrain periaqueductal gray (PAG) produced analgesia strong enough to permit surgery.¹³ Over the years, further work showed that stimulation of several brain sites, including the sensory cortex, thalamus, hypothalamus, midbrain, pons, and medulla, produced inhibitory effects on spinal nociceptive processing, suggesting an integrated network of brain regions that produce anti-nociceptive influences in situations where it is desirable to not be behaviourally diverted due to the noxious input. Such situations could include those where there is high arousal as in sports or battle or during placebo analgesia. Many electrophysiological, anatomical, and pharmacological studies determined that these descending influences on spinal nociceptive processing relied on relays in the rostroventral medulla (RVM), including the medial nucleus raphe magnus and it is now accepted that the RVM is the final common output for descending influences from rostral brain sites (for an excellent review see Gebhart 2004¹⁴ and Porreca et al, 2002¹⁵).

The focus of research on these descending influences was on inhibitory anti-nociceptive effects, with surgical applications for the treatment of chronic pain using periventricular electrical stimulation.¹⁶ In the 1990s, pioneering work by Gebhart and colleagues established without doubt the additional presence of descending facilitatory influences on spinal nociceptive processing. Independent work led by Fields et al had focused on characterising the response properties of cells within the RVM and showed that two major types of neurones were present: “ON” and “OFF”.¹⁷ This work by Fields et al paved the way for understanding better the role of the RVM in relation to processing and “top-down” modulation of pain. “OFF” cells are thought to comprise a descending inhibitory system that attenuates nociceptive information directly at the level of the spinal cord (anti-nociception)^17,18 whereas “ON” cells have a facilitatory influence on nociceptive processing through descending systems projecting to the spinal cord (pro-nociceptive).^17,18 The functional role of anti-nociception in everyday life and its importance is relatively easy to grasp; stress, fear, intense exercise, or escaping from a predator when injured are a few situations where control of pain has an obvious value to the behaving animal. However, to understand the functional role of pro-nociception in everyday life is less easy. Again, pioneering work by Gebhart and other colleagues unequivocally established a role for the RVM in the maintenance of hyperalgesic states following peripheral nerve injury. When tissue is injured there is increased input and sensitivity to stimulation at the site of injury; this is called primary hyperalgesia and is caused by increased excitability of peripheral nociceptors. More importantly, there is also increased sensitivity to stimuli from uninjured tissue either surrounding or distal to the site of injury; this is called secondary hyperalgesia and is caused by changes in excitability of neurones in the CNS or “central sensitisation”. Most work has focused on spinal cord changes during generation of central sensitisation but the work described above on pro-nociceptive influences provided an alternative explanation for the generation and maintenance of secondary hyperalgesia. Work from the laboratories of Gebhart, Porreca, and others have clearly established the importance of such “spino-bulbo-spinal” loops in hyperalgesia and possibly the development of chronic pain.^14,15,19 Indeed, recent work from our laboratory using FMRI in a human model of somatic secondary hyperalgesia has shown for the first time brainstem involvement in the generation and maintenance of hyperalgesia in humans.²⁰ Many animal studies have now been published that unequivocally show that pro-nociception via the RVM plays a key role in generating and maintaining a cardinal symptom of chronic pain; hyperalgesia. Why? One obvious answer is to prevent further damage to an already damaged area of the body. However, there are numerous other examples of “functional disorders” (for example, irritable bowel syndrome, fibromyalgia, etc) that are associated with discomfort and pain but where tissue pathology is often lacking. What role pro-nociceptive influences have in these conditions, and the potential of higher brain centres influencing RVM pro-nociceptive output, has yet to be thoroughly established. One thing is certain however, which is that the enormous potential pro-nociceptive influences may play in human disease is only just being realised.

Concurrent with these developments in our understanding of spino-bulbo-spinal loops for generating and maintaining exaggerated pain behaviours in chronic pain conditions (that is, pain making the pain worse via RVM pro-nociceptive influences) has been observations of a different phenomenon in which a spino-bulbo-spinal loop produced analgesia after acute simultaneous noxious input (pain makes less pain): this phenomenon is called diffuse noxious inhibitory control (DNIC) and is commonly known as counterirritation.²¹ This phenomenon goes back to the Hippocratic aphorism: “If two sufferings take place at the same time, but at different points, the stronger makes the weaker silent”. Experimental results during the 1980–early 1990s showed that counterirritation or counterstimulation has a well defined neural substrate in animals and humans. The mechanism to account for this phenomenon, DNIC, relies on spino-bulbo-spinal loops involving ascending pathways in the anterolateral spinal columns, integration in the lower brain stem, and descending influences or anti-nociception reaching the dorsal horn neurones. Neurones in the dorsal horn of the spinal cord were found to be inhibited when a nociceptive stimulus is applied to any part of the body that is distinct from the excitatory receptive fields of the dorsal horn neurones that are inhibited (hence the origin of the term “diffuse” as opposed to the similarly observed hypoalgesic effects of painful stimulation to the same segmental region of the body, as for instance during transcutaneous electrical nerve stimulation). DNIC generated inhibitions on dorsal horn neurones are potent and affect all the activities of the convergent neurones and persist after removal of the conditioning stimulus. This “pain inhibiting pain” effect is well known from folk medicine across many cultures. Indeed, in early surgical procedures on humans and animals this concept was harnessed without realisation of the underlying mechanisms (for example, use of the twitch in horses and nasal forceps in cattle during caudectomies or castrations which are both potentially very painful procedures). Reproducing this phenomenon in the human or animal laboratory has generated an enormous literature, and one simple example shows that ischaemic pain in the arm elevates the threshold for heat pain on the forehead and the threshold for painful stimulation of the tooth pulp.²² Many stressful stimuli are able to produce such counterirritation which has led to the term “stress induced analgesia”. Studies have used ischaemic pain, cold pressor pain, noxious heat, and even long distance running as conditioning stimuli.^22–25 Several surgical, pharmacological, and electrophysiological experiments on DNIC support the idea there is a spino-bulbo-spinal loop with minimal input from the PAG-RVM or other supraspinal pathways.^21,26 This implies that a system exists distinct from the anti- and pro-nociceptive systems described above, possibly within more caudal regions of the brainstem, and this system is what drives DNIC.^27,28

To date, these significant changes of phasic pain perception have been found by various psychophysical techniques, human reflex studies, and pain related evoked cerebral potentials²⁹ but none has been reported using brain imaging methodologies in combination with visceral stimulation.

The work reported by Wilder-Smith and colleagues³⁰ in this issue of Gut extends current observations considerably by combining brain functional magnetic resonance imaging (FMRI) with DNIC and rectal balloon distension in IBS patient subgroups and healthy controls (see page 1595). They show that compared with rectal balloon stimulation alone, the median rectal pain scores to balloon distension in healthy controls during heterotopic stimulation (foot with ice water) were decreased significantly. This robust psychophysical finding was not found to be the case in IBS patients or subgroups. This psychophysical observation implies dysfunction of the systems subserving DNIC and has been shown in a previous study on fibromyalgia patients.³¹ It is interesting to note that these two different patient populations (IBS and fibromyalgia) that are notoriously difficult when it comes to finding a “cause” of the pain, display an apparent common dysfunction of DNIC mechanisms. The study by Wilder-Smith and colleagues,³⁰ however, extends the observations to include functional brain imaging data, using FMRI, as a way of possibly explaining these psychophysical observations. They observed clear differences in brain activation patterns between controls and patient subgroups to rectal balloon distension both with and without heterotopic stimulation, highlighting an additional change in brain activation due to the dysfunctional DNIC system in patients. Combining brain imaging in patients with psychophysical manipulations provides a powerful approach for increasing our understanding of the brainstem structures involved in producing a DNIC effect. In addition, it has the capability of identifying higher cortical regions linked to such brainstem structures that as a consequence of disease might play a role in disabling the DNIC system. Their study has not provided all the answers, but neither should we expect it to, as many further experiments are required to establish the precise location and role of such brainstem and cortical regions. However, their study is the first of its kind and paves the way towards a better understanding of DNIC and how this acute experimental manipulation with its long history integrates with the above mentioned current theories of anti- and pro- nociception in the generation of hyperalgesia and chronic pain.

Work from our laboratory using counterstimulation to a noxious stimulus in conjunction with human FMRI showed that attention does play a major factor in the perceptual modulation of pain produced by counterstimulation.³² However, there are several studies suggesting that distraction does not account completely for the analgesia produced by activation of DNIC mechanisms,^24,26,29,33 including a fairly recent electrophysiological study in humans that supports the suggestion that the analgesic effect of heterotopic noxious stimulation in humans is based on activation of a specific inhibitory pain control system where release of endogenous opioids via the PAG-RVM system is unlikely to be involved.²⁹ The reciprocal issue as to whether hypervigilance to the acute pain stimulus during heterotopic stimulation causes disruption of DNIC effects has not been addressed, particularly in patients where the acute pain stimulus has a potential meaning, for instance rectal balloon distension in IBS patients. Of course, brain regions involved in maintaining a hypervigilant state could well be connected to brainstem structures involved in the DNIC system, and therefore this apparent confound could be the precise trigger for DNIC dysfunction. Again, this is speculation and further work is needed.

It is interesting to note that many of the changes Wilder-Smith and colleagues³⁰ observed were in brain regions involved in controlling emotional, autonomic, and classic PAG descending pain modulatory systems, suggesting involvement of the PAG-RVM system. It is also possible that brainstem pro-nociceptive plastic changes have occurred in these IBS patient subgroups as part of the evolution towards chronicity and it is this change that contributes partly to their heightened sensitivity and pain report. This would fit with current theories of pro-nociception.¹⁹ Therefore, when acute experimental manipulations are subsequently performed on such patients to trigger DNIC effects, it could be that possible raised pro-nociceptive influences nullify or counteract any descending pain modulatory effects normally produced by DNIC, thereby accounting for a lack of behavioural pain relief. This is speculation and clearly further work must be done to further interrogate these possibilities. However, the study by Wilder-Smith and colleagues³⁰ does provide a basis for generating hypotheses that can be tested in such future experiments. This will ultimately lead to a fuller understanding of the pain component of IBS patients and hopefully better targeted treatments.