The Role of Autonomic Nervous System in Pain Chronicity

Abstract

The role of the autonomic nervous system (ANS) in chronic pain (CP) and in its chronicity is considered secondary and reactive to the nociceptive processes in the somatic nervous system (SomNS). However, research and clinical data strongly suggest the opposite. The ANS is an ancient, complex and ample part of the nervous system. It serves and controls visceral organs and somatic tissues. The ANS takes part in all aspects of all types of pain and influences its mechanisms at both peripheral and central levels. In this chapter we bring together the evidence from biomedical disciplines and clinical practice to support an alternative theory which contradicts the traditional views on the subject. We also raise questions which require further research to consolidate facts, advance our knowledge and improve treatment strategies for CP. The importance of this topic is difficult to overestimate because of the significant impact of CP on society and the lack of understanding, efficient therapy or cure.

1. Introduction

Chronic pain (CP) burdens a significant proportion of the population with pooled estimates for prevalence of 18–43% worldwide [1, 2, 3]. The vital role of the somatic nervous system (SomNS) in all types of pain is well recognised, but the importance of the autonomic nervous system (ANS) is mainly acknowledged in visceral or in ‘sympathetically mediated’ pain. The SomNS is perceived to be involved in all CP mechanisms and major dimensions of pain: physiological, sensory, affective, cognitive, behavioural, and sociocultural [4]. This is also true for the ANS, which is involved in major pain mechanisms and domains of all types of pain, not only visceral. A complete profile of ANS capability in CP formation has not been outlined, despite the accumulation of a sufficient body of evidence.

We planned this chapter as a brief conceptual narrative of a new notion of a comprehensive role of the ANS in CP. We will not didactically review the basic anatomy and physiology of the ANS, as we assume that anyone can find relevant information with its interpretation in current medical textbooks. Unfortunately, sometimes textbook authors present a simplified version of ANS structure and function. Without challenge, these deeply rooted views have been propagated from edition to edition or referenced in other publications. For example, it is widely considered that a leading role in CP development belongs to the SomNS; the ANS only responds to acute or already established CP. This conclusion frequently follows the outcomes of experimental studies [5]. We propose the opposite: the ANS plays a primary role in any type of CP and in pain chronicity. Table 1 summarises and compares our suggestions with traditional beliefs on the subject:

Traditional views	Our suggestions
Primary role in development of musculoskeletal, neuropathic, and visceral chronic pain belongs to SomNS. Visceral CP is mainly sensed via somatic structures by the ‘referred’ mechanism. No CP develops without SomNS participation. The ANS plays only a secondary, reflective role in chronic pain.	The ANS plays a global role in CP, possibly more important than those of the SomNS. No CP develops without an essential ANS input. ‘Referred pain’ is only one of the mechanisms of visceral pain perception.
CP normally involves the sympathetic division of the ANS (e.g. CRPS, Fibromyalgia). The parasympathetic division of the ANS produces mostly an anti-inflammatory, anti-stress and, in general, a positive effect in CP conditions.	Sympathetic and parasympathetic divisions of the ANS have complex anatomical and physiological relationships, and both participate in CP development.

Table 1.

A brief summary of authors suggestions versus traditional views on the role of ANS in chronic pain development.

Our view challenges current understanding of ANS involvement in pain chronicity and opens new avenues for diagnosis, treatments, and outcome monitoring. Our opinion draws on basic facts and advanced knowledge of different fields including, but not limited to, evolutionary biology, anatomy, epidemiology, pathophysiology, diagnostics and western and traditional medicine (i.e., effects of treatment). Therefore, the structure of this chapter follows the above list of biomedical disciplines and encompasses illustrative examples of medical treatment, interventions, and investigations. Surprisingly, there are still many gaps in CP theory. By highlighting them and asking appropriate questions we hope to encourage independent thinking and to stimulate future research. This is aimed to improve an evidenced based approach to refractory CP conditions which burden our society.

2. Evolutionary biology

Acute pain is one of the essential phenomena in biology because it helps organisms to survive. Therefore, it must have emerged early in phylogenesis, and since then it has evolved along with the growing complexity of the nervous system. Our knowledge about nervous system evolution lacks satisfying clarity [6]. For example, it is not clear when the central nervous system (CNS) appeared or whether it debuted independently more than once in the history of animal life on Earth [7]. Also, we do not know for certain if the SomNS arrived before the ANS, or whether the sympathetic division of the ANS developed earlier than the parasympathetic division. Some embryological studies report that the oldest autonomic structures were unmyelinated vagal fibres from the dorsal motor nucleus of the vagus [8].

Scientists face significant difficulties answering the above questions as nervous tissue does not preserve well in fossils. To support any of these conflicting points, authors sometimes use a ‘common sense’ approach by asking what is more important, ability to ‘fight or flight’, or control of the internal milieu in a precise way. One theory proposes that the “evolutionary origin of brainstem parasympathetic motor neurons out of branchial motor neurons, and spinal sympathetic motor neurons out of spinal motor neurons” [9].

However, comparison of ancient (but still living) species with modern organisms gives us essential facts for better understanding of evolutionary puzzles. For instance, sympathetic systems early in history employed acetylcholine (ACh) in postganglionic efferent neurons, and only later the majority of these fibres switched to noradrenaline (NA), except for sudomotor fibres. Certainly, the neuromediator change was reflected in sympathetic influence on some target organs with dual (sympathetic and parasympathetic) supply, when the stimulating effect mediated by ACh was passed over to the parasympathetic system.

ANS centres are located in phylogenetically ancient areas of the brain (hindbrain and midbrain) where autonomic, as well as the old somatic pain pathways (i.e., paleospinothalamic and archispinothalamic), terminate or make connections to; while the new somatosensory pain pathway (neospinothalamic) travels to the neocortex (forebrain).

Conclusion: SomNS and ANS have similar peripheral nociceptors, use the same neurotransmitters, and often share anatomical pathways and central connections. Both systems are of similar phylogenetic age [10, 11] and both should have been equally involved in pain processing, analysis, and responses.

3. Anatomy

Vast anatomical data help us to appreciate the fundamental role of the ANS in any type of CP. One can observe this in the complexity of the peripheral ANS: intricate structure and autonomy of local reflexes; abundancy (present in somatic and visceral peripheral nerves); diversity (variety of neuron types with different functions) [12]; and phylogenetic age of its spinal cord tracts and brain centres. Specificity in anatomical organisation of the ANS is the reason for precise homeostatic control: the efferent ANS (by its diversity of function and size) significantly “outweighs the somatic efferent pathways” [13]. Throughout this chapter we will continue comparing the ANS with the SomNS and draw your attention to their close interactions and inseparable activity. In this section we discuss the afferent, central, and efferent parts of the ANS and their significance for CP, but we will not cover the enteric part of the ANS.

3.1 Afferent ANS

Practically all somatic nerves contain autonomic fibres, and all of them are considered to be of efferent type (sudomotor, pilomotor and vasomotor). However, not all nerves contain somatic fibres; visceral nerves consist only of autonomic fibres. Therefore, any damage to peripheral nerve(s) will affect the performance of the peripheral ANS. This is also true for any type of damage at the level of the spinal cord as autonomic pain (sensory) tracts are in close proximity to somatic pathways. In consequence, when interventions treat peripheral nerves, nerve roots or epidural space, or spinal cord targets, the therapeutic effect on somatic and visceral/autonomic structures often cannot be differentiated.

Some publications advocate that primary visceral (sensory) neurons do not belong to the ANS, or if they do, they are not divided into sympathetic or parasympathetic fibres (despite visceral afferents travelling along sympathetic or parasympathetic nerves). This concept is often oversimplified in the literature. In order to understand the matter, one has to explore: the definition of the ANS; types of fibres in the peripheral nervous system (somatic and autonomic) and connections their primary sensory neurons make in the spinal cord; ascending tracts (with their targets, number of neurons involved and their functions).

Definitions of the ANS which are currently used by various dictionaries, institutions, publications, and other sources of information, typically declare that it controls involuntary functions of the body (internal organs and glands). Some interpretations might add a conflicting statement, for example, characterising the ANS as a part of the peripheral nervous system only (a network of peripheral nerves and ganglions), or suggest that the ANS has only motor or efferent fibres. This ambiguity undermines the functional intricacy and capacity of the ANS, which consists of various afferents, spinal cord and brain tracts and a network of analysing and executing centres.

Afferent innervation for internal organs comes from vagal (85% of vagal fibres) and spinal (50% of splanchnic nerves fibres) visceral afferent neurons [14]. The neuron cell bodies of these afferents are located in dorsal root ganglions (DRG) or in cranial nerve (IX and X) ganglions. Vagal afferent neuron projections are organised viscerotopically within the solitary tract in the medulla and spinal visceral afferent neurons connect to Rexed laminae I and V and deeper layers of the spinal cord in a segmental order [14].

Different types of neurons outside and inside CNS carry molecule and transcription factor signatures. Vagal afferent and efferent neurons, sympathetic post-ganglionic and autonomic neurons in the CNS are defined by homeodomain transcription factor Phox2b [15, 16], but visceral spinal afferents are not. The latter, in contrast to somatic sensory neurons, do not typically target Rexed lamina II but give rise to different pain pathways within the spinal cord.

General visceral afferents which travel with sympathetic peripheral nerves have their cell bodies in DRGs and their axons synapse with the second-order sensory neurons predominantly in laminae I and V as well as deeper laminae (VII, VIII and X) of the spinal grey matter. Spinal afferents which project to viscera comprise only a few percent of all sensory neurons in DRGs, the vast majority of neurons there are of somatic nature. Visceral afferents which ascend along vagal nerves have their cell bodies in inferior (nodosum) ganglions and some in superior (jugular) ganglions. They project to the nucleus tractus solitarius (NTS) of the brainstem. We are not going to discuss autonomic afferents of VII (facial) and IX (glossopharyngeal) cranial nerves here.

Spinal visceral afferents are not morphologically different from somatic afferent neurons with cell bodies in DRGs. However, they might differ by their spinal cord pathways and a number of additional functions. These are local efferent and trophic roles, both related to antidromic transport and release of chemicals and mediators via the afferent terminals to the innervated cells to influence visceral activity. Vagal visceral afferents show a great deal of diversity and coding strategies in respect to the organs these neurons serve [17, 18, 19, 20]. The conventional view is that nociceptive information does not get transferred via fibres within the vagus nerve. However, there are data supporting participation of vagal afferents in pain directly and indirectly. The latter might include interaction with sympathetic afferents at the cervical level [21], by inhibition of nociceptive dorsal horn activity, or by mediation of unpleasant symptoms (like nausea and bloating) which can exacerbate the pain experience. It is important to note that pelvic organs receive afferent innervation from two sources, both lumbar and sacral outputs. This list of possible mechanisms of visceral afferents in pain transmission and modulation is not exhaustive. The majority of evidence is based on animal studies, but due to high level of phylogenetic conservation, the majority of anatomical and physiological data could be applied to humans.

3.2 Ascending spinal cord tracts and vagal projections

The complexity of the afferent part of ANS is not fully discovered. New research emerges every year clarifying some and giving start to new questions. However, the situation with ascending autonomic spinal cord pathways is even more perplexing. The confusion comes from traditional descriptions of the ascending sensory pathways, including:

• name and phylogenetic age of a particular tract and its position within spinal cord;

• number of neurons and synaptic connections involved;

• laterality (ipsilateral, contralateral, or bilateral);

• destination(s) and branches to other brain centres;

• communication to somatic sensory pathways;

• connection(s) to motor tracts (and reflex activity);

• descending modulating and inhibitory effects exerted by neurons of interest.

For the purpose of this chapter, we allocate the highest significance to destination of the tracts and interconnections to other pathways. The latter feature enhances sensory experience and responses, including neuromodulating functions. It would be also useful to pay attention to evolutional order of appearance. Here we are not going to talk in detail about anatomical position and laterality. We will briefly mention this information only for selected pathways, as it is important in relation to accessibility by pain relieving interventions (their successes or failures).

Studies of nociceptive ascending spinal cord pathways confirmed existence of a large group of tracts. Not all of them end up in the brain cortex, many relay information to various areas of phylogenetically older parts of brain. Activation of these areas together with cortex centres contributes to multidimensional pain experience and its chronicity.

Fibres making the shortest (oligosynaptic) way to the cerebral cortex belong to relatively young structures (found in higher mammals), hence, forming the lateral spinothalamic (neospinothalamic) tract. It is monosynaptic on the segment to thalamus and, therefore, the fastest one. It brings sharp and well-localised sensation (small receptive fields) with a definitive quality (burning, stinging etc) of various intensity. These signals reach out to the somatosensory cortex; therefore, alert and warn consciousness. Neospinothalamic fibres are somatotopically organised (in all connections and at all levels), crossing to the opposite side in the spinal cord, and carry only somatosensory (not visceral) nociceptive information.

The older parallel tracts, named paleospinothalamic and archispinothalamic, include one or more synapses before thalamus. They make extensive connections to brainstem and other brain structures, lack somatotopic arrangements, target internuclear thalamus nuclei, and start subconscious autonomic and descending neuromodulating reflexes. These pathways tap into the affective dimension of pain experience.

Autonomic nociceptive signals travel via older tracts. Primary autonomic sensory neurons converge their input on the next neuron together with somatosensory afferents. This happens in the grey matter of the spinal cord. Viscero-somatic wide dynamic range neurons take input from large diameter (myelinated skin afferents) and smaller (myelinated and non-myelinated skin and deep tissue afferents) and primary visceral afferents. Visceral, as well as somatic, nociceptive information (via convergent neurons) could be transmitted via multiple pathways:

• Spinobulbar—targeting Ventrolateral Reticular Formation (VRF), Dorsal Reticular Nucleus (DRt), Nucleus Tractus Solitarii (NTS), Rostral Ventromedial Medulla (RVM);

• Spinopontine—targeting most studied Parabrachial Nucleus (PBN);

• Spinomesencephalic—targeting Periaqueductal Grey (PAG);

• Spinodiencephalic—targeting nuclei of Lateral and Medial Thalamus, Hypothalamus.

Majority of the above anatomical discoveries were done with fine antero−/retrograde tracing techniques on animals (rats, cats, and monkeys) but often the results are transferrable to humans because of high level of phylogenetical conservation. The more recent studies revealed the presence of direct tracts connecting spinal cord with cortex and subcortical telencephalon bypassing thalamus.

The significance of simultaneous activation of parallel oligo- and polysynaptic ascending nociceptive pathways is not fully researched; however, we can appreciate its contribution to vivid reality of pain or pain relief in everyday life. This is also important prediction of outcome (and duration) of pain-relieving ablative procedures [22].

The above-mentioned ascending (relaying visceral nociceptive information) tracts project signals further by connecting to RVM, DRt, pontine noradrenergic groups, the hypothalamus, amygdala, the ventrolateral medulla VLM, the NTS, the rostral ventromedial medulla, PBN, the PAG, the thalamus and cortex (parietal somatosensory, prefrontal, frontal motor, orbital and cingulated). Majority of these destinations are parts of ANS. Some of the descending circuits which originate from PAG, DRt and some other centres exhibits suppression and facilitation at spinal cord synaptic sites of ascending tracts.

Visceral pain is also transferred by midline postsynaptic dorsal column (PSDC) pathway. The axons of PSDC neurons transmit pelvic visceral nociception, they travel uncrossed in the dorsal column. The primary termination of the visceral input of the PSDC cells is the dorsal column nucleus. Pelvic visceral cancer pain responds to limited or punctate surgical midline myelotomy, thoracic visceral pain—to a lesion at the lateral edge of the gracile fasciculus, and experimental pancreatic pain to complete bilateral lesion of the gracile fasciculus [23, 24].

3.3 Central ANS

As discussed earlier, ascending ANS spinal cord tracts project to many brain locations and, therefore, are capable of production of multiple effects including descending pain control. Although, autonomic nociceptive pathways do not directly influence somatic pain, both systems meet at the spinal cord level (when primary afferents converge). Non-discriminative somatosensory nociception shares (phylogenetically old) pathways with visceral afferents and, therefore, highlight the same areas of the brain. Central and, therefore, efferent parts of ANS might be involved in visceral and at the same time in somatic pain due to such overlap. This is also reflected in the fact that “Pain Matrix”, network of brain centres responsible for pain processing shares key areas with the ANS.

The traditional take on these relationships is that the ANS passively responds to acute or chronic pain. Observers measure a shift of autonomic balance between sympathetic or parasympathetic tone using heart rate variability (HRV) or other tests (sudomotor activity, muscle sympathetic activity etc) or simply vital signs (heart rate, rate of breathing). We consider this topic is largely uncovered and deserves more attention.

The ANS governs body functions and influences mental state, emotions (conscious and subconscious phenomenon) and feelings (conscious phenomenon). Its centres include those of forebrain (insular and anterior cingulate cortex, amygdala, hypothalamus) and brainstem (PAG, PBN, NTS, VLM and some other parts of medulla). There is a fast-growing body of publications showing the complex relationship between pain conditions and activity of ANS centres [25]. There are many examples of this co-existence in people with diseases of various systems: cardiovascular [26]; respiratory [27, 28]; digestive [29, 30]; genitourinary [31, 32]; immune [33]; thermoregulation [34]; cerebral circulation and headaches [35, 36, 37]; sleep and circadian rhythms [38, 39]. Many of the above publications and similar are observational studies (and rarely prospective) or reviews. It is difficult to say if chronic pain was the reason for recorded changes or if pre-existing disturbances in ANS functions created vulnerability to chronic pain.

We would argue that in real life, disturbing and disabling chronic pain cannot develop without disorder of ANS control. Disturbed autonomic functions and chronic pain are in reciprocal relationship often with positive feedback: chronic pain might be a reason for autonomic symptoms and developed symptoms might reinforce and facilitate duration of pain, and they usually trigger and exacerbate each other. Therefore, frequently pre-existing autonomic derangement (even mild) due to lifestyle or any other reasons makes people more vulnerable to development of chronic pain.

3.4 Efferent ANS

In many sources, the ANS is considered a binary structure with sympathetic and parasympathetic divisions. For simplicity of teaching, the efferent output of the ANS is divided into craniosacral (parasympathetic) and thoracolumbar (sympathetic), with opposite effects on target organs. However, this is true only for a few targets; many internal organs, glands, skin structures and blood vessels are innervated only by one division. If both divisions are involved, they do not produce opposite responses (stimulation vs. suppression) or each branch functions under different conditions. So, the correct view is that ANS divisions work synergistically to provide stability of the internal environment and provide with the adaptive responses for internal organs and somatic structures.

The anatomy of the efferent ANS is more complex than that of the somatic motor system. It has preganglionic segments, ganglions, and postganglionic motor neurons. Detailed structure of the efferent ANS is well described in the literature. We will touch only upon its relevance for CP development and perpetuation.

A vital point supporting our view is based on the involuntary reflexes which are delivered by autonomic efferent fibres to visceral and somatic targets. The afferent information for this activity comes from visceral or somatic sources, but the central nuclei belong to the ANS. Physiological reflexes might change under pathological conditions as well as the homeostatic control of internal organ functions. This could lead to a variety of symptoms and painful conditions. These changes might replace the original programs and become chronic through learning and neuroplasticity mechanisms. Altered function (i.e., bowel contraction, acid production or abnormal blood supply) might cause more pain and unpleasant sensations perceived through the ANS.

Pain inherently boosts pathological neuroplasticity through reinforcement learning where the insula plays an important role [40, 41]. The underlying mechanisms could negatively affect physiological training-induced neuroplasticity in physical tasks [42]. However, the precise effect of pain (acute, experimental, or chronic) on motor skill learning is the subject of debate as no strong evidence has been provided by research [43]. The longer the duration of reflex changes, and of pain, the more complex the situation becomes and the more challenging and less successful treatment is. At some point, the pain condition reaches an irreversible phase [44], when treatment pursues palliative outcomes.

Examples of altered reflexes affecting visceral organs could be Irritable Bowel Disease (IBS), where pain is linked to abnormal gut motility, or urinary bladder conditions, when disturbing symptoms of urgency, incontinence and spasms convolute with pain. As for somatic organs affected by pathologically-changed autonomic reflexes, (e.g., skin thermoregulation) the afferent part is mediated by somatosensory afferents, but the central control (hypothalamus) and efferent output (sudomotor, vasomotor and pilomotor nerve fibres) is provided by the ANS [45].

Recent research has revealed an interesting relationship between autonomic and somatic neurons, which might explain muscle weakness in certain painful conditions with altered sympathetic outflow to muscles, for example CRPS. Sympathetic efferent fibres innervate neuromuscular junctions and are vital for maintenance and function of synapses between somatic motor nerves and muscles [46, 47].

In The Senses: A Comprehensive Reference, Second Edition (2020), chapter 5–21 [48] possible means by which the sympathetic nervous system could influence CP in somatic tissues are summarised. These are as follows: sympathetic-somatic afferent coupling; sensitisation of somatic nociceptors; neurogenic inflammation; and central changes in sympathetic pathways with the release of a variety of neuroactive substances and participation of neuroendocrine system. Earlie, Prof Jänig [49] discussed sympathetic (other than sudo−/pilomotor or vasoconstrictor) innervation of skin and deep somatic tissues, including muscles (vasodilators) and bones (peptidergic neurons, probably affecting mineralisation).

ANS involvement in musculoskeletal pain was investigated on the model of delayed onset muscle soreness (DOMS). Fleckenstein et al set out to discover to what extent sympathetically mediated pain (SMP) is responsible for exercise-induced acute muscle pain or damage in the upper limb [50]. They found that sympathetic regional (stellate ganglion) blockade causes pain relieving and anti-inflammatory effects. They suggested mechanisms for these effects which outlasted local anaesthetic block. These could be due to interruption of the vicious cycle of pain and local reflexes [51], allowing a reboot or a change in cytokine profile from pro- to anti-inflammatory [52]. There is also a possibility of changing of sympathetic and parasympathetic balance secondary to regional sympathetic outflow interruption. In fact, these effects might follow any peripheral neural injection as autonomic fibres will always be affected by local anaesthetic due to the abundance of peripheral ANS fibres, as it was mentioned above.

Conclusion: Complexity (of all parts) and ample presence of the ANS; interconnections within and with the SomNS; active control of body functions (involved in pain mechanisms), emotions and behaviour; neuroplasticity of reflexes support evidence of global and fundamental role of the ANS in pain development and chronicity.

4. Epidemiology

Epidemiological studies have highlighted the prevalence of different pain conditions and their risk factors. Despite ongoing research, some of these facts (e.g., uneven gender distribution, drug sensitivity with therapeutic response, associations with other diseases) are difficult to explain. ANS imbalance might be one of the reasons we overlook.

Gender difference in pain prevalence, sensitivity and analgesic response has been reported in the literature [3, 53]. Women experience more severe pain and report it more frequently. They develop pain in more anatomical sites and for a longer time, with a higher prevalence across the majority of pain conditions. But some painful disorders are strikingly more frequently seen in women: fibromyalgia; pelvic and musculoskeletal pain; and temporo-mandibular joint pain amongst others. This is routinely attributed to genetic factors, sex hormone profile and cyclical changes in serum concentrations, tissue nerve density, and psychological factors, but rarely to ANS input.

Previous research has shown that women have a prevailing parasympathetic tone whereas men have a prevailing sympathetic tone. This difference disappears after the age of 55 years [54]. Sympathetic system activation has been reported in CP states. We do not know how the parasympathetic division contributes, but this certainly involves complex, multilevel and non-linear interrelationships between the sympathetic division and other determinants.

One of the conditions which is three to four times more common in females is complex regional pain syndrome (CRPS). Reported risk factors for CRPS include: history of migraine; osteoporosis; asthma; and angiotensin converting enzyme (ACE) inhibitor therapy. The latter two are associated with parasympathetic predominance, but osteoporosis is considered more related to sympathetic activation [55].

Neurotransmitters are used in experimental research to obtain strong evidence on the mode of activity of ANS structures in question. Drugs which we prescribe for treatment of any illness might intentionally (indication) or unintentionally (side-effect) shift the balance between autonomic divisions. This is mediated through direct or indirect effects on adrenal and acetylcholine receptors in peripheral or central ANS. For instance, many antihypertensives suppress sympathetic outflow, whilst some antidepressants block and some opioids stimulate cholinergic pathways. Thus, treatment of comorbidities might affect pain conditions and vice versa. This is also important when pain-relieving drugs are chosen for a particular individual.

Time course of chronic diseases corresponds to constant changes which the ANS undergoes due to ageing, adjustment to climate, food habits, physical activity and many other factors. For example, asthma is more prevalent in boys, but later in life becomes more prevalent in women. This probably is due to a shift in autonomic balance which affects ANS airway control.

Conclusion: When we assess a case of CP it is essential to understand how it is related to excess or insufficient activity in each ANS division. This might influence our choice of drugs, interventions and other treatment methods, as well as help prognosticate. Coexisting medical conditions might give us a clue about autonomic balance and its dynamics, however, future research with appropriate questions and a fresh view on the problem might shed more light on the matter. The role of the parasympathetic system in developing CP has not been fully elucidated.

5. Pathophysiology

Practice makes perfect. This maxim is fully applicable to ANS design and functioning, but it requires a few clarifying comments. The ANS functions according to inherited programs (reflexes), and by learned behaviour patterns, which are established and maintained since birth and childhood [56]. This is achieved through learning by continuous feedback from internal and external environments, and neuroplastic changes which strengthen the neural circuits. Training of the nervous system is ongoing; it happens with or without our conscious acknowledgement and regardless of its value for the individual: regularly used activity gets reinforced; unused gets forgotten. For instance, one can develop insomnia when sleep routine is regularly disrupted by shift work or chaotic lifestyle. However, reintroduction of sleep hygiene will assist restoration of normal night sleep patterns. The same principle is employed in biofeedback bowel [57] or bladder training [58] for certain ANS disorders. In this subsection we discuss a few important consequences of autonomic dysfunction which impact on chronic pain development.

Any medical condition is associated with disturbed function of one or another organ, and therefore, with the disturbance of autonomic regulation of the corresponding physiological system. The opposite statement is also true: disturbed autonomic regulation will cause symptom development (into a medical condition) or prevent recovery from a condition-inducing event. This could be applied to acute pain as a symptom of a condition in question, or to chronic pain as a disease on its own.

A degree of autonomic disturbance might vary with different types of pain, the part(s) of nervous system involved, and anatomical region(s) affected. ANS dysfunction can be of local or global significance, and of mild or more severe presentation. We can associate diseases limited by anatomical region with the corresponding typical pain picture, but systemic medical conditions (e.g., diabetes, cardiovascular and lung diseases, rheumatoid arthritis, sickle cell disease) contribute to many chronic pain states. That is why the situation with diagnosis of disease causation and with recognition of factors leading to pain chronicity is not straightforward. Traditionally, abnormal ANS function and its diagnosis is overlooked in many (especially somatic) pain states, therefore the prescribed treatment often addresses only local symptoms, rather than pathophysiology of the underlying mechanisms.

For many chronic diseases we should recognise a reversible preliminary phase with subtle signs, which are usually below the threshold of current medical tests. Over a period of time autonomic regulation becomes progressively abnormal, but due to built-in robustness of the ANS, clinically significant deviation from medical norms might manifest years after. The preliminary phase is not usually identified, and underlying issues are not corrected. Partially this is because subclinical signs do not fall into pathological zones, but rather into domains of fitness or risk factors. This is a field of preventive medicine which, unfortunately, is largely unfamiliar to the general public. The quality of life at this stage deteriorates slowly and patients usually adapt to these changes without noticing the ongoing problem.

The important question at this stage is whether a single organ autonomic dysfunction develops in isolation in a particular chronic pain state, or whether it is always a part of the more systemic trend. The diagnostic value of many available tests of ANS status in pre-clinical phase and in even in mild cases is questionable. Their results are frequently reported as negative (or mildly abnormal) as often subjective severity of symptoms of ANS dysfunction do not match objectively measured parameters [59].

From our clinical observations when patients are convinced that their symptoms fit into a picture of Postural Orthostatic Tachycardia Syndrome (POTS), interstitial cystitis or CRPS but investigations do not support their perceptions our attempts to reassure them often fail. At that point our misunderstanding of the situation, broken relationship with patient, and lack of tests with higher resolution or sensitivity (they define disease criteria) leads to delayed diagnosis and treatment. On the other hand, labelling patients with the above diagnoses without sufficient evidence may medicalise them for life and prevent recovery. This unfortunate dilemma is one of the innate weaknesses of medical practice. It is triggered by the patient’s suffering from severe presenting symptoms.

The definition of suffering according to the Oxford English Dictionary is as follows: “the state of undergoing pain, distress, or hardship”. In CP all three entities—pain, distress and hardship—are intertwined, making it difficult to address them. The relationship between chronic pain, distress and hardship is well recognised. It dwells in emotional, social, and behavioural domains, and often is maintained by general symptoms (fatigue, chest tightness, mental fog and memory disturbances, sleep disorders and many others).

Sometimes the above constellation of symptoms is explained as an affective component of pain (linking it to the use of the old somatic nociceptive pathways), which is only partially true, as in fully developed CP we deal with neuroplasticity of nervous system where the ANS is responsible for many of these consequences [60, 61, 62]. Supporting evidence from research shows sympathetic hyperactivity in mental fatigue [63], significant and substantial ANS role in memory consolidation during sleep [64], association of mental fog and autonomic hyperarousal [65], abnormal autonomic sleep regulation in CP [66], and activation of autonomic pathways for chest pain and dyspnoea [67].

When medical professionals meet distressed patients who do not have clearly visible pathology which could explain the high degree of suffering, they often refer to these cases as those with functional (neurological) symptoms. However, many of these ‘unexplained’ symptoms could be due to disorganised activity of the ANS. We support the idea that suffering in CP could not happen without inherent participation of the ANS. This is because of a few reasons. ANS reaction to any pain or insult is inseparable to pain, even if pain is of somatic origin. Chronicity of pain and suffering is always at least partially driven by local or global dysfunction of the ANS. This includes control by emotional and behavioural centres, and often is not related to the severity of the index trauma.

CP patients develop maladaptive emotions and demonstrate changed behaviour. This includes poor coping and passive [68] strategies, fear-avoidance, and lack of motivation to invest efforts for their recovery, and social withdrawal. The ANS plays an important role in these changes. An experimental study [69] demonstrated that visceral pain response might relate to personality type, and it discovered sympathetic and parasympathetic co-activation in response to somatic and visceral pain. It is well known that the longer the chronic pain condition lasts and the more prominent are the patient’s passive approach, sick role, and other maladaptive psychological trends, the worse these features become, and patients with these symptoms are less likely to improve.

Finally, we should not forget that autonomic dysfunction in control of inflammation and immunity [33, 70, 71, 72], endocrine system [48], circadian rhythms [39, 73, 74, 75, 76, 77], tissue regeneration [78, 79, 80], including ANS itself [81] also contributes to chronic pain development and its chronicity.

Conclusion: The reciprocal relationship between pain and ANS control of involuntary body functions, emotions, behaviour and body regeneration makes the ANS an integral and indispensable player in a drama of CP. The earliest phase of autonomic dysfunction is not recognised and corrected.

6. Diagnostics

In this subsection we review diagnostic investigations currently available for assessment of ANS activity and discuss their limitations. We also describe potential tests (based on autonomic features) which might be applicable for pain assessment.

Heart rate variability (HRV) is one such non-invasive tool which is used in lab research, for diagnostic purposes, or in everyday life to monitor cardiorespiratory fitness. HRV employs electrocardiography (ECG) or plethysmography (PPG) for measuring distances between electrical heart complexes (or beats) over a period of time. The raw data obtained from ECG or PPG are calculated into various indexes which (as per convention) might describe activity of ANS branches.

Although HRV uses cardiac electrical activity for calculations, it shows not only good predictability of mortality in the heart conditions, but also demonstrates abnormalities in ANS performance in many diseases and pain states. However, HRV is not condition-specific. Additionally, there is no validated scale that can diagnose a degree of autonomic dysfunction in a particular illness.

HRV is a cheap, easy to use and widely accepted tool. For a full analysis it requires only a budget peripheral wearable device and a smartphone application. The analysis is based on mathematical calculations: descriptive statistics for time domain and spectral analysis for frequency domain. The latter uses the term “power” in relation to the energy within a particular frequency band, which should not be confused with biological “strength” of ANS divisions.

We do not know what the power of the ANS is and how to physically measure it. With HRV we might see a snapshot of the balance between sympathetic and parasympathetic activity. Whether this balance is on proportionally suppressed or enhanced divisions it is not possible to say. Furthermore, autonomic activity might be disturbed only in one organ or system, or in case of global autonomic failure, different organs might be affected unequally. These points should be taken into consideration when interpreting a HRV report in relation to a particular pathology.

There have been attempts to match HRV with organ-specific physiological activity. For instance, by parallel measurement of HRV and of high-resolution manometry of colon [82]. This experiment showed parasympathetic activation and sympathetic withdrawal during triggered propulsive colonic activity.

Clinical tests require laboratory conditions for measurements, calibrated and medically certified equipment, and professional interpretations. These tests investigate a single organ or a system specific autonomic dysfunction, but they are often invasive and might require anaesthetic input. For example, those used in cardiovascular medicine (e.g., tilt table test with plasma catecholamine concentration measurement), in urology (e.g., urodynamic tests), neurology (e.g., skin biopsy for nerve fibre density, nerve conduction studies, sudomotor activity and recordings of muscle sympathetic nerve activity), gastroenterology (e.g., gut motility, food transit, bacterial overgrowth).

Non-invasive options include disease specific questionnaires for organ function and thermography, a measurement of the surface temperature from the distance by thermal camera. The latter is useful in diseases with local change of blood supply, like in vascular abnormalities (vessel stenosis or arterio-venous malformation), regional sympathetic activity suppression (disease or local anaesthetic injection) or its excess (Raynaud’s or iatrogenic). CP conditions which manifest with skin temperature changes include CRPS, neuropathic pain with neurogenic inflammation, ischaemic pain and some others [83].

Pupillometry (PPM) is another window into ANS activity. The size of the pupil depends upon the rhythmical activity of a sphincter (parasympathetic control) and dilator (sympathetic innervation), triggered by the amount of light reaching the retina. Despite a growing body of research in anaesthesia and acute perioperative pain management which use PPM for assessment of pain [84, 85] and drug effects, the utility of this non-invasive method has not been fully established.

Facial expressions (FE) and emotion recognition is a complex field (the ANS plays a major role in it) where stable prediction is not technically achieved. FE have been used in acute pain assessment for a long time, but not in CP. Computer vision techniques often employ facial action coding systems (FACS) which detect face geometry and movement patterns [86, 87].

We suggest that in CP sufferers, FE could be used to assess the effect of pain-relieving intervention. According to clinical observations (unpublished data of the first author—DK), successful interventions in cancer pain dramatically change the quality of FE. For example, if a patient smiles before a procedure, it looks unnatural, forced or laboured. When the pain is relieved by intervention the smile becomes more natural with genuine facial mimic. This is a promising area for research of the role of the ANS in CP and pain relief with an objective and quantitative outcome.

Parameters of voice and of speech change under stress, emotional and cognitive load, pathological conditions and via ANS influence [88, 89]. A few voice-forming and modulating muscles are innervated by autonomic motor nerve fibres. Voice analysis could be used for monitoring of therapy and prediction of deterioration during the course of disease [90]. It is becoming more popular for pain assessment with the arrival of Artificial Intelligence (AI) based software [91].

Conclusion: Testing ANS state is essential in CP management; it demonstrates universal autonomic participation in CP. Clinical tests could be condition-specific, but invasive and demanding (equipment, staff etc.). Non-invasive methods are becoming more available for personal use (HRV, thermography), but some are still under-developed (pupillometry, facial and voice analysis). The resolution of existing tests is still low for the early recognition of pathology. Tests give a cross-sectional view (snapshot) on the condition but cannot provide longitudinal data for the evaluation of underlying pathophysiology. The latter could be addressed with the use of wearable multi-modal biosensing systems [92, 93].

7. Treatment

This section bears a dual purpose. It speculates on how the ANS could shape the outcomes of conventional pain-relieving procedures, and highlights the potential therapeutic interventions for disturbed ANS control. We discuss peripheral nerve blocks, epidural blocks and neuromodulation.

Local anaesthetic (LA) of sufficient concentration blocks nerve conduction allowing painless surgery. Unfortunately, pain returns if nerve blockade fades away. However, for post-operative analgesia significantly lower concentrations of LA than for surgery are required as there is no ongoing tissue damage.

In CP an injection of Lidocaine (LA) could provide a relieving effect of significantly longer duration (sometimes for several months or years) than the length of the nerve block [94].

First, we would like to describe thoracic differential epidural (TDE) blockade which is used as a diagnostic tool for abdominal pain to discriminate between somatic, visceral or central pain, and to predict response to visceral nerve block [95]. This intervention exploits two facts: smaller diameter visceral nociceptive afferents are blocked by a lower concentration of LA; and pain relief in visceral pain lasts longer than anaesthetic block duration [96]. This intervention showed that in many patients with pancreatitis, pain is of somatic nature [97].

Using the example of TDE, we might generalise that the therapeutic effect of epidural or peripheral nerve injection (beyond the LA duration) for musculoskeletal (somatic) pain could be due to concomitant sympathetic blockade. Epidurals, nerve root injections and peripheral nerve blocks produce sympathetic blockade in corresponding dermatomes or nerve distributions [98, 99].

Spinal cord stimulation (SCS) can provide pain relief and improvement of other symptoms in visceral and somatic pain by neuromodulation of various targets within the spinal cord. For example, SCS for refractory abdominal pain can improve chronic nausea and vomiting [100]. For neuropathic visceral abdominal pain, clinicians target the upper-mid thoracic level where splanchnic nerves emerge from the spinal cord. Improvement of gastroparesis and intestinal motility is highly suggestive of sympathetic blockade produced by SCS. However, available studies do not demonstrate consistent ANS reaction to SCS in sudomotor activity [101], heart rate variability (HRV), baroreceptor reflex sensitivity (BRS) and muscle sympathetic nerve activity (MSNA). Nor do they provide a plausible hypothesis for mechanisms of pain relief related to autonomic control [102]. This is another important area for future research.

Similarly to SCS, sacral nerve stimulation (SaNS) for pelvic organ dysfunction (bladder and rectal control) can also result in improvement in pain control [103] related to treated conditions.

Percutaneous tibial (somatic peripheral nerve) nerve stimulation (PTNS), which is used to improve urinary bladder control in Overactive Bladder, is an effective and minimally invasive technique [104]. It requires multiple sessions to achieve prolonged effect. PTNS also relieves chronic pelvic pain [105]. The mechanism of action is unknown but clinically it improves autonomic reflexes of the targeted organs. We can speculate about two possibilities:

• Somatic afferent stimulation affects autonomic efferent output (this could be at the level of spinal cord, brain, or both);

• Antidromic stimulation of sympathetic fibres which supply skin and blood vessels produces this effect.

Percutaneous or surgical vagal nerve stimulation has been suggested for many conditions caused by or associated with autonomic dysfunction [106, 107, 108, 109, 110, 111, 112], but the main indications are refractory epilepsy and certain mood disorders.

Non-medical options to maintain healthy ANS activity, which frequently involves parasympathetic stimulation and sympathetic withdrawal, include:

• Meditation, controversial reports [113, 114, 115]

• Slow diaphragmatic breathing [116]

• Acclimation to cold exposure [117]

• Exercises [118, 119, 120]

Conclusion: When planning pain-relieving interventions, healthcare professionals should consider treatment of underlying ANS dysfunction.

8. Traditional medicine (TM)

Traditional medical practices always acknowledge the complex relationship between organs and somatic tissues. It is reflected in diagnostic methods and in the holistic approach to treatment.

One of the unique methods used by many systems is the pulse diagnostic tool. It requires years to master but is claimed to provide invaluable information about any organ in the body. In general, it is probably an ancient equivalent of HRV, but much more sophisticated in the amount of detailed information it might provide an experienced practitioner.

Many traditions use a quasi-anatomical system: a whole-body map covered with lines or meridians (channels or vessels) with named points with very precise locations. The ‘vital energy’ freely flows through these structures controlling the activity of different physiological systems (lungs, bowel, liver, stomach etc.). There are 12 principal meridians in Traditional Chinese Medicine (TCM). There is no equivalent concept in western medicine.

The 24-hour biorhythm cycle allocates time of the highest and of the lowest activity to each channel. For example, the first meridian (Lungs) is the most active between 3 and 5 AM and 12 hours later (3 and 5 PM) it is in the lowest energy state. This circadian clock schedule correlates with clinical observations. For instance, maximum activity in the lung meridian corresponds to the peak of nocturnal asthma attacks. Maximum activity of the second meridian (Bowel) falls between 5 and 7 AM when people wake up after the night sleep and open their bowel. The next meridian (Stomach) is the most active when people normally have their breakfast, between 7 and 9 AM, and so on.

Abnormal flow of ‘energy’ (deficiency or excess and blockage) in one or more channels is the reason for symptoms of disease. Needling of the points according to the acupuncture recipe restores ‘energy’ flow and cures the disease and relieves pain (Yuan 2015). The choice of acupoints depends on the diagnosis, biorhythms and relationship between meridians. For example, neck and shoulder pain could be related to abnormal energy situations in gallbladder, bladder or large intestine meridians. Points could be chosen on these meridians or on others (via laws of relationship), around the painful area or distant to it; and they vary on a different time or day.

There are, of course, other than acupuncture treatment methods in TM: medications, breathing practices, postures and movements. The latter two could be organ specific, and they are synchronised with breathing to optimise vital energy and its circulation.

Conclusion: TM uses a holistic approach. It operates at the levels of aetiology and pathophysiology rather than symptoms as western medicine does. Similarly to the ANS, the Meridian system functions according to biological rhythms and connects organs and somatic tissues (skin, muscles, bones, ligaments and joints).

9. Conclusion

CP is considered incurable as per current beliefs, personal experience of medical professionals and statistics. So, in a pain clinic, in the media and in professional literature, patients once diagnosed with CP receive the same message, they have to live with it. The present situation is maintained by the ignorance of already known facts. However, progress is being fuelled by breakthroughs in related fields across multiple disciplines. We propose that pain becomes chronic through significant input from the ANS. Autonomic dysfunction (subclinical or apparent, local or global) provides the background for suboptimal organ activity and subsequently leads to the development of chronic symptoms, including pain, or the transition from existing acute pain into CP.

We would like to bring attention to the mind-blowing complexity and ample presence [48, 49, 121] of the ANS in our lives, as well as to its important role in CP and pain chronicity. This chapter serves only to outline a topic which could easily fill a whole book and warrants ongoing research. Current evidence from evolutionary biology, anatomy, epidemiology, pathophysiology, diagnostics, and pain medicine (western and traditional) supports our view and paves the way for future work. However, even now, people might change their view on the topic and this could lead to improved outcomes in CP management. Therefore, the main takeaway message is that we have to seek the signs of ANS dysfunction in any CP condition and address the underlying mechanisms.

Conflict of interest

The authors declare no conflict of interest.