The challenges of chronic pain and fatigue

Hypermobility

Joint hypermobility is a common variant of connective tissue affecting up to 20% of the general population and in symptomatic individuals associated with a wide variety of extra-articular manifestations.^27–30 Hypermobility is associated with pain, fatigue, gastrointestinal disturbance and headache.^31–34 It is frequently missed in clinical practice but may provide an important aetiological connection between fibromyalgia, ME/CFS and multiple systemic symptoms, including those described above.^35,36

Recent work indicates that symptomatic hypermobility may be present in ∼80% of fibromyalgia and ME/CFS patients, yet have only been diagnosed in one out of four cases previously.¹⁵ Differences in connective tissue matrix (eg collagen) throughout the body could explain disparate symptoms including irritable bowel and bladder, headache and dysfunctional uterine bleeding. Variant connective tissue also provides a putative mechanistic link for brain–body symptoms, whereby differences in connective tissue could lead to impaired vasoconstriction and, subsequently, impaired cardiovascular autonomic function.³⁷ Joint hypermobility is firmly associated with dysautonomia with some authors suggesting that at least 50% of hypermobile patients meet diagnostic criteria for PoTS and up to 80% all forms of orthostatic intolerance.^38–40 There is a phenomenological overlap with panic and anxiety, and those with hypermobility are over-represented in panic and anxiety populations by a factor of seven.⁴¹

Small fibre neuropathy

Small fibre neuropathy (SFN) is due to structural injury selectively affecting sensory and/or autonomic axons of a small diameter. The clinical presentation is dominated by peripheral neuropathic pain, in particular, but postural hypotension and other symptoms of autonomic dysfunction can occur, and there is clear clinical overlap with the fibromyalgia, in association with anxiety, hypermobility and autonomic dysfunction (FAHA) syndrome which we describe earlier.⁴² Autonomic symptoms may predominate in some patients in particular. The more ‘typical’ picture is of a ‘length-dependent’ neuropathy, with symptoms starting in the feet and progressing proximally, eventually resulting in a ‘glove and stocking’ picture. As in fibromyalgia, the pain may be chronic, but there may be a marked stimulus-dependent component (eg a sufferer may not be able to tolerate wearing tights or socks, or complain of pain associated with the weight of a duvet). In related conditions, increased temperature may cause ‘attacks’, but this type of presentation is more common in certain rare ‘channelopathies’ (eg erythromelalgia and paroxysmal extreme pain disorder (PEPD)).

SFN may be associated with diabetes, HIV (and certain reverse transcriptase inhibitors), alcohol use, Sjogren's syndrome and lupus, among others. Amyloid polyneuropathy can present as SFN. However, at least 50% of cases are idiopathic but, of this group, around one-third have variants of the SCN9A gene, encoding the Nav1.7 sodium channel, a voltage-gated ion channel that is selectively expressed in sensory and autonomic neurones.⁴³ These Nav1.7 variants cause increased excitability of sensory neurones and eventual degeneration of small fibres. SFN may also be associated with variants in the SCN10A gene (encoding the Nav1.8 channel), which enhance the excitability of dorsal root ganglion cells.

There are two main reasons that the real incidence is unknown. Firstly, there are no internationally agreed diagnostic criteria for SFN and, more importantly, the diagnosis is not easy to make. Bedside clinical neurological examination is usually normal, including tendon reflexes, coordination and large fibre sensory function (such as light touch, vibration sensation and proprioception). There may be a distal loss of pinprick or thermal sensation. In reality, accurate testing of thermal sensitivity is rarely performed and, in inexpert hands, unlikely to be reliable. Conventional electrophysiological testing is also normal. The gold standard for diagnosis is quantification of intra-epidermal nerve fibre density in a skin biopsy (usually a punch biopsy from a distal site) by histochemical staining for an antigen called PGP 9.5 (an ubiquitin hydrolase) that is found in all nerve fibres. This is a highly specialised technique, only performed in specialist centres.

In summary, it is certainly possible that significant numbers of patients with clinical presentations similar to Patient A (illustrated in Box 1) may have SFN. The presence of a truly a ‘length-dependent’ neuropathy, should alert a physician to this possibility, but the variation in the clinical presentation of the condition, and the difficulty in establishing a firm diagnosis, mean that many patients are, in all probability, currently ‘missed’.

Mast cell activation

There is increasing interest in the role of mast cells (MCs) in chronic disease. There are several patient groups in the UK, USA and elsewhere that actively espouse the importance of ‘mast cell activation disorders’ (MCAD) in patients with chronic fatigue, and dysautonomia, particularly in the context of hypermobility (hypermobile type Ehlers–Danlos syndrome (hEDS)). Is this plausible? If so, what, if any, is the clinical relevance?

Mature differentiated MCs are found only within tissues, where they congregate around nerves, blood vessels and lymphatics. Two main types of MCs have been described. Found in connective tissue, skin and the peritoneal cavity) they contain tryptase (MCT) in their granules and express interleukin-5 (IL-5) and IL-6. MCs which are home to the respiratory mucosa and gut contain tryptase and chymase (MCTC), and express IL-4.⁴⁴ Activated, mast cells produce a range of pre-formed and newly synthesised mediators. Within a few minutes of activation, pre-formed mediators (histamine and proteases) are released, following which, there is de novo synthesis of membrane-derived lipid mediators (prostaglandins and leukotrienes) and a range of pro- and anti-inflammatory cytokines and chemokines.

In the context of disease, MCs are typically recognised for their role in immediate immunoglobulin (Ig) E-mediated, allergic responses in anaphylaxis, physical urticarias, angioedema, food allergy, venom allergy and asthma. However, pathological MC activation has also been implicated in nonallergic disorders, including headache syndromes, irritable bowel syndrome, non-celiac gluten enteropathy, autoimmune syndromes and neuropsychiatric disorders, in addition to rare ‘primary’ mast cell disorders such as systemic mastocytosis (SM) and monoclonal mast cell activation syndrome. There are agreed diagnostic criteria for SM but not for other types of MCAD.⁴⁵

Symptoms of MCAD include urticaria, pruritus, flushing, intestinal hurry, abdominal cramping, osteoporosis, headache, postural hypotension and neuropsychiatric symptoms. Particularly relevant to this review are the recently described associations with hypermobility and hEDS.⁴⁶

Immunohistochemical analysis identified an increased content of chymase-positive MCs in the undamaged skin of patients with signs suggestive of ‘true’ connective tissue disorders.⁴⁷ Other investigators have described symptoms compatible with a non-IgE-mediated MCAD in patients with joint hypermobility. The most common symptoms reported are naso-ocular symptoms, and a history of anaphylaxis or asthma. As long ago as 2015, a possible new syndrome was suggested involving mast cell activation syndrome (MCAS), PoTS and hypermobility (EDS). In one early cohort study, screening of patients with PoTS and EDS revealed that two-thirds reported symptoms consistent with MCAS. The association has recently been reviewed in depth by Kohn and Chang.⁴⁸

So, how is MCAS diagnosed and managed? Testing relies on the measurement of MC mediators. It is suggested that ideally, serum tryptase and chromogranin A, plasma histamine, prostaglandin (PG) D2 and heparin, as well as urinary (random and 24-hour) histamine, N-methylhistamine (NMH), 11-β-PGF2α, and leukotriene (LT) E4 should be assessed. These tests are not widely available, but some can be organised through specialist centres. If MCAS is suspected in the context of hypermobility, PoTS and/or other suggestive symptoms, one pragmatic approach is to initiate a trial of therapy with a combination of H1 and H2 blockers, in combination with cromoglycate, or a leukotriene receptor blocker.

Like SFN, discussed earlier, physicians are limited by the lack of readily available tests, clear diagnostic criteria or treatment protocols. However, it seems likely that mast cell disorders will be increasingly recognised in the future and MCAS will become a more ‘mainstream’ diagnosis.

Inflammatory reactivity

There is evidence of raised inflammatory markers in fibromyalgia and ME/CFS and proteomic analysis suggests that upregulated inflammation can play a major role in the aetiology of fibromyalgia.^49,50 Preliminary results from our recent study into autonomic and inflammatory mechanisms of pain and fatigue suggest correlation between baseline inflammatory markers and pain and fatigue symptoms.⁵¹

However, in ME/CFS conventional markers of inflammation (eg CRP and ESR) are normal. Non-steroidal anti-inflammatories are rarely of long-term benefit, and the associated gastrointestinal side effects and enhanced cardiovascular risks mitigate against the use of these drugs. Corticosteroids are also not indicated, and may be positively harmful, resulting in altered mood, osteoporosis, impaired glucose tolerance and hypertension, as well as ‘steroid dependency’ as a consequence of adrenal suppression.

Interoceptive processing

Differences in interoceptive processing (relating to the internal sensation of what is happening in the body) maybe key in delineating the underlying brain–body mechanisms in such conditions, including hypermobility.^52,53 Recent work suggests fatigue may be related to disrupted interoceptive mechanisms, equally, pain is also linked to differences in interoception.^54–57 Mismatch between such observed and expected sensory signals representing bodily states may be important in symptom generation and maintenance, and such models have been extended to fatigue.^58,59 Thus interoceptive mismatch may be a factor in generation and maintenance of symptoms in such conditions, and may provide possibilities for treatment targets (eg https://doi.org/10.1186/ISRCTN17018615).⁶⁰).⁶⁰

What about long COVID?

The UK and much of the rest of the world are still battling the consequences of the ‘second wave’ of SARS-CoV-2 infection. In the UK, the severity of this is perhaps greater than anticipated, primarily due to lack of a general desire to support effective public health measures in a timely fashion, sociopolitical and commercial imperatives prevailing over high-quality scientific advice. However, while clinicians’ capacity to manage the acute illness associated with infection has improved greatly over the last 9 months, mortality is still high in vulnerable groups. There is currently an increasing focus on the longer-term consequences of infection. Chronic illness, particularly affecting the respiratory system, after infection was well-described in the original SARS outbreak many years ago affecting up to 30% of patients, and Carfi and colleagues first identified a similar problem with SARS-CoV-2 infection in a large group of Italian patients, with a reported persistence of at least one symptom, particularly fatigue and dyspnoea, in nearly 90% patients who had recovered from acute infection.^61,62 This is now widely recognised and was the subject of an authoritative review published recently by the National Institute for Health Research (NIHR).⁶³ Common symptoms include breathlessness, cough and palpitations, but also chronic fatigue (which can be severe), myalgia and joint pain, anxiety, low mood, cognitive impairment (‘brain fog’) and atypical chest pain. In the NIHR review, the authors conclude:

Indeed there is a possibility that the symptoms described may be due to a number of different syndromes (eg post-intensive care syndrome, post-viral fatigue syndrome and long-term COVID syndrome). Some people may be suffering with more than one syndrome at the same time.

A proportion of patients presenting with ill-health post-COVID will be found to have evidence of residual ‘physical’ damage affecting the lungs or cardiovascular system. This does not appear to be the case in the majority of patients, however, nor is there any objective evidence of an on-going ‘inflammatory’ process nor the development of autoimmunity or a conventional ‘reactive’ illness. The non-cardiopulmonary symptoms described map closely onto those which constitute the subject of this review. This raises the possibility that long COVID should, in the absence of proven residual tissue injury, be regarded as part of the CFS spectrum, and managed as such. Acute SARS-CoV-2 infection is an exogenous ‘stressor’, which has precipitated the condition. Individual physiological and psychological responses to such stressors are highly variable, as discussed earlier. Concomitant psychosocial stress factors, related to economic uncertainties, employment issues, and disruption of normal social and familial structures and processes, both at the level of the individual and within a wider societal context, mean that it may be very difficult to develop the appropriate coping strategies that allow full recovery from acute illness. We should not, therefore, be surprised that the condition that has been labelled as long COVID appears to be increasingly common. It is also true that, in acute SARS-CoV-2 infection, there is, of course, a need to make a specific ‘virological’ diagnosis to inform management and to aid efforts to reduce the spread of the virus in the population. The condition is now relatively ‘common’ and still increasing in prevalence. No such specific virological diagnosis is usually made in the majority of patients with ‘post-viral’ syndromes.

Long COVID is, at the time of writing, still very poorly understood, however, and there are a number of current centrally-funded research initiatives designed to address (among other things) whether there is objective evidence of chronic nervous system damage (central, peripheral or autonomic) and whether there are specific genetic factors that may pre-dispose an individual to chronic illness after acute infection.

We discuss later the importance of identifying other conditions that may present with chronic fatigue, pain or related symptoms, and this is, of course, no less important in the context of long COVID. Throughout this article we have, however, emphasised the need for a holistic, multi-professional approach to our patients. It is encouraging to hear that the NHS is investing in 40 centres precisely for this purpose. We hope, however, that these centres will not restrict their remit exclusively to the care of long COVID sufferers. Encouraging news on vaccine development may mean that COVID-19 is ‘defeated’ and long COVID may hopefully be of little clinical importance in 5 years’ time. Patients, such as our index case (Patient A), will, on the other hand, always be there.

Am I missing something?

Outpatient assessment of a patient with fatigue and chronic pain can be challenging. This is particularly the case when the patient is polysymptomatic and they have had many interactions with healthcare professionals before. It is all too easy to ‘admit defeat’ as in Dr A's case, or initiate a test or another referral in ‘response’ to every symptom described by the patient during the course of their illness (Dr B). How can this be avoided?

One approach is to adopt the ‘snapshot’ approach and ask the patient to identify a maximum of three ‘live’ clinical problems (eg in the previous 2 weeks), highlighting which is the most significant one. If the primary problem is fatigue or pain (as in our exemplar Patient A), it can be highly informative to identify the single most important additional symptom: the ‘plus one’ approach. This should raise a specific subset of diagnostic possibilities, often limited in number, which will prompt further, more focused ‘direct’ questioning, and may inform an initial investigation strategy, if considered appropriate.

All physicians are trained to identify ‘red flag’ symptoms, such as unexplained weight loss, which, in older patients particularly, may point to a diagnosis of malignancy, malabsorption, endocrine disease or depression. However, there is a number of conditions that are frequently missed, and easily identified. Some of these are summarised in Table 1. We focus on the ‘plus one’. The average time it takes to diagnose Behçet's, for example, may be up to 7 years and there is frequently a delay in the diagnosis of Sjogren's syndrome; few doctors ask about sicca.⁶⁴ Inflammatory bowel disease or endometriosis can also easily be missed, as can HIV, especially in older patients.^65,66.

What is rarely helpful or informative in a busy clinic setting is to rehearse in detail with a patient the detailed chronology of what may be several years of ill-health. This can all too often result in a ‘memory test’ for the patient, with the physician seeking to corroborate dates and details in the notes or electronic patient record. This wastes time, and can be frustrating for all concerned. However, it may be useful before, or more often after, the consultation to look back at previous letters, imaging or pathology results to help substantiate or refute a presumptive diagnosis. In patients with a ‘long history’ and multiple medical interactions, it is, however, important to avoid ‘medical transference’ and give undue credence to diagnoses made by other doctors, often years before. Firstly, the patient's recollection of the outcome of a previous consultation may be far from accurate, and secondly, the professor may well have been wrong. It does happen.

The future

In this short review we have highlighted what we feel to be a real association between chronic pain and fatigue, hypermobility, anxiety and dysautonomia, and suggested how these things may be linked pathophysiologically. We feel that the explicit recognition of this phenotype is of potentially of great clinical utility, particularly in planning care at the level of the individual patient. In addition, we would strongly espouse the development of effective multiprofessional clinical services as are planned, for example, for long COVID sufferers, to serve the true clinical needs of such patients, addressing both physical and mental health needs, avoiding unnecessary investigations, and maximising the opportunity to make a clinical diagnosis of other associated conditions.