Chronic fatigue syndrome reflects loss of adaptability

In this issue, Van Oosterwijck et al. [1] report that physical exercise lowered pain thresholds and was associated with exacerbation of symptoms in patients with myalgic encephalomyelitis (ME)/chronic fatigue syndrome (CFS) whereas, on the other hand, postexercise activity levels did not significantly decrease. Based on these and similar findings in patients with CFS, we present a conceptual framework that might provide a better understanding of the key features and pathophysiological mechanisms of CFS, and thus improve its diagnosis and treatment. Van Oosterwijck et al. [1] correctly note the frequent cooccurrence of a chronic ‘fatigue–pain’ symptom cluster, usually diagnosed as CFS and/or fibromyalgia. Recently, it has been proposed that this cluster should be classified under the unifying label of ‘central sensitivity syndromes’– a broad range of functional somatic disorders mainly characterized by common sensory abnormalities (i.e. widespread pain, hyperalgesia, allodynia and hypersensitivity to noise, bright light and certain chemical substances) [2]. However, ‘stress intolerance and pain hypersensitivity syndromes’ may be a more appropriate umbrella term for these syndromes because it reflects these patients’ inability to adequately adapt to all kinds of physical and mental stressors, including pathological pain processing [3]. Within the innovative neurobiological stress paradigm of ‘allostasis’– the need for stability through continuous change [4] – this general loss of adaptability may be understood as a failure of allostasis. Although the mechanisms underlying this failure are still unclear, they may include complex and interrelated disturbances of different components of the stress system, (i.e. the hypothalamic–pituitary–adrenal (HPA) axis), the sympathetic nervous system and various neurotransmitters that modulate perceptual–cognitive and affective brain circuits, all of which operate in intimate connection with the immune system and central pain mechanisms [5]. We and others have hypothesized that the pathophysiology of CFS might include a ‘switch’ from HPA axis hyperfunction to hypofunction following a period of chronic physical and/or psychosocial stress in vulnerable persons resulting in inadequate cortisol reactivity which may in turn, via low glucocorticoid signalling, increase inflammatory activity [5]. This assumption is consistent with the relatively low basal cortisol levels and blunted diurnal cortisol rhythm frequently observed in CFS patients [5], but recent data suggest that a decrease in glucocorticoid receptor sensitivity might play a role as well [6]. Abnormal activation of innate immunity involves the release of pro-inflammatory cytokines that influence the brain and give rise to ‘sickness behaviour’. This evolutionary, physiological and behavioural reaction normally occurs during infection or severe injury and its purpose is to optimally fight bodily threats by reorganizing priorities, saving energy and promoting healing and recovery. Characteristic symptoms are profound lethargy, feelings of malaise, concentration difficulties, headache, mild fever, sensory hypersensitivity and generalized pain. In CFS patients, however, this ‘flu-like’ symptom complex may be typically provoked by any kind of stressor (e.g. physical effort, mental pressure, strong emotions) and lead to a motivational shift by urging the patient to withdraw from activities [7]. Yet, the situation may be more complex. Not only is there evidence for basal hyperfunction of the sympathetic nervous system in CFS [8] and fibromyalgia [9], but dysfunctional descending pain-inhibiting pathways [10] and various psychological mechanisms may also contribute to abnormal pain perception [11]. The data presented by Van Oosterwijck et al. [1] fit within the stress adaptability hypothesis, which includes immune-related central pain sensitization, and thus make a strong case for refining current diagnostic criteria of CFS [12] to incorporate – as a mandatory criterion –patients’ maladaptive postexertional response. Novel clinical diagnostic criteria have meanwhile been developed [13] but it remains to be seen whether these criteria will empirically prove to be appropriate in identifying the key features of the illness. Although several randomized controlled trials have provided evidence for the efficacy of graded exercise therapy in CFS, no consensus exists about its value because effect sizes are generally modest, drop-out rates have been fairly high, and posttrial effects are not reflected by an objective increase in daily activity levels [14]. Van Oosterwijck et al. [1], on the basis of their findings, go one step further and caution that even moderate, self-paced and physiologically limited exercises might be potentially harmful. This does not mean that exercise therapy should be totally avoided in CFS. Indeed, it may still be useful for some patients to interrupt inactivity-related negative spiralling that may reinforce disability. Moreover, exercise may contribute to restoring stress system function, including normalizing cytokine–HPA axis feedback [15]. In addition, and perhaps more importantly, exercise may positively influence self-esteem, mood, perceived control, cognitive functioning and symptom focusing [14]. However, clinicians should always keep in mind that even minimal over-exertion may be counterproductive by increasing pain and symptoms, thus prolonging recovery. Therefore, the time may have come to replace the concept of ‘graded exercise therapy’ with ‘paced exercise therapy’ in CFS, using flexible and individually tailored exercise programmes that not only augment patients’ physical condition but also help them to better recognize ‘bodily signals’ and connect these to realistic goals and aspirations [14, 16]. When implemented in this way, such programmes may contribute to a reduction of abnormal fatigue and fatigability, less pain and improved physical and mental resilience. Although the results of the study by Van Oosterwijck et al. [1] are challenging, more research is needed to investigate the complex neurobiological and psychological mechanisms that might explain abnormal postexertional reactions in CFS. For example, in the context of the stress adaptability hypothesis, the precise role of decreased glucocorticoid signalling and high basic sympathetic tone in pain hypersensitivity should be further unravelled. From a therapeutic point of view, the assumption that ‘listening to their body’ during exercise and daily activities might contribute to restoring allostasis (i.e. normalizing stress system functioning, immune functioning and central pain processing) in CFS patients should be tested by prospective investigations. Finally, as the conceptual framework discussed above may lead to better diagnostic criteria rooted in modern neurobiological research, it may definitively put to rest persistent dualistic claims by some authors and patient support groups (e.g. see 17) that a true ‘neurological’ ME should be distinguished from a ‘psychiatric’ CFS. No conflict of interest was declared.