Our comparison of Harvey and Wessely's model with a model more biologically based suggests that a biological model based upon IO&NS pathways is more appropriate to describe this complex organic disorder. Activation of the IO&NS pathways induces important characteristic ME/CFS symptoms, forms of chronic fatigue and F&S symptoms in general. The IO&NS pathways can be instigated by infections, viral and bacterial, psychosocial and physical stressors as well as medical disorders such as (auto)immune disorders, which all function as precipitating and/or maintaining factors for ME/CFS.
The IO&NS model presented here may explain the spontaneous improvement or resolution of illness that occurs in those with an acute onset and particularly in adolescents with an infectious onset. Spontaneous resolution of the symptoms may occur when these IO&NS responses are diminished once the initial infection is eradicated. However, in predisposed persons, the initial infection may induce extensive and long-term sequelae in the IO&NS pathways. Our pathophysiological model could also explain why F&S symptoms may persist in the absence of increased cytokine levels as is documented in post-infection fatigue studies [76
]. The primary infections and subsequent responses in some specific cytokines might have resolved while the damage caused by O&NS to lipids, proteins or DNA and consequent autoimmune responses may persist and disable the patients, thus explaining residual F&S symptoms [19
]. Moreover, it is always possible that neuroinflammation, other cytokines or proinflammatory products are involved which had not been measured in these studies.
It is important to note that the above-mentioned IO&NS pathways also offer a plausible explanation for the earlier mortality due to cardiovascular disorder in ME/CFS, which is described in detail elsewhere [77
Based on their model, Harvey and Wessely [1
] recommend clinicians to avoid spending too much time chasing 'rare or unlikely diagnoses', or in their own words: 'not to spend too much time looking for zebras among the horses', and they propose to limit the organic investigations to a small set of blood tests. In contrast, we suggest that clinicians should examine the IO&NS pathophysiology, their sequelae, and the possible precipitating and maintaining factors (for example, infections) in any given patient. In our opinion, not investigating the IO&NS pathways could lead to inappropriate diagnosis of the underlying pathophysiology and thus possible inappropriate treatment for the patient.
The set of blood tests proposed by Harvey and Wessely [1
] includes some widely accepted tests for inflammatory diseases, such as C reactive protein and antibody assays for bacterial and viral infections and, in our opinion, should be conducted when assessing patients with ME/CFS. In addition, we suggest additional tests (outlined in Appendix 1) that are also good candidates and that can be performed as part of routine laboratory investigations; some of these are already approved by regulatory agencies. We think that regulatory agencies should objectively evaluate and approve the other tests, so that ME/CFS patients are reimbursed and these assays can be used on a regular basis in these patients.
Further, to identify the most accurate therapeutical approach it is likely necessary to define subtypes according to the IO&NS pathophysiology and the precipitating and predisposing factors [4
]. Thus, a distinction should be made between the type of IO&NS disorders the patients suffer from. Does the individual patient suffer from gut-derived inflammation, T helper (Th) 1-like or Th 2-like immune responses, monocytic activation, dysregulation of RNase L pathway, channelopathy, mitochondrial dysfunction, a low coenzyme Q10 syndrome, damage to fatty acids, functional proteins or DNA, autoimmunity, and so on, or combinations of these aberrations? It is also important to uncover potential triggers and maintaining factors, such as bacterial and viral infections, psychological stressors, overexertion, and the rare 'zebras', such as (auto)immune disorders, that may explain the IO&NS pathophysiology of F&S symptoms and ME/CFS. Finally, it is important to pinpoint the predisposing and maintaining factors, such as IgG subclass deficiencies and immunosuppression with recurrent infections, respectively. When bacterial or viral infections have been shown to be important maintaining factors, antibiotics and antiviral medications are essential [4
]. Intravenous immunoglobulins are an evidence-based treatment option for ME/CFS accompanied by common variable hypo-γ-globulinaemia or IgG subclass deficiencies, and recurrent infections or autoimmunity [20
]. Treatment with carnitine, coenzyme Q10, and so on, may be advised in subjects with depleted mitochondrial functions [7
]. Gut-derived inflammation responds favourably to treatment with glutamine [22
]. Mouse models of ME/CFS demonstrate that F&S symptoms can be induced by peripheral and central IO&NS pathways, including lipid peroxidation and depleted antioxidant defences, and that those symptoms may be reversed by specific anti-inflammatory and antioxidant therapies targeting the IO&NS pathways [31
]. Needless to say that there are still many IO&NS pathways in ME/CFS for which no adequate treatment exists, such as severe damage to lipids and proteins, and autoimmune responses.
We do not agree with the statement of Harvey and Wessely [1
] that a mental state examination remains the best investigation in persons with 'unexplained fatigue' because the 'fatigue' could be explained by a mental condition, such as depression. First, because we would propose that a biological investigation is likely to be a better indication of the underlying causes that may account for many F&S complaints. Second, patients with ME/CFS can be discriminated from those with depression by using a characteristic symptom profile [64
] or biological markers [72
]. Third, as described above, the co-occurrence between depression, F&S symptoms and ME/CFS is complex. It appears that ME/CFS and major depression are symptomatic manifestations of shared IO&NS pathways. Based on the above, we propose that the F&S symptoms of ME/CFS are not caused by depression and that ME/CFS does not cause depression. This implies also that both disorders are distinct diagnostic categories that should be treated differently.
We propose that future research should use high throughput, high quality screening, as made possible by translational research employing genotyping microarrays and functional genomics (assays of IO&NS genes), novel IO&NS animal models of ME/CFS, including transgenic mouse models (IO&NS overexpression or knockouts), and promoter induction based indicator cell lines that are specific to the brain (for example, neuroinflammation, damage by O&NS), muscles (for example, damage by O&NS, mitochondrial dysfunctions) and the gut (for example, gut inflammation and gut-derived inflammation) in order to further delineate novel drug targets in the IO&NS pathways and develop new drugs to treat this complex medical disorder. Multivariate pattern recognition studies should be carried out in order to (a) define clinical subtypes of ME/CFS and chronic fatigue and their associations with co-occurrent depression, and (b) examine the shared IO&NS pathways versus those that discriminate both disorders. Finally, interventional studies should be carried out to test the clinical efficacy of (novel) drugs in treating the biological causes of ME/CFS subgroups, defined by biomarkers such as inflammatory profiles or gene expression.