A Critique of the CFS Guidelines (Revised Draft 2001)
Abhijit Chaudhuri DM MD MRCP(UK)
The publication of the RACP document, "Chronic Fatigue Syndrome Guidelines" is timely. However, there are several shortcomings in this document that must be addressed. In the following paragraphs, I have drawn attention to these shortcomings.
Comments on the Guidelines
1.1 What is CFS:
The RACP should adhere to the modified CDC definition of CFS instead of offering its own definition in the first sentence of this paragraph, that should read as:
"CFS is defined as unexplained, persistent or relapsing chronic fatigue of new onset that substantially reduces previous levels of activity. Fatigue in CFS is usually not relieved by rest and cannot be attributed to any known medical or psychiatric diseases".
1.2 Diagnosis of CFS: (paragraph 2):
The word "restricted" should be changed to "appropriate", i.e. and an appropriate set of laboratory investigations.
1.3 Investigations :
Recommended screening investigations should also include:
serum C-reactive protein (CRP)
serum creatine kinase (to exclude mitochondrial and other muscle diseases)
The paragraph under additional investigations must be changed to draw attention to the following:
|"In specific clinical circumstances, following tests may be indicated: Autoantibodies (rheumatology screen) if arthralgia or other symptoms of connective tissue diseases are present; antibodies to endomysium (or gliadin) to screen for coeliac disease in patients with bowel symptoms
Infectious disease screening (Hepatitis B/C, Lyme disease or HIV) in the presence of appropriate risk factors
Urinary cortisol and Short Synacthen Test if Addison's disease is a differential diagnosis;
MRI brain scan if multiple sclerosis is a possibility
Muscle biopsy if serum creatine kinase is elevated"
1.4 Management (concluding paragraph):
The last sentence is seriously misleading (These strategies, together with promotion of a clear understanding of the illness are sometimes termed cognitive behaviour therapy). This particular sentence should be deleted.
The RACP seems to suggest that cognitive behaviour therapy (CBT) provides a clear understanding of CFS. This claim is unfounded and lacks evidence. With our current state of knowledge, it would be erroneous to claim that any therapy for CFS, organic or behavioural, provides a clearer insight into the cause of the problem. Additionally CBT is not a specific strategy for CFS where its claimed benefit is still questionable. CBT has been used successfully in the management of painful symptoms of rheumatoid arthritis but one cannot claim that its use in rheumatoid arthritis implies that the symptoms of the arthritic pain are largely behavioural.
1.5 Physical activity:
The last sentence of the first paragraph should be deleted (Graded exercise programmes have been shown to be safe for people with CFS, and can improve both aerobic capacity and functional status) since it is plainly inaccurate due to the following reasons:
Firstly, no study of graded exercise programme has been carried out in the modified CDC defined population with CFS. Secondly, graded exercise programmes are not entirely safe for patients with CFS since some patients may experience serious deterioration of their symptoms after exercise. Thirdly, no long term study has established that graded exercise programme can significantly improve aerobic capacity in CFS.
It appears that the RACP has failed to recognise that post-exertional malaise is a valid CFS symptom. There is no evidence that patients with CFS demonstrate avoidance behaviour to physical activity as claimed. Even if graduated exercises were effective in CFS, it does not imply that CFS symptoms are contributed by an avoidance behaviour to physical activities. Graded exercises benefit patients with mitochondrial muscle disease, which is not known to be caused by avoidance to physcial activities. Thus, the second paragraph of this section (beginning with: It is important...) is a mixture of imagination and half-truths and should be entirely deleted. The UK experience of graded exercise in CFS has shown that as a single intervention, graded exercise was associated with the highest negative grading for its effect on fatigue by the patients (see the overview).
2. What is Chronic Fatigue Syndrome
2.1 The spectrum of fatigue disorders:
One cannot issue a broad statement like "At present, CFS cannot be considered a disease" because of the following reasons:
Patients with postviral neuromyasthenia or myalgic encephalomyelitis (ME) fulfil the modified CDC-criteria for CFS. Both neuromyasthenia and ME are classified as neurological diseases by the World Health Organisation in the International Classification of Diseases (ICD-10). If ME is accepted by the RACP to be same as CFS (Page 18: What other terms are commonly used for CFS), then ME/CFS is a neurological disease according to the WHO classification. On the assumption that CFS is a heterogeneous disorder, one may only state that "there is no disease marker that characterises CFS population as a whole although the post-viral fatigue subgroup (ME/neuromyasthenia as defined in the ICD-10) may have a neurologic basis".
It is also difficult to justify the statement that "CFS is best regarded as an illness". In addition, use of the term illness does not always imply a subjective state as claimed since illnesses may be caused by diseases, e.g. "terminal illnesses". RACP appears to be making a subtle attempt in this paragraph to suggest that patients with CFS have a subjective problem or an illness behaviour. The second and the third sentences of this paragraph should be deleted.
I believe it is not the purpose of this document to establish whether CFS is a disease or an illness since it is designed to provide guidelines for care.
2.2 What is CFS:
RACP is confusing the issues of CFS and unevaluated chronic fatigue. If two thirds of unselected referrals of patients with chronic fatigue have another medical or psychiatric disorder, then these patients do not have CFS. Thus, the second paragraph of this page is misleading and should be deleted.
2.3 What other terms are commonly used for CFS (second paragraph) :
It is incorrect to propose that neurasthenia is same as CFS. It is neuromyasthenia (and not neurasthenia) that is similar to CFS. The discussion on neurasthenia (a psychiatric disorder in ICD-10) should be replaced by a note on neuromyasthenia (a neurologic disorder in ICD-10) for which was widely reported in the medical literature between 1930-60s. Put simply, neuromyasthenia was the term used to describe epidemic outbreaks of what would be currently called CFS following viral infections ("epidemic neuromyasthenia") .
The suggestion that neurasthenia was an alternative term for CFS is to be found largely in the psychiatric literature. Most neurological texts of the 19th century make it clear that neurasthenia was an inaccurate and confusing term and should not be used in clinical practice.
2.4 How common is CFS:
The RACP must clarify the diagnostic definitions used in the various population prevalence studies of CFS. For example, the higher prevalence of CFS in the primary care was based on a broader diagnostic definition for CFS (the Oxford Criteria) that would be inappropriate.
2.5 Who is at risk of CFS (second paragraph):
I am unclear what is meant by the first sentence of the second paragraph (It is unlikely that common, non-specific viral infections trigger the onset of CFS, but specific infections, such as mononucleosis, quite commonly do so) . Is RACP is suggesting that only uncommon, specific viral infections cause CFS? There is no evidence to this claim. Indeed, worldwide carefully planned epidemiological studies have shown that there is no specific viral infection(s) that is more likely to trigger the CFS symptoms in the susceptible individuals when compared with the non-CFS population exposed to the same infection in the community. We have recently reviewed this topic for a chapter in a major textbook and on the basis of my experience, I would suggest that this paragraph should be entirely rewritten with appropriate references and correct information.
Among non-viral infections, there is a higher incidence of chronic fatigue after Lyme disease that should have been mentioned.
2.6 Does chronic fatigue overlap with other illnesses
In my opinion, this section is the most seriously flawed part of the RACP guidelines.
The first paragraph does not discuss organic diseases where comparable symptoms of chronic fatigue are well recognised. While the document takes care to mention somatoform disorders and chronic fatigue, it does not mention the following diseases with overlap symptoms of chronic fatigue, e.g. post-polio fatigue, multiple sclerosis, dysautonomic states, orthostatic intolerance or postural-tachycardia syndrome, cardiogical Syndrome X, post-Guillain Barre syndrome fatigue, chronic hepatitis and so on. Many of these syndromes may present with chronic fatigue and normal routine physical examination and screening laboratory tests (e.g. Syndrome X and orthostatic intolerance). The reader of the document may be led to believe that the overlap of chronic fatigue is only restricted to the functional disorders.
In the second paragraph, the number of psychiatric references cited to support the psychiatric co- morbidity in CFS is the highest in the entire document. However, analysisof these wide number of references also show how uncritical and biased the selections are. Firstly, many of these references belong to the pre-1994 era and deal with a population not selected by using the current diagnostic criteria (modified CDC) that were only published in December 1994. Secondly, the most recent reviews which clearly show that psychiatric co-morbity in CFS has been over-exaggerated in the old literature are omitted. Overall, the psychiatric co-morbidity in CFS is probably seen in about a quarter (25%) of the patients and certainly in no more than a third (33%); this figure is comparable with the psychiatric co-morbidity seen with other chronic medical illnesses (multiple sclerosis, diabetes or cancer). Thirdly, the RACP has failed to take note of the most recent prospective study of people exposed to a common viral infection (infectious mononucleosis) which clearly confirmed that there is no higher incidence of pre-morbid psychiatric diseases in those who eventually develop CFS after the viral illness.
In the third paragraph, the document confuses the issue of the somatoform disorders and CFS. Fewer than 5% of CDC-defined CFS population would fulfil the DSM-IV diagnostic criteria for somatoform disorders as shown in a published study. However, this has not been referred to. Somatoform disorders are seldom confused with CFS where symptoms of fatigue are usually acute and of new onset, with a previous history of normal health. Only psychiatrists may find this distinction difficult, as appears from the cited references.
In my opinion, this entire section should be scrapped for its abysmal quality of potentially misleading information.
2.7 What are the outcome of fatigue states:
The term "prolonged fatigue" should be defined while suggesting that the rate of spontaneous recovery in the first six months is high after a viral infection since it may be confused with chronic fatigue. The fact that complete recovery of CFS in adults is uncommon should be clearly stated.
This section makes selective references to the studies looking at the unselected population with chronic fatigue (rather than CDC-defined CFS) and their perception of the illness. This section should be amended with references to some of the longitudinal studies of CFS patients. Studies that have examined functional status and the quality of life measures in the CFS population clearly confirm that the scale of impairment across a range of physical and mental activities in CFS can be comparable to those seen in many other chronic medical conditions.
2.8 What is known about the pathophysiology of CFS:
This subsection is confusing, poorly written and is an inappropriate reflection of the general standards of the literature review followed in the guidelines. I find it strange that the latest reference used in the discussion of this topic in a year 2001 document is dated 1995. This suffices to show that the authors of the RACP document have not desired to look beyond 1995 for the pathophysiology of CFS. There are overlaps even on the stated hypotheses, the earliest one of which is taken from an ancient article before CFS was known as an entity presumably because it supported the psychological hypothesis (1961). However, non-psychological hypotheses of both pre-and post-CFS era have been largely ignored.
The pathophysiology of CFS, in my view, can be stated very simply. An unbiased account is offered below as an example that the RACP may choose to follow:
(1) The psychiatric hypothesis: CFS is a functional somatic syndrome and/or an illness behaviour linked to the depressive disorders. In this view, it is held that CFS develops when a psychologically vulnerable individual acquires an infection and perceives himself/herself to be physically ill even after recovery from this disease. The illness is perpetuated by a combination of depression, dysthymia, physical deconditioning, avoidance behaviour for physical activities and somatisation. The relative contributions of the individual components of these perpetuating factors may vary among patients. This hypothesis has been claimed to have been supported by the outcome of some of the intervention studies of graded exercises and CBT in CFS.
(2) The neurobiologic hypothesis: CFS is primarily considered to be a disorder of the central nervous system (central fatigue) and the symptomatology reflects the outcome of a complex interaction between the cognitive, physical and neuropsychological changes induced by the initial disease process (commonly a viral infection, rarely a toxin like ciguatera fish poison). The physical changes may affect the immune system, neuroendocrine regulation, ion channels, metabolic rates, autonomic responses and the cerebral perfusion. Studies have shown that patients with CFS may have some of these changes although the relative contributions of the individual components may differ in patients and at present, there is no consistent biological marker for CFS.
The reported primary care prevalence of CFS at 2.5% is not based on the CDC-definition.
2.10 Phenomenoa associated with CFS:
The statement that "premorbid and concurrent depression are common in CFS" must be amended appropriately in the light of the recent data and restated as: "pre-morbid and concurrent depression in CFS are probably as common as other chronic medical conditions of comparable duration and disability". The higher prevalence of depressive disorders in CFS was observed in the studies that did not apply the CDC-case definition and included patients with depression and chronic fatigue as "CFS".
3. Evaluating people with fatigue
3.1 I think we should be stating "Evaluating patients with fatigue".
3.2. What psychological evaluation is required?
The document states that evaluation by a specialist psychologist or psychiatrist "may be useful both for the diagnostic and treatment purposes" (paragraph 1, line 3). This is an inappropriate advice and should be reviewed. Referral to a psychologist or psychiatrist in CFS is only indicated for concurrent or suspected psychological problem in the same way as a neurological referral would be indicated if a disease like multiple sclerosis needs to be excluded. The level of co-morbid psychiatric disease in CFS is not significantly higher than any other chronic medical disease. Furthermore, psychiatric intervention for CFS has been shown to be ineffective in the long term and is poorly accepted by the patients.
4. Managing CFS
4.1 A rehabilitative model:
I am surprised by reading the following statement (last paragraph, last line): "This is in sharp contrast to the widely held but inappropriate belief that prolonged rest and social withdrawal should be advocated". Where did the RACP discover this "widely held belief"? Studies have shown that an average patient with CFS only takes the equivalent of "prolonged rest" for 3 days in a typical month. Most CFS patients would like to avoid social exclusion (preference for social exclusion is uncommon in CFS and common in major depression). I believe this is a good example where the authors of the RACP guidelines have made an erroneous statement solely based on their personal belief rather than the evidence-based science. How would they categorise this statement by the Level of Evidence in their own classification? At the very least, this sentence should be deleted as a mark of respect to the intelligence of the CFS patients.
4.2 Unproven therapies in CFS
The document cites references that indicate as a group, CFS patients are high responders to placebo (30-50%). More recent and randomised controlled pharmacologic trials have shown that the rate of placebo response in CFS is modest or low (arond 20-25%). This section needs to be rewritten.
4.3 (Bulleted list)
Design of any intervention trial of CFS should incorporate a follow up duration of at least a year (12 months) since shorter follow -up periods (3-6 months) might confuse the issue of regression to mean phenomenon with true therapeutic benefit. I have fewer comments on the remaining sections of the guidelines. I must, however, point out that not all children and young adults recover from CFS and severe CFS symptoms have been known among pubertal adolescents (Section on CFS in Adolescents) .
It is understandable that it may be difficult to encompass the current views on a complex and heterogeneous disorder like CFS in a critical and unbiased approach. The efforts of the authors in preparing the RACP guidelines deserve praise and there are some areas in this document that are generally well-written . Sadly, this document contains many flawed statements and observations. In some areas as already pointed out, the accounts appear biased and inaccurate . In addition, I have deep concerns about the selectivity of the literature review in a few places. As an example, while great effort has been taken to discuss the psychiatric co-morbidity in CFS with particular reference to depression, the stated account does not bring home the point that the rate of psychiatric co-morbidity in CFS is comparable to many other medical and neurological conditions and that as a group, CFS patients can be reliably distinguished from depressed patients by clinical assessment and also by using widely validated neuroendocrine tests (urinary cortisol and buspirone test) if required. The document has overemphasised the psychologically driven cognitive-behavioural model of CFS and has failed to review the appropriate literature on the neurology, neuroendocrinology, and neuropsychiatric changes in CFS. This is rather surprising given the fact that "ME" (an alternative term for CFS) is a neurological disease in the current classification of the WHO. The document fails to draw the parallel between CFS and comparable chronic fatigue seen in post-infective neurological disorders like post-polio fatigue, multiple sclerosis and post-Guillain-Barre syndrome fatigue. Instead, the paper has devoted much of its clinical discussion on the comparison of psychiatric disorders like depression with CFS. It is perhaps unforgiving that the document does not mention CFS after ciguatera fish poisoning, an Australian discovery that has provided an important concept in the current understanding of the CFS pathophysiology (ion channel dysfunction).
Other notable omissions in this document are:
- neuroendocrine studies distinguishing depression and CFS (buspirone test)
- orthostatic intolerance (neurally mediated hypotension)
- vestibular, gait and balance disorders in CFS subgroup (Pedersen's syndrome)
- metabolic changes in muscle (e.g. lactate response in sub-anaerobic exercises)
- the nature of cognitive impairment in CFS
- objective evidence of deterioration of physical and mental fatigue after exertion
- higher incidence of new-onset asthma and/or allergy after CFS
The quality of references and review on the neurobiological aspect of CFS is very poor with several important omissions of research carried out by the international groups in the past three or four years (e.g. Natelson, Lane, Dinan and our own). The cited references show a skewed representation of a group of psychiatrists (notably Wessley and Hickie). These failures clearly undermine not only the quality of science but also the quality of advice presented in the guidelines. The section on CFS management fails to recommend pacing, which is the most effective coping strategy for the management of CFS symptoms. No mention is made on the effects of stress, surgery, anaesthesia or pregnancy on CFS. With the exception of the role of analgesics and antidepressants, the pharmacologic treatment of CFS symptoms have been dealt with relaively poorly (e.g. fatigue, chronic pain relief, neurally mediated hypotension etc.)
The guidelines also show a preference to the CBT and graded exercise therapy and ignore the criticisms and the flawed designs of the trials upon which their success has been claimed. CBT as an intervention is no more effective than a purposeful physician-patient relationship where both the physician and the patient are fully informed and educated about our current understanding of CFS. I also have serious reservations in recommending the graded exercise programme to every CFS patient. Although a flexible schedule of physical activity is important in the management of CFS, one must be cautious in advocating the overzealous use of the graded exercise therapy on the basis of our experience in UK. Three national support groups- The ME Association, Action for ME and the 25% Group recently carried out a treatment audit among their membership involving self-reported outcomes of a wide range of pharmacological, non-pharmacological and alternative approaches to management of CFS/ME. Results of these questionnaires indicate that there are major concerns about the use of graded exercises. Analysis of the 2,338 respondents in the largest questionnaire showed that nearly half of these responders (1,214) had tried graded exercise. Among these, 417 patients believed it was helpful and another 187 patients reported no change. However, the disturbing fact was that the highest number of respondents (610; nearly 50% who tried the exercise) reported that the graded exercise had actually made their condition worse. No other treatments being assessed achieved such a high negative rating in any of the questionnaires.
The RACP guidelines on CFS have several shortcomings. Many areas of the text appear highly opinionated in favour of the psycho-behavioural model of CFS. In its current form, this document cannot be recommended for acceptance since it does not reflect the cumulative base of knowledge on CFS.
I would urge the RACP to appropriately revise the guidelines.
I shall be happy to provide clarifications on my comments should this be required.
Abhijit Chaudhuri DM MD MRCP(UK)
Clinical Senior Lecturer in Neurology
Clinical Senior Lecturer in Neurology
University of Glasgow
Consultant Neurologist Institute of the Neurological Sciences
South Glasgow University Hospitals NHS Trust
1345 Govan Road
Glasgow G51 4TF
Alison Hunter Memorial Foundation
PO Box 6132 North Sydney 2059 Australia
Phone/Fax +61 2 9958 6285
All material on the site © AHMF | Sitemap