Ellen M Goudsmitb PhD FBPsS
Charles Shepherdc MBBS
a Former co-editor of the ME and CFS
b Health Psychologist (retired)
c Hon Medical Adviser ME Association
Myalgic encephalomyelitis (ME) is a disabling condition characterised by profound fatigue following minimal exertion and a delay in recovery after exertion ends. In 1988, when specialists introduced the concept of chronic fatigue syndrome (CFS), it was assumed that the illness in question was identical to ME but the lack of consensus criteria for the latter has prevented the testing of this assumption. In this article, we propose criteria and guidelines which can be used to study ME and determine whether it is a synonym for CFS or one of several subgroups contained within the CFS construct. If clinically meaningful differences are identified, the criteria may help to increase diagnostic precision and facilitate further research into the cause, course and treatment of ME.
Chronic fatigue syndrome (CFS) is a disabling condition characterised by profound fatigue of new onset, plus four or more listed symptoms, namely lymph node pain, sore throat, muscle and/or joint pain, post-exertional malaise, headaches, unrefreshing sleep and impaired memory or concentration.
The term was introduced in 1988 to replace a number of different names for what appeared to be the same symptom complex, previously referred to as post-viral fatigue syndrome, epidemic neuromyasthenia, myalgic encephalomyelitis (ME), Royal Free Disease and Icelandic Disease.[2,3] A working case definition for the new construct was devised for the Centers for Disease Control and Prevention (CDC) in order to provide more diagnostic consistency and improve the comparability and reproducibility of clinical research. However, a subsequent meeting of the specialists advising the CDC concluded that the criteria did not effectively distinguish CFS from other types of unexplained fatigue and the case definition was revised.
The new diagnostic algorithm quickly became the gold standard, both in research and clinical practice  but there has been increasing concern about the sensitivity, specificity, and reliability of the various criteria. For example, two studies found that people who met the 1988 case definition reported a larger number and more severe symptoms than those who fulfilled the 1994 revision.[7,8]
Marked differences were also identified within the samples selected using the same criteria. For instance, Kennedy et al.  assessed three groups of patients who all met the 1994 criteria but whose symptoms had a different mode of onset. The people whose CFS had developed after service in the first Gulf War were found to have significantly more physical symptoms (fatigue and pain) and significantly reduced muscle power in the lower limbs than those who had become ill following exposure to organophosphate insecticides, and the group with other aetiologies.
An additional flaw was identified by King and Jason  who compared patients with CFS and major depressive disorder and were not able to differentiate between the two based on the prevalence of seven out of the eight minor criteria. These findings not only demonstrated the heterogeneity of the population with CFS as currently defined, but also revealed a strong potential for individuals with another disorder to be misclassified as having CFS. Likewise, they challenge the assumption of equivalence, i.e. whether CFS can still be compared to conditions such as ME and post-viral fatigue and accordingly, if interventions devised for the former are appropriate and equally effective for the latter.[5,10,11]
ME is a condition which was first described in the literature in the 1930s. However, it first came to prominence after a series of outbreaks, the most famous of which closed the Royal Free Hospital for three months in 1955. For many years, the illness was thought to be caused by an enteroviral infection [13-16] but to date, research has not been able to differentiate between ME and syndromes triggered by other viruses (e.g.17,18].
The onset is often but not always acute and symptoms indicative of an infection may be accompanied by myalgia, lymphadenopathy, or a gastro-intestinal upset. . However, instead of recovering, patients begin to experience profound fatigue following activities which were previously completed without difficulty. According to Ramsay  who studied ME from 1955 until his death in 1990, muscle fatiguability “is the dominant factor and without it a diagnosis should not be made”. Also typical is a prolonged delay in the restoration of muscle power which Ramsay described as “a striking and possibly unique feature of this type of fatigue”.[20,21] The muscle fatigue following minimal activity plus delayed recovery is not recognised in the criteria for CFS but is a core symptom of classic ME.[2,22]
As in CFS, patients report a variety of other symptoms. Prominent are a flu-like ‘malaise’ and neurological symptoms such as disequilibrium and vertigo.[19,23,24] Patients may also experience sensory disturbances such as paraesthesiae, tinnitus and hyperacusis as well as visual abnormalities such as photophobia  and/or increased sensitivity to certain patterns. Neuropsychological symptoms associated with ME include cognitive problems including loss of shortterm memory, an inability to concentrate and difficulties with word finding i.e. anomia and dysnomia. The deficits are often severe and a major source of disability. In one of the few studies where all participants were diagnosed using criteria for ME rather than CFS, the performance on a selection of tests revealed scores that fell between those of patients with mild and moderate Alzheimer's Disease.
Aside from the fatigue, muscle weakness and evidence of central nervous system (CNS) dysfunction, there may also be symptoms associated with impaired circulation. This manifests itself in cold extremities, perceived low temperatures and a sudden facial pallor.  In addition, many patients report a disturbance in thermoregulatory control e.g. feeling weak after a hot bath [2,19,29] as well as an intolerance to alcohol.[19,30] All these symptoms show a marked diurnal and cyclical variability in their intensity, and although it is not always possible to identify a specific cause for the exacerbations, reports suggest that the condition often worsens as a result of over-exertion, concurrent infections, and in some cases, 'stress'.[17,23].
Studies of the symptoms reported by patients indicate that ME and CFS cannot be differentiated in terms of the prevalence or severity of fatigue. Moreover, there appears to be no significant difference in the percentage of individuals with myalgia, headaches, sleep disturbance, gastro-intestinal symptoms, or emotional distress.[7,23,32]. There are no data for multi-joint pain to allow a meaningful comparison, but there is evidence suggesting that people with ME experience a greater number and more severe neurological and cognitive symptoms. [28,33,34] In contrast, research has indicated that post-exertional fatigue, a characteristic symptom of ME, is not as common or as disabling in patients with CFS.[32-35]
Aside from the subtle differences in the symptom complex, there are other considerations which challenge the assumption of equivalence. For instance, ME/post viral fatigue is much less common than CFS: 1 per 1000 versus 10-26 per 1000 in the UK, respectively.[36,37] Also noteworthy are the epidemics of ME which provide strong evidence of a shared, infectious aetiology. In contrast, research into CFS has identified a more heterogeneous population with subsets showing associations between their symptoms and reports of childhood trauma, over-active lifestyles and chronic stress.[5,38,39]
It has been posited that traumatic events and a failure to cope with recurring stressors makes individuals more vulnerable to infection and undermines recovery. However, research has not identified a higher incidence of ME or post-viral fatigue in countries with a history of long term conflict (e.g. Israel), or in populations with a higher prevalence of trauma-related disorders, e.g. the children of Holocaust survivors (Glassman, personal communication). More significantly, research on CFS has not yet replicated the finding of the delayed recovery of muscle strength after minimal exertion. To summarise, the evidence suggests that infection and immune dysfunction could play a greater role in the aetiology of ME [34,40] while more cases of CFS may be attributed to both the physiological effects of chronic stress and psycho-social factors.
The heterogeneity of the population with CFS was acknowledged by the International Study Group advising the CDC as far back as 1994 and resulted in the recommendation to stratify samples and compare of subsets. This led to several longitudinal studies following patients with a variety of infectious diseases as well as other illnesses resembling ME.[40,42-45] However, further research into ME has been limited by the lack of standardised criteria.
To explore whether ME is a discrete disorder or subgroup of CFS and to answer many other questions relating to this condition, a number of scientists from the United Kingdom formulated guidelines based on many years of clinical experience and at the time, the most consistent descriptions of the illness [e.g. 2,22]. The first version was devised for two national patient groups in the 1990s and used in a number of studies on ME [e.g.46,-49]. Following new evidence, particularly in relation to the link of symptoms with exertion, they were revised and updated to permit research to continue.
The only alternative is the International Consensus Criteria or ICC. The ICC initially appeared to be resolve a lot of problems relating to heterogeneity of the population, describing many of the symptoms of classic ME and avoiding the vagueness of the term ‘malaise’ used in case definitions for CFS. The ICC are more specific, referring to a pathological inability to produce sufficient energy on demand, “postexertional symptom exacerbation”, “postexertional exhaustion”, “a marked, rapid physical and/or cognitive fatigability in response to exertion”, a prolonged recovery period and “a low threshold of physical and mental fatigability (lack of stamina) resulting in a substantial reduction in pre-illness activity level”. Postexertional neuro-immune exhaustion (PENE) as they named this characteristic feature, had to cause marked disability and was compulsory for diagnosis. However, it's not easy to assess in practice. In a recent study by Jason et al. , more than 10% of the patients selected using the ICC did not report that minimal exercise made them tired. This is possibly because the researchers required only one of the characteristics of PENE to be present, rather than all.
A second study from Jason and his colleagues also compared the CDC criteria with the ICC and showed that while the latter identified a subset of patients with more functional impairments and physical, mental and cognitive problems, they also had a higher rate of psychiatric illness. Indeed, 61.5% had a current psychiatric diagnosis compared to 27% who fulfilled the CDC criteria. Again, this may be a result of patients reporting evidence of post-exertional worsening but not meeting every characteristic of PENE. This 'short-cut', also likely to occur in clinical practice, might have led to misclassification.
For example, another interesting finding was that more than a half had a gradual onset. This plus the higher rate of psychiatric illness are at odds with descriptions of individuals with classic ME, more of whom report an acute infectious onset and symptoms such as muscle weakness and an intolerance to alcohol.[19 p.147, 23,50] They also tend to have lower rates of psychiatric illness.[22,34,53] According to Jason et al. , the requirement of a larger number of symptoms (at least eight for the ICC versus five for the CDC criteria) might “inadvertently increase the rate of psychiatric morbidity” and they recommended further refinement of case definitions focusing on a small set of core symptoms. Finally, based on a review of all criteria, they noted some of the disadvantages of polythetic case definitions that include patients on the basis that they experience a given number of listed symptoms, many of which are ubiquitous and do not allow clinicians to differentiate between CFS and other disorders.[6,33,50]
All patients should fulfil the following five
1. A new onset of significantly abnormal levels of muscle fatiguability and/or muscle weakness, precipitated by relatively minor levels of activity. Symptoms typically worsen during the next 24-48 hours.
2. The presence of symptoms indicating the involvement of the brain and central nervous system (e.g. impaired short-term memory and concentration, disturbed sleep patterns, balance problems).
3. Periods of impaired circulation compatible with autonomic dysfunction (e.g. facial pallor, disturbances in thermoregulation including inappropriate sweating and sensitivity to both heat and cold; postural hypotension and/or orthostatic intolerance).
4. Fluctuation of symptoms, from hour to hour and day to day.
5. These symptoms must have been present during the past three months (to exclude patients with the debility which often follows illnesses such as influenza).
Many symptoms experienced by people with ME are also reported by people with other disorders. The most prevalent of these include pain – which can be muscular, arthritic or neuropathic in character; hyperacusis and tinnitus; photophobia and blurred vision; frequency of micturition and hypersensitivity to chemicals and drugs. Also common are symptoms suggestive of immune system dysfunction and/or persisting infection, such as episodes of low-grade fever (not exceeding an oral temperature of 38.6C) combined with feeling feverish; sore throat which may be persistent or recurrent, and arthralgia.
While the presence of these symptoms are not discriminative, the marked diurnal fluctuations in severity are typical of ME. Exacerbations are frequently triggered by physical or mental exertion and this association should always be sought whilst taking the history.
Characteristic physical signs are sometimes seen in ME and in combination with the symptoms above also contribute to the validity of the diagnosis. Nevertheless their absence does not exclude the condition. They are as follows:
2. Tenderness and possible enlargement of lymph nodes.
3. A positive Romberg test or Fukuda test.
Although ME often follows an infection, usually a
viral illness (which may be sub-clinical), it may also be triggered by other
factors such as immunizations, trauma and exposure to chemicals. Furthermore, in
a minority of patients, ME has a gradual onset with no apparent triggering
factor. For these reasons, and in line with the criteria for CFS, evidence of a
preceding viral illness is not a prerequisite for diagnosis or inclusion in a
Because it is vital that the samples used in research are as 'pure' as possible, the presence of certain co-morbid disorders would be grounds for disqualification. Some of the more common alternative diagnoses to be borne in mind before selecting a participant for a study on ME are listed in Shepherd and Chaudhuri[54,p.15]
It is of particular importance to identify cases of chronic fatigue which can be largely explained by psychological factors. For example, if there are signs of persistent anhedonia, apathy, low self-esteem, feelings of worthlessness and guilt, the possibility of primary depressive illness should be considered and, if there is any doubt, the participant should be excluded from the study. Similarly, if the patient has had any other illness or undertaken any treatments - orthodox, complementary or nutritional - in the previous three months, their inclusion may introduce additional variables which could confound the results.
We recommend that the Profile of Fatigue-Related Symptoms (PFRS) developed by Ray et al.  may help support the diagnosis and suggest that a person should score at least 2 out of six on the two items relating to the presence of muscle weakness. This is a core symptom and the score provides additional evidence of its presence and severity. The testing of muscle fatigue on two occasions, at least 24 hours apart, may also be of value.
A protocol was developed by Dr Parish, a rheumatologist who has studied ME since 1955. Using criteria almost identical to those proposed above (Wood, personal communication), Paul et al.  found a significant decline in quadriceps strength which persisted after 24 hours in all the participants. Moreover, there were significant differences between the patients and controls during the recovery phase, 200 minutes following exercise, when the results of the latter had returned to normal, and after 24 hours. This protocol is different to others that have evaluated post-exertional fatigue in that it demands a continually increasing energy cost throughout the duration of the exercise period, and therefore differs from the steady levels achieved in cycle ergometry used in other studies.
However, one cycling exercise test that deserves further consideration was recently described by Snell et al. Although the patients were diagnosed with CFS, all reported a worsening of symptoms after physical activity. Various tests using a cycle ergometer were conducted to evaluate work efficiency, i.e., oxygen consumption and work output at the ventilatory/anaerobic threshold. The results revealed lower workloads and oxygen consumption at peak exercise compared to sedentary controls but none of the individual tests helped to discriminate between the groups on day 1. However, data from the second session showed a decrease in performance in patients that was not seen in controls and identified two measures which contributed most to the discrimination between the groups (workload at ventilatory threshold and peak workload). The researchers noted that their findings may have been affected by the heterogeneity of the sample but the results helped to correctly classify over 95% of the participants and demonstrated the importance of a second test.
Formulating a definition of ME and CFS has proven a challenging task. Without a definitive test, all one can do is regularly review the current working definitions and build a consensus based on clinical experience and best evidence. These criteria were devised to improve diagnosis and allow researchers to compare and contrast ME and CFS. This will not only enable scientists to test the assumption that the two conditions are identical, but they should also help to clarify whether interventions which have been shown to be effective in one can be safely used to treat the other. Finally, if evidence indicates that ME is a discrete entity or a distinct subgroup, the criteria may not only prove to be of value in research, but also in clinical practice.
The authors wish to thank all our colleagues who commented on the previous versions and the patients who participated in our studies on ME.