Dear Sir,First and foremost, we would like to thank Drs van der Meer and Lloyd  for their careful and thorough review of the International Consensus Criteria (ICC) paper . We support this type of open discussion and strongly agree that only in this way will the basic and clinical science of myalgic encephalomyelitis (ME)/chronic fatigue syndrome (CFS) evolve soundly. This being said the content of their feedback also leads us to believe that several elements have been taken out of context or that we may not have articulated these components as well as we had hoped. As a result, we fully appreciate this opportunity to rectify any such miscommunication.
We certainly agree with Drs van der Meer and Lloyd that this is a controversial field of investigation into a very complex and poorly understood illness that also suffers from significant social stigmatization. Their statement that, ‘Quite a bit of mediocre research has been done in relation to CFS’ is unduly harsh in our opinion; as with any field of investigation, the case definitions and laboratory assays evolve appreciably over time. We would argue that this field has advanced significantly over the past years and continues to do so, at least in part owing to this open exchange of views. Consistent with this, our ICC document was submitted as part of an ongoing development, which will include clinical guidelines and an International Symptom Scale (ISS). The ISS was not referenced because, as indicated, it is a future project. As stated in the paper, ‘the primary goal of this consensus report is to establish a more selective set of clinical criteria that would identify patients who have neuroimmune exhaustion with a pathological low threshold of fatigability and symptom flare in response to exertion’. Although, as indicated, the ICC is directed primarily at clinical assessment in its current format, we certainly look forward to the International Symptom Scale being fully developed so that these elements may be deployed in the selection of more homogeneous sets of research subjects.
The International Consensus Panel clearly states that the scope of the paper entitled, ‘Myalgic Encephalomyelitis: International Consensus Criteria’ (ICC) is ‘limited to the criteria of ME and their application’. A discussion of therapies such as cognitive behaviour therapy (CBT) and graded exercise therapy (GET) raised by Drs van der Meer and Lloyd was not within the scope of the ICC paper. The cardinal symptom of ME is a pathological low level of threshold of fatigability that is characterized by a specific pattern of inability to produce sufficient energy on demand associated with measurable, objective, adverse responses to exercise. This is a refinement of the much more inclusive criteria of fatigue that goes by the label of CFS and may or may not include ME. The panel did not intend to dismiss the broader range of components involved in fatiguing illnesses but instead focused on a patient population best described as having pathophysiological abnormalities that arise in response to exertion. It is precisely the issue of patient cohort heterogeneity raised by Drs van der Meer and Lloyd themselves that we hope to address with a more specific criteria and efforts such as this one directed at improving the clinical identification of ME patients specifically. One only has to question why 5 of 15 CFS patients in recent phase II clinical trials of rituximab were unresponsive to treatment to appreciate the need for more specific criteria .
We believe that a prime cause of inconsistent and confusing research findings in this heterogeneous patient population has been the use of overly inclusive criteria for CFS such as the Oxford  and Reeves criteria , because each study may include very different patient sets. This dilutes the results, makes them difficult to confirm, leads to scepticism and thus inhibits progress. Jason et al.  found the Reeves empirical criteria to be flawed because it was possible to meet the CFS criteria without any physical symptoms, only 10% actually had ME/CFS and 38% of patients with Major Depressive Disorder were misclassified as having CFS. The increased estimates of CFS from 0.24% (Fukuda) to 2.54% (Reeves, USA) and 2.6% (UK) confirm that the Reeves and Oxford criteria select patient sets that are approximately ten times larger and more inclusive than those selected by the Fukuda criteria . Whilst some general information about fatigue may be ascertained using these inclusive criteria, it is not productive to try to study the mechanisms of ME if up to 90% of the research patient set do not meet the criteria for ME. The Fukuda criteria, in which postexertional malaise is an optional criteria, is based on what was known in the early 1990s and has been used in the majority of research studies. In comparison with the Fukuda criteria, the Canadian Consensus Criteria , which made postexertional malaise compulsory, differentiated ME patients from those who were depressed and selected patients with greater physical and cognitive functional impairments . Following the publication of the 2003 Canadian Consensus Criteria, studies confirmed their utility in this regard. The WHO International Classification of Diseases (ICD) 2010 further clarifies this issue of general fatigue. This falls under WHO ICD R53 – malaise and fatigue. However, ME is classified as a neurological disease, ICD G93.3. The WHO stipulates that the same condition cannot be classified under more than one rubric because individual categories and subcategories must remain mutually exclusive by definition. Therefore, ME (ICD G93.3) must be removed and kept separate from the overly inclusive criteria for chronic fatigue and malaise (ICD R53), which actually states that G93.3 is an exclusion of R53.
The 2011 ICC for ME continues the direction of refinement undertaken in the Canadian Consensus Criteria and incorporates more recent knowledge and understanding. More specifically, the rationale of the ICC document advances the successful strategy of the Canadian Consensus Criteria of not viewing symptoms isolated in a nominal list, but rather as coordinated patterns of symptoms that directly reflect the regulatory interactions of the underlying systems involved. If the same symptoms consistently flare in response to exertion, they are more likely to share a common cause. For example, if a patient consistently experiences flu-like symptoms, sore throat and tender lymph nodes in response to exertion, it suggests that immune activation is a component of their underlying pathophysiology, which then can be studied scientifically. Given that the Oxford criteria are based on general chronic fatigue and it is possible to meet the Reeves criteria without having any physical symptoms, it is surprising that the ICC was admonished for its use of subjective symptoms. Symptoms by definition are subjective; however, the clinician can observe visible signs that accompany symptoms such as swollen lymph nodes, crimson crescents in tonsillar fossae, abnormal accommodation responses of the pupils, abnormal body temperature or blood pressure, abnormal tandem gait and positive Romberg test, thus confirming symptoms. In most neurological clinics, muscle weakness and ataxia are valid symptoms/signs that are used as part of a diagnosis for a variety of conditions. Further, objective serial stress tests recorded by heart monitors and magnetic resonance spectroscopy (MRS) can confirm many symptoms, e.g., postexertional exhaustion and pain, decreased cerebral oxygen and significantly reduced anaerobic threshold heart rate, VO(2) peak and peak work, prolonged recovery period, and loss of capacity to recover from acidosis [10–19]. Tilt tests can identify orthostatic intolerance (OI) [20–22]. The use of 24-h Holter heart monitors and cardiac MRS can identify abnormalities in heart function [23–26]. SPECT scans can identify cerebral hypoperfusion, and other brain imagery tests can identify additional brain abnormalities [27–31].
The ICC identify patients who have measurable and reproducible pathophysiological abnormalities in response to exertion, which people with general fatigue or depression do not have. In contrast, the Reeves and Oxford criteria do not require postexertional malaise or exhaustion in their definitions, and it is optional in the Fukuda criteria. The more general and stereotypic the criteria, the less useful they become because lumping together patients whose chronic fatigue is an integral part of many other diseases skews both clinical and research findings. It is prudent to study patients who have various kinds of adverse pathophysiological reactions to exertion to isolate various pathogeneses responsible for them and determine appropriate treatments. The use of patient sets selected by relevant, well-defined and consistent ICC for ME in research will thus advance science, provide greater clarity of understanding and elicit more reproducible scientific results. This will also allow a selection of more homogeneous sets of ME patients, given the current knowledge, which can then be compared with other populations. Whether patients with less severe conditions represent a continuum, faulty diagnosis or different disease entities can only be determined by future studies.
Another element of debate concerned the use of the name ‘myalgic encephalomyelitis’. We submit that this name had been in existence for decades before the coining of the label of ‘chronic fatigue syndrome’. The suffix ‘itis’ has not resulted in research being restricted to inflammatory mechanisms and the efficacy of various anti-inflammatory drugs. Although the name ‘myalgic encephalomyelitis’ may not be perfect, it is the most accurate and appropriate name available and indicates underlying pathophysiology. Obviously, the ethical implications and medical risks prevent brain and spinal biopsies, thus limiting direct evidence. However, spinal autopsies have identified neuroinflammation of the dorsal root ganglia (Chauduri A. Royal Society of Medicine Meeting 2009). There is simply too much evidence of pathophysiologic neurological and immune dysregulation, immune activation and an imbalance between inflammatory and anti-inflammatory mediators to be ignored [32–56]. Unfortunately, the name ‘CFS’ and its hybrids ‘ME/CFS’ and ‘CFS/ME’ have been used to refer to both ME and general chronic fatigue. The best way to end the resulting confusion is to only use the name ME for those who meet the more restrictive ICC criteria for this very serious disease, which is consistent with the WHO ICD neurological classification.
The significant misalignment between the focus of this consensus paper, which is limited to ME, and the commentary of Drs. van der Meer and Lloyd appears to be centred on the use of the much more inclusive criteria for CFS or general fatigue. This poignantly establishes the urgent need to clarify this issue by distinguishing patients who meet the ICC criteria for ME from those that satisfy the broader and more inclusive Oxford, Reeves empirical criteria and the NICE criteria. Not only it is common sense not to mix diseases classified under various rubrics in one heterogeneous pot, but it is in keeping with the WHO classification rule that a disease cannot be classified under more than one rubric. Research on other fatiguing illnesses, such as cancer or multiple sclerosis (MS), is carried out on patients who have those diseases. It is imperative that research for ME be carried out on patients who actually have ME. When advances in scientific technology are applied to patients who meet the more specific case definition of the ICC for ME, the current urgent need for identifying and confirming specific biopathological mechanisms and biomarkers will be facilitated, and our improved understanding of the pathophysiology can then be directed towards enhancing treatment efficacy.
Scientific debate about the relative merits of various hypotheses is constructive and healthy. However, it is important to avoid confusing patient stratification with scientific bias. To do so is to ignore a well-justified and extensively used strategy for the study of complex disease populations. The distinguished clinical, research and pedagogical status of our panellists has been clearly laid out in the document. Clinicians and researchers who had extensive experience in this area were asked to donate their time and expertise to the panel, and they in turn were asked to suggest others. The authors’ opinions did vary in some areas, opinions were discussed, and only those criteria that were agreed upon by all authors were included in the final manuscript. The panel does not claim to represent the entirety of the field. However, we suggest that consensus amongst a panel of investigators of this size, geographical distribution and diverse specialties and backgrounds has merit and represents a valuable contribution to the field.