Chronic Fatigue Syndrome

Chronic fatigue syndrome (CFS), like fibromyalgia and multiple chemical sensitivity, comprises a number of poorly understood signs and symptoms, and whether a patient receives the diagnosis for one or another of these symptom clusters may depend on the specialty of the physician making the diagnosis.¹ Patients with CFS also often fulfill case definitions for these other illnesses.¹ This overlap suggests that these "functional somatic illnesses" may be variants of one another.² However, this does not necessarily mean that these syndromes share the same pathobiological processes or causes. For example, patients with fibromyalgia have been found to have elevated levels of substance P in spinal fluid³ and reduced pain thresholds, while patients with CFS have not.⁴ Similarly, the fatigue reported by patients with fibromyalgia may be secondary to chronic sleep disruption because of pain, while fatigue may be primary in CFS.

The case definitions for CFS^{5 - 6} reflect the observation that the severe fatigue and influenza-like symptoms, which often begin suddenly, were initially thought to represent an underlying viral infection. Thus, the diagnosis requires at least 6 months of new-onset symptoms of fatigue accompanied by infectious, rheumatological, and neuropsychiatric symptoms, and which cannot be explained by other medical diagnoses. In the United States, CFS has a prevalence of 0.52% in women and 0.29% in men.⁷ Patients with CFS may experience severe disability; one study reported that patients with CFS have lower self-reported functional status than a group of similar patients with congestive heart failure.⁸ In this article, I review the evidence for several proposed hypotheses about the etiology of CFS.

The initial hypothesis that CFS represents a form of chronic Epstein-Barr virus (EBV) infection was abandoned when it was found that elevated EBV titers, which reflect prior infection, are not uncommon in healthy people without CFS.⁹ Similarly, there are no convincing data that other infectious agents are associated with CFS. On the other hand, acute infections can evolve into CFS and an influenza-like onset of CFS is more common in the winter than in other seasons.¹⁰ A postviral syndrome that appears similar to CFS is not uncommon following infectious mononucleosis¹¹ and severe viral infection,¹² but persistence of an infectious agent in CFS has not been demonstrated.

Another hypothesis proposes that acute infection may lead to CFS via sustained immunological activation or dysregulation, rather than through persistent infection. Although a number of studies have found some form of immune activation in patients with CFS, these results have been inconsistent.¹³ Rather than causing CFS, such changes may reflect other consequences of CFS, such as inactivity, disturbed sleep, and chronic stress. For instance, 1 study found no evidence of immunological dysfunction in a group of patients with CFS who had been matched with a control group that was equally sedentary.¹⁴ Despite this negative result, some data do support an underlying immunological problem.¹⁵ For instance, some patients with CFS have a dysregulated 2,5 ribonuclease-L antiviral defense pathway¹⁶ and treatment with an immune active agent, mismatched RNA, may reduce disability.¹⁷

A different set of hypotheses proposes that CFS may reflect depression or other psychiatric illness. Patients with CFS and major depression, however, appear to have different profiles from patients with only major depression. Patients with CFS have less self-reproach and more somatic symptoms than do depressed patients,¹⁸ less personality disturbance,¹⁹ and different immunological profiles.²⁰ They are also more likely to display down-regulation of the pituitary-adrenal axis²¹ rather than the up-regulation that often occurs in depression.

Another explanation for CFS is that it is a form of somatization disorder, which may be diagnosed in patients with many years of multiple system complaints without an obvious organic basis. However, this diagnosis is subject to the bias of the examiner. Clinicians who do not believe that CFS has an organic basis are more likely to diagnose somatization disorder among such patients even if they do not meet Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria for somatization disorder. In fact, as few as 5% of patients with CFS may meet these strict criteria for somatization disorder.²² This implies, for instance, that a physician could appropriately diagnose somatization disorder in a patient with lifelong multisystem symptoms, but not in the more common case of a previously asymptomatic patient who complains of persisting and debilitating systemic symptoms following an acute influenza-like illness.

Another study²³ used factor analysis of self-reported symptoms to define CFS and found approximately that one fourth of patients with CFS reported a large number of symptoms and had a high probability of also having other major psychiatric disorders. The remaining patients had both fewer symptoms and less psychopathology. These and similar results have led to suggestions that the case definition for CFS be changed to include only patients with relatively few somatic symptoms.²⁴ Although this strategy would exclude patients who endorse multiple somatic symptoms, it is not known whether it would also exclude patients with somatization disorder or CFS.

Regardless of the diagnostic label, many patients with CFS have psychiatric comorbidity. Clinical experience suggests that treating comorbid psychiatric conditions may relieve these psychiatric symptoms, but does not alter the somatic symptoms; nonetheless, patients with CFS who receive antidepressants for at least 6 months have reduced costs in medical care.²⁵

Wessely et al²⁶ have described a "cognitive behavioral" model that emphasizes the importance of psychological factors in influencing recovery from CFS. They argue that personality factors can interact with illness triggers and subsequent deconditioning to affect the duration of illness. Thus, patients with a tendency to have mood problems or to amplify somatic sensations might become worried about activity-related symptoms following a viral illness and thus further reduce their activity. In terms of beliefs about illness, the prognosis is less favorable when patients make catastrophic statements about their condition or attribute their illness solely to physical factors, such as viral infection, rather than to stress or psychological factors.²⁶

If beliefs, attributions, and fears are important in maintaining a state of chronic fatigue, then treatments aimed at altering these thought processes should improve the health of patients with CFS. Sharpe et al²⁷ and Deale et al²⁸ have reported that cognitive behavioral therapy (CBT) led to improved clinical outcomes in patients with CFS.

While these results support the role of beliefs in maintaining illness in CFS, this does not mean that CFS is necessarily psychological in origin. For instance, CBT can improve the symptoms of patients with other chronic diseases such as rheumatoid arthritis.²⁹ Furthermore, CBT is not always effective in treating CFS^{30 - 31} and its efficacy may be greater in treating patients who are at a relatively low functional level because of a possible "ceiling effect" for high-functioning patients.³² Thus, while some patients may have cognitive and behavioral factors that impede recovery, this is not necessarily true for every patient with CFS.

Although an initial study using tilt testing found that a majority of patients with CFS developed delayed orthostatic hypotension that could be reversed with fludrocortisone, β-blockade, or disopyramide,³³ a subsequent placebo-controlled trial did not find a therapeutic effect of fludrocortisone.³⁴ Furthermore, 2 controlled studies found no difference in orthostatic intolerance between unmedicated patients with CFS and sedentary healthy control subjects.^{35 - 36} One of these studies³⁵ found that patients with CFS had lower cardiac stroke volumes, even in baseline conditions. Whether this finding might indicate covert cardiac dysfunction³⁷ or reduced blood volume³⁸ remains unknown.

Although tilt testing is not a sensitive way to diagnose orthostatic intolerance in CFS, it may be reasonable to monitor heart rate and blood pressure after 5 minutes of supine rest and then every minute for 5 to 10 minutes of standing, as some patients may show a dramatic postural tachycardia or other orthostatic change within this brief time frame.^{39 - 40} If either change is present, it may be reasonable to observe if treatment with β-blockade, an α-agonist, or intravascular volume expansion relieves the patient's fatigue symptoms. Tilt table testing may be a better diagnostic tool for adolescents than for adults. A recent controlled study found that 11- to 19-year-old patients with CFS were highly sensitive to orthostatic challenge.⁴¹

Patients with CFS appear to have significant cognitive abnormalities⁴² and those without psychiatric comorbidity may have the greatest degree of cognitive dysfunction.⁴³ The degree of cognitive abnormality in such patients also appears to be directly related to the amount of impairment in functional status.⁴⁴ A follow-up study found that among patients with CFS and no psychiatric codiagnosis, 66% had abnormalities on a magnetic resonance image of the brain vs 30% of the patients with CFS and a psychiatric codiagnosis, and vs 22% of those in the control group.⁴⁵ The abnormalities were generally subtle and the most common finding was small T2-weighted lesions in the frontal lobes.

These abnormalities appeared to have functional significance in that the patients with these abnormalities reported significantly poorer physical functioning on the Medical Outcomes Study Short-Form 36 Health Status Survey, a common disability assessment tool.⁴⁶ As a whole, these studies suggest that some patients with CFS, principally those without concurrent psychopathology, may have subtle encephalopathy.

Advances in understanding CFS are impeded by at least 3 factors. First, researchers tend to test their favorite theories and rarely perform confirmatory studies. Second, studies often compare data from inactive patients to healthy subjects who might be extremely active and thus differences between groups could reflect activity level or fitness rather than underlying disease. Finally, patients with CFS are a heterogeneous group with frequent comorbid diagnoses (eg, depression, fibromyalgia, irritable bowel syndrome) whose presence could alter study outcomes. Use of stratification strategies to control for such comorbid conditions would address the inherent heterogeneity of such samples⁴⁷ and would help to isolate possible organic etiologies in subgroups of patients with CFS symptoms.