Chronic Fatigue Syndrome and Herpesviruses: the Fading Evidence

THERE IS ACCUMULATING EVIDENCE AGAINST HERPESVIRUSES BEING AETIOLOGICAL AGENTS OF CHRONIC FATIGUE SYNDROME (CFS). THE APPARENT UNUSUAL SEROLOGICAL RESPONSE TO HERPESVIRUSES IN CFS PATIENTS IS MORE LIKELY TO REPRESENT AN EPIPHENOMENON OF A MORE GENERALIZED IMMUNOLOGICAL NATURE

SUMMARY
Herpesviruses, in particular Epstein-Barr virus (EBV), cytomegalovirus (CMV) and human herpesvirus 6 (HHV-6), have, for the past two decades, come under considerable scrutiny as aetiological agents of chronic fatigue syndrome (CFS). However, virological findings of herpesviruses in CFS have not been consistent between different studies, and the unusual patterns of serological responses to EBV, CMV and HHV-6 have not been specific for CFS, being observed also in asymptomatic individuals. In addition, patients with symptomatology suggestive of CFS do not appear to have an increased frequency of these herpesviruses, as detected by culture or polymerase chain reaction, compared with controls, which argues against an ongoing active herpetic infection. Studies have also shown that the presumable elevation of antibody titres to EBV, CMV or HHV-6 in CFS are not observed only with these viruses, but also with other organisms such as herpes simplex virus and measles.

Historical Background
FATIGUE HAS, FOR CENTURIES, been a concern and matter of speculation among patients, practising clinicians and scientists. Hippocrates recognized the muscular ‘fatigue pains’ that could result from the slightest exercise in persons who are deconditioned.1 In the 18th Century the occurrence of chronic, clinically significant fatigue was first articulated, and by the mid-19th Century the term ‘neurasthenia’ was introduced to reflect the belief that exhaustion derived from the loss of nervous energy.2 Fatigue following acute febrile illnesses was noted in military personnel serving in the US Civil War (1861–1865) and in World War I (1914–1918).3 In the 1930s, chronic fatigue was associated with recovery from brucellosis, while fatigue following infectious mononucleosis and influenza was described in the late 1940s to the mid-1950s.2 During these decades, reports of epidemics, in Iceland, Britain and the USA, of a febrile illness that resulted in protracted fatigue were described. The cumulative experience demonstrated striking similarities both in clinical presentation and in the demographics of those affected with chronic fatigue, encouraging further speculation and research on aetiological agents. The illness was not lethal; it was a morbid relapsing syndrome, which mostly affected young women.1 The inability to assign a unifying cause of the disorder led to its being named ‘chronic fatigue syndrome’ (CFS). A case definition was proposed in 1988,4 and redefined in 1994.5 Despite multiple studies of theoretical causative agents or markers, the aetiology of this complex syndrome remains unknown. The case history in the shaded panel (below) illustrates the complexity of symptoms referred to by patients with CFS: This case history exemplifies a common presentation of a patient with CFS who meets the revised Centers for Disease Control and Prevention (CDC) case definition (Table 1). The disorder is characterized by self-reported, severe disabling fatigue, impairments in concentration and short-term memory, sleep disturbances and musculoskeletal complaints. The diagnosis can be made only after alternative medical and psychiatric causes of chronic fatigue have been excluded. No pathognomonic signs or diagnostic tests for this condition have been validated. An extensive history covering medical and psychosocial circumstances, a mental examination and a thorough physical examination are essential in the evaluation of these patients. A minimum battery of laboratory screening tests should be obtained at least once, to include the tests performed on the 35-year-old woman described above. Routine application of other screening tests is of no value, although they may be indicated on an individual basis to confirm or exclude other diagnoses. This paper reviews the data and the controversies relating one putative set of aetiological agents to CFS, the herpesviruses – in particular, Epstein-Barr virus (EBV), cytomegalovirus (CMV) and human herpesvirus type 6 (HHV-6).

EBV Infection
Epstein-Barr virus is a ubiquitous herpesvirus, which infects over 90% of the population worldwide. When EBV infections occur during childhood, they are largely asymptomatic. EBV causes infectious mononucleosis in about one-third of the cases in which primary infection develops during adolescence or adulthood. After recovery from a primary infection, a latent, persistent EBV carrier state is established in B lymphocytes. Frequently thereafter, EBV may reactivate from latency and result in oropharyngeal excretion.6 A triad of lymphadenopathy, fever and pharyngitis characterizes classic infectious mononucleosis. Other symptoms may include malaise, headache and a transient splenomegaly in over 50% of patients. The diagnosis of acute EBV infectious mononucleosis is based on this clinical presentation, together with the associated haematological findings of atypical lymphocytosis and the presence of heterophile antibodies. Epstein-Barr virus infection stimulates the development of antigen-specific antibodies which are useful targets for serological assays, particularly in the setting of acute heterophile-negative presentations (around 10% of cases). Immunoglobulin M antibodies to the EBV viral capsid antigen (IgM anti-VCA) appear quickly after infection, persist for a period usually lasting around 1–4 months and disappear, constituting the best indicator of a recent primary infection, while anti-VCA immunoglobulin G (IgG) antibodies remain for life, providing definitive evidence of past infection. Antibodies to early antigen diffuse pattern (anti-EA-D) are detected in about 70% of people late in the course of infectious mononucleosis and usually disappear after recovery. Antibodies to early antigen restricted pattern (anti-EA-R) are occasionally seen in infectious mononucleosis, arise after anti-EA-D antibody titres peak, and often remain detectable for a few years. Antibodies to EBV-encoded nuclear antigens (anti-EBNA) appear late in primary infection, so that its absence from the serum of a patient with symptoms of infectious mononucleosis suggests a recent EBV infection. In addition to it being the major cause of infectious mononucleosis, EBV is associated with Burkitt’s lymphoma, nasopharyngeal carcinoma, a posttransplantation lymphoproliferative disorder and an often fatal X-linked lymphoproliferative disorder, among other rarer disorders.7,8 Its association with CFS has been controversial.

EBV and CFS
In 1975, Horwitz et al.9 described three patients, among a group of 14 individuals with infectious mononucleosis, who remained ill for up to 28 months and persistently manifested the presence of antibodies to EA-R and elevated levels of IgG anti-VCA antibody titres, suggesting continued EBV activity (Table 2). In 1982, Tobi et al.10 reported seven patients suffering from (among other complaints) fever, malaise, lymphadenopathy, myalgia, hepatomegaly, elevated IgG anti-VCA and anti-EA geometric mean antibody titres compared with healthy controls. Eight of 67 individuals had detectable IgM anti-VCA, and 13 of 67 lacked antibodies to EBNA. These serological findings were considered comparable to those occurring in various immunodeficiencies which, by that time, had been related to active or reactivated infection and an inadequately contained EBV infection. Finally, in an investigation of a cluster of patients complaining of prolonged or recurrent fatigue in Nevada, USA, the CDC reported that 15 affected individuals had significantly higher antibody titres against various components of EBV compared with a matched control cohort.14,15 By 1986, these case series had stimulated the widespread application of EBV serological testing in patients with chronic fatigue. Fortunately, longer prospective studies began to enable a more definitive assessment of the implications of the virus. It was already apparent from the early reports that there was considerable variability in serological test results, with a wide range of values in the general population, indicating that ‘normal’ responses to EBV antigens can vary substantially.11,13–15 In particular, the CDC investigators commented on the highly variable results provided by different commercial laboratories, raising doubts about the reliability and the precision of serological tests for EBV.14,15 For example, raised titres of IgM-VCA antibodies were reported by Tobi et al.,10 but were infrequently detected in other studies.11–13 Although there has been some consistency in the literature regarding the presence of anti-EA antibodies, some studies have reported elevated titres of antibodies against EA-R,9,10 whereas other studies have documented either raised titres of EA-D antibodies only12 or raised titres of both.13 Absence of serological responses to EBNA was noted in few individuals12,13 and not in all studies.9–11 Taken together, these data revealed a disconcerting lack of consistency. The initial reports linking EBV with chronic fatigue encouraged Horwitz et al.16 to assemble the data that had been collected during 8 years of prospective assessment of patients following recovery from infectious mononucleosis. He noted high anti-VCA IgG (>1:320) and restricted anti-EA (>1:10) antibody titres in 27% and 13% of subjects, respectively, into the range considered to suggest a chronic EBV infection. Elevated anti-EA titres had been demonstrated in up to 55% of asymptomatic pregnant women,17 and were noted to be more frequent in older adults.18 Contrary to the initial notion that EA-R antibody titres rarely arise with infectious mononucleosis, Fleisher et al.19 documented their presence in nearly half of a cohort of college students with infectious mononucleosis. Buchwald et al.20 reported a high frequency of unexplained fatigue in an unselected group of patients in a general practice, and a lack of correlation between fatigue and abnormal EBV serological findings. There are two reports of a group of patients with chronic fatigue and high titres to IgG anti-VCA or anti-EA who were clinically indistinguishable from a group of patients with chronic fatigue and low titres.20,21 In addition, field studies showed the presence of antibodies to EA and IgM-VCA even in some entirely asymptomatic persons.22 Despite the early data associating unusual serological responses to EBV antigens and chronic fatigue, no studies showed a correlation between a specific antibody titre and the presence or severity of any symptoms.12,13,23 Moreover, the improvement in symptoms reported by some CFS patients in a randomized, double-blind placebo-controlled aciclovir trial was not related to treatment arm. More importantly still, EBV serological titres were remarkably stable throughout the study, and no clinical, immunological or serological features distinguished the patients who improved from those who did not.24 Finally, EBV can be recovered from throat washings and peripheral blood mononuclear cells in well over 50% of those with acute infectious mononucleosis, organ transplant recipients or AIDS. In CFS, however, the frequency of isolation of EBV from the oropharynx and blood has been found to be far lower, and similar to that of asymptomatic seropositive controls.13,25 Although the cumulative data do not support a role for ongoing EBV infection in CFS, they do not exclude the possibility that primary EBV infection can trigger CFS. A prospective cohort study conducted by White et al.26 determined a relative risk of 2.7–5.1 for CFS 6 months after glandular fever or a glandular fever-like illness, compared with chronic fatigue following an ordinary episode of an upper respiratory tract infection.

CMV and CFS
Cytomegalovirus is another herpesvirus for which infection is widespread and usually asymptomatic. Clinically, CMV may cause a congenital syndrome in newborns and, in normal immunocompetent adults, it is recognized as the causative agent of acute benign infectious mononucleosis. In the immunocompromised, e.g. organ transplant recipients or individuals with AIDS, CMV can cause its most grave syndromes – affecting the lung, the gastrointestinal tract, brain and eyes. As with other herpesviruses, after resolution of a primary infection, CMV remains latent in tissues for life. There have been occasional reports of low-positive IgM anti-CMV titres in patients with CFS.11 Higher CMV-IgG titres or geometric mean titres in CFS patients, compared with control groups, have also been reported in several studies.14,15,24 In general, however, there is no evidence of active CMV infection in individuals with chronic fatigue. The isolation of a CMV-like organism from a patient with CFS has also been reported. Virus particles suggestive of CMV were detected on electron microscopy, and sequence analysis for these virus-like organisms revealed a partial homology with a CMV gene.27 These observations have not been reproduced or validated.

HHV-6 and CFS
In 1986, Salahuddin et al.28 reported the isolation of a novel human herpesvirus in peripheral blood mononuclear cells (PBMCs) from patients with diverse lymphoproliferative disorders. The agent was originally named the human B-lymphotropic virus because it was believed to infect only B cells. Subsequent studies demonstrated infections in other cell types29 and latent carriage in macrophages and PBMCs.30 As the sixth human herpesvirus to be discovered, it is now known as HHV-6, denoting its position. Because of its novelty and its ability to establish life-long infection it, too, became a putative agent to consider in CFS. Human herpesvirus 6 infects almost everybody during childhood, usually without causing any illness. It causes exanthem subitum (also known as roseola infantum), a benign illness of infants. Exanthem subitum is clinically manifested by fever, sometimes accompanied by mild upper respiratory symptoms and cervical lymphadenopathy, and a characteristic maculopapular rash on the trunk and neck that appears following defervescence.31 Meningitis, febrile seizures and even encephalitis may complicate exanthem subitum. In addition, mild mononucleosis-like illnesses,32 pneumonitis and hepatitis have been associated with HHV-6 infection. There is also speculation that HHV-6 is an important pathogen in organ-transplant recipients.33 Shortly after the discovery of HHV-6 in 1986, reports emerged arguing for an association between the virus and a chronic ‘post-infectious’ fatigue syndrome or CFS.34–37 Some of these studies showed a relationship between CFS and a higher prevalence or higher titres of antibodies against HHV-6. Strikingly, one of these studies reported a greater proportion of positive cultures for HHV-6 from PBMCs of CFS patients compared with controls.34 In some studies, the identity of the virus recovered was supported by the use of monoclonal antibodies against HHV-6 antigens and by polymerase chain reaction (PCR) assays specific for HHV-6 DNA. Subsequent work and a closer appraisal of the above reports suggest that the association between HHV-6 and CFS is no stronger than that for EBV and CFS. The initial serological studies suggested that HHV-6 infection had occurred in most people with lymphoma,38 HIV infection39 and CFS,40 but in few healthy adults. However, these studies involved extremely insensitive assays for HHV-6 antibodies. When HHV-6 antibody assays were improved further, it was found that the true seroprevalence of HHV-6 is much higher than previously thought.41,42 It is now known that the virus infects over 80% of the population worldwide, usually before the age of 2 years. These findings suggest that the initially perceived increase in HHV-6 seroprevalence in conditions such as CFS, if accurate, merely reflects higher mean antibody titres, which were detectable with insensitive assays. A few studies have found higher titres of HHV-6 in CFS patients compared with the control group; however, there have also been several studies indicating the contrary. Several investigators reported identical seroprevalence42,43 or similar titres25,44,45 in CFS patients and control group. Studies also showed higher levels of EBV antibodies in patients with chronic fatigue and elevated titres of HHV-6.34 The virus has also been frequently detected in the saliva and salivary glands and from PBMCs of healthy individuals.42,46,47 Other studies have reported detection (by PCR) of HHV-6 in saliva in only 3% of healthy persons, whereas HHV-7 was detected in saliva in over 50% of healthy individuals. In contrast, HHV-6 could be found in salivary gland tissue in over 50% of these individuals.48 Wallace et al.49 did not find any significant differences in the detection of HHV-6, as well as EBV, CMV and HHV-7 from PBMCs, analysed by PCR, between CFS patients and control populations. Conclusions Epstein-Barr virus, CMV and HHV-6 are ubiquitous pathogens whose known abilities to persist and reactivate in humans made them particularly appealing to study in association with CFS. Initial reports suggest that CFS patients had abnormal serological antibody responses to the EBV, CMV and/or HHV-6 antigens. Further research could not confirm those previous findings, or has found that those ‘abnormal’ antibody responses were also seen with a variety of other organisms, including multiple viruses.14,15,24 These later data suggest that the apparent unusual serological responses to EBV, CMV and HHV-6 described by some investigators may merely reflect an epiphenomenon associated with a broader immunological impairment. This assumption is also supported by the observation of abnormal responses consistent with higher in vitro natural killer cells and suppressor activity and lower interleukin-2 production in response to phytohaemagglutinin stimulation in CFS patients compared with controls.13,25 The complexity of the immune system and immunological assays call for further research and a detailed review of the data to study a possible association of an immune splenomegaly, weight loss and lymphocytosis. Only one had a history of infectious mononucleosis, but all had detectable IgM anti-VCA antibody titres, and four had positive EA-R antibody titres, leading the investigators to suspect a chronic or reactivated EBV infection. Shortly thereafter, DuBois et al.11 described 14 similar patients who were diagnosed with chronic mononucleosis or chronic active EBV (CEBV) infection. Most reported significant fatigue and the absence of pronounced lymphadenopathy, pharyngitis, high fever, abnormal liver enzymes, splenomegaly and heterophile responses. The patients had elevated IgG anti-VCA and anti-EA antibody (D and R) titres compared with normal controls. Two similar reports, in 1985, associated persistent or recurrent fatigue, fever, headache, tender lymph nodes, myalgia, arthralgia, sore throat, depression and decreased ability to concentrate with an unusual EBV