Chronic fatigue syndrome (CFS) affects 400,000 to 900,000 adults in the United States 12. At least a quarter of those suffering from CFS are unemployed or receiving disability because of the illness; the average affected family forgoes $20,000 annually in lost earnings and wages (approximately half of the average United States household income); and, the annual value of lost productivity in the United States is approximately $9 billion 234. Despite the public health burden imposed by CFS, diagnostic, treatment and prevention strategies have proven difficult to devise because the etiology, pathophysiology and risk factors of CFS remain unknown [reviewed in 5].

Because the symptoms resemble those of infectious diseases and because CFS may follow an acute infectious disease, a considerable body of work has attempted to identify persistent or reactivated infection in patients with this illness [reviewed in 5]. These efforts have included analysis of seroprevalence, evaluation of viral RNA and DNA (i.e., latent/active infection), and most recently, molecular identification procedures for unique, previously uncharacterized pathogens 6789101112. None of these reports have developed convincing evidence for a significant association between any infectious agent and CFS.

However, negative, inconclusive, and conflicting case control studies do not mean that infectious etiologies for CFS should be dismissed. A recently discovered flavivirus, GB virus-C/HGV (GBV-C) 1314151617, has many properties that demanded a study assessing its possible association with CFS. GBV-C preferentially replicates in peripheral blood mononuclear cells (PBMC), primarily B and T lymphocytes, and in bone marrow in vivo 1819. Of note, some flaviviruses activate the classic complement pathway 20 and we have found a high prevalence of split products (C3a, C4a, and C5a) in patients with CFS compared to controls 21. We have also examined mononuclear cell populations for gene expression patterns and have found differences between patients and control subjects 2223. GBV-C viremia may persist for several years following primary infection and serologically antibodies are generated against the envelope protein, E2 [reviewed in 1518]. The antibodies are also long-lived and may protect against re-infection 24. GBV-C viremia and presence of antibodies are usually mutually exclusive, and only a small percentage of exposed individuals exhibit both viremia and anti-E2 antibody 25. As yet, no illness has been associated with GBV infection.

Therefore, thinking that complement activation and altered mononuclear cell gene expression might reflect continued infection with this virus, we chose to evaluate banked serum specimens from subjects enrolled in a case control study for GBV-C RNA, an indication of persistent active infection that might be associated with CFS and for specific antibodies associated with clearance of a GBV-C infection (anti-E2) as an indicator of past infection with this virus.

This study adhered to human experimentation guidelines of the U.S. Department of Health and Human Services and the Helsinki Declaration. The CDC Institutional Review Board approved study protocols. All participants were volunteers who gave informed consent.

Two cases (16.7%) and 1 control (4.8%) were seropositive (2-tailed Fisher's exact test p = .54). Only one person (a control subject) had detectable GBV-C RNA. Although the odds ratio of this difference is 5.14, the confidence interval (CI 0.32–49.59) demonstrates the difference was statistically insignificant.

This exploratory study did not identify significant differences in active or cleared GBV-C infections between individuals fulfilling a clinical definition of CFS or in control subjects. These data would not exclude GBV-C infection in small percentages (<20%) of CFS patients.

Since previous studies that used seropositivity or gene expression alone have failed to demonstrate an association between specific infections and CFS, we reasoned that a combined serological and virological approach would be more appropriate since only ongoing infection with GBV-C is associated with viremia. These results are consistent with values reported in a variety of population studies regarding the prevalence of GBV-C. Previous studies have addressed Herpesviridae, enteroviruses, retroviruses, hepatitis C, HTLV-II 5 and recently parvovirus 30 and Borna virus 31, but no definitive causative links have been made. Thus, another attempt to link a majority of CFS cases with a specific infectious agent did not identify a specific association.

CFS remains a clinically identifiable, but daunting, medical and psychological problem.

The author(s) declare that they have no competing interests

JJ and SB conceived the study. JJ was responsible for specimen identification, evaluation of results and preparation of the manuscript. PK performed the antibody and PCR procedures and participated in the preparation of the manuscript. SB supervised the laboratory procedures and participated in the preparation of the manuscript. WR was responsible for the primary patient study and participated in the design and evaluation of the study, and participated in the preparation of the manuscript.