Diverse etiologies for chronic fatigue syndrome.

Sir—Koelle et al. [1] recently studied 22 pairs of identical twins discordant for chronic fatigue syndrome and concluded that there was no major contribution for viral infections in the perpetuation of chronic fatigue syndrome (CFS). The authors should be commended for their methodology and the use of well-matched control subjects. However, the study raised several issues. First, similar to previous studies, the approach of Koelle et al. [1] was to look for statistical differences among the well-matched pairs with respect to the presence of viral antibodies and, more specifically, the presence of DNA of the viruses studied. Although these viruses were no more prevalent among the patients with CFS than among their healthy twins, one cannot conclude that these viruses are not the cause of CFS in a small subset of patients. CFS has been described in a small number of patients who had had well-documented acute Epstein-Barr virus (EBV), cytomeg-alovirus (CMV), and parvovirus B19 infections [2–4], and many of the patients responded to specific antiviral therapy. Of the first 200 patients with CFS who we evaluated for viral etiologies (table 1), only ∼10% had etiologies that were attributed to the viruses studied by Koelle et al. [1]. Chlamydia pneumoniae infection, an uncommon , although treatable, cause of CFS, was also dismissed in a previous, smaller study [5]. Second, latent EBV DNA and EBV viruses were often found in the blood and saliva, respectively, of asymptomatic, sero-positive individuals [6, 7], and, therefore, by themselves, are not ideal markers of active viral infection or the resultant symptoms of CFS. Would detection of virus specific mRNA be more indicative of smoldering infection in the PBMCs, and would a positive response to antiviral therapy increase the specificity of the finding? Third, the tissue localization and persistence of viruses may be responsible for the symptoms of CFS, and viruses may not be detected by viral assay of PBMCs or plasma. We have seen 2 patients with CFS who, after acute viral infection, had urine samples positive for EBV DNA and urine cultures persistently positive for CMV growth during a period of 6 months, but whose blood samples tested negative for EBV DNA and CMV DNA, respectively. Both patients improved after receiving intravenous cidofovir therapy. About one-half of our first 200 patients with CFS had significantly elevated levels of neutralizing antibodies to coxsackie-virus B and echoviruses, compared with control subjects from the community. On repeat …