Serum cytokine levels in postinfective fatigue syndrome.

To the Editor—[Q1]Previous studies have sought evidence for a role of abnormal cytokine activity in patients with chronic fatigue syndrome and have had conflicting results [1–3]. These ambiguous results may reflect heterogeneity in groups of patients considered to have chronic fatigue syndrome and variations in assay systems. We established postinfective fatigue syndrome as the only well-characterized model of the onset and evolution of chronic fatigue syndrome in a prospective cohort of individuals followed up from the onset of acute infection (Dubbo Infection Outcomes Study [DIOS]) [4]. Longitudinally collected clinical data and blood samples from participants in DIOS provide a unique opportunity for nested case-control studies examining the path-ophysiology of chronic fatigue syndrome. We previously reported the lack of association between cytokine production from cultured peripheral blood mononuclear cells and the postinfective fatigue syn-drome–related illness in participants in DIOS [5]. We now report a masked analysis of a longitudinal case-control series from DIOS that extended the number of cytokines tested and focused on serum levels. Twenty patients with acute infection were selected, including 5 patients with serologically confirmed acute Epstein-Barr virus (EBV) infection followed by postin-fective fatigue syndrome lasting у6 months, 5 patients with acute infection (not primary EBV but seropositive for EBV) followed by postinfective fatigue syndrome, and 10 matched control subjects with acute EBV infection followed by prompt recovery. Serum samples and clinical data from baseline and from 3–6 months and 9–12 months after onset of infection were analyzed. Serum samples were coded according to case-control status before transfer to the cytokine analysis laboratory. Thirty-five analytes were measured in serum samples with use of a multiplex im-munoassay, including the chemokines [Q2]epithelial cell–derived neutrophil-activating peptide 78, eotaxin, growth-regulated oncogene a, interleukin (IL)–8, in-terferon (IFN)–inducible protein 10, monocyte chemotactic protein 3, mon-okine induced by gamma IFN, macro-phage inflammatory protein 1a, macro-phage inflammatory protein 1b, and regulated upon activation normal T cell expressed and secreted; the cytokines IFN-factor a, tumor necrosis factor b; and the growth factors nerve growth factor, plate-let-derived growth factor b, transforming growth factor b, vascular endothelial growth factor, fibroblast growth factor b, granulocyte colony-stimulating factor, and granulocyte-macrophage colony-stimulating factor. All of the study groups were predominantly female and were matched for both sex distribution (by x 2 test,) and P p .670 age (by analysis of variance,). P p .597 Cytokine data were analyzed by 2-way analysis of variance examining the effects of time and type of case …