Coxsackie B enteroviruses have been implicated repeatedly as agents associated with chronic fatigue syndrome (CFS). The objective of this study was to compare the serological evidence for the presence of Coxsackie B virus neutralising antibody, with the polymerase chain reaction (PCR) detecting a portion of the 5' nontranslated region (NTR) of the enterovirus genome. Serum samples from 100 chronic fatigue patients and from 100 healthy comparison patients were used in this study. In the CFS study group, 42% patients were positive for enteroviral sequences by PCR, compared to only 9% of the comparison group. Using the neutralisation assay, 34% of study patients were positive, compared to 41% of comparison patients. In the study group, 66/100 patient results correlated, i.e., they were either positive/positive or negative/negative for both tests. Of those that did not correlate, the majority were PCR-positive/Coxsackie B antibody-negative (21/34). In the comparison group, 58/100 patient results correlated. Of those that did not, the majority were PCR-negative/Coxsackie B antibody-positive (37/42). The Coxsackie B antibody neutralisation assay was not able to differentiate the CFS study group from the healthy comparison group, and thus the clinical relevance of this assay may be questioned. The PCR assay did differentiate the two groups with significantly more CFS patients having evidence of enterovirus than the comparison group.