Comparison between LightCycler Real-Time Polymerase Chain Reaction (PCR) assay with serum and PCR-enzyme-linked immunosorbent assay with whole blood samples for the diagnosis of human brucellosis.

BACKGROUND: To overcome some of the limitations of conventional microbiological techniques, polymerase chain reaction (PCR)-based assays have been proposed as a useful tool for the diagnosis of human brucellosis.
METHODS: A single-blinded comparative study was undertaken that compared 2 different PCR assays: a SYBR Green I LightCycler-based Real-Time PCR assay (LC-PCR; Roche Diagnostic) with serum samples and a PCR-enzyme-linked immunosorbent assay (ELISA) with whole blood samples. Both assays amplify a 223-bp sequence of a gene that codes for the synthesis of an immunogenetic membrane protein specific for Brucella genus (BCSP31). We analyzed the diagnostic yield of these assays with 60 samples obtained from patients with active brucellosis and 37 samples obtained from a control group composed of patients with febrile syndromes of other defined etiologies, asymptomatic subjects with past brucellosis or exposure to Brucella infection who had persistently high titers of anti-Brucella antibodies, and healthy subjects.
RESULTS: The sensitivities of LC-PCR with serum samples, PCR-ELISA with whole blood samples, and blood cultures were 93.3%, 90%, and 65%, respectively. Three control samples (8.1%) had a positive PCR-ELISA result, and 2 of these samples (5.4%) also had positive LC-PCR results. The specificity and positive likelihood ratios were 94.6% and 17.3, respectively, for LC-PCR and 91.9% and 11.1, respectively, for PCR-ELISA.
CONCLUSIONS: The diagnostic yield of LC-PCR with serum samples was higher than that of PCR-ELISA with whole blood samples. The speed and technical simplicity of LC-PCR in serum samples make it a useful alternative to blood cultures for patients with suspected brucellosis and negative or doubtful serological test results.