The purpose of this study was to evaluate the clinical usefulness of methicillin-resistant Staphylococcus-specific polymerase chain reaction (MRS-PCR) and broad-range universal PCR (U-PCR) for diagnosing PSI.
A prospective diagnostic study.
Thirty-two clinically suspect PSI patients and six control patients who underwent computerized tomography-guided biopsy and/or surgical treatment were enrolled.
Tissue samples were examined by microbiological culture, histopathology, and real-time PCR (MRS-PCR and U-PCR). The diagnostic accuracy of real-time PCR was analyzed based on the definitive diagnosis of infection, defined as a positive result from microbiological culture or histopathology.
All six control subjects were negative for PSI for all analyses. Twelve clinically suspect PSI subjects received definitive diagnoses (PSI group). The non-PSI group consisted of six control subjects plus the remaining 20 patients from the PSI clinically suspect group. MRS-PCR results were positive for all MRS-cultured PSI subjects. U-PCR was positive for all subjects in the PSI group with one discrepancy between real-time PCR and microbiological culture results in differentiation between gram-positive and gram-negative bacteria. In the non-PSI group, MRS-PCR and U-PCR were positive in three and seven cases, respectively. The sensitivity, specificity, positive predictive value, and negative predictive value of MRS-PCR for diagnosing MRS infection were 1.00, 0.91, 0.57, and 1.00, respectively; those for the diagnosis of bacterial infection with U-PCR were 1.00, 0.73, 0.63, and 1.00, respectively.
Identification of MRS infection and ability to differentiate between gram-positive and gram-negative bacteria is rapidly achieved using MRS-PCR and U-PCR. Real-time PCR provides a sensitive molecular diagnosis of PSI and may contribute to antibiotic selection.