ABSTRACT
INTRODUCTION: In addition to antibiotic treatment, slow-growing and non-cultivable bacteria can lead to false-negative results for sterile body site infections. In this study, we investigated the efficacy of 16S rRNA polymerase chain reaction (PCR) for such infections. METHODOLOGY: Following routine culture procedures, 16S ribosomal RNA (16S rRNA) PCR was performed for samples collected from sterile body sites between July 2017and September 2018. The samples were separated into two groups for likely (group 1) and unlikely infections (group 2) based on clinical and laboratory findings, as well as clinician opinion. Sequence analysis was performed for PCR-positive samples using 16S rRNA primers. Mixed chromatograms were analyzed with the RipSeq Mixed program, and Stata 15.1 was used for statistical analysis. RESULTS: Eighty-seven of 139 samples collected from 116 patients were placed in group 1, and 52 were placed in group 2. Compared with culture as the reference method, the sensitivity, specificity, positive predictive value, and negative predictive value for 16S rRNA PCR were 89.8%, 85.6%, 77.2%, and 93.9%, respectively. 16S rRNA PCR identified infections in 13 culture-negative samples. Among these, three had Bartonella quintana, Mycoplasma salivarium, and Mycobacterium avium complex infections, which cannot be detected with commercial multiplex PCR kits. CONCLUSIONS: Our study demonstrates that 16S rRNA PCR is effective for the diagnosis of sterile body site infections, especially for cases of meningitis and infective endocarditis where routine cultures fail.