Standard blood cultures require at least 24–120 h to be reported as preliminary positive. The objective of this study was to compare the reliability of Gram staining and fluorescent in-situ hybridization (FISH) for detecting bacteria in otherwise negative blood culture bottles. Ninety-six sets were taken from patients with a diagnosis of sepsis. Six incomplete blood culture sets and eight blood cultures sets demonstrating positive growth were excluded. We performed Gram stain and FISH on 82 sets taken from post-operative septic patients: 82 negative aerobic blood cultures, 82 anaerobic blood cultures, and 82 blood samples, as well as 57 blood samples taken from healthy volunteers. From the eighty-two blood sets analyzed from the septic patients, Gram stain visualized bacteria in 62.2% of blood samples, 35.4% of the negative aerobic bottles, and in 31.7% of the negative anaerobic bottles. Utilizing FISH, we detected bacteria in 75.6%, 56.1%, and 64.6% respectively. Among the blood samples from healthy volunteers, FISH detected bacteria in 64.9%, while Gram stain detected bacteria in only 38.6%. The time needed to obtain the study results using Gram stain was 1 h, for FISH 4 h, and for the culture method, considering the duration of growth, 5 days. Gram stain and FISH allow quick detection of bacteria in the blood taken directly from a patient. Finding phagocytosed bacteria, which were also detected among healthy individuals, confirms the hypothesis that blood microbiome exists.

The aim of the study was to evaluate particular polymerase chain reaction primers targeting selected representative genes and the influence of a preincubation step in a selective broth on the sensitivity of group B Streptococcus (GBS) detection by nucleic acid amplification techniques (NAAT). Research samples were vaginal and rectal swabs collected in duplicate from 97 pregnant women. They were used for enrichment broth culture-based diagnostics, bacterial DNA isolation, and amplification, using primers based on species-specific 16S rRNA, atr and cfb genes. To assess the sensitivity of GBS detection, additional isolation of samples preincubated in Todd-Hewitt broth with colistin and nalidixic acid was performed and then subjected to amplification again. The introduction of the preincubation step increased the sensitivity of GBS detection by about 33–63%. Moreover, NAAT made it possible to identify GBS DNA in an additional six samples that were negative in culture. The highest number of true positive results compared to the culture was obtained with the atr gene primers, as compared to cfb and 16S rRNA primers. Isolation of bacterial DNA after preincubation in enrichment broth significantly increases the sensitivity of NAAT-based methods applied for the detection of GBS from vaginal and rectal swabs. In the case of the cfb gene, the use of an additional gene to ensure the appropriate results should be considered.