The use of the polymerase chain reaction (PCR) to identify pathogens from blood cultures of patients with device-associated infections has been evaluated. Shachor- Meyouhas et al. (2013) report an evaluation of 70 patients with central venous catheters to determine PCR efficacy in identifying bacteria in blood samples. This study found that the sensitivity of PCR is low (46%); however, the specificity of PCR is significantly higher (98%) than the specificity of culture or immunology-based methods. PCR has the advantage of providing a rapid diagnosis, an important benefit due to the increasing incidence of antibiotic resistance, and need for prompt and appropriate treatment with wide-spectrum antibiotics for the affected patients.
As reported in 2015, another study collected blood samples from patients with signs of sepsis and performed multiplex PCR and microarray assays to evaluate their clinical utility (Ljungstrom et al., 2015). Multiplex PCR allows simultaneous detection of multiple targets by amplifying genomic segments of different infectious agents, including viruses (Edwards and Gibbs, 1994; Elnifro et al., 2000). An advantage of this method is that it allows the detection of organisms that could be missed by culture methods.
In the study reported by Ljungstrom et al. (2015), microarray assays showed a sensitivity of 62% and a specificity of 99% in addition to high degree of agreement to culture method results. Previous studies using microarrays for rapid identification of bacterial organisms show similar results (Tissari et al., 2010). However, this methodology appears not to be clinically relevant for testing since the time to obtain results is as long as the time required for culture results (about 2 days). Other molecular techniques that could be applied for the identification of bacteria involved in medical device infections include denaturing gel electrophoresis, terminal restriction fragment length polymorphism (TRFLP), single-strand conformation polymorphism and DNA sequencing (Xi, 2014). Although these techniques might not be relevant for routine clinical use, they represent tools to provide information for clinical prevention. Song et al. (2013) have utilized denatured gradient gel electrophoresis to study the bacterial diversity of biofilms on endotracheal tubes removed from neonates to obtain information for clinical prevention, diagnosis and treatment of bacterial infections. The denatured gradient gel electrophoresis profiling of 16S rDNA gene amplicons from 35 endotracheal tubes were used to estimate the diversity of the bacteria attached to the tubes. Cloning and sequencing was used to identify and determine the distribution of bacteria. Interestingly, dominant bands showed that 35 samples contained Klebsiella spp., 28 samples carried Pseudomonas spp., 27 samples had Streptococcus spp., and 32 samples had uncultured bacteria (Song et al., 2013). Undeniably, this confirms that nucleic acid- based methods could provide more specific and comprehensive information regarding bacteria attached on medical devices when compared to culture methods.
Another study, aimed at characterizing the dynamics of indwelling urinary catheter colonization from eight patients, used TRFLP to profile microbial communities based on their DNA content (Liu et al., 1997).
Nucleic acid-based methods possess high specificity for detection that could represent a downside as some of the findings could have unknown significance and lead to mismatched clinical diagnoses. In addition to poor sensitivity, molecular techniques do not differentiate live from dead cells and can be inhibited by the biological components of the biofilm matrices (Xi, 2014).
The combination of culture, microscopic and molecular methods offers an advantageous alternative for the diagnosis of device-related infections. Sonication of infected devices to enhance culture positivity can be supplemented with PCR and fluorescence in situ hybridization (Percival et al., 2015).