The application of High Resolution Melting Analysis (HRMA) for rapid detection of bacteria responsible for bloodstream infections
[Thesis]
Ozbak, Hani
Dark, Paul; Warhurst, Geoffrey
University of Manchester
2013
Thesis (Ph.D.)
2013
Background: The diagnosis of bloodstream infection is a significant challenge for healthcare providers and is often associated with severe illness (sepsis) and poor outcomes. Rapid detection and identification of pathogens followed by characterisation of antibiotic resistance could help direct early treatment and improve patient care. Standard blood culture methods, which usually take 2-5 days to complete, can confirm if there is a bacteraemia or not in suspected patients. However, molecular approaches have been developed and are being increasingly investigated to overcome disadvantages of culture. One of the main potentials of molecular techniques is that they should be able to identify pathogens within a short time which could help clinicians treat patients earlier with rational antimicrobial therapy and limit overuse of antibiotic exposure. Objectives: To present the development and optimisation of a simple, rapid and cost-effective Real Time PCR methods combined with a High Resolution Melting Analysis (HRMA) approach, to detect and identify common bacteria associated with bloodstream infections. Approach: 16S rRNA and Gram classification primers were used on a broad range real-time PCR for molecular Gram typing and HRMA in a single run. Differentiation of bacterial species was achieved using a multi-parameter, decision-tree approach based on Gram typing, grouping according to melting temperature (Tm) and sequential comparisons of melting profiles (Curve shapes) against reference organisms. Findings: A preliminary validation was undertaken by blinded analysis of 53 consecutive bloodstream isolates from a clinical microbiology laboratory. 50 isolates contained organisms present on the panel and 96% of these were identified correctly at genus or species level. A correct Gram classification was reported for all 53 isolates. The strategy of amplification of the bacterial signal to an appropriate level using a short term pre-culture system (STPCS) for up to 12 hours prior to HRMA analysis significantly improved the overall sensitivity of the assay in spiked blood. Conclusion: This study suggests that a PCR-HRMA approach could be used as an alternative cheap approach to other molecular approaches for rapid detection and identification of bacteria responsible for >95% of bloodstream infections especially when combined with a Short Term Pre-Culture System (STPCS). Such development together with the current standard culture-based methods could allow clinicians to establish more effective management and treatment of patients with suspected bloodstream infection at an earlier stage than is possible with only current culture-based approaches.