This SBIR Phase I project aims to develop and demonstrate the feasibility of a system that can identify a broad range of bacteria in < 1 hour directly from blood. Existing diagnostic systems require 2-6 days to identify the bacteria, which delays treatment using appropriate antibiotics. Sepsis, which is often caused by a bacterial infection, contributes to up to half of all hospital deaths in the US largely due to complications arising from delays or inappropriate treatment (which is mainly because of lack of information regarding the infection-causing bacteria). It is the most expensive hospital-treated condition in the United States ? representing $20.3 billion in healthcare costs. The proposed system aims to sharply reduce the time to provide information needed by physicians, enabling them to initiate treatment using the appropriate antibiotic rapidly. This is expected to lead to significant reductions in costs, complications, and mortality. Further, by eliminating the need for culturing, new insights into bacterial behavior can be acquired contributing to additional fundamental knowledge of bacteria. This knowledge will aid antibiotic discovery efforts, improve understanding of mechanisms of drug resistance, support new biomanufacturing processes through faster detection of bacterial contamination of pharmaceutical products and foods, and characterization of the microbiome.
The challenges of rapid bacterial identification are primarily in two areas ? sample preparation, that eliminates the need for culturing, and multiplex identification, to cover the diverse range of bacterial species capable of causing infection. The proposed work is based on leveraging the differential response between bacteria and the cells in a sample matrix to selectively break down the sample matrix, but not bacteria, such that the largest lysis debris is smaller than intact bacteria. The debris is subsequently separated from intact bacteria using a size-based technique enabling rapid isolation of a broad range of bacteria directly from the sample. A novel Fourier-transform infrared microspectroscopy system is then used to identify the isolated bacteria at clinically relevant concentrations. No bacteria-specific reagents are employed. The proposed Phase I work addresses the highest risk technical hurdles in the process. Isolating and detecting low concentration of bacteria directly from blood is accomplished through the three objectives of this project that optimize the purification and bacterial recovery from the sample, identification accuracy, and performance of the integrated system relative to the traditional culturing-based method.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
