In-vitro waveguide system for the detection of Escherichia coli in drain fluid after colorectal anastomotic leakage

Abstract

Colorectal anastomotic leakage is one of the most feared complications after colorectal surgery and is associated with increased morbidity, prolonged hospital stay and risk of local cancer reoccurrence. One out of ten anastomoses leak, leading to infections. Colorectal anastomotic leakage is currently diagnosed by sniffing and watching drain fluid, which takes six to eleven postoperative days. Clinical research shows the presence of Escherichia coli (E. coli) bacteria in the drain fluid indicates leakage from postoperative day one, while the concentration grows from day four. Therefore measuring E. coli can reduce the time to diagnosis to one day. In this thesis an E. coli measurement system is proposed, which enables early detection and therefore greatly decrease morbidity, mortality, hospital stay and healthcare costs. METHODS An E. coli detection system was designed, simulated and assembled. The system was built around an evanescent waveguide device, which features a reference and sensing path to enable differential measurements. The surface of the sensing path is functionalised to capture E. coli. Light is coupled in and out of the waveguide device by fibres. Evanescent light in the waveguide device is absorbed by captured E. coli. By measuring the light absorption, the concentration of E. coli bacteria is calculated. The proposed system is portable, user-friendly and is tested in clinically relevant simulations. The proposed solution features a microcontroller, a constant current source circuit, an infrared LED, two photodiodes, a transimpedance differential amplifier, a display and computer software. It is powered by USB, communicates over USB and rejects power line noise. Subsystem performance was measured in clinically relevant simulations, i.e. 0.1 dB light absorption measurements. This resembles 10^6 CFU/ml E. coli on postoperative day five. RESULTS The subsystems were assembled on experimental PCB and measured. The light source outputs 2.655 mW of light and shows no drift after a warm-up period of five minutes. Electrical measurements show a stable LED driving circuit with no oscillations. Light coupling from the LED to a fibre was not measurable. The photodiode monitoring circuit can measure light absorption in high detail. A drift of 6.92·10^-5 dB per minute was measured. The detection time for individual measurements is 60.8 ms, or 6.08 seconds for an averaged measurement. Electrical measurements show oscillations of 2 kHz in the monitoring circuit, which may limit the functioning. The waveguide device was coupled with 40 dB of light loss. It confines light in the vertical plane, but not in the horizontal plane. Therefore the waveguide device could not be embedded in the measurement system and no biological measurements could be performed. CONCLUSIONS Measurements show a stable infrared light source and precise monitoring circuit were constructed. The proposed measurement system has a detection limit of 9,700 CFU/ml E. coli, thus anastomotic leakage can be detected from postoperative day four. The detection time is 6.08 seconds and allows quick measurements. The designed electronics and waveguide devices are portable, although the complete measurement system is not. The system can calculate the bacteria concentration automatically, adding user-friendliness. In future research the LED can be replaced by an fibre-coupled laser, the differential amplifier stage can be replaced by an instrumentation amplifier and the waveguide device’s functioning can be improved.