A minimally invasive chip based near infrared sensor for continuous glucose monitoring

Assessment of glycaemia in diabetes is crucially important for prevention of both, acute and long term complications. Continuous glucose monitoring (CGM) is certainly the most appropriate way for optimizing the glycaemic control, since it prevents or delays the progression of complications associated with hypo- or hyperglycaemic events, reducing morbidity, mortality, and overall costs in health care systems. In this paper we describe the concept and first in vitro results of a minimally invasive, chip-based NIR-Sensor for continuous glucose monitoring. The sensor concept is based on difference infrared absorption spectroscopy, which was evaluated within laboratory measurements of D+-Glucose dissolved in water. The laboratory measurements revealed a linear relationship between glucose concentration and the integrated difference spectroscopy signal with a coefficient of determination of 99.6% in the concentration range of 0-500 mg/dL. Suitable wavelength bands were identified in which the correlation is preserved and commercial light sources are available for realisation of a spectrometer-less, integrated NIR-sensor. In the designed sensor the component area (non-disposable) is separated from the detection area (disposable, low-cost). The disposable part of the sensor is fluidically connected to a micro-dialyses needle, accessing glucose subcutaneously via the ISF (interstitial fluid) or intravascularly. The non-disposable part contains all the optical elements, like LED's and photo-detectors. The in-and out-coupling of the optical signal is achieved across the plane of the chip by using total internal reflection on mirrors integrated into the fluidic chip. The glucose is continuously measured by considering the difference signals of light at the corresponding wavelengths, as a function of time or in defined intervals if the light sources are modulated. The in-vitro measurements show an absolute error of about 5 mg/dL with a relative error of 5% for glucose concentrations larger than 50 mg/dL and about 12% in the hypoglycemic range (< 50 mg/dL).