High resolution laser doppler imaging for microvascular blood flow measurements

Sammanfattning: Techniques for functional assessment of blood flow dynamics in individual microvessels are likely to become tools of increasing importance, for example in the evaluation of new vasoactive drugs. Laser Doppler perfusion Imaging (LDI) is a non- contact method for visualization of spatial and temporal blood flow dynamics. In this thesis the LDI technique was adapted for investigation of flow dynamics in separate microvessels. The technique was evaluated and used for flow studies during microdialysis in human skin, in the rabbit tenuissimus muscle and in the hamster cheek pouch model. Furthermore, an algorithm was introduced, relating the technique to quantitative flow assessment.For investigations of individual structures of microvascular networks, a resolution higher than that in standard LDI is required. This can be achieved by focusing the laser beam and reducing the step length between adjacent measurement sites to a degree determined by the requirement of the application. In High Resolution LDI (HR-LDI) the laser spot is reduced to a diameter of 100 μm in the focal plane situated 7 cm beneath the scanner head. The laser beam is moved in steps of 125 or 250 μm in this plane, resulting in full format images of 8 mm x 8 mm or 16 mm x 16 mm. Enhanced High Resolution LDI (EHR-LDI) reduces the diameter of the laser beam to 40 μm in the focal plane, positioned 40 mm beneath the scanner head. The step length in this plane is 25 or 50 μm, which corresponds to full format images of 1.5 mm x 1.5 mm or 3 mm x 3 mm respectively. Thus blood flow variations within an area of a single measurement site in standard LDI are visualized by EHR-LDI.Inside separate micro vessels a high concentration of blood cells is present and the prerequisite of a low density of moving scatterers usually assumed in laser Doppler theory is not valid. Many of the detected photons originating from such vessels can thus be expected to have suffered multiple Doppler shifts. To further investigate this situation, an in vitro model of plastic tubes perfused with human blood was used. The LDI signal was found to scale linearly with average flow velocity (0 - 9 mm/s) and was not influenced by haematocrit variations (16 - 44 %). An underestimation of the LDI signal was observed as the tube diameter increased, interpreted as being caused by the limited depth of laser light penetration. By an algorithm considering both the Doppler shifts and the tube diameter, this effect could be compensated for. Furthermore, volume blood flow in tubes of diameters 500 μm, 750 μm and 1.4 mm were predicted by correlation coefficients of 0.994, 0.993 and 0.996.In applications based on high resolution LDI, the textures of the vascular structures in the rabbit tenuissimus muscle and the hamster cheek pouch preparation were rendered in good agreement with corresponding microscopic images. The vascular structures showed their sensitivity to vasoactive substances by changes in flow. The spatial heterogeneity in these changes and in flow redistribution occurring after a period of ischemia were visualized by the images, whereas the temporal flow changes inside separate microvessels were captured by single point recordings. The muscle tissue reacted to ambient air oxygen tension by a reduction in blood flow. Further, blood flow changes in the human skin following microdialysis probe insertion could be followed in detail. 60 minutes after probe insertion skin blood perfusion had returned towards pre-insertion levels. Individual variations, which might be of importance for the interpretation of microdialysis data were, however, observed.