Mieke Bus

8 117 NBI and IMAGE1 S™, enhance contrast of the mucosal surface and vascular structures, improving tumour detection rate. A first in-vivo study on NBI showed promising results. No studies on IMAGE1 S™, had been published so far. PDD uses fluorescence to improve tumour visualization. However, due to the acute angle of the ureterorenoscopes there is an increased risk of false-positives. OCT produces cross-sectional high-resolution images, pro- viding information on tumour grade and stage. A pilot study showed promising diagnostic accuracy. CLE allows ultra-high resolution microscopy of tissue resulting in images of the cellular structure. CLE could not be applied in-vivo in the upper urinary tract yet, due to tech- nical limitations. Before implementation of these techniques in the routine management of UTUC, more research has to be conducted. In the second part of this thesis, we focused on the use of OCT in the upper urinary tract. OCT is the optical equivalent of ultrasonography, using back-scattered light instead of back-reflected sound waves to produce high-resolution, cross-sectional images in which layered tissue anatomy can be distinguished. Light scattering decreases the OCT signal magnitude with depth, and limits the imaging range to approximately 2 mm. This signal decrease is quantified by the attenuation coefficient (µ oct [mm -1 ]) that allows in-vivo differ- entiation between tissue types. During carcinogenesis, changes occur in cellular architecture resulting in an increased nuclear-cytoplasm ratio and increased amount of mitochondria. Physically, this sub-cellular organization of tissue determines light scattering properties. Therefore, we hypothesized that lesion stage is obtained from image-based assessment of lesions; and that lesion grade correlates with µ oct . In Chapter 4 we did an in-vivo pilot study to assess the feasibility of OCT in UTUC. Eight consecutive patients underwent URS for UTUC suspicion or follow-up. OCT datasets were intra-operatively obtained from the ureter and pyelum. All patients eventually underwent nephroureterectomy. OCT staging was performed by visual inspection of lesions found on OCT images and OCT grading by quantification of the OCT signal attenuation on lesions and compared with histopatholog- ical diagnosis. Seven in-vivo OCT diagnoses on staging were in accordance with histology. In the eighth patient tumour thickness transcended OCT imaging depth range and was, therefore, inconclusive on tumour stage. For grading, median µ oct for grade 2 lesions was 1.97 mm -1 and 3.52 mm -1 for grade 3 (p-value <0.001). Healthy urothelium was too thin to reliable determine µ oct . These study results warranted further research, to confirm in a larger sample size grading and staging accuracy of OCT. We, therefore, investigated OCT in UTUC in a larger patient population in Chapter 5 . In total 26 patients underwent diagnostic URS including biopsies and OCT imaging, followed by nephroureterectomy. The ability of OCT to differentiate low-, and high-grade UTUC was assessed. Secondly, sensitivity, specificity, negative predictive value and positive predictive value of UTUC grading and staging by OCT were assessed.