Mieke Bus

118 Chapter 8 In 83% staging of lesions was in accordance with histopathology. Sensitivity and specificity analysis for non-invasive and invasive tumours were 100% and 92%, respectively. Tumour size >2 mm and inflammation were risks for false-positives. For grading of low-, and high-grade lesions, median μ oct was 2.1 mm -1 and 3.0 mm -1 (p-value <0.01), respectively. ROC-analysis showed sensitivity of 87% and specificity of 90% using a cut-off value μ oct of 2.4 mm -1 . This study confirmed the ability of OCT to visualize, grade and stage lesions in the upper urinary tract. Although the studies described above showed high resolution cross-sectional images of the ureter using OCT, its imaging depth is limited to 1-2 mm. Tumour sizes that transcended the OCT imaging depth range resulted in false positives for tumour invasion. To overcome this important limitation, combining OCT with an imaging technique with an increased imaging depth range could be a solution. Therefore, we combined OCT with Endoluminal Ultrasound (ELUS) in Chapter 6 . Endoluminal Ultrasound (ELUS) produces lower resolution images compared to OCT, but has increased imaging penetration depth. To enable co-registration of OCT and ELUS in human nephroureterectomy specimens, CT was used to obtain overall anatomy and exact probe localization. OCT and ELUS imaging catheters were introduced via the ureteral orifice up to the renal pelvis in five nephroure- terectomy specimens and pulled back following each measurement. The OCT and ELUS systems scan a cylindrical trajectory of 54 mm by 10 mm and 890 mm by 10 mm, respec- tively. Visualization software (AMIRA®) was used for reconstruction and co-registration of CT, OCT and ELUS data from the complete renal pelvis and ureter. Co-registered OCT and ELUS datasets were compared with histology. 3D-CT enabled exact dataset fusion to compare histology with OCT and ELUS findings in this ex-vivo setting. In the high resolution OCT images of normal appearing renal pelvis and ureter, the urothelium, lamina propria and muscularis were clearly visible. OCT permits high-resolution imaging of normal ureter and ureter lesions whereas ELUS provides more depth information. However, in ELUS images of normal appearing ureter and anatomical layers could be distinguished, but the urothelial layer could not be identified. Using ELUS, images of suspected lesions were visible, although exact staging and differentiation between non-invasive and invasive lesions was not possi- ble with ELUS due to low-resolution images. For this reason, ELUS was no beneficial addition to OCT in this ex-vivo setting. In chapter 7 a reflection is given on the work presented in this thesis. This thesis demon- strated that optical diagnostics have the potential to improve the diagnosis of upper urinary tract urothelial carcinoma (UTUC). However, optical diagnostics still need comprehensive investigation before implemented in clinical practice. In this chapter a plea is given to start well-conducted studies the further evaluation of diagnostic technologies in the upper uri- nary tract. In addition several potential clinical applications of optical diagnostics in the upper urinary tract are discussed.