|Keywords:||colonoscopy; adenoma detection; adenoma miss rate; colorectal cancer; risk factors|
|Full text PDF:||http://dspace.library.uu.nl:8080/handle/1874/312310|
Colorectal cancer is a major cause of cancer related morbidity and mortality in the Western world. The development of colorectal cancer from precursor lesions, i.e. adenomas, takes approximately 10 years. This relatively slow transition allows for screening, detecting and removing adenomas before they become malignant. Colonoscopy is the most sensitive method for detecting colorectal adenomas but is not perfect as approximately 25% of adenomas are missed. In the first part of this thesis new modalities that specifically may help to increase adenoma detection and thereby improve quality of colonoscopy in general. A prediction score was developed to identify subjects at risk for inadequate bowel preparation. The discriminative ability of the prediction score was well (AUC 0.63) and may specifically be suitable to identify subjects that could benefit from intensified bowel preparation. We further investigated two novel technologies that are designed to improve visualization at the proximal sides of folds and inner curves of flexures: Full Spectrum Endoscopy (Fuse) colonoscopy and EndoRings colonoscopy. Both Fuse colonoscopy (7% vs. 41%) and EndoRings colonoscopy (10% vs. 48%) had significantly lower adenoma miss rates when compared to standard colonoscopy. Finally, in a pilot study we investigated the feasibility of eye tracking technology to measure viewing behavior of endoscopists during real-time and self-performed colonoscopies. Gaze patterns were successfully measured in 90% of the procedures and it appeared that gaze patterns across the endoscopy monitor may serve as a good measure for colon inspection. In the second part of this thesis four epidemiological studies are described in which factors that might be associated with colorectal cancer development and survival were studied. In one of these studies the use of antibiotics appeared to be associated with an increased risk of developing colorectal cancer. The results of study in a prospective cohort including more than 500.000 subjects (EPIC cohort) showed no significant associations between the consumption of coffee and tea and the risk of developing colorectal cancer. In the same cohort we found that the consumption of dairy is neither associated with disease-specific nor with all-cause death in patients with colorectal cancer. In another cohort study we observed that in colon cancer patients high socioeconomic status is associated with a lower risk of undergoing open or converted laparoscopic surgery, anastomotic leakage or abscess formation, and a lower 30 days mortality risk following surgical resection of the primary tumor compared to low socioeconomic status.