부산대학교 도서관

Introduction: Currently in the UK, 11-14000 patients are diagnosed with rectal cancer each year sixteen percent of whom will have a locally advanced primary rectal cancer requiring exenterative pelvic surgery (extending beyond the conventional anatomic boundaries of anterior resection surgery). A further 10-20% of patients suffer local recurrence that may be amenable to exenterative surgery. The surgical management, the radiological assessment and clinical outcomes have been poorly understood and largely limited to self-reported personal series. The aim of this thesis was to evaluate surgical optimisation and staging assessment in this group of patients. Methods: The role of lateral pelvic sidewall lymphadenectomy in optimising outcomes for patients with locally advanced and recurrent rectal cancer was examined through a meta-analysis of the literature and a prospective database of patients undergoing exenterative surgery at the Royal Marsden. The safety and feasibility of exenterative surgery were evaluated using as endpoint the diagnostic accuracy of modern imaging techniques (PET-MRI fusion, diffusion weighted MRI and high resolution MRI) in this patient undergoing exenterative surgery compared to histopathology. The learning curve of exenterative surgery was evaluated using a Risk Adjusted Cumulative Sum (RA-CUSUM). Results: Extended lymphadenectomy showed no significant benefit. MRI was overall very accurate in predicting tumour invasion within the pelvic compartments (ROC0.877). Patients with invasion of the anterior above the peritoneal reflection compartment were shown to have poorer prognosis (HR=4.621, p=0.012). Diffusion weighted MRI did not improve overall accuracy. PET MR image fusion was feasible with very good diagnostic accuracy for all the compartments. Finally, a colorectal surgeon required a minimum of 14 cases under supervision to improve the morbidity following exenterative pelvic surgery for locally advanced primary and recurrent rectal cancer. Future studies: The findings need to be tested prospectively in multicentre studies to validate the results of the learning curve, staging and lymphadenectomy.