Colorectal Cancer Surveillance after Resection

Abstract

Background: Multiple guidelines have recommended performing surveillance colonoscopies after resection of colorectal cancer. However, there has been debate about the optimal frequency of endoscopic surveillance and its yield. Objective: The objective of this study was to explore the adherence and the yield of surveillance colonoscopy in our population. Design: This was a retrospective cohort study. Setting: This study was carried out in two tertiary care centers in Riyadh of all patients who were found to have colon cancer on colonoscopy. Patients and Methods: All patients who were found to have colon cancer on colonoscopy at one center from January 2016 to July 2019 and the second center from April 2012 to April 2019 were included in the study. The main outcome measures were as follows: rate of adherence to surveillance colonoscopies as well as the rate of synchronous and metachronous tumors identified. The sample size was not calculated. Results: A total of 280 patients were identified with a mean age of 64.5 years (standard deviation: 13.1), 61.4% were male, and the majority (72%) of the original cancers were left sided (rectum – 25.45%, rectosigmoid – 13.45%, sigmoid – 25.82%, and descending colon – 7.27%). Surveillance was performed in 55.7%, 17.9% did not attend their scheduled procedures, 2.1% refused, while there was no clear reason in 8.9%. Surveillance was not applicable to 15.3% of the study population. Over the course of three rounds of surveillance, the timing of the procedures was in line with recommendations in 66.8% in the first round while it was 20% in the second and third rounds. The procedures were normal in 69%, 71.9%, and 90% of the first, second, and third rounds, respectively. Metachronous adenocarcinoma was seen in 2.6% and synchronous adenocarcinoma in 0.6% during the first round. Adenocarcinoma was found in 5 out of 75 patients in the second round and 1 out of 10 in the third round. Conclusion: The attrition rate and deviation from timing suggested by guidelines are high. Furthermore, the yield of surveillance colonoscopies for important findings is clinically relevant. The limitations of this study were as follows: retrospective, probability of unmeasured confounders, as well possibility of attrition bias.