S0777 Time Trend Analysis of Pre-Operative Characteristics and Post-Operative Pouch Outcomes in UC Patients Undergoing Ileal Pouch Anal-Anastomosis

Abstract

INTRODUCTION: Proctocolectomy and ileal pouch anal anastomosis (IPAA) is the treatment of choice for patients with medically resistant ulcerative colitis. In this retrospective cohort study, we aimed to observe trends in pre-operative characteristics and post-operative pouch outcomes experienced by patients undergoing IPAA based on the decade of the surgery. METHODS: We obtained electronic medical record data from patients at our tertiary Inflammatory Bowel Disease Center. Patients with a pre-operative diagnosis of UC who underwent an IPAA procedure between January 1990 and May 2017 were included in this study and were followed until December 31, 2019. We stratified the data according to the decade of surgery and assessed differences between groups using t-tests for continuous data and Chi-square tests for categorical variables. We used Kaplan Meier curves to visualize the probability of pouch survival based on the decade of surgery. We also created a Cox proportional hazards model and adjusted for covariates to determine factors associated with pouch failure (defined as pouch excision). RESULTS: A total of 322 patients were included in the study and the median length of follow-up was 8.9 years. 10% of patients experienced pouch failure. Over time, there was a significant increase in the percentage of patients treated with anti-TNF therapies before surgery, the number of IPAAs conducted with stapled anastomoses, and the number of three-stage surgical procedures (Table 1). Pouch survival analysis stratified by decade of surgery demonstrated a significant decrease in the probability of pouch survival in more recent years (Figure 1). A Cox proportional hazard model adjusted for covariates showed that patients having an IPAA in the 2000s or 2010s had 5-times (AHR: 4.68; 95% CI: (1.43–15.34), P < 0.05) and 17-times (AHR: 17.23 95% CI: (3.51–84.57), P < 0.001) greater likelihoods of pouch failure, respectively, than patients who had their surgeries in the 1990s (Table 2). CONCLUSION: This study demonstrates that the likelihood of pouch failure in patients who underwent surgery in the most recent decade is greater than those who underwent surgeries in prior decades. Increased exposure to anti-TNF therapies in this population suggests that these patients may have more refractory disease. Additional studies should be conducted in order to further elucidate the underlying drivers of this association.Table 1.: Study PopulationTable 2.: Adjusted Cox Proportional Hazard Model For Pouch Failure. A Cox proportional hazards model was used to determine factors associated with pouch failure (defined as pouch excision). The multivariable model included all variables found to be statistically significant in univariable analyses along with covariates thought to be associated with pouch failure. A P-value less than 0.05 was considered to be statistically significantFigure 1.: Pouch Survival Probability For UC Patients Based on Decade of IPAA. Data were stratified according to the decade that surgery was performed. For patients who underwent multi-stage procedures, this was defined as the decade that the first stage was performed. Statistical differences between groups were determined using a log-rank test and a P-value less than 0.05 was considered statistically significant.