Assessment of Functional Outcomes after Restorative Proctectomy for Rectal Cancer - A Population-Based Cohort Study

Date of ISAC Approval: 
22/05/2019
Lay Summary: 
Colorectal cancer is the second leading cause of cancer-related death in North America, and one-third of these cases are rectal cancer. The treatment of rectal cancer most commonly involves radiotherapy to the rectum, followed by surgery. After the rectum is surgically removed, the large intestine (colon) from above is connected to the lower portion of the remaining rectum, or the anus, to maintain normal transit of stool from below. While this operation avoids a permanent bag at the skin level, many patients experience bothersome bowel symptoms as a consequence of losing their rectum (e.g., "going" often, not emptying their bowels fully, accidental leakage of liquid and gas). Additionally, some individuals may notice changes in their urination and sex life as a consequence of pelvic surgery and radiotherapy. The objectives of this study are to determine the frequency of bowel, urinary, and sexual symptoms after surgery for rectal cancer, and to understand factors associated with their development. We will use a large hospital-admissions database (HES) to identify patients relevant to our study, and a large primary-care database (CPRD) to identify bowel, urinary, and sexual symptoms. We will use advanced statistical methods to understand factors most associated with each set of symptoms.
Technical Summary: 
The most common operation performed for rectal cancer is a restorative proctectomy. While this operation avoids a permanent ostomy, many patients (~70%) are left with significant bowel dysfunction (e.g., frequency, urgency, and incontinence) that impairs their quality of life. Furthermore, patients may experience urinary and sexual dysfunction, both of which can be consequences of pelvic surgery and radiotherapy. The primary objective of this study is to estimate the incidence of bowel dysfunction after rectal cancer surgery, and to identify factors associated with its development. Secondary objectives are to estimate the incidence of, and risk factors for, sexual and urinary dysfunction, as well as to describe the correlation between all three outcomes. This will be a large cohort study making use of two linked databases. Cohort inclusion criteria will be based on relevant rectal surgery procedures codes in the HES database, and outcomes will be defined using symptom read-codes recorded in CPRD. The cohort will be described using demographic, patient, and disease characteristics. A nested case-control design will be used to identify factors associated with the development of bowel dysfunction. The principal exposures of interest will include treatment (the use of preoperative radiotherapy, the use of postoperative chemotherapy), operative (operative approach, anastomotic height, neorectal reservoir, use of a diverting ostomy, diversion time), and postoperative (anastomotic leak) characteristics. A conditional logistic regression model will be used to adjust for potential confounders. Sensitivity analyses will be performed to test the robustness of our results, focusing primarily on outcome classification.
Health Outcomes to be Measured: 
The following outcomes will be measured in this cohort: (1) bowel dysfunction; (2) sexual dysfunction; and (3) urinary dysfunction. Each of these outcomes will be operationally defined using read-codes for relevant symptoms recorded in CPRD GOLD.
Collaborators: 

Samy Suissa - Chief Investigator - McGill University
Dr Marylise Boutros - Collaborator - McGill University
Paul Brassard - Collaborator - McGill University
Dr Richard Garfinkle - Collaborator - McGill University
Samy Suissa - Corresponding Applicant - McGill University

Linkages: 
HES Admitted