OBJECTIVE: There are wide variations in local recurrence rate following curative surgery for rectal cancer and there are substantial inconsistencies among surgeons regarding the method of defining curative resection. This paper seeks to explore whether defining criteria is one of the important factors driving the variations in outcome. METHOD: A literature review was undertaken to find all UK-based studies that had data on curative resection and local recurrence rates. The studies were divided into groups with distinct definitions of curative resection for rectal cancer. Meta-analyses were performed to pool the risks of local recurrence by group definition. Statistical tests were used to explore the variation in local recurrence by group. Confounding relationships of age, sex, Dukes stage, length of follow-up and year of study were explored as far as possible given the limitations of the available data. RESULTS: For rectal cancers significant differences were found between the pooled local recurrence risks by group definition (P < 0.01). Meta-regression tests including all the studies indicate that the definition of curative resection is an important predictor of local recurrence. CONCLUSION: It is suggested that a standardized approach towards defining curative resection and local recurrence may have a significant effect on outcomes in colorectal cancer surgery and would enable comparisons to be made between different series.