Outcomes following completion and salvage surgery for early rectal cancer: A systematic review

Helen J. S. Jones, Chris Cunningham, Gary A. Nicholson, Roel Hompes

Research output: Contribution to journalReview articleAcademicpeer-review


Objectives To establish outcomes after completion and salvage surgery following local excision in literature published since 2005, to inform decision-making when offering local excision. Background Local excision of early rectal cancer aims to offer cure while maintaining quality of life through organ preservation. However, some patients will require radical surgery, prompted by unexpected poor pathology or local recurrence. Consistent definition and reporting of these scenarios is poor. We propose the term “salvage surgery” for recurrence after local excision and “completion surgery” for poor pathology. Methods Electronic databases were searched in February 2016. Studies since 2005 describing outcomes for radical surgery following local excision of rectal cancer were included. Pooled and average values were obtained. Results A total of 23 studies included 262 completion and 165 salvage operations. Most completion operations were done within 4 weeks; local recurrence rate was 5% and overall disease recurrence rate was 14%. The majority of salvage operations for local recurrence were within 15 months of local excision, often following adjuvant treatment. Re-do local excision was used in 15%; APR was the most common radical procedure. Further local recurrence was uncommon (3%) but overall disease recurrence rate was 13%. Estimated 5-year survival was in the order of 50%. Heterogeneity was high among the studies. Conclusions Patients undergoing local excision must be informed of risks and expected outcomes, but better data on completion and salvage surgery are required to achieve this. Systematic review registration number CRD42014014758.
Original languageEnglish
Pages (from-to)15-23
JournalEuropean journal of surgical oncology
Issue number1
Publication statusPublished - 2018

Cite this