BACKGROUND

This review of cancer outcomes is based on key literature searches of the medical databases and meta-analysis of short-term benefits of laparoscopy in rectal cancer treatment.

METHODS

We carried out a systematic review of randomized clinical trials (RCTs) and prospective non-randomized controlled trials (non-RCTs) published between January 2000 and September 2013 listed in the MEDLINE and EMBASE databases (PROSPERO Registration number: CRD42013005076). The primary endpoint was clearance of the circumferential resection margin. Meta-analysis was performed using a fixed-effect model, and sensitivity analysis by a random-effect model; subgroup analysis was performed on subsets of patients with extraperitoneal cancer of the rectum. Relative risk (RR) and mean difference (MD) were used as outcome measures.

RESULTS

Twenty-seven studies (10,861 patients) met the inclusion criteria; eight were RCTs (2,659 patients). The RCTs reported involvement of the circumferential margin in 7.9 % of patients who underwent laparoscopic and in 6.9 % of those undergoing open surgery; the overall RR was 1.00 (95 % confidence interval 0.73-1.35) with no heterogeneity. Subgroup analysis of patients with extraperitoneal cancer showed equivalent involvement of the circumferential margin in the two treatment groups. Although significantly more lymph nodes were retrieved in the surgical specimen after open surgery, the MD of -0.56 was of marginal clinical significance. The sensitivity and subgroup analyses revealed no other significant differences between laparoscopic and open surgery in the rate of R0 resections, distal margin clearance, mesorectal fascia integrity, or local recurrence at 5 years.

CONCLUSIONS

Based on the evidence from RCTs and non-RCTs, the short-term benefit and oncological adequacy of laparoscopic rectal resection appear to be equivalent to open surgery, with some evidence potentially pointing to comparable long-term outcomes and oncological adequacy in selected patients with primary resectable rectal cancer.