Learn more about the benefits of registering on the new BJS website

Meta‐analysis of the impact of surgical approach on the grade of mesorectal excision in rectal cancer. BJS 2017; 104: 1609-1619.

Published: 17th October 2017

Authors: B. Creavin, M. E. Kelly, E. Ryan, D. C. Winter


The subspecialization of colorectal surgeons, and improvements in the quality of mesorectal excision have revolutionized rectal cancer surgery. With the increasing use of minimally invasive techniques, the completeness of the mesorectal excision has been questioned. This study aimed to assess the pathological outcomes of open versus laparoscopic rectal resection.


A meta‐analysis of RCTs was undertaken. The primary endpoint was the adequacy of the mesorectal excision. Secondary endpoints included circumferential resection margin and distance to resection margins.


Four studies were included, reporting on 2319 patients; 972 (41·9 per cent) had open and 1347 (58·1 per cent) had laparoscopic resections. Meta‐analysis of adequacy of the mesorectal excision showed a small difference in achieving an intact mesorectum in favour of open surgery (risk ratio (RR) 1·06, 95 per cent c.i. 1·02 to 1·10; P = 0·001). Superficial defects were more common in laparoscopic surgery (RR 0·70, 0·54 to 0·89; P = 0·004). Deep mesorectal defects (RR 0·78, 0·51 to 1·20; P = 0·256), circumferential margin (CRM) positivity (RR 0·85, 0·62 to 1·16; P = 0·310), and distance to radial (mean difference (MD) –0·06, 95 per cent c.i. –0·10 to 0·23; P = 0·443) and distal (MD 0·03, –0·06 to 0·12; P = 0·497) margins were all similar. A complete resection (intact mesorectum, negative CRM and distal margin) was achieved in 350 of 478 patients (73·2 per cent) in the laparoscopic group and 372 of 457 (81·4 per cent) in the open group (risk difference (RD) 8 (95 per cent c.i. 3 to 13) per cent; P = 0·003). However, an acceptable mesorectum (intact or superficial defects only) was present in 1254 of 1308 (95·9 per cent) and 916 of 949 (96·5 per cent) in the laparoscopic and open groups respectively (RD 1 (–1 to 3) per cent; P = 0·263).


Small differences in mesorectal quality were evident between open and laparoscopic rectal resections. This may be attributable to use of laparoscopic instruments; however, to date minor defects have not affected oncological outcomes.

Full text

Your comments