Back to 2008 Program and Abstracts
The Learning Curve of Laparoscopic Rectal Resection for Cancer: a Single-Center Experience
Marco Montorsi*, Matteo Rottoli, Stefano Bona, Paolo P. Bianchi, Riccardo Rosati
General Surgery, University of Milan, Istituto Clinico Humanitas IRCCS, Milan, Italy
Laparoscopic rectal resection (LRR) has been shown to be feasible but challenging. The role of the learning curve in this surgery is not fully elucidated. To evaluate its impact, we prospectively collected, in two consecutive periods, operative and outcome data of pts submitted to LRR for cancer at a single institute performed by four surgeons equally experienced in laparoscopic surgery.
Methods: From November 1999 to May 2007, 141 patients with rectal cancer were treated by laparoscopy. Learning curve was evaluated in two consecutive periods, 1999 to 2003 (first period) and 2004 to 2007 (second period). The evaluated variables were: operative data (operative time, conversion rate, intraoperative complications), short-term outcomes (postoperative complications, mortality, lenght of hospital stay, readmission rate), and oncological outcomes (site of the tumor, number of lymphnodes, resection margins, port recurrence)
Results: Number of patients (71 and 70), demographic data and oncological stage were similar in the two periods. No differences were found in operative time (274 and 294 minutes, p 0.12), intraoperative (7% and 12.9%, p 0.25) and postoperative complications rate (19.7% vs 17.1%, p 0.69). Anastomotic leakages occured in 8 patients, equally in the first (11.3%) and in the second period (11.4%, p 0.97). Lenght of hospital stay decreased in the second period (9 vs 8 days, p 0.18). There were no readmission in hospital after discharge in both groups. No differences were observed among the four surgeons in operative data and outcomes. The number of resected lymphnodes per patient (18) was the same in the two periods. There were statistically significant differences in the distribution of tumor site (percentage of the tumors located in the mid and lower rectum was 45.1% in the first period and 72.9% in the second period, p 0.01) and in conversion rate (23.9% vs 11.4%, p 0.05). There were 2 microscopical infiltrations of the distal margin (2.8%) and 1 port site metastasis (1.4%), all in the second period. Five yrs overall and disease free survival rates were 82.1% and 75.6%. Disease free survival rate was significantly lower when conversion to open surgery was required (78.7% vs 61.8%, p 0.04).
Conclusions: When performed by experienced surgeons, LRR for cancer is feasible, safe and oncologically effective since the beginning of the experience. The parameters which significantly changed during the learning curve were conversion rate and the anatomic site of the rectal tumors. Operative time and morbidity did not show an improvement, probably due to a different case-mix in the second period (more distal tumours).
