Prospective comparison of laparoscopic vs. open resections for colorectal adenocarcinoma over a ten-year period.

PURPOSE: The aim of this study was to define the long-term oncologic outcomes of laparoscopic resections for colorectal cancer.

METHODS: We analyzed our experience via a prospective, nonrandomized, longitudinal cohort study. The period of study extended from April 1991 to May 2001. Laparoscopic resection was offered selectively in the absence of a large mass, invasion into abdominal wall or adjacent organs, and multiple prior abdominal operations. Every laparoscopic resection performed with curative intent for adenocarcinoma was included. Twenty percent of patients whose procedures were converted to open resection were included in the laparoscopic-resection group because of intention to treat. Oncologic outcome measures of this group were compared with a computerized, case-matched, open-resection group, the case-matching variables being age, gender, site of primary tumor (colon vs. rectum), and TNM stage. The laparoscopic-resection group was followed up prospectively, and data were updated regularly. The follow-up techniques consisted of a combination of office visits, telephone calls, and the United States Social Security Death Index database.

RESULTS: The laparoscopic-resection group consisted of 172 patients with a mean age of 67 (range, 27-85) years. The open-resection group consisted of 172 patients with a mean age of 69 (range, 30-90) years. Mean follow-up was 52 (range, 3-128) months. Complete (100 percent) follow-up data were available. The TNM stage distribution was 63 Stage I (37 percent), 51 Stage II (30 percent), 47 Stage III (27 percent), and 11 Stage IV (6 percent) tumors for the laparoscopic-resection group and 65 Stage I (38 percent), 48 Stage II (28 percent), 51 Stage III (29 percent), and 8 Stage IV (5 percent) tumors for patients in the open-resection group (P = 0.75, not significant). Thirty-day mortality was 1.2 percent (2 deaths) in the laparoscopic-resection group and 2.4 percent (4 deaths) in the open-resection group (P > 0.05, not significant). Early and late complication incidences were comparable. Local recurrence was observed in three patients (1.7 percent) in the laparoscopic resection group with the primary tumor in the colon and in three patients (1.7 percent) with the primary tumor in the rectum, for a total incidence of local recurrence in the laparoscopy group of 3.5 percent (6 patients). In the open-resection group, local recurrence was observed in two patients (1.2 percent) among those with primary tumor site in the colon and in three patients (1.7 percent) in the group with primary tumor in the rectum, for a total incidence of local recurrence in the open-resection group of 2.9 percent (5 patients). One of the local recurrences in the laparoscopy group occurred in the port/extraction site, for an incidence of 0.6 percent. Metastasis occurred in 18 patients (10.5 percent) in the open group and in 21 (12.2 percent) in the laparoscopy group. Stage-for-stage overall five-year survival rates were similar in the two groups. The Kaplan-Meier statistical analysis performed for colonic vs. rectal primary adenocarcinoma confirmed that TNM stage for stage-overall survival was similar in the laparoscopic and open-resection groups (log-rank P = 0.22).

CONCLUSIONS: Notwithstanding the drawbacks of a nonrandomized study, no adverse long-term oncologic outcomes of laparoscopic resections for colorectal cancer were observed in a single center's experience during a ten-year period.