Traumatic handling of the tumor independent of pneumoperitoneum increases port site implantation rate of colon cancer in a murine model

Background: Reports of port site tumor recurrences after laparoscopic-assisted resection of colon tumors have raised concerns about the safety of laparoscopic cancer surgery. Tumor cell suspension studies in animals have implicated the CO2 pneumoperitoneum (pneumo) in the etiology of port tumors. Unfortunately, in several ways, the cell suspension model is unrealistic and does not permit assessment of how tumor cells become liberated from the primary tumor. The purpose of this study was to establish a more realistic splenic tumor model and to determine the relative importance of the CO2 pneumo and excessive surgical manipulation in the development of port site and incisional tumor recurrences. Methods: Splenic tumors were established in female Balb/C mice (n = 134) via a subcapsular injection of 10(5) C-26 colon adenocarcinoma cells (0.1 ml volume) via a left-flank incision at the initial procedure. Ten days later, the animals were reexplored via a 1-cm left subcostal incision. Those with isolated splenic tumors (95%) were randomized into one of four groups: (a) control, (b) CO2 pneumo, (c) crushed tumor, or (d) crushed tumor with pneumo. Ports were placed in the left lower, light lower, and right upper quadrants of each mouse. In groups 1 and 2, the mice underwent a meticulously performed splenectomy; in groups 3 and 4, the tumor capsule was crushed intraabdominally prior to splenectomy. In groups 1 and 3, the subcostal incision was closed and the ports were removed after 15 min of anesthesia. Following splenectomy, group 2 and group 4 mice underwent closure of the subcostal incision and a 15-min CO2 pneumo (4-6 mm Hg) after which the ports were removed. Twelve days later, the mice were killed and examined for abdominal wall tumor implants. Results: Significantly more animals in group 3 (crushed tumor) developed port site and incisional tumors than those in group 1 (control) (p < 0.002 for both comparisons). The same results were found when group 4 (crush plus pneumo) was compared to group 2 (pneumo) (p < 0.002 for both comparisons). Regarding the port wounds, when the ports are considered individually (number of ports with tumors/total number of ports for each group), there were significantly more port tumors in the two crush groups than in the noncrush groups. No significant differences were noted when the port site and incisional tumor rates for group (control) and group 2 (pneumo) were compared or when the results for group 2 (crush) and group 4 (crush pneumo) were compared. Conclusions: A splenic tumor model was successfully established. When compared to meticulous technique, pur posefully traumatic handling of the splenic tumor before resection resulted in significantly more port wound and incisional tumors. In contrast, the addition of a pneumo after splenectomy did not significantly influence the incidence of port tumors in either the "good" or the "poor" technique groups. These results suggest that surgical technique plays a larger role in the development of port site tumors than the CO2 pneumoperitoneum.