Review of investigations regarding the etiology of port site tumor recurrence.

Abstract

The animal and human studies presented, at first glance, present a confusing and conflicting story. In regards to the animal studies, much of this confusion can be traced to the use of a variety of different models, none of which truly reproduces the human situation. Nonetheless, there is much to be gleaned from these efforts. The authors present conceptualization of the port wound tumor dilemma follows. In order for wound tumors to develop, viable tumor cells must be liberated from the primary tumor and find transport to a wound. Rarely, patients with colon tumors will present with or will develop widespread intraabdominal carcinomatosis. These tumors have the ability to spontaneously shed considerable numbers of viable cells which have the ability to implant on uninjured peritoneal surfaces. Unfortunately, the surgeon has little chance for success in these patients with either open or minimally invasive methods. Fortunately, most colon adenocarcinomas do not spread in this manner. Differences in the clinical behavior and manifestations of colon tumors most likely reflect the genetic makeup of individual tumors. Colonic neoplasm's ability to invade and metastasize varies considerably from tumor to tumor. Thankfully, as mentioned, the vast majority of colon tumors are not prone to cause carcinomatosis. Despite this fact, the human data available suggests that tumor cells can be found in the peritoneal cavity using sophisticated methods in about half of the patients after colectomy. If this is the case, then why aren't more wound tumors seen? Logic dictates that there must be a critical number of free intraabdominal cells above which successful wound seeding is likely. It makes sense that traumatization of the tumor will result in increased numbers of liberated cells. Therefore, surgical approach and technique should impact considerably on outcome. For the majority of colon tumors, if the lesion is assiduously avoided during mobilization and resection, it is unlikely that enough tumor cells will be shed to result in port site tumors. The recent interim results of the Cleveland Clinic's and the Barcelona randomized trials certainly support such a hypothesis. With over 300 patients enrolled (combined series) and with an average follow up of over 2 years, in neither trial has a port site tumor been noted. Similarly, with an average follow up of just under 3 years, Franklin et al noted that there were no port site tumors in their prospective trial of 191 consecutive laparoscopic colectomies for cancer. In the clinical setting, experience and surgical expertise seem to be the best predictor of outcome, in regards to wound tumors. The few animal studies that allow assessment of the impact of technique (i.e. those that utilize an intraabdominal solid tumor model which allows tumor excision) indeed support this hypothesis. In these studies poor technique resulted in significantly more wound tumors. Furthermore, it has been shown that for laparoscopic procedures, there is a definite learning period during which the incidence of wound tumors is considerably higher than that of open resection. With experience the laparoscopic incidence falls to that of open resection. Furthermore, a number of recent studies suggest that is possible to lower the incidence of wound tumors via peritoneal and wound irrigation with a variety of agents. These animal study results are in keeping with the recent clinical results. Both would suggest that given proper and adequate training and with sufficient attention being paid to avoid tumor handling that the incidence of wound tumors will be as low as that following open colectomy. How large a part, if any, does the CO2 pneumoperitoneum play in the port wound tumor story? Certainly, the results of the bulk of the animal studies performed, to date, have suggested that the CO2 pneumoperitoneum plays a critical role in the development of port wound tumors. With few exceptions, these studies have utilized tumor cel