We read with interest the paper from Drs. Bouvy et al. [2], who have provided some helpful insights into the behavior of malignant tumor cells during laparoscopic surgery. As with work previously reported by Allendorf et al. [1], they have demonstrated that laparoscopic procedures may be associated with reduced growth of tumor cells in a smallanimal model. This may be due to reduced physiological stress in animals undergoing laparoscopic surgery, resulting in less perioperative suppression of the immune system. However, we would caution readers not to draw the conclusion that laparoscopic approaches to the resection of abdominal malignancy are safe. In Dr. Bouvy's study, cancer cells were not introduced into the peritoneal cavity until after the surgical procedure was concluded. Therefore the potential for cancer cells to be redistributed by the insufflation of C0 2 gas and cause wound metastases was not tested.Our own recent experience [ 4] with laparoscopic and open surgical laceration of an implanted adenocarcinoma in a rat model has demonstrated a fivefold increase in the incidence of metastases in the abdominal access wounds of rats undergoing laparoscopic surgery despite the advantage of reduced growth of the implanted' 'primary'' tumor. Similar results have been reported by Jones et al. [3] following the introduction of free tumor cells into the abdominal cavity at the commencement of either laparoscopy with C0 2 insufflation or laparotomy in a hamster model. These studies suggest that if cancer cells are liberated during prolonged C0 2 insufflation, metastatic implantation and growth in abdominal wall wounds is more common following laparoscopic surgery. Interestingly, a further experiment using the same implanted tumor model in our institution during laparoscopic tumor laceration at both gasless and conventional laparoscopy demonstrated a threefold reduction in the incidence of wound metastases following gasless laparoscopy (unpublished data).It would be interesting to repeat the experiments per-