Incidental diagnosis (malignancy not suspected before surgery);
PORT-SITE METASTASIS AFTER LAPAROSCOPIC SURGERY: WHAT CAUSES THEM AND WHAT CAN BE DONE TO REDUCE THEIR INCIDENCE?
Version of Record online: 11 FEB 2009
© 2009 THE AUTHORS. JOURNAL COMPILATION © 2009 BJU INTERNATIONAL
Volume 103, Issue 9, pages 1150–1153, May 2009
How to Cite
Sooriakumaran, P., Kommu, S. S., Anderson, C. and Rane, A. (2009), PORT-SITE METASTASIS AFTER LAPAROSCOPIC SURGERY: WHAT CAUSES THEM AND WHAT CAN BE DONE TO REDUCE THEIR INCIDENCE?. BJU International, 103: 1150–1153. doi: 10.1111/j.1464-410X.2009.08363.x
- Issue online: 7 APR 2009
- Version of Record online: 11 FEB 2009
- Accepted for publication 31 October 2008
Much of oncological surgery has turned in favour of laparoscopy and urology is no exception; indeed, for nephrectomy and prostatectomy it is becoming the standard in most centres. However, while laparoscopic uro-oncology continues to be effective with favourable outcomes confirmed, reports of metastases at the site of trocar insertion are being reported. Such port-site metastases (PSMs) were first reported in general surgery 30 years ago , with the first urology report only 14 years ago ; this most probably represents the later uptake of laparoscopy in urology rather than any lower predilection for PSMs in laparoscopic urology than in general surgery. Ramirez et al. estimated that the overall incidence of PSMs after laparoscopic surgery was 1–2%. The precise incidence of PSMs has not been well-defined for urological laparoscopy. A recent review  identified 17 English-language studies reporting 29 cases of PSMs or tumour seeding secondary to laparoscopic urological surgery in the past 20 years. Table 1 illustrates the reported cases of urological port site metastases.
|Procedure||Tumour||TNM/Grade||Access||Retrieval bag||Metastases: trocar only or widespread||Time to PSM, months|
|Robotic radical cystectomy||TCC||pT3bN0/G3||Trans||Yes||Trocar||10|
|Bladder mass biopsy||TCC||T3/G2||Trans||No||Widespread||U|
|Radical Nx||RCC||pT3/G4w/sarcomatoid||U||Yes||Trocar alone||25|
|Radical Nx||RCC||pT1/G2||Trans||Yes||Trocar (+ ascites)||5|
|Radical Nx||RCC||pT1N0/G2||Trans||Yes||Trocar (+ ascites)||12|
|Radical Nx||RCC||pT2N0/G3||Trans + hand-assisted||No||Hand port||9|
|Radical Nx||RCC||pT3a/G2||Retro||No||Trocar (extraction)||39|
|Radical Nx||RCC||pT2N0/G3||Trans||Yes||Trocar (3)||8|
|NU||TCC||pT3/G3||Trans + hand-assisted||Yes||U||3|
|Simple Nx||TCC*||pT3 N1/G4||Retro||No†||Widespread||3|
|RPLND||NSGCT (testis)||II (postchemo)||U||Yes||Widespread||U|
Micali et al. described 13 patients with PSMs of 10 912 patient records assessed in an international multicentre retrospective audit, giving a rate of only 0.12%. This rate compares favourably with the reported incidence of scar metastases in open radical nephrectomy, at 0.4%. However, given the sparse published data, and the few events of PSM overall, it is difficult to make accurate comparisons between laparoscopic urological procedures and their open counterparts, and indeed, to know whether urological surgery is in any way different from other surgical disciplines for PSMs.
There are several postulated causes for developing PSMs; of these, tumour aggressiveness appears to be the most favoured. It is a truism of cancer biology that the more aggressive the tumour in terms of grade and stage, the more likely that tumour is to metastasize. Thus the phenomenon of PSMs might simply reflect the biological aggressiveness of the primary tumour. High-grade TCC accounts for most PSMs reported in urology. In an international survey of tumour seeding in urological laparoscopy, seven of 13 PSMs were from TCC; six of these were high-grade disease . Studies have reported that port-site recurrences of RCC [6,8–10] after radical nephrectomy have been associated with a high Fuhrman grade, and recently the first PSM after a partial nephrectomy was reported in a 2.5-cm high-grade (grade 3) clear-cell renal tumour .
It seems fairly intuitive that metastases would be more common in more biologically aggressive disease, but why port sites are targeted is far from clear. In an experimental study on rats using intraluminal radiotracer, Polat et al. showed that trocar-site contamination was more common when the primary tumour was manipulated traumatically than atraumatically. Hence, it might be that violation of the primary tumour boundaries promotes tumour-cell seeding, and that the use of radioactive tracers might aid in the intraoperative detection of port site contamination by tumour cells. However, as yet, such radioscintigraphic methods have not been used in the in vivo setting, and thus future data on this method of potentially quantifying the risk of PSMs is eagerly awaited. Murthy et al. also showed in experimental models that tumour cells adhere to wound margins by connecting to fibrin deposits during the stages of wound healing. Hofstetter et al. reported that abdominal insufflation does not increase haematogenous spread of disseminated intraperitoneal tumour cells to trocar sites in a colon-cancer model. These studies again reinforce the notion that PSM occur as a result of direct seeding of disseminated tumour cells rather than due to haematogenous spread.
However, tumour aggressiveness is unlikely to be the full story; it does not explain why some patients are more likely to have a port-site recurrence independent of tumour grade and stage. Yildirim et al. reported a case of PSMs after laparoscopic cholecystectomy in a renal transplant patient, suggesting that immune depression might be a factor predisposing to metastases, including of the port sites. It is well recognized that the trauma of surgery itself can depress immune function  and thus can, at least in theory, predispose to the spread of tumour metastases . Animal models have shown that immune function is significantly more depressed after open surgery than after laparoscopy , and this might explain why PSMs appear less commonly than wound metastases from open surgery, although as explained above, direct comparisons of this are problematic.
Some authors investigated the role of the pneumoperitoneum in the phenomenon of PSMs, and as yet no consensus can be reached from published reports. What is clear is that it appears that the trauma of surgery itself disseminating tumour cells to port sites, especially in an immunocompromised patient, is far more important than the intraperitoneal pressures generated or the gases used in creating the pneumoperitoneum. Ludemann et al. showed, in an elegant experimental study using cultured cancer cells, that tumour implantation into port sites depended more on the size of the port-site wounds and thus the amount of wound healing generated, than on the type of insufflation used for the pneumoperitoneum.
Given therefore that contamination of port sites by disseminated tumour cells might predispose to future metastases (although this is still unconfirmed), an appropriate prevention strategy might be to minimize tumour handling and the trauma caused. This would all come under the principles of good surgical technique. Perhaps then also it would seem reasonable to try to avoid spillage of tumour cells during retrieval. Hewett et al. and Gutt et al. independently concluded that tumour cells are transferred by surgical manipulation and surgical instruments that contaminate the port site. The most obvious means of direct spread are via the specimen or its removal using contaminated instruments through an unprotected wound . Hence, it would seem judicious to extract the specimen intact using an impermeable laparoscopic retrieval bag, using devices such as the LapSac or Endocatch. However, some surgeons morcellate the specimen before extraction; this practice is common after transabdominal laparoscopic radical nephrectomy. The specimen is morcellated inside an impermeable bag under laparoscopic control; assuming care is taken to avoid spillage of the bag contents, PSMs should not occur any more frequently than after intact specimen removal [22,23]. However, Castilho et al. reported PSMs in two of 32 patients after morcellation following a laparoscopic radical nephrectomy; the authors have subsequently changed their practice and now remove the specimen intact. Fentie et al. reported PSMs 25 months after surgery in one of 57 patients after specimen morcellation for RCC. Unfortunately, given the paucity of published data, it is not possible to make direct comparisons of the rates of PSMs after intact specimen removal vs morcellation.
From the discussion above it would seem sensible that to minimize the risk of PSMs, laparoscopic surgery should be done with meticulous tissue handling, avoidance of unnecessary surgical trauma, use of the smallest trocar wounds that are possible, and adequate care in specimen removal. Beyond this, if the patients’ immune status can be optimised, then this should also be attempted.
Various cytotoxic agents and anti-adhesion agents (5-fluorouracil, hyaluronate, methotrexate or cyclophosphamide) used for intraperitoneal or port-site irrigation have been studied in animals, but there has been no clear benefit confirmed clinically . Umpleby et al. recommend peritoneal lavage with large volumes of saline after laparoscopy; Goldstein et al. showed that adding 500 U/L of heparin to the saline decreased the adherence of malignant cells to a denuded surface. In a recent study, Jurczok et al. showed that in a murine model, tumour implantation and PSMs after laparoscopic surgery might be prevented by the intraperitoneal administration of specific oligopeptides or cytotoxic agents such as mitomycin.
In summary, although tumour aggressiveness is probably the most significant factor involved in PSMs, poor surgical technique with traumatic handling of the tumour is the risk factor over which surgeons have the most control. Emphasis on adhering to proper oncological surgical principles is the best method of prevention.
CONFLICT OF INTEREST
- 12Effect of types of resection and manipulation on trocar site contamination after laparoscopic colectomy: an experimental study in rats with intraluminal radiotracer application. Surg Endosc 2008; 22: 1396–401, , , .