Comment on "Oncological Outcomes After Anastomotic Leakage After Surgery for Colon or Rectal Cancer: Increased Risk of Local Recurrence".

Abstract

To the Editor, We read with interest the article by Koedam et al.1 demonstrating anastomotic leakage as an independent risk factor for local recurrence after rectal cancer surgery. We congratulate the authors and agree that COLOR trial2 was not powered to be able to extrapolate data on anastomotic leaks and also how in a pre-FOLFOX era the outcomes of stage III colon cancer remained the same irrespective of chemotherapy. However, with respect to the COLOR II observations,3 we have the following concerns. The authors have noted that an ileostomy was performed in 52.4% (44/84) of patients with an anastomotic leak but we observe that “ileostomy” was not included as a factor in the univariate analysis for local recurrence at 5-years follow-up. In COLOR II, performing a diverting ileostomy was left to the discretion of the operating surgeon without being part of a protocol. As seen in COLOR II, preoperative radiation may lead to the increased rate of anastomotic leaks3,4 and not doing a diverting ileostomy can exaggerate sub-clinical leaks which would have not manifested otherwise.5,6 If anastomotic leaks increase the local recurrence,1,7 then the very purpose of neoadjuvant radiation to improve local control in rectal cancers would be defeated. In today’s era, to facilitate the timely administration of the adjuvant treatment and hence further decrease the recurrence rates, an ileostomy may at times prove beneficial. The authors, in the discussion, propose that increased recurrences after anastomotic leaks may be attributed to a delay in adjuvant chemotherapy (although rightly stated as not applicable to COLOR II). This reasoning may have been applicable if there were an increase in distant recurrences in addition to the local recurrences, which is not the case. The authors have referred to a hypothesis of spillage of cancer cells into the peritoneal cavity during the surgery and adhesion of tumor cells due to the inflammatory response of the leak. If this hypothesis is to be believed then the incidence of distant peritoneal failure should also be seen along with or apart from local recurrences. Anastomotic leaks in this analysis were not seen to cause seeding of distant peritoneal sites. Also, we observe that 5 of 23 patients (25.2%) with positive resection margins had a local recurrence at 5 years and 9 of 84 patients (13.3%) with anastomotic leaks had a local recurrence at 5 years. However, it would be interesting to note how many patients with positive resection margins had anastomotic leaks. We are not sure what could be the likely explanation for the increased recurrences in patients who had anastomotic leaks but this finding gives grounds to further evaluate total neoadjuvant treatment to treat rectal cancers,8 other than evaluating this important observation in a trial powered to study the same.