Preoperative Chemoradiotherapy Effects on Anastomotic Leakage After Rectal Cancer Resection

In Brief Objective: To assess the effects of preoperative chemoradiotherapy (CRT) on anastomotic leakage (AL) after rectal cancer resection, using propensity score matching. Background: Conflicting data have emerged over the last decade regarding the effect of preoperative CRT on AL. Methods: We reviewed 1437 consecutive patients with rectal cancer who underwent low anterior resection (LAR) at our institution between 2005 and 2012. AL evaluated as grade C was the primary endpoint, as proposed by the International Study Group of Rectal Cancer in 2010. The patients were treated with (n = 360) or without (n = 1077) preoperative CRT. The total radiation dose was 50.4 Gy in 28 fractions. Multivariate and propensity score matching analyses were used to compensate for the differences in some baseline characteristics. Results: The preoperative CRT group contained more patients with the following characteristics, older age, male sex, smoker, advanced stage tumor, lower/mid rectal tumor location, ultra-LAR, and diverting stoma, than the non-preoperative CRT group (all Ps < 0.05). Postoperative AL occurred in 91 patients (6.3%). Before propensity score matching, the incidence of AL in patients with or without preoperative CRT was 7.5% and 5.9%, respectively (P = 0.293). After propensity score matching, the 2 groups were nearly balanced except for the initial stage and the length of the surgeon's career, and the incidence of AL in patients with or without preoperative CRT was 7.5% and 8.1%, respectively (P = 0.781). Conclusions: We did not observe that preoperative CRT increased the risk of postoperative AL after LAR in patients with rectal cancer, using propensity score matching analysis. We reviewed 1437 patients with rectal cancer who underwent low anterior resection with or without preoperative chemoradiotherapy (CRT). In our analysis, we did not observe any significant increase of anastomotic leakage rates in preoperative CRT patients when studying not only the entire study cohort but also the propensity score–matched cohort.