Purpose: In stage II colorectal cancer (CRC), high-risk patient selection is required, but no candidate markers have been elucidated. Our concern was whether anastomotic leakage (Lk) is a potential available clinicopathological factor for selecting high-risk stage II.
Methods: Two hundred seven patients with stage II CRC who underwent curative resection were analyzed. Clinical variables were tested for their relationship to survival.
Results: The 5-year disease-free survival rate (DFS) was 87.0%. The univariable prognostic analyses indicated that Lk (P = 0.003) was the only significant factor. The multivariable prognostic analysis revealed that Lk remained to be potently independent [hazard ratio (HR), 4.21, P = 0.021), and the DFS was 58.3% in cases with Lk, while 88.7% in the counterpart. The multivariable logistic regression analysis revealed perioperative blood transfusion (P = 0.001) was independently associated with Lk. Intriguingly, Lk was closely associated with hematogenic recurrence (P = 0.003) rather than peritoneal or local recurrence. Although sustained increase of the serum C-reactive protein at 2 weeks after operation predicted poor prognosis, the mutitivariable analysis including the C-reactive protein level revealed that Lk still indicated the prognostic potential (HR, 3.70, P = 0.075).
Conclusions: The findings concluded that Lk may be a high risk for systemic recurrence in stage II CRC.