ABSTRACT
BACKGROUND: Monitoring surgical quality has been shown to reduce locoregional recurrence (LRR). We previously showed that the arterial stump length (ASL) after complete mesocolic excision (CME) is a reproducible quality instrument and correlates with the lymph-node (LN) yield. We hypothesized that generating an LRR prediction score by integrating the ASL would predict the risk of LRR after suboptimal surgery. METHODS: 502 patients with curative resections for stage I-III colon cancer were divided in two groups (CME vs. non-CME) and compared in terms of surgical data, ASL-derived parameters, pathological parameters, LRR and LRR-free survival. A prediction score was generated to stratify patients at high risk for LRR. RESULTS: The ASL showed significantly higher values (50.77 mm ± 28.5 mm) with LRR vs. (45.59 mm ± 28.1 mm) without LRR (p < 0.001). Kaplan-Meier survival analysis showed a significant increase in LRR-free survival at 5.58 years when CME was performed (Group A: 81%), in contrast to non-CME surgery (Group B: 67.2%). CONCLUSIONS: The prediction score placed 76.6% of patients with LRR in the high-risk category, with a strong predictive value. Patients with long vascular stumps and positive nodes could benefit from second surgery to complete the mesocolic excision.
