Predictive Model For Anastomotic Leakage After Colorectal Cancer Surgery

Simplified Predictive Model For Anastomotic Leakage After Colorectal Cancer Surgery: A Pilot Retrospective Study

This pilot study aimed to develop a simplified early postoperative model for AL risk stratification using readily available clinical and laboratory variables.

This retrospective single-centre cohort included 117 patients who underwent colorectal cancer resection with primary anastomosis between January 2024 and December 2025. Patients with protective stomas or incomplete data were excluded. Demographic, clinical, surgical, and first-postoperative-day laboratory variables were analysed. Multivariable logistic regression was used to evaluate factors associated with AL and to assess model performance.

Inclusion Criteria:

1. colorectal cancer resection
2. construction of a primary anastomosis
3. availability of clinical and laboratory data from the first postoperative day

Exclusion Criteria:

1. staged procedures without primary anastomosis
2. Hartmann's procedure
3. protective stoma at the index operation
4. incomplete data for the variables required for model construction.