Background: Colorectal anastomotic leak is a serious postoperative complication. Existing scoring systems have shown limited sensitivity and lack robust external validation. This study aimed to develop and validate a novel early postoperative scoring system - BADCAL (Blood loss, ASA grade, Duration of surgery, and levels of CRP, Albumin and Leukocyte count on postoperative day 3) for predicting anastomotic leak.
Methods: A prospective study was conducted involving 109 patients who underwent colorectal surgery. The BADCAL score was developed using 6 parameters: intraoperative blood loss, American Society of Anesthesiologists (ASA) grade, duration of surgery, C-reactive protein (CRP) level on postoperative day 3, albumin level, and leukocyte count. Each variable was assigned 0 to 2 points based on predefined risk thresholds, yielding a total score range of 0 to 11. Patients were stratified into low-risk (0-3), moderate-risk (4-6), or high-risk (7-11) groups. The primary outcome was radiologically or surgically confirmed anastomotic leak within 30 days postoperatively.
Results: Anastomotic leaks occurred in 13 of 109 patients (11.9%). The BADCAL score demonstrated excellent discriminatory ability (area under the receiver operating characteristic curve, 0.97 [95% CI, 0.93-1.00]). No patient with a score of ≤3 developed a leak (sensitivity of 100% and negative predictive value of 100%). High-risk scores (≥7) were associated with a 71.4% leak rate and 95.8% specificity. All 6 parameters were significant predictors, with the strongest associations observed for CRP of >60 mg/L, ASA grade ≥ III, intraoperative blood loss of >200 mL, and operative duration of >4 h.
Conclusion: The BADCAL score is an accurate and clinically practical tool for early postoperative risk stratification of colorectal anastomotic leaks. External validation is needed to confirm its generalizability.
Keywords: Anastomotic leak; BADCAL score; C-reactive protein; Colorectal surgery.
Copyright © 2025 Society for Surgery of the Alimentary Tract. Published by Elsevier Inc. All rights reserved.