Postoperative PLR Slope for Early Warning of Colorectal Anastomotic Leakage

Anastomotic leakage (AL) after colorectal resection is often confirmed between postoperative days (PODs) 4 and 8. We evaluated whether the POD1-to-3 platelet-to-lymphocyte ratio (PLR) "rising slope" in routine blood counts provides an earlier warning of ISREC grade B/C AL.

In a prospective single-center cohort (2021 to 2025), analyses used a POD3 landmark among patients who were alive and without an AL diagnosis through POD3. PLR slope was derived from POD1 to 3 blood draw timestamps and modeled at 10 PLR units/day. Thirty-day International Study Group of Rectal Cancer (ISREC) grade B/C AL was adjudicated blinded using CT with IV contrast (± rectal contrast when feasible) and/or endoscopy/operative findings. First logistic regression with bootstrap optimism correction estimated the association and AUC. A sensitivity-first operating threshold, selected using nested resampling and CRP/NLR comparators, was evaluated.

A total of 101 participants developed 30-day AL after the POD3 landmark assessment. PLR slope was associated with AL (OR=1.31 per 10 units/day; 95% CI: 1.02-1.89; P=0.039) with AUC 0.74 (95% CI: 0.58-0.86) and optimism-corrected AUC 0.72. The sensitivity-first threshold achieved sensitivity 83.0%, specificity 61.0%, PPV 12.0%, and NPV 98.3%, with a median lead-time of 2 days (IQR=1 to 3) from POD3 assessment to radiologic confirmation. In the CRP subset, AUCs were 0.67 for CRP POD3, 0.70 for CRP slope, and 0.62 for NLR POD3 (all P>0.20 vs. PLR slope). Decision-curve analysis favored PLR-slope triage over imaging-all-or-none triage across 5% to 12% threshold probabilities.

In our study, the POD1-to-3 PLR slope is a low-cost, POD3-available rule-out adjunct that may provide earlier warning before radiologic confirmation of clinically relevant AL. Multicenter validation is required to confirm generalizability, finalize operating thresholds, and quantify downstream imaging and outcome trade-offs.