Kunlin Guo, Toufeng Jin, Guangzhe Pian
Anastomotic leakage (AL) is a serious complication after colorectal cancer (CRC) surgery, and early postoperative risk stratification remains clinically challenging. The Geriatric Nutritional Risk Index (GNRI) integrates serum albumin and body weight information, but its role as a postoperative marker and its incremental value beyond albumin alone remain uncertain. This single-center retrospective cohort study included 302 patients who underwent laparoscopic radical colorectal cancer surgery at Yanbian University Hospital between January 2022 and October 2024. Postoperative day 3 (POD3) GNRI and POD3 albumin were evaluated using receiver operating characteristic analysis, paired DeLong testing, correlation and variance inflation factor assessment, exploratory multivariable logistic models, model comparison, 1000-resample bootstrap internal validation, and colon versus rectum subgroup and interaction analyses. AL occurred in 16/302 patients (5.30%). POD3 albumin had a numerically higher area under the receiver operating characteristic curve (AUC) than POD3 GNRI (AUC: 0.744, 95% confidence interval [CI]: 0.617-0.870 vs AUC: 0.691, 95% CI: 0.597-0.785; DeLong P = .331). POD3 GNRI and POD3 albumin were correlated (Pearson R = 0.774; Spearman ρ = 0.761). The apparent AUCs were 0.766 for the clinical model, 0.793 for clinical plus POD3 albumin, and 0.801 for clinical plus POD3 GNRI; corresponding optimism-corrected AUCs were 0.680, 0.706, and 0.724. Bootstrap warnings about fitted probabilities approaching 0 or 1 indicated model instability. Low POD3 GNRI was associated with increased AL risk and may help identify patients for postoperative risk enrichment and heightened surveillance. However, POD3 GNRI should not be interpreted as having established added value beyond albumin alone, and its incremental value remains uncertain. These exploratory findings require external validation.