Abstract
Background/Objectives: We compared preoperative complete blood count (CBC)-derived indices for in-hospital postoperative complications after incisional hernia repair. Methods: This single-center retrospective study (January 2020–March 2025) included 106 of 173 eligible patients; an expanded cohort of 115 permitted same-day samples. The first qualifying admission CBC was used. Pretreatment eligibility was established during clinical-record review, but treatment timestamps were unavailable in the analytical extract. Five markers were compared using ROC-AUC, average precision (AP), Firth regression, and nested cross-validation. Results: The strict cohort had 11 complications, nine major outcomes (Clavien–Dindo ≥ III), and five deaths. For any complication, red-cell distribution width (RDW) had the highest numerical AUC (0.757), versus 0.718 for the cumulative inflammatory index (IIC). The major-outcome AUCs were 0.836 for IIC and 0.829 for RDW; the AP difference was 0.209 (95% CI: −0.186 to 0.497), and capture curves crossed. The adjusted IIC odds ratio per standard deviation was 2.96 (95% CI: 1.16–9.08; Holm p = 0.052). Adding IIC to age and emergency presentation increased the out-of-fold AUC from 0.726 to 0.804, but the increment remained uncertain (95% CI: 0.000–0.170). Conclusions: RDW ranked highest numerically for the primary endpoint; IIC showed numerical enrichment for major outcomes without established superiority. Selection, uncertain pretreatment timing, and few nonfatal major events limit generalizability. These exploratory markers require prospective independent validation before clinical use.