Sternal Infection Burden Index (SIBI-S) for risk stratification of sternal wound infection after isolated coronary artery bypass grafting: an international multicenter cohort study

Background sternal wound infection remains a serious complication following coronary artery bypass grafting (CABG), associated with substantial morbidity and mortality. Existing risk stratification models rely predominantly on clinical and procedural factors and incompletely capture the perioperative inflammatory and immunonutritional milieu that underlies impaired wound healing. Objectives To evaluate the association between the Sternal Infection Burden Index–Surgical (SIBI-S), a composite inflammatory and immunonutritional index, and the risk of sternal wound infection following isolated CABG. Methods This international, multicenter retrospective cohort study included 6,790 adult patients who underwent isolated CABG between 2010 and 2025 across four centers. SIBI-S was calculated using the C-reactive protein–to–albumin ratio combined with the neutrophil-to-lymphocyte ratio. The primary endpoint was clinically diagnosed sternal wound infection. Time-to-event analyses were performed using the Kaplan–Meier methods and Cox proportional hazards models stratified by center and adjusted for established clinical and procedural risk factors. SIBI-S was analyzed as a continuous variable, per 1–standard deviation (SD) increase, per 10-unit increase, and using restricted cubic spline modeling to assess non-linear associations. Results During follow-up, 283 patients (4.2%) developed sternal wound infection. Mean SIBI-S values were significantly higher among patients who developed infection compared with those who did not (17.1 ± 17.8 vs 10.4 ± 14.1; P < 0.001). In adjusted Cox models, each 1-SD increase in SIBI-S was associated with a more than fourfold increased risk of sternal wound infection (HR 4.29, 95% CI 3.50–5.27), and each 10-unit increase conferred a nearly threefold higher risk (HR 2.76, 95% CI 2.39–3.18; both P < 0.001). Restricted cubic spline analyses demonstrated a strong non-linear, threshold-dependent association between SIBI-S and infection risk (P for nonlinearity = 0.0002). Empirical cumulative distribution analyses revealed a rightward shift of the entire SIBI-S distribution among patients with sternal wound infection, indicating a consistently higher inflammatory burden across the full spectrum of values. Conclusions SIBI-S is a strong and independent predictor of sternal wound infection following isolated CABG, demonstrating a non-linear, graded relationship with infection risk. By integrating complementary inflammatory and immunonutritional pathways, SIBI-S may enhance perioperative risk stratification beyond conventional clinical models.17AFor image description, please refer to the figure legend and surrounding text.17BFor image description, please refer to the figure legend and surrounding text.