Systemic inflammatory indices and clinical outcomes after emergency coronary artery bypass grafting for ST-elevation myocardial infarction: a retrospective observational study

Background Emergency coronary artery bypass grafting (CABG) after ST-segment elevation myocardial infarction (STEMI) is performed in a high-risk population in whom the relevance of complete blood count-derived inflammatory indices remains uncertain. We assessed their associations with in-hospital outcomes. Methods This single-center retrospective study included 143 of 194 adults who underwent emergency CABG after STEMI between 2010 and 2024. The systemic inflammatory response index (SIRI), systemic immune-inflammation index (SII), aggregate index of systemic inflammation (AISI), neutrophil-to-lymphocyte ratio (NLR), and systemic immune-inflammatory complex index (SIICI) were calculated from the last preoperative and first postoperative blood samples; the latter was generally obtained within approximately 30 min after skin closure. The primary endpoint combined in-hospital mortality, stroke, sepsis, low cardiac output syndrome, or re-exploration for bleeding; mortality was also analyzed separately. False discovery rate correction was applied across the five indices. Within-patient changes and time-by-group interactions were assessed. Each index was added separately to a base model comprising age, ejection fraction, and preoperative creatinine; Firth penalized logistic regression was used for mortality. Results The composite endpoint occurred in 25 patients (17.5%), and 12 (8.4%) died. After false discovery rate correction, preoperative SIRI, AISI, and SIICI were higher in the composite-outcome group, whereas all five early postoperative indices were lower. All indices increased after surgery, but the increase was smaller in the composite-outcome group, with significant time-by-group interactions for all indices except NLR. In standardized univariable models, preoperative SIRI, SII, AISI, and NLR were associated with the composite endpoint. Single-index discrimination was limited (area under the curve, 0.579–0.677), and no index improved discrimination beyond the base clinical model. For mortality, preoperative AISI showed the highest discrimination (area under the curve, 0.836; 95% confidence interval, 0.751–0.913), although adjusted estimates were unstable with extremely wide confidence intervals. Conclusions Several preoperative indices were higher in patients with adverse outcomes, whereas the early postoperative increase was smaller for several indices. None provided additional discrimination beyond the base clinical model. These findings are exploratory and hypothesis-generating. Postoperative indices should be interpreted as concurrent markers of perioperative clinical severity rather than validated predictors of subsequent events.