Preoperative SASP burden predicts perioperative vulnerability in older patients undergoing elective cardiac surgery: a prospective cohort study

Abstract Cellular senescence and its associated secretory phenotype (SASP) represent a fundamental mechanism of biological aging. SASP-related inflammation has been implicated in cardiovascular disease. However, its role in perioperative risk in cardiac surgery remains largely unexplored, and established risk scores such as EuroSCORE II and the STS Score do not regard biological aging or senescence. In this prospective single-center cohort study, 131 patients aged ≥ 65 years undergoing elective cardiac surgery were enrolled. A panel of 19 preoperative SASP biomarkers was quantified by a multiplex immunoassay. Of those, 17 markers were evaluated as predictors of a composite adverse outcome, an adapted multimorbidity score with 21.4% event rate, with MACCE (36.6%) as a secondary endpoint. Logistic regression, false discovery rate correction, bootstrap validation, and incremental value analyses (DeLong test, IDI, NRI) were used. Five markers showed nominal univariable significance (IGFBP-1, GDF-15, TNF-RI, TNF-α, and IGFBP-4), though none survived FDR correction. Adjusted for procedure type, IGFBP-1 (OR = 1.675, 95% CI = 1.175–2.389, p = 0.004) and IGFBP-4 (OR = 1.432, 95% CI = 1.044–1.966, p = 0.026) emerged as independent predictors of the composite adverse outcome (Model M3, AUC corrected = 0.746), numerically comparable to EuroSCORE II (AUC = 0.727). Model M3 improved net reclassification over EuroSCORE II (NRI + 0.528, p = 0.008) and the STS Score (NRI + 0.418, p = 0.040), with consistent estimates under Firth-penalized sensitivity analysis. Preoperative SASP biomarkers, particularly IGFBP-1 and IGFBP-4, independently predict adverse outcomes, namely multimorbidity, after elective cardiac surgery. They improve individual risk stratification beyond standard clinical scores in older patients, suggesting a measurable dimension of perioperative vulnerability that warrants external validation in larger cohorts.