PerfSCORE as a Predictor of Early Mortality in Open-Heart Surgery: Comparative Evaluation with NIRS and S100B

This study aims to investigate the correlation between NIRS and S100B, and the effects of NIRS, S100B, and PerfSCORE parameters on morbidity and mortality in patients undergoing open-heart surgery who had their cardiopulmonary bypass (CPB) flow volume adjusted using NIRS.

In this prospective study, 60 adult patients undergoing open-heart surgery between January and June 2024 were enrolled. Intraoperative NIRS values were recorded at baseline, at the initiation of cardiopulmonary bypass (CPB), at CPB termination, and at the end of surgery for both cerebral hemispheres. Serum S100B levels were measured preoperatively (T1), and postoperatively at 6 hours (T2) and 24 hours (T3). Correlation analyses were performed between NIRS values and S100B levels. In addition, the effects of S100B, NIRS, and PerfSCORE on mortality and morbidity were compared.

A total of 73 procedures were performed in 60 patients. A weak negative correlation was found between early NIRS values and postoperative S100B levels ( r = -0.163 to -0.282). No significant correlation was observed between NIRS2 values and S100B, while a weak positive correlation emerged between NIRS3 and S100B1. No significant association was found between S100B levels and the incidence of early mortality or neurological complications. In contrast, PerfSCORE parameters were significantly associated with early mortality (Nagelkerke's R ² = 0.765), showing a prediction accuracy of 83.3% for deceased patients and 98.1% for survivors.

Although weak correlations exist between cerebral oxygenation and S100B levels, S100B protein alone does not appear to be a reliable marker for early mortality. PerfSCORE parameters demonstrated greater predictive value and clinical utility.