Perfusion Monitoring During Cardiopulmonary Bypass in CABG: Temperature, Lactate, and NIRS

Abstract

Objectives: During cardiopulmonary bypass (CPB), temperature, lactate, and NIRS-derived cerebral oxygen saturation (rSO₂) are recorded simultaneously; however, whether they provide complementary or interchangeable information remains unclear. In adults undergoing elective on-pump coronary artery bypass grafting (CABG), we examined their within-patient relationships across consecutive CPB phases and tested whether temperature and lactate remained independently associated with rSO₂ after adjustment for hematocrit, mean arterial pressure (MAP), and PaCO₂. Methods: This prospective, single-center, observational study included 85 adults undergoing elective isolated on-pump CABG. MAP, esophageal temperature, bilateral frontal rSO₂, lactate, hematocrit, and PaCO₂ were recorded simultaneously at seven CPB-related phases, yielding 595 phase-level observations. Within-patient associations were evaluated using repeated-measures correlation. Adjusted associations were assessed with a linear mixed-effects model and supported by active-CPB, hemisphere-specific, and interhemispheric-agreement analyses. Results: Cerebral rSO₂ decreased after CPB onset, lactate increased progressively, and hematocrit declined. Within-patient correlations among temperature, lactate, and rSO₂ were statistically significant but negligible (all r²≤0.025). In the mixed-effects model, hematocrit (standardized β=1.66), MAP (β=1.05), and PaCO₂ (β=0.85) were independently associated with cerebral rSO₂ (all p≤0.003), whereas temperature and lactate were not. This pattern was preserved in active-CPB and hemisphere-specific analyses. Interhemispheric agreement was high, supporting the use of the bilateral mean for this population-level analysis. Conclusion: During CPB in elective CABG, temperature, lactate, and cerebral rSO₂ behaved as complementary, non-redundant signals rather than interchangeable perfusion indices. Cerebral oxygenation was independently associated with hematocrit, MAP, and PaCO₂, supporting cautious multiparameter interpretation and multicenter outcome-based confirmation.