Abstract
BACKGROUND: The human brain is a highly complex and nonlinear system, nonlinear complexity measures such as approximate entropy (ApEn) and sample entropy (SampEn) can better reveal characteristics of brain dynamics. However, no studies report complexity of perioperative physiological signals to reveal how brain complexity associates with age, varies along with the development of surgery and postoperative neurological complications. AIM: This study examined the complexity of intraoperative regional cerebral oxygen saturation (rSO(2)), aiming to reveal brain dynamics during surgery. METHODS: This retrospective cohort study enrolled patients who scheduled for robot-assisted urological surgery. Intraoperative rSO(2) was continuously monitored throughout the surgery. Postoperative delirium (POD) was diagnosed by the Confusion Assessment Method. ApEn and SampEn were used to characterize the complexity of rSO(2). Pearson correlation coefficients were used to measure the correlation between complexity of rSO(2) and age. The association between complexity of rSO(2) and POD was examined using T tests. RESULTS: A total of 68 patients (mean [SD] age, 63.0 (12.0) years; 47 (69.1%) males) were include in this analysis. There was a significant reverse relationship between the complexity of rSO(2) and age (The correlation coefficients range between - 0.32 and - 0.28, all p < 0.05). Patients ≥ 75 years showed significantly lower complexity of rSO(2) than the other two groups. Older age remained an independent factor influencing complexity of rSO(2) after adjusting for a number of covariates. Six patients (8.8%) developed POD, and POD patients had lower complexity of rSO(2) compared with non-POD patients. CONCLUSIONS: The complexity of rSO(2) may serve as a new candidate marker of aging and POD prediction.