INTERPRETING AND EXPANDING ON THE OBSERVED CORRELATION BETWEEN THE AMBULATORY ARTERIAL STIFFNESS INDEX AND THE LOWER LIMIT OF CEREBRAL AUTOREGULATION DURING CARDIAC SURGERY

Abstract

Abstract Objective: The lower limit of cerebral autoregulation (LLA) refers to the mean arterial pressure below which cerebral blood flow becomes pressure dependent, resulting increased stroke risk. Thus, markers correlated with the LLA are highly desired. The LLA measured during cardiac surgery, correlates with the Ambulatory Arterial Stiffness Index (AASI) determined from intraoperative continuous radial blood pressure (BP) before cardiopulmonary bypass (‘baseline’). Using these data the present study investigated factors underlying this association and its potential refinement. Design and method: The study population included 181 patients undergoing cardiac surgery (age 71 ± 8 years, 68% males). Given the linear relationship between systolic BP (SBP) and diastolic BP (DBP), the AASI is defined as 1 minus the DBP-on-SBP regression slope. The analytic expression of AASI is 1-Rsd/K, where Rsd is the SBP-DBP correlation coefficient, and K = SD (SBP)/SD (DBP), where SD = standard deviation. K measures the relative increase in arterial stiffness during systole (‘stiffening’). Factors affecting the LLA were investigated using multilinear regression models with backward-stepwise procedure. Each model included age, sex, diabetes mellitus, heart rate, and one BP factor (average SBP, DBP, and pulse pressure) during baseline, called ‘group1’, and one or more of the factors including body mass index (BMI), Rsd, 1/K, AASI (equivalent to Rsd/K), and BMI∗Rsd/K (equals BMI∗(1-AASI), called ‘group2’. Results: BMI, Rsd and 1/K, alone or taken together, adjusted to group 1 factors, were found to be the only significant explanatory predictors. Results were insensitive to the BP factor used. These three predictors were either weakly- or not-correlated with one another. The table below shows the correlation coefficient r of a single predictor from group2 with the LLA and its P-value, and the P value adjusted for group1 predictors. Conclusions: Results explain the likelihood of an elevated LLA by a single- or combined predictors associated with the known effects on cerebral vasculature of obesity (lower BMI); elevated arterial stiffening (greater K) and reduced beat-by-beat BP control (lower Rsd). Thus, combining these predictors may be useful markers for cerebral autoregulation.