Cerebral Autoregulation Monitoring with Ultrasound-Tagged Near-Infrared Spectroscopy in Cardiac Surgery Patients

Abstract

Background: Individualizing mean arterial blood pressure (MAP) based on cerebral blood flow (CBF) autoregulation monitoring during cardiopulmonary bypass (CPB) holds promise as a strategy to optimize organ perfusion. The purpose of this study was to evaluate the accuracy of cerebral autoregulation monitoring using microcirculatory flow measured with innovative ultrasound-tagged near-infrared spectroscopy (UT-NIRS) noninvasive technology compared with transcranial Doppler (TCD).

Methods: Sixty-four patients undergoing CPB were monitored with TCD and UT-NIRS (CerOx™). The mean velocity index (Mx) was calculated as a moving, linear correlation coefficient between slow waves of TCD-measured CBF velocity and MAP. The cerebral flow velocity index (CFVx) was calculated as a similar coefficient between slow waves of cerebral flow index measured using UT-NIRS and MAP. When MAP is outside the autoregulation range, Mx is progressively more positive. Optimal blood pressure was defined as the MAP with the lowest Mx and CFVx. The right- and left-sided optimal MAP values were averaged to define the individual optimal MAP and were the variables used for analysis.

Results: The Mx for the left side was 0.31 ± 0.17 and for the right side was 0.32 ± 0.17. The mean CFVx for the left side was 0.33 ± 0.19 and for the right side was 0.35 ± 0.19. Time-averaged Mx and CFVx during CPB had a statistically significant "among-subject" correlation (r = 0.39; 95% confidence interval [CI], 0.22-0.53; P < 0.001) but had only a modest agreement within subjects (bias 0.03 ± 0.20; 95% prediction interval for the difference between Mx and CFVx, -0.37 to 0.42). The MAP with the lowest Mx and CFVx ("optimal blood pressure") was correlated (r = 0.71; 95% CI, 0.56-0.81; P < 0.0001) and was in modest within-subject agreement (bias -2.85 ± 8.54; 95% limits of agreement for MAP predicted by Mx and CFVx, -19.60 to 13.89). Coherence between ipsilateral middle CBF velocity and cerebral flow index values averaged 0.61 ± 0.07 (95% CI, 0.59-0.63).

Conclusions: There was a statistically significant correlation and agreement between CBF autoregulation monitored by CerOx compared with TCD-based Mx.

Figures

Figure 1

Linear correlation and Bland-Altman bias analysis results between time averaged Mx and CFVx values obtained during CPB (A). Similar statistical analyses between optimal blood pressure (ABPopt) measured by CFVx and Mx respectively, are shown (B). The dashed lines represent the 95% confidence interval of the regression line and the 95% limits of agreement for the bias analysis. Data that had the same value among patients are presented as one data point.

Figure 2

Average Mx and CFVx during cardiopulmonary bypass in 5mmHg bins. Both Mx and CFVx shows increase in their value as mean arterial pressure moves away from the optimal blood pressure indicating trends towards pressure dependent changes in cerebral blood flow. Mean arterial pressure at lowest Mx or CFVx were defined as the optimal blood pressure (Black arrow). In this example, the optimal blood pressure based on MAP at which Mx is the lowest is 80 mmHg. Similarly, MAP at which CFVx is at the lowest is 80mmHg. MxR=Mx Right; CFVx1=CFVx Right; MxL=Mx Left; CFVx2=CFVx Left;

Figure 3

Histogram showing the percentage of patients versus the differences in optimal blood pressure (ABPopt) measured during cardiopulmonary bypass, as defined by Mx and CFVx respectively.