Deviation From Personalized Blood Pressure Targets Is Associated With Worse Outcome After Subarachnoid Hemorrhage

Abstract

Background and Purpose- Optimal blood pressure (BP) management during the early stages of aneurysmal subarachnoid hemorrhage remains uncertain. Observational studies have found worse outcomes in patients with increased hemodynamic variability, suggesting BP optimization as a potential neuroprotective strategy. In this study, we calculated personalized BP targets at which cerebral autoregulation was best preserved. We analyzed how deviation from these limits correlates with functional outcome. Methods- We prospectively enrolled 31 patients with aneurysmal subarachnoid hemorrhage. Autoregulatory function was continuously measured by interrogating changes in near-infrared spectroscopy (NIRS)-derived tissue oxygenation-a surrogate for cerebral blood flow-as well as intracranial pressure (ICP) in response to changes in mean arterial pressure using time-correlation analysis. The resulting autoregulatory indices were used to identify the upper and lower limit of autoregulation. Percent time that mean arterial pressure exceeded limits of autoregulation was calculated for each patient. Functional outcome was assessed using the modified Rankin Scale at discharge and 90 days. Associations with outcome were analyzed using ordinal multivariate logistic regression. Results- Personalized limits of autoregulation were computed in all patients (age 57.5±13.4, 23F, mean World Federation of Neurological Surgeons 2±1, monitoring time 67.8±50.8 hours). Optimal BP and limits of autoregulation were calculated on average for 89.5±6.7% of the total monitoring period. ICP- and NIRS-derived optimal pressures strongly correlated with one another (P<0.0001). Percent time that mean arterial pressure deviated from limits of autoregulation significantly associated with worse functional outcome at discharge (NIRS, P=0.001; ICP, P=0.004) and 90 days (NIRS, P=0.002; ICP, P=0.003), adjusting separately for age, World Federation of Neurological Surgeons, vasospasm, and delayed cerebral ischemia. Conclusions- Both invasive (ICP) and noninvasive (NIRS) determination of personalized BP targets after aneurysmal subarachnoid hemorrhage is feasible, and these 2 approaches revealed significant collinearity. Furthermore, exceeding individualized limits of autoregulation was associated with poor functional outcomes.

Keywords: Cerebral autoregulation; Cerebral blood flow; Cerebrovascular disease; Subarachnoid hemorrhage.

Figure 1.. Generation of MAP_OPT and personalized limits of autoregulation.

Panel A shows the characteristic parabolic curve that is generated when plotting an autoregulatory index against a range of binned blood pressures during a 4-hour recording. The vertex of this U-curve is located at the point with the lowest autoregulatory index and thus represents the optimum blood pressure (MAP_OPT). By setting a threshold for impaired autoregulation at TOx=+0.30, intersecting MAP values provide estimates of lower and upper limits of autoregulation (LLA, ULA). These intersections correspond with the inflection points of Lassen’s autoregulatory curve, as shown by the dotted vertical lines between Panels A and B. In Panel C, a continuous time trend of these limits can be displayed in real-time, while superimposing the patient’s actual BP in black, to provide clinicians with a dynamically updating BP target. Panel D shows another patient’s 12-hour recording, during which the patient’s blood pressure frequently oscillated outside personalized limits of autoregulation (regions highlighted in dark red above ULA).

Figure 2.. Correlation between ICP- and NIRS-derived limits of autoregulation.

ICP- and NIRS-derived autoregulatory parameters significantly correlated in bivariate non-parametric analysis (Panels A–C). Both ICP- and NIRS-derived %time outside LA also correlated with one another (Panel D). Best-fit lines and corresponding 95% confidence intervals are displayed to illustrate linear relationships. LA: limits of autoregulation.

Figure 3.. Percent time MAP spent outside limits of autoregulation associates with discharge and 90-day functional outcome following aSAH.

A&B: Deviation from ICP-derived autoregulatory limits associated with discharge and 90-day outcome. C&D: Deviation from NIRS-derived autoregulatory limits associated with discharge and 90-day outcome. P-values shown demonstrate statistical significance in univariate ordinal regression. Error bars represent 95% confidence intervals. OR: odds ratio; CI: confidence interval.