Dynamic cerebral autoregulatory response to blood pressure rise measured by near-infrared spectroscopy and intracranial pressure

Objectives. Noninvasive near-infrared spectroscopy (NIRS) continuously monitors changes in cerebral hemoglobin saturation (Hb(Diff)) and content (Hb(Total)). It may allow visualization of the dynamic cerebral autoregulatory response to rapid blood pressure increases without relevant contamination of the MRS signal from extracerebral hemoglobin. Design: Prospective cohort study. Settings: Multidisciplinary pediatric intensive care unit. Patients: Six consecutive children in coma due to severe encephalopathy (head trauma, five patients; mumps encephalitis, ne patient) requiring artificial ventilation, invasive arterial blood, and intracranial pressure monitoring. Interventions: Frontotemporal recording of Hb(Diff) and Hb(Total) while rapidly elevating blood pressure by bolus injection of phenylephrine. Measurements and Results: During an increase of blood pressure of 13 +/- 1 mm Hg with a "rise time" of 16 +/- 1 secs (mean of a total of 31 injections +/- SEM), a significant linear correlation was found between Hb(Diff) and intracranial pressure signals (mean coefficient, 0.46 +/- 0.04) but not between Hb(Total) and intracranial pressure. Three response patterns were observed. First, Hb(Diff) and intracranial pressure reduction, corresponding with vasoconstriction and normal dynamic autoregulation (n = 3); second, Hb(Diff) and intracranial pressure increase, corresponding with persistent vasodilation and abolished autoregulation (n = 11); and third, transient Hb(Diff) and intracranial pressure increase followed by a decrease at peak blood pressure elevation, called impaired autoregulation (n = 15). In one patient with fatal brain swelling, phenylephrine testing showed no effect on NIRS signals (n = 2). Furthermore, there were significant correlations between 31 pooled interindividual pairs, of HbDiff changes with intracranial pressure changes (values at baseline averaged over 60 secs subtracted from values at peak blood pressure elevation averaged over 5 secs), with a correlation coefficient of .82 (p < .001). Conclusions. NIRS represents a new and promising technique for bedside determination of dynamic cerebral autoregulation during acutely induced blood pressure rise. The significant correlations found between NIRS signals and intracranial pressure excluded relevant extracerebral contamination of the NIRS signals. In our patients with severe encephalopathy, dynamic autoregulation was in most instances not fully preserved.