Detection of impaired cerebral autoregulation using spectral analysis of intracranial pressure waves

Successful resuscitation following severe traumatic brain injury (TBI) requires rapid evaluation of intracranial pressure (ICP), cerebrovascular reactivity (autoregulation), and cerebral metabolism. During impaired autoregulation, inadequate cerebral blood flow (CBF) can lead to ischemia while excessive CBF can result in elevated ICP. Without information regarding the state of autoregulation, treatment of either situation may ameliorate one problem but exacerbate the other. It has been hypothesized that fast Fourier transform (FFT) analysis of arterial blood pressure (BP) and ICP waves can differentiate states of intact and impaired autoregulation. BP and ICP waves were recorded in canines before and after ischemic injury during arterial normotension, hypertension, and hypotension induced with dopamine or nitroprusside infusion. Transfer functions (TFn) were calculated from FFT spectra as ratios of ICP and BP harmonic peak amplitudes to distinguish states of vasoreactivity. During normotension and hypertension, autoregulation was intact and TF1 averaged 0.05. During hypotension, TF1 averaged 0.22 (8 x baseline, p < 0.010). During impaired autoregulation following ischemic injury, TF1 averaged 0.50 (18 x baseline, p < 0.010; 2 x nitroprusside levels, p < 0.01). This large difference in TF relative to baseline extended over a large range of BP (60 < BP < 180 mm Hg). Based on these data and previous results, it was estimated that TF can differentiate impaired autoregulation from effects solely related to elevated ICP or active vasodilation for ICP < 30-40 mm Hg. This suggests that for specific, but widely applicable physiologic conditions, spectral analysis can identify states of impaired autoregulation and, as an adjunct to traditional monitoring techniques, aid in acute resuscitation and prevention of secondary injury in TBI.