Temporal evolution of cerebral computed tomography perfusion after acute subarachnoid hemorrhage : a prospective cohort study

Abstract

Background: Changes in cerebral perfusion occur in subarachnoid hemorrhage that possibly relate to clinical presentation and complications. Purpose: To evaluate changes in computed tomography perfusion (CTP) parameters between the acute and subacute stage of subarachnoid hemorrhage. To analyze correlation of these parameters to SAH severity and delayed cerebral ischemia. Material and Methods: Cerebral CT perfusion was assessed in a prospective cohort of 44 patients with acute subarachnoid hemorrhage at < 72 h (CTP1) and 8–10 days (CTP2), using the mean of all regions of interest. Regions of interest were located at arterial territories of the anterior, middle, and posterior cerebral artery and basal ganglia and midpons cerebellar hemispheres. Linear regression models (univariable and multivariable) were used to explore the association between changes in perfusion parameters (absolute and relative differences) and relevant clinical data. Results: Worse perfusion parameters on the first 72 h were correlated with poor admission clinical scores: cerebral blood flow positively correlated with Glasgow Coma Scale (rS ¼ 0.398, P ¼ 0.008), and negatively correlated with Hunt & Hess scale (rS ¼ 0.348, P ¼ 0.020) and World Federation of Neurosurgeons scale (rS ¼ 0.384, P ¼ 0.010). Cerebral blood volume positively correlated with Glasgow Coma Scale (rS ¼ 0.332, P ¼ 0.028) and negatively correlated with World Federation of Neurosurgeons scale (rS ¼ 0.353, P ¼ 0.019). Mean transit time negatively correlated with Glasgow Coma Scale (rS ¼ 0.415, P ¼ 0.005) and positively correlated with Hunt & Hess scale (rS ¼ 0.471, P ¼ 0.001) and World Federation of Neurosurgeons scale (rS ¼ 0.386, P ¼ 0.010) scores. There were no differences between absolute CTP1/CTP2 parameters. Patients with delayed cerebral ischemia had DTmax mean decrease of 2.08 s (95% CI ¼ 4.04–0.12; P ¼ 0.038). Conclusion: Early cerebral hypoperfusion correlates with poor clinical grade at admission in subarachnoid hemorrhage and with higher amounts of blood. Tmax was decreased at 8–10 days, in patients with delayed cerebral ischemia, which may favor the application value of Tmax in signaling delayed cerebral ischemia.