Cerebral watershed hypoperfusion in subarachnoid hemorrhage: computed tomography perfusion analysis

Abstract

Object A better understanding of the pathophysiology of vasospasm-induced delayed cerebral ischemia and earlier detection of hypoperfusion before ischemic injury are needed to guide therapy in subarachnoid hemorrhage (SAH). The cerebrovascular physiology of the major arterial territories differs from that of the watershed zones (WZs) in a way that would suggest a differential topographic sensitivity of the brain to vasospasm. The primary end point of the study was to investigate the vasospasm-induced hypoperfusion in relation to cerebrovascular topography and timing from the onset of SAH. Methods Forty-one patients were prospectively enrolled and scheduled for perfusion-weighted (PW) CT at 3 time points (≤ 3 days, Days 4–8, and Days 9–15 after SAH). Perfusion-weighted CT maps were visually assessed for side-to-side perfusion asymmetry. The PW CT topographic pattern was categorized into absence of asymmetry, WZ, and vascular territory hypoperfusion. Perfusion-weighted CT revision was performed by investigators blinded to clinical information. The null hypothesis for the primary end point was that there would be no difference in hypoperfusion space-time distribution among the different vascular territories. Multivariate logistic regression and Cox proportional hazards modeling were used for statistical analysis. Results Delayed cerebral ischemia occurred in 26 patients and its predicting variables were increasing age (p = 0.045), Fisher grade (p = 0.007), and hypoperfusion on the PW CT performed within the 1st 72 hours after SAH (p = 0.004). The timing of the PW CT with respect to the day of SAH affected the topographic pattern of hypoperfusion: watershed-zone hypoperfusion was more common within the first 3 days after SAH (p = 0.018), while the proportion of territorial hypoperfusion increased subsequently. Among the different covariates, a young age was independently associated with a higher risk of developing hypoperfusion in the WZs (p = 0.02). Conclusions This study suggests the existence of a cerebral topographic heterogeneity to the hemodynamic effects of SAH and differential pathogenetic mechanisms of hypoperfusion according to timing, age, and brain topography. Hypoperfusion in the WZs may be an early precursor to more profound ischemic events. The PW CT detection of such brain-sensitive zones could offer a warning signal of the early hemodynamic effects of SAH and cerebral vasospasm.