CT and MR perfusion can discriminate severe cerebral hypoperfusion from perfusion absence: evaluation of different commercial software packages by using digital phantoms

Abstract

Computed tomography perfusion (CTP) and magnetic resonance perfusion (MRP) are expected to be usable for ancillary tests of brain death by detection of complete absence of cerebral perfusion; however, the detection limit of hypoperfusion has not been determined. Hence, we examined whether commercial software can visualize very low cerebral blood flow (CBF) and cerebral blood volume (CBV) by creating and using digital phantoms. Digital phantoms simulating 0-4% of normal CBF (60mL/100g/min) and CBV (4mL/100g/min) were analyzed by ten software packages of CT and MRI manufacturers. Region-of-interest measurements were performed to determine whether there was a significant difference between areas of 0% and areas of 1-4% of normal flow. The CTP software detected hypoperfusion down to 2-3% in CBF and 2% in CBV, while the MRP software detected that of 1-3% in CBF and 1-4% in CBV, although the lower limits varied among software packages. CTP and MRP can detect the difference between profound hypoperfusion of <5% from that of 0% in digital phantoms, suggesting their potential efficacy for assessing brain death.