Comparison of Perfusion Weighted Magnetic Resonance Imaging and Single Photon Emission Computed Tomography for Assessment of Cerebrovascular Reserve in Symptomatic Carotid Territory Stenosis

Background: Perfusion Weighted Magnetic Resonance Imaging (PW-MRI) and HMPAO Single Photon Emission Computed Tomography (SPECT) are both cerebral perfusion measurement techniques. Imaging before and after acetazolamide administration can assess cerebrovascular reserve in symptomatic haemodynamic cerebrovascular disease. We compared SPECT and PW-MRI parameters in this patient group. Methods: We identified 10 patients with haemodynamically induced symptoms and intraor extra-cranial arterial stenoses with back-to-back acetazolamide challenge SPECT and PW-MRI, 4 of whom had resting studies. Regions of interest (ROIs) were applied to parameter maps using an ASPECTS template and perfusion parameters expressed relative to contralateral ROIs, giving 118 challenge and 48 resting ROIs. Results: SPECT relative cerebral blood flow (rCBF) correlated with PW-MRI time to peak (TTP) (r = −0.568), mean transit time (MTT) (r = −0.317), regional cerebral blood flow (rCBF) (r = 0.299) and cerebral blood volume (CBV) (r = 0.224). Bias between SPECT and PW-MRI rCBF was small (−0.018) with wide limits of agreement and a systematic measurement error. Preto post-acetazolamide PW-MRI rCBF change showed poor sensitivity and specificity for detecting change in SPECT rCBF. SPECT and PW-MRI rCBF had stronger correlation and smaller bias in unilateral stenosis than with bilateral stenosis. Conclusion: Systematic bias between techniques limits interchangeability in cerebrovascular reserve measurement in patients with cerebrovascular stenosis.