Calibrated MRI to evaluate cerebral hemodynamics in patients with an internal carotid artery occlusion

Abstract

The purpose of this study was to assess whether calibrated magnetic resonance imaging (MRI) can identify regional variances in cerebral hemodynamics caused by vascular disease. For this, arterial spin labeling (ASL)/blood oxygen level-dependent (BOLD) MRI was performed in 11 patients (65±7 years) and 14 controls (66±4 years). Cerebral blood flow (CBF), ASL cerebrovascular reactivity (CVR), BOLD CVR, oxygen extraction fraction (OEF), and cerebral metabolic rate of oxygen (CMRO2) were evaluated. The CBF was 34±5 and 36±11 mL/100 g per minute in the ipsilateral middle cerebral artery (MCA) territory of the patients and the controls. Arterial spin labeling CVR was 44±20 and 53±10% per 10 mm Hg ▵EtCO2 in patients and controls. The BOLD CVR was lower in the patients compared with the controls (1.3±0.8 versus 2.2±0.4% per 10 mm Hg ▵EtCO2, P<0.01). The OEF was 41±8% and 38±6%, and the CMRO2 was 116±39 and 111±40 μmol/100 g per minute in the patients and the controls. The BOLD CVR was lower in the ipsilateral than in the contralateral MCA territory of the patients (1.2±0.6 versus 1.6±0.5% per 10 mmHg ▵EtCO2, P<0.01). Analysis was hampered in three patients due to delayed arrival time. Thus, regional hemodynamic impairment was identified with calibrated MRI. Delayed arrival artifacts limited the interpretation of the images in some patients.

Figure 1

Hemodynamic maps. (A) Mean maps of all patients (P) without delayed arrival artifacts (DAAs) are shown (left hemisphere), mean maps of their corresponding controls (C, right hemisphere). (B) Mean maps of the patients with DAA (P, left hemisphere) and their corresponding controls (C, right hemisphere). Delayed arrival artifacts are visible on the ASL images of the patients with DAA and this propagates to the OEF and CMRO₂ maps. Note that the images of the control subjects of the patients with DAA (B, right hemispheres) seem to have higher values compared with the control subjects of the patients without DAA (A, right hemispheres). We hypothesize this to be caused by the fact that the right hemisphere in (B) shows the mean image of three controls while the right hemisphere in (A) shows the mean image of 11 controls which decreases the amount of variation on which the mean perfusion images are based. BOLD CVR, blood oxygen level-dependent cerebrovascular reactivity (% per 10 mm Hg ▵EtCO₂); CBF, cerebral blood flow (in mL/100 g per minute); CBF-HC, cerebral blood flow during hypercapnia (in mL/100 g per minute); CMRO₂, cerebral metabolic rate of oxygen (in μmol/100 g per minute); OEF, oxygen extraction fraction (in %).

Figure 2

Patient example. (A) Example images and quantitative data of a 54-year-old male patient with a right-sided internal carotid artery (ICA) occlusion. The oxygen extraction fraction is increased in the gray matter of both MCA territories and the ability of the vasculature to modulate its tone in response to a hypercapnic stimulus (ASL CVR) is exhausted in both hemispheres. (B) Example images and quantitative data of a 70-year-old male patient with a bilateral ICA occlusion. The OEF is higher and the CMRO₂ are lower in the gray matter of the left compared with the right MCA territory. ASL CVR, arterial spin labeling cerebrovascular reactivity; BOLD, blood oxygenation level-dependent; CBF, cerebral blood flow; CMRO₂, cerebral metabolic rate of oxygen; MCA, middle cerebral artery; OEF, oxygen extraction fraction.

Figure 3

(A) Demonstration of the error in the OEF estimate introduced by errors in the CBF and ΔBOLD measurements performed at hypercapnia and hyperoxia breathing level. (B) Demonstration of the OEF error introduced by errors in the α, β, ΔCMRO₂, and the ΔC_aC_v. CBF_CO2, cerebral blood flow measured at hypercapnia breathing; CBF_O2, CBF measured at hyperoxia breathing; ΔBOLD_CO2, change in blood oxygenation level-dependent signal change from baseline to hypercapnia breathing; ΔBOLD_O2, change in BOLD signal change from baseline to hyperoxia breathing; OEF, oxygen extraction fraction; ΔCMRO₂, the relative change in the CMRO₂ as a result of hypercapnia breathing; ΔC_aC_v, the ratio between the arterial to venous concentration of oxygen before and during hyperoxia breathing.