Quantification of cerebral blood flow in healthy volunteers and type 1 diabetic patients: comparison of MRI arterial spin labeling and [(15)O]H2O positron emission tomography (PET)

Abstract

Purpose: To compare cerebral blood flow (CBF) values measured using magnetic resonance imaging (MRI) arterial spin labeling (ASL) with those obtained with [(15)O]H2O positron emission tomography (PET), the gold standard for measuring CBF in vivo.

Materials and methods: Data were collected in 11 healthy men and in 20 age- and body mass index (BMI)-matched type 1 diabetic men. Pseudo-continuous ASL (PCASL) data were acquired at 3 T and [(15)O]H2O PET scans were acquired using a high-resolution PET scanner. Input functions were obtained using on-line arterial blood sampling. Whole brain and regional CBF values were compared.

Results: For both modalities, whole brain CBF was similar in both subject groups. In groups combined, average whole brain CBF was 0.30 ± 0.05 mL · cm(-3) · min(-1) for [(15)O]H2O PET and 0.34 ± 0.05 mL · cm(-3) · min(-1) for ASL MRI (P < 0.01). A significant correlation between methods was observed for whole brain, gray and white matter. In 12 out of 33 brain regions a significant difference between methods was observed.

Conclusion: PCASL provides CBF values that correlate with [(15)O]H2O PET-derived values, but is less accurate. PCASL may be an attractive alternative when absolute quantification is not needed.

Keywords: [15O]H2O positron emission tomography (PET); arterial spin labeling (ASL); cerebral blood flow (CBF); diabetes.