Arterial spin labeling for motor activation mapping at 3T with a 32-channel coil: Reproducibility and spatial accuracy in comparison with BOLD fMRI
Highlights
► We evaluated functional ASL for motor activation mapping compared to BOLD fMRI. ► fASL revealed activation in ipsilateral secondary motor areas. ► In terms of activation location, fASL reproducibility was comparable to BOLD. ► In terms of activation quantification, fASL reproducibility was higher than BOLD. ► In terms of spatial accuracy, fASL had higher specificity than BOLD.
Introduction
Functional MRI (fMRI) is the tool of choice for mapping variations in neuronal activity during motor or cognitive tasks or during sensitive or sensory stimulation. BOLD fMRI is considered to be the gold standard for clinical activation fMRI studies. The BOLD signal is based on capillary/venous hyperoxygenation in activated brain areas leading to a minor signal decrease on T2* images (Bandettini et al., 1992; Belliveau et al., 1991). The spatial accuracy of BOLD fMRI for motor activation mapping has shown good agreement with perioperative electrical stimulation (Lehericy et al., 2000) and PET (Ramsey et al., 1996). However, BOLD fMRI presents several limitations. Firstly, the BOLD signal results from complex interactions between simultaneous variations in cerebral blood flow (CBF) (Stefanovic et al., 2006), cerebral blood volume (Stefanovic et al., 2006), and the cerebral metabolic rate of oxygen (Tuunanen and Kauppinen, 2006). Incomplete knowledge of the respective contributions to neuronal activity impairs the robustness of BOLD fMRI and its reproducibility. Secondly, the sensitivity of BOLD fMRI to signals originating from macroscopic veins (Mangia et al., 2004) and to local modulation of vasoreactivity, impairs its accuracy. False-positive and false-negative activations were observed in areas such as neuro-oncology (Bartos et al., 2009) and ischemic pathology (Krainik et al., 2005, Rossini et al., 2004), and were attributed to vasoreactive modifications inherent to these specific pathological contexts. Functional arterial spin labeling (fASL) is an innovative biomarker of neuronal activity based on direct measurement of the local perfusion variations and offers an alternative to BOLD fMRI. Arterial spin labeling allows non-invasive imaging and quantification of brain perfusion using magnetically labeled arterial protons in the brain-feeding arteries as an endogenous tracer (Detre et al., 1992). Recently proposed for activation-fMRI, fASL eflects only the vascular component of the vascular coupling (Roy and Sherrington, 1890) and is potentially a more direct biomarker of neuronal activity than BOLD fMRI. fASL offers two main advantages in comparison with BOLD fMRI. The first advantage is the direct and absolute quantification of activation-related CBF (Detre et al., 1992). This potentially yields higher intra-individual reproducibility (Leontiev and Buxton, 2007) and is essential for longitudinal studies. The second advantage is more accurate spatial localization of neuronal activity (Jin and Kim, 2008, Luh et al., 2000) notably due to lower sensitivity to venous contamination (Aguirre et al., 2002, Luh et al., 2000). This is essential for presurgical planning because it conditions excision margins. fASL has shown high agreement with PET for CBF quantification and activation location during visual stimulation (Chen et al., 2008). Moreover, fASL has demonstrated a correlation between stimulation intensity and CBF increase not shown by PET (Chen et al., 2008). Yet fASL has a low signal-to-noise ratio (SNR) due to the minor difference between the label and control images (around 1% of label image magnitude) (Mildner et al., 2003). This is partially compensated for by using long acquisition times and encourages imaging at higher field strength. Some experimental studies have already used fASL to investigate visual, motor and language functions (Aguirre et al., 2002, Ances et al., 2008, Chen et al., 2008, Kemeny et al., 2005, Leontiev and Buxton, 2007, Tjandra et al., 2005, Wang et al., 2003) and indicated that, at 3T, fASL showed some activations not seen at 1.5 T (for example in the supplementary motor area (Yongbi et al., 2002)). High-field MRI (Golay and Petersen, 2006) can be associated with the use of phased-array coils (Duyn et al., 2005, Wang et al., 2005) and parallel imaging. The increase of number of elements of phased-array coils offers higher SNR and higher perfusion contrast while parallel imaging reduces geometric distortion artifacts. Only a few studies focused on the intra-individual reproducibility of fASL. These were mostly conducted at 1.5 T (Aguirre et al., 2002, Hermes et al., 2007, Kemeny et al., 2005) or without comparison to BOLD fMRI (Borogovac et al., 2009). The only studies comparing fASL with BOLD fMRI at 3T included a limited number of subjects and/or separate sessions (Leontiev and Buxton, 2007, Tjandra et al., 2005) and focused only on quantification of neuronal activation. To our knowledge, no study has compared fASL and BOLD fMRI intra-individual reproducibility of neuronal activation location.
The aim of this study was to evaluate fASL for motor activation mapping in comparison with BOLD fMRI, at 3T with a 32-channel coil and dedicated post-processing tools. The evaluation was performed in terms of involved anatomical area location, intra-individual reproducibility of location and quantification of neuronal activation, and spatial accuracy.
Section snippets
Subjects and paradigm
Twelve healthy subjects were included: 7 women and 5 men, mean age 28.6 (± 2.7) years old who were strongly right-handed (92.5%, ± 13.4% based on the Edinburgh Handedness Inventory). Exclusion criteria were psychiatric or central nervous system disorders, history of brain surgery or trauma and regular medication, and besides contraceptives. Written and informed consent was obtained from all subjects. Three fASL sessions were performed 7 days apart at the same time of day to minimize the effect of
Individual functional maps
Activation in the left primary hand motor area was observed for all 36 sessions with both fASL and BOLD fMRI. The prevalence of activation observed in primary and secondary motor areas is represented in the graph in Fig. 4. fASL individual functional maps demonstrated activation in ipsilateral secondary motor areas, i.e. in the preMA, SMA and PAA, more often than BOLD fMRI maps, with anatomical depiction as shown in Fig. 5. Mainly, activation was observed in right preMA for 92% of sessions with
Discussion
fASL and BOLD functional motor maps were obtained for all 36 sessions performed on 12 healthy right-handed subjects, all showing contralateral activation in the primary sensorimotor, premotor and supplementary motor areas. Individual and group functional ASL maps showed ipsilateral activation in the premotor and parietal association areas that was not shown by BOLD fMRI. A high degree of intra-individual reproducibility of fASL maps was observed, comparable to BOLD fMRI for activation
Conclusion
Functional ASL is an innovative biomarker of neuronal activation providing motor activation mapping with high intra-individual reproducibility. Functional ASL seems to detect neuronal activation closer to the involved neurons, with high spatial accuracy, and scores higher positive predictive values than BOLD fMRI. The performance and advantages of fASL shown in this study need to be confirmed in pathological subjects. The growing availability of high-field MRI and phased-array coils along with
References (64)
- et al.
Experimental design and the relative sensitivity of BOLD and perfusion fMRI
NeuroImage
(2002) - et al.
Regional differences in the coupling of cerebral blood flow and oxygen metabolism changes in response to activation: implications for BOLD-fMRI
NeuroImage
(2008) A fast diffeomorphic image registration algorithm
NeuroImage
(2007)- et al.
Hemispherical asymmetry in human SMA during voluntary simple unilateral movements. An fMRI study
Cortex
(2003) - et al.
Comparison of pulsed arterial spin labeling encoding schemes and absolute perfusion quantification
Magn. Reson. Imaging
(2009) - et al.
Caffeine dose effect on activation-induced BOLD and CBF responses
NeuroImage
(2009) - et al.
Technical aspects and utility of fMRI using BOLD and ASL
Clin. Neurophysiol.
(2002) - et al.
CASL fMRI of subcortico-cortical perfusion changes during memory-guided finger sequences
NeuroImage
(2005) - et al.
Arterial spin labeling: benefits and pitfalls of high magnetic field
Neuroimaging Clin. N. Am.
(2006) - et al.
Cortical layer-dependent dynamic blood oxygenation, cerebral blood flow and cerebral blood volume responses during visual stimulation
NeuroImage
(2008)