Cerebral blood flow (CBF or cerebral perfusion) is a well-characterized physiological measure that mirrors cerebral metabolic rate and neuronal function. As such, it may provide a useful parameter for the evaluation of brain physiology and function during both healthy and pathological states. Existing measurements of CBF include nuclear medicine approaches that utilize radioisotope tracers (eg, H2O15 PET) or contrast agents that alter radioactivity (eg, Xe CT*), or dynamic MR imaging methods that capture the first pass of an exogenous contrast agent (eg, Gd-DTPA). These methods have limitations in certain patient populations. For instance, there are ethical concerns related to the exposure to radioactivity in children even when patients are ill. Recent evidence suggests a potential link between the administration of gadolinium-containing contrast agents and nephrogenic systemic fibrosis/nephrogenic fibrosing dermopathy (NSF/NFD) in patients with renal failure. For both nuclear medicine and dynamic contrast MR approaches, repeated measurements during a single scanning session remain difficult due to cumulative effects.

Figure 1. Diagram showing the labeling scheme of ASL. Click on image for larger view.

Arterial spin labeled (ASL) perfusion MRI is an alternative and emerging noninvasive method to directly measure CBF by using the water molecules in arterial blood as a natural tracer. Its methodological scheme is analogous to that used in PET and SPECT (see Figure 1 at right), where water molecules in inflowing arteries are magnetically “labeled”(MR signal manipulation through inversion or saturation) using radiofrequency pulses proximal to the tissue of interest and function as a diffusible tracer. Image acquisition is usually carried out after a delay time (~1 second) that allows the labeled blood water to flow into the imaging slices. Perfusion can be determined by pair-wise comparison with separate images acquired without labeling (control). Repeated measurements of interleaved label and control acquisitions are carried out to improve the signal-to-noise ratio (SNR) of perfusion images, often requiring a few minutes of scan time. By taking into account the tracer half-life of blood T1—T1 refers to the time constant for the recovery of excited nuclear spins to original magnitude along the longitudinal axis—absolute CBF can be quantified in well-characterized physiological units of mL/100 g/min. Because ASL does not require administration of contrast agent or radioactive tracer, it is a safe, economical, and convenient alternative to current clinical approaches. Also, ASL scans can be repeated as often as required during the same scanning session without cumulative effects.

Although ASL methods have the aforementioned potential advantages, the primary weakness that limits its widespread clinical applications is the relatively small labeling effect (<1% raw signal). This limitation arises from the fact that the volume of arterial blood available for labeling is only on the order of 1% to 2% of total brain volume. The labeled blood further relaxes during the transit from the labeling region to the brain, resulting in a net labeling effect of less than 1% of raw MRI signal in brain tissue. Because the arterial transit time from the labeling region to the brain is comparable to the tracer half-life (blood T1), ASL techniques are very sensitive to transit effects, which can produce focal artifacts in perfusion images corresponding to intravascular signal. On the other hand, information on delayed arterial transit times may have diagnostic or mechanistic significance in cerebrovascular disorders.

During the past decade, theoretical and experimental studies have been conducted to improve the image quality and accuracy of CBF quantification using ASL. One major advance that has greatly facilitated the applications of ASL is the increasing availability of high magnetic field MRI scanners. Performing ASL at high magnetic field not only offers an increased SNR that is proportional to the main field strength, but it also improves the labeling effect due to prolonged tracer half-life (T1) at high field. Recent ASL methods also take advantage of the array imaging technique that provides a high sensitivity through small element receiver coils. Quantitative CBF measurements with ASL perfusion MRI have been shown to agree with results from 15O-PET and dynamic contrast MR approaches. ASL perfusion measurements also have been demonstrated to be reproducible (variable ~10%) across intervals varying from a few minutes to a few days. All these developmental and validation studies have brought ASL to the frontier of practical clinical usage and the verge of commercialization.

Pediatric ASL Perfusion MRI

Because of the noninvasive nature of ASL and several characteristics of pediatric neurophysiology, ASL is particularly promising for detecting pathological changes in neurologic and psychiatric disorders in children. First, blood flow rate is higher in children compared to adults, which tends to enhance ASL perfusion contrast. Similarly, the water content of the brain is higher in children than adults, which means that the concentration and half-life of the tracer are greater. This factor increases the equilibrium MR signal and spin-lattice, spin-spin relaxation times (T1, T2), improving ASL image quality. Second, data from Doppler ultrasound studies suggest that blood flow velocities in carotid arteries are higher in healthy children when compared to adults, with the peak velocity occurring within the age range of 5 to 8 years. This effect can be translated into reduced arterial transit time for labeled blood to flow from the labeling region to the brain, which results in reduced relaxation of the labeling effect and reduced transit effect (focal intravascular signal) in pediatric ASL perfusion images.

On average, a 70% improvement in the SNR of perfusion images in healthy children as compared to healthy adults has been demonstrated. Figure 2 shows mean CBF maps of children (ages 5 to 10 years), adolescents (ages 11 to 16 years), and young adults (ages 18 to 30 years) acquired on 3 Tesla MRI scanners. The mean global CBF value decreases from 81.2 mL/100 g/min in the child group to 64.4 mL/100 g/min in the adolescent group and to 43.5 mL/100 g/min in the young adult group. The figure also demonstrates the growing fraction of white matter and declining fraction of gray matter during brain development. Such quantitative developmental CBF data may provide a benchmark for future ASL studies on neurodevelopmental disorders. To date, only limited nuclear medicine literature has documented such developmental changes in CBF and cerebral metabolism across age groups.

Figure 2. Mean CBF images for the child group (age 5 to 10 years, left), adolescent group (age 11 to 16 years, middle) and young adult group (age 18 to 30 years, right). Click on image for larger view.
Figure 3. A representative pediatric arterial ischemic stroke (AIS) case (6-year-old boy); initial scan was performed 31 hours following symptom onset. CBF map shows hypoperfusion with delayed transit (blue arrow) in the left middle cerebral artery territory, infarcts on initial DWI, and follow-up (3 months) T2 images indicated by green arrows. Click on image for larger view.
Figure 4. A 59-year-old male with glioblastoma, showing markedly increased blood flow in ASL CBF maps and aggressive-appearing ring enhancement on postcontrast 3D T1-weighted SPGR image. Grading of CNS neoplasms using continuous arterial spin labeled perfusion MR imaging at 3 Tesla. J Magn Reson Imaging. Vol. 22, No. 4, 2005. 475?482. Copyright ? 2005; Jiongjiong Wang, PhD. Reprinted with permission of Wiley-Liss Inc, a subsidiary of John Wiley & Sons Inc. Click on image for larger view.
Figure 5. CASL perfusion maps TOF MRA (axial MIP) images of an AVM patient. Bright signal in draining veins adjacent to nidus is evident on ASL images, indicating AV shunting with superficial drainage. Click on image for larger view.

Another major application of ASL in pediatric neuro-imaging is in the diagnosis of cerebrovascular disorders in children. Stroke affects two to eight per 100,000 children per year in Europe and North America, and ranks among the top 10 causes of death in this age group. ASL may assist the management of pediatric stroke patients, in conjunction with diffusion and other conventional MRI modalities. One representative case of pediatric arterial ischemic stroke (AIS) (male, 6 years) is shown in Figure 3. The perfusion deficit is consistent with restricted diffusion in the left middle cerebral artery (MCA) territory. The CBF in the affected MCA territory is 17% lower than in the unaffected side. The presence of delayed arterial transit effects (focal intravascular signals) indicates collateral blood supply to the affected region. The follow-up T2 weighted MRI shows a relatively small chronic infarct when compared to the volume of hypoperfused brain tissue on acute imaging. This may be attributed to the mild to moderate degree of perfusion deficit in addition to the potential collateral blood flow.

Other pediatric applications of ASL include the evaluation of CBF in neonates with severe congenital heart defects (CHDs) and children with sickle cell disease (SCD). A diminished baseline CBF level was observed in neonates with CHD primarily due to a poor blood supply from the heart. Increased inspired carbon dioxide stimulated an average increase in CBF of 100% and may have therapeutic benefits in these neonates. In contrast, children with SCD generally demonstrate increased CBF as compared to healthy children, as a means to compensate for the reduced hemoglobin level in this population. Cognitive function and IQ have been shown to inversely correlate with CBF in children with SCD.

Representative ASL Applications in Adult Neurologic and Psychiatric Diseases

Many ASL perfusion MRI studies have been carried out in a variety of neurologic and psychiatric disorders in adults, including brain tumors, stroke, epilepsy, neurodegenerative disorders, and cerebral effects of drug manipulation. Brain tumors have been an active area for physiological neuroimaging given the potential link between high blood flow/volume and tumor angiogenesis. The majority of existing MRI studies to measure blood flow and volume in brain tumors have used dynamic contrast MRI. However, the absolute quantification of blood flow or volume remains challenging in dynamic contrast MRI because of the leakage of contrast agents into brain tissue through the damaged blood-brain barrier (ie, BBB permeability change). ASL uses water molecules as a freely diffusible tracer and is much less sensitive to the effect of permeability changes of the BBB. Theoretically, ASL is able to provide accurate quantification of blood flow in brain tumors. Recent studies using ASL and dynamic contrast MRI approaches have demonstrated the relationship between high blood flow/volume and high-grade brain tumors. A representative case of glioblastoma (GB) is show in Figure 4. ASL perfusion maps demonstrate markedly increased blood flow in the peripheral parts of GBM, corresponding to aggressive ring enhancement on postcontrast T1 weighted images.

ASL perfusion MRI studies in neurodegenerative disorders, epilepsy, and pharmacological manipulations have yielded expected CBF variations. For instance, decreased perfusion was demonstrated in temporoparietal regions in patients with Alzheimer disease, while the perfusion deficit was more widespread in frontal dementia patients. Interictal mesial temporal hypoperfusion can be detected using ASL in temporal lobe epilepsy patients, concordant with FDG-PET findings of hypometabolism. ASL perfusion MRI has been used successfully to demonstrate the varying patterns of CBF changes that occur after acetazolamide administration in patients with cerebrovascular diseases.

Besides providing quantitative tissue perfusion, ASL is useful for evaluating vascular disorders, such as arterio-venous malformation (AVM). AVMs are characterized by the direct “shunt” between the artery and vein that are normally connected by a network of capillaries. The labeled blood water directly flows into the veins through the arteriovenous shunt in an AVM, resulting in focal intravascular signals in ASL perfusion images similar to delayed arterial transit effects, except that here the effect is primarily related to rapid transit so that intravascular signal is primarily venous (see Figure 5). Through quantifying the fraction of labeled signal in the vasculature and tissue, it is possible to estimate the percentage of blood supply that is directly shunted into the veins. ASL therefore may become a diagnostic tool to evaluate the degree of AVM as well as the efficacy of therapy.

Looking Ahead

Because CBF is a vital physiologic parameter that is coupled with metabolism and neuronal function, it is a valuable measurement in the evaluation and management of a wide spectrum of neurologic and psychiatric disorders. During the past decade, ASL already has proven its value in several important clinical applications as listed above. With the latest technical advances and improved awareness among clinicians, ASL should have a great impact on blood flow imaging in neuroradiology. Future technical improvements will further reduce the acquisition times for ASL perfusion MRI, while increasing the slice coverage, resolution, image quality, and stability of the measurements. These techniques have a broad range of potential applications in clinical and basic research in brain physiology, as well as in the vascular physiology of other organs.

Jiongjiong Wang, PhD, is a research assistant and professor of radiology at the University of Pennsylvania School of Medicine. Daniel J. Licht, MD, is assistant professor of neurology and pediatrics, and director of the Neurovascular Imaging Lab at the Children’s Hospital of Philadelphia. Ronald L. Wolf, MD, PhD, is assistant professor of radiology at the Hospital of the University of Pennsylvania.