Introduction
The brain is an integral part of any living organism. The brain receives blood from the heart and this blood contains oxygen and nutrients that nourish it. At the same time, it carries away any metabolic wastes that result from the breakdown of the nutrients. Of all the organs in the body, the brain receives a bulk of the blood pumped out by the heart accounting for up to 20% of total cardiac output. Conditions that alter the perfusion of the brain have wide-ranging implications eventually resulting in death.
Medical conditions such as stroke and cardiovascular accidents are a direct outcome of impaired or defective perfusion of the brain. Quantification of cerebral blood flow is an essential undertaking in the determination of various disease states, prognosis, and assessment of the success of treatment (Martin, Friston, Colebatch & Frackowiak, 1991).
Cerebral Blood Flow (CBF) is a function of the total metabolic and perfusion functions of the brain at any one given time, whether in the resting or working state. By assessing the blood flow in different areas of the brain, it is possible to extrapolate the results in determining the state of the brain in terms of its function in the respective areas. The measure of perfusion can be made quantitatively by determining the amount of blood circulating in a specific region of the brain (Figueiredo, Clare & Jezzard, 2005).
Assessment entails factoring in the volume of blood delivered to the region under investigation, the size of the region, and the time of delivery of blood to the specific region. Brain imaging in the pediatric population is relatively a new field with little research on proper and safer imaging methods available. Earlier on, invasive techniques that entailed the use of radioactive contrast agents were the gold standard for pediatric brain imaging. The use of these agents in the pediatric population is subject to ethical and moral issues due to the myriad side effects associated with such techniques.
In earlier methods of determination of brain perfusion, researchers utilized tracers that could cross the blood-brain barrier in and out of the brain. Most of these methods were invasive and they entailed injecting a radiolabelled tracer into the general circulation and monitoring its distribution in the tissues. Agents used included, deuterated water (D2O) and radio-labeled Fluorine gas. Older methods such as Positron Emission Tomography (PET) entailed injection of radioactive water molecules into general circulation. After injection, the tracer would commence its decay by releasing positive charges called positrons.
With the help of special machines, the positrons would then be spotted as they circulated and penetrated. The amount of positrons in a given region of the brain is proportional to the amount of blood that a specific region receives. Furthermore, other methods such as the use of endogenous tracers like gadolinium-diethyltriaminepentaacetic acid (Gd-DTPA) have been used in the quantification of cerebral blood flow in children with brain injury via a process called dynamic contrast-enhanced MRI (DCE-MRI). All these methods have ethical and moral issues especially when applied to children, hence the need for a method that is non-invasive and safer to use in the pediatric population (Takahashi, Shirane & Sato, 1999).
Modern methods of assessing cerebral blood follow to utilize the same principles of tracking a labeled compound as it circulates in the blood. The difference arises from the non-invasive nature of these modern methods when compared to earlier methods that were invasive. Furthermore, these methods encompass advanced techniques in assessing cerebral blood flow and as such are friendly to the pediatric population. An example of such a technique is the use of the arterial spin labeling technique (ASL) in neuroimaging studies.
Arterial Spin Labelling
Arterial Spin Labelling (ASL) is one of the modern methods that are widely used in the determination of cerebral blood flow (CBF). This technique entails the use of tracers just like in the earlier methods only that in ASL the tracer is magnetically labeled rather than radioactively labeled. ASL has been used in the determination of perfusion of various body organs such as the kidney, liver, and brain. ASL has been widely used in the determination and quantification of blood flow to the brain since perfusion of the brain is a major determinant of one’s health. The tracer in ASL is magnetized arterial blood water. Arterial blood water is subjected to magnetizing forces produced by a radiofrequency pulse that is strategically aligned at an angle of 1800 to the region of interest (slice). The applied radiofrequency is analogous to the frequencies used in Magnetic Resonance Imaging (MRI).
The magnetized arterial water acts as the tracer only that it is endogenous and not exogenous (Detre, 2007). Magnetization of the arterial water is carried out just below the area of interest (slice). The magnetized water, after some time (transit time), migrates into the area of interest then into the tissues where it exchanges with water in the tissues. The water that is innate to the tissues acquires magnetization from the inflowing labeled water and as such alters the resultant magnetic resonance and the intensity of the image of the tissue water. As the process occurs, an image is obtained. This image is referred to as the tag or labeled image.
Thereafter, the process is repeated but without inversion of the arterial blood, water to obtain what is referred to as the control image (Silver & Joseph, 1985). Once the two images are obtained, a calculated subtraction is carried out to produce a perfusion image (signal). The image represents the amount of blood that is distributed within the particular voxel in the specific slices under a specified transit time. Subtraction entails a pixel-by-pixel elimination process to achieve a true representation of the voxel’s perfusion status (Barbier, Lamalle, & Decorps, 2001).
The perfusion image is in essence a signal difference between magnetized arterial blood water entering the tissue and the tissue water that is unmagnetized. To achieve this perfusion signal, a systematic delay (inversion delay) is initiated between the time of inversion of arterial blood and the time of acquiring the image. The perfusion signal is under the influence of several factors such as the transit time, the flow, and the T1 of the blood and tissue. Due to probable variations in readings, multiple readings are obtained to eliminate the influences of the signal-to-noise ratio.
Furthermore, ASL has been employed in the determination of the prognosis of certain disease states such as malignancies since the rate at which a tumor is perfused is a direct measure of its aggressiveness. By using ASL, it is possible to quantify perfusion and in turn decide on modalities of management of the malignancies (Tofts, 2003). There are two approaches of Arterial Spin Labelling based on how the magnetizing signal is used to invert the arterial blood water. These are the Pulsed Arterial Spin Labelling (PASL) and the Continuous Arterial Spin Labelling (CASL), depending on the mode of application of the radiofrequency pulse.
Efficiency in labeling plays a major role in determining the success of ASL in the measurement of cerebral perfusion. In CASL, the efficiency in labeling is a function of the speed of blood and the orientation of the labeling plane. This efficiency may be reduced in certain medical pathologies that may result in false readings. The PASL method has a wide labeling area covering a massive capacity of blood vessels even though the length of the labeling slab is limited. As such any arterial blood water outside the labeling slab will not be inverted, a limitation that multiplies when a multislice imaging system is employed since the amount of unlabelled arterial blood water is elevated. Other errors called model errors have an impact on the type of results obtained in the determination of cerebral blood flow.
Various parameters have errors that are unique to them and these errors have an overall effect on the perfusion readings obtained by the use of ASL. The error of propagation elaborates the impact of each parameter on the resultant error in the perfusion measurements. The motion of blood results in a measurable parameter referred to as the T1 (Bokkers, Laar & Ven, 2008). Error results from the challenging task of determining this value when blood is in motion.
Furthermore, errors that occur during inversion result in the occurrence of multiple errors across the perfusion image. Parameters such as the arrival time may have errors that result from the fore mentioned inversion errors and the error is magnified in neurological pathologies of the brain such as transient ischaemic attacks and stroke (MacIntosh et al, 2008). To determine arrival time with the highest precision modifications need to be made to the two Arterial Spin techniques. This entails obtaining measurements at very unalike delay times when using CASL. When using PASL, the inversion times are modified to reduce the occurrence of the error. These two approaches reduce the anticipated error as they result in elongation of the imaging times (Deibler et al, 2008).
Continuous Arterial Spin Labelling (CASL)
The basic principle behind the use of CASL entails the application of a radiofrequency that is continuous just below the slice of interest. The frequency follows an adiabatic system that is continuous. As the radiofrequency is applied, a magnetic field is applied parallel to the flow of arterial blood water resulting in the acquisition of a magnetic field by the protons found in the arterial blood water. A continuous radiofrequency signal is applied to obtain the tag image.
Since some of the arterial spins are in constant motion, a gradual variance of the resonance frequency leads to the inversion of the applied magnetic field. At the same time, the static frequencies will undergo saturation. In most cases the slice of interest is demarcated from the start of the circle of Willis or close to the common carotid artery and any spin that goes past this point is continuously inverted due to the presence of a continuous radiofrequency (Adel et al, 2008).
As the spins pass through the circulation, they encounter a magnetic field that leads to the increased resonance of the spins as they undergo magnetization. At the same time, the magnetic field is subjected to an armor frequency sweep, when a constant radiofrequency is applied. The spins of the blood are in tandem with the overall magnetization and undergo inversion at the tagging plane. Inversion and uninversion occur as the blood spins transverse the proximal and distal planes respectively (Buxton et al, 1998).
In CASL, the efficiency of labeling approaches 95% (Buxton et al, 1998). Due to the constant magnetization of the spins in CASL, a process called magnetization transfer (MT) occurs. This may be because of using a single coil for producing the image that creates a strong off-resonance pulse that in turn results in a massive magnetization transfer. The resultant magnetization transfer in turn decreases the strength of the signal emanating from the tissues under investigation resulting in an overestimation of the observed perfusion because of T1 relaxation. As such, the magnetization of the sluice is decreased. To obtain a quality image, a delay is initiated to allow for the taking of the image.
To achieve a high-quality image with a low noise-to-signal ratio, the magnetization transfer is set such that the tag image and control images have a similar magnetization transfer such that on subtraction the effects because of the two cancel out. CASL is a useful method but it is limited due to the extended periods of labeling that are magnified especially when using high magnetic field strengths. Furthermore, the transit time in CASL is long such that in some disease states there is decay long before the signal reaches the desired voxel resulting in false readings.
Pulsed Arterial Spin Labelling (PASL)
PASL is a form of arterial spin labeling that entails the use of short radiofrequency pulses to label enormous spins in the determination of cerebral blood flow. Like other ASL methods, PASL is non-invasive and produces perfusion signals that are under the influence of a myriad of determinants such as transit time, perfusion, and period width of the specific label. The radiofrequency is usually applied in pulses just before the slice of interest (Calamante, Thomas, Pell, Wiersma & Turner, 1999).
Just like in CASL, the pulse in PASL gives an adiabatic inversion that is dependent on a rapid sweep that is intrinsic to the radiofrequency pulse. In the majority of the cases, a hyperbolic secant pulse brings about inversion.PASL is entailed three principle methods. EPISTAR (echo-planar imaging and signal targeting with alternating radiofrequency) involves the utilization of a radiofrequency pulse oriented at 1800 in the same plane as an applied magnetic field gradient to facilitate the inversion of the longitudinal magnetization of the protons present in a slab that is below the slice of interest. A calculated delay is then allowed and immediately after inversion, a timed echo-planar image snapshot is taken. The resultant signal is a culmination of the contribution of labeled arterial blood that is flowing into the slice.
The control image is obtained by strategically positioning the slab above the image slice, an area that is devoid of inflowing arterial blood. This is of value in eliminating interference that may occur because of magnetization transfer. Variation in readings from the different slices may be minimized through the application of what is called a saturation pulse just before the application of the tag and image pulses (Zhang, Silva, Williams & Koretsky, 1995). This aims at minimizing the static magnetization of the tissues under investigation to alleviate in totality any interference that might occur during image extraction. The resultant perfusion image is a representation of the extent of perfusion of the voxel under a specified period determined by the predetermined transit time.
FAIR (flow-sensitive alternating inversion recovery) is another PASL method that entails the tagging of all arterial blood water external to the region of the imaging slice to minimize the delays in transit that may result in loss of signal strength. The working geometry of the inversion slab in FAIR is altered to encompass contributions of blood entering the slice of interest from the top. The pulse used in obtaining the label is non-selective followed by a selective gradient pulse to obtain the control image. Because of this sequence, interferences brought about by magnetizing transfer are eliminated (Zoccoli, Lucchi & Andreoli, 1996).
QUIPSS II (quantitative imaging of perfusion using a single subtraction) is another PASL technique that has the advantage of versatility since it can be used with almost all labeling techniques. The inverted area is taken through a saturation process immediately after inversion to slice off the appendage of the inflowing bolus. As such, the time labeled arterial blood water takes to migrate from the plane of labeling to the image slice is shortened increasing the sensitivity of the technique. QUIPPS II has a factor of independence that results from obtaining the image immediately after the whole of the bolus flows into the image slice since the time taken for the bolus to migrate is no longer a factor influencing the sensitivity of the technique.
Other forms of Pulsed Arterial Spin Labelling have found use in modern imaging studies. These methods are divided further depending on their symmetry in the application. Asymmetric PASL methods include the Transfer Insensitive Labelling Technique (TILT), Double Inversion with Proximal Labelling of both tagged and control images (DIPLOMA), and Signal Targeting with Alternating Radiofrequency – Half-Fourier Single Shot Turbo spin-Echo (STAR-HASTE).
Symmetric labeling sequences entail the use of methods such as Uninverted Flow Alternating Inversion Recovery (UNFAIR), Flow Alternating Inversion Recovery Extra Radiofrequency pulse (FAIRER), Flow Alternating Inversion Recovery Excluding Radiation damping (FAIRER), and unprepared Basis and Selective inversion (BASE). Other PASL methods have been named based on the labeling method (Hendrikse, Petersen & Laar, 2007). These include Pulsed Star labeling of Arterial Regions (PULSAR) and Quantitative Star labeling of Arterial Regions (QUASAR). All these methods have found use in advanced modern imaging techniques but most of them have not been studied in the pediatric population (Bode & Wais, 1988).
Literature Review
Arterial Spin labeling has become the main standard of determination of cerebral blood flow in adults. Since this method is non-invasive, it provides a desirable platform for determining cerebral blood flow without a conflict of moral and ethical aspects encountered in other methods that entail injection of tracers (Chen, Mai, Pippa & Edelman, 2001). As such, arterial spin labeling is an appealing method in the pediatric population.
Several studies have been carried out to determine the efficacy of ASL in children. Wang et al. (2002) carried out a study to determine the efficacy and accuracy of arterial spin labeling as a method of determining cerebral perfusion in pediatrics. In this study, the authors set out to determine the longitudinal reproducibility (precision) and accuracy of arterial spin labeling when compared to a phase-contrast MRI, a method commonly used in the pediatric population (Wang & Licht, 2006).
By using the two forms of ASL, the researchers found that reproducibility of the results was, achieved using the pseudo-Continuous Arterial Spin Labelling (pCASL) when compared to Pulsed Arterial Spin Labelling (PASL) in the pediatric population under study. Furthermore, the study showed that with an increase in age there is a corresponding increase in accuracy and reproducibility of pCASL. The authors attributed this to unstandardized variables in the pediatric population such as the elevated velocity of blood in children as compared to adult populations (Chiron, Raynaud & Maziere, 1992). In addition, the positioning of the head during tagging may be the cause of low accuracy observed when using PASL, since in PASL the head position is varied while in pCASL the position is fixed.
Age is a factor that is implicated in determining the accuracy of measurement of cerebral blood flow. Developmental changes in the brain as one grows older result in varied proportions of perfusion of the brain with increasing age, such that in determinations of cerebral blood flow age becomes a factor to be considered (Chugani, 1998). A study carried out in children and adults that involved mapping of cerebral blood flow patterns showed a distinct difference in perfusion between children and adults especially in the grey matter (Biagi et al, 2007). In this study, Continuous Arterial Spin Labelling was used to determine cerebral perfusion in children and adults.
The study showed a decrease in perfusion among adults when compared to children up to the age of sixteen when there was a rapid drop in cerebral perfusion (Biagi et al, 2007). This was attributed to developmental processes occurring in the pediatric population such as increased neuron production and myelination that require a constant supply of oxygen and nutrients to facilitate proper and timely development. Increased perfusion in the grey matter in the two subsets understudy was attributed to an increased rate of gyrification and synapse formation as one grows older. In another study carried out, other factors modulating the results such as gender were incorporated to assess variations brought about by such factors.
It was observed that in general people of the female gender had relatively lower cerebral perfusion when compared to males (Biagi et al, 2007). Furthermore, perfusion in the grey matter followed a similar trend. The observed differences were linked to differences in hematocrit concentration between the two genders. In addition, hormonal factors such as the presence of estrogen were implicated in the observed differences (Biagi et al, 2007).
Limited studies on cerebral blood flow have resulted in the slow adaptation of Arterial Spin Labelling as a method of determining cerebral perfusion. Studies have been carried out to determine the feasibility of ASL in measuring cerebral perfusion in pediatric populations. In one study, Pulsed Arterial Spin Labelling at 1.5T was used in determination determining cerebral perfusion in children, due to its non-invasive nature, a property that is friendly to this population (Wang et al, 2006).
Furthermore, the relation of age to values of the Signal-to-Noise ratio was assessed in this study. The researchers found out that the volume of white matter was decreased while that of the grey matter was elevated in all the children under study. Furthermore, it was observed that the signal was clearer in images obtained by PASL in children than in adults especially in the cortical and sub-cortical areas of the grey matter. The researchers conclude that PASL was a method that was feasible in measuring cerebral blood flow in children with desired precision and Signal to Noise Ratio (Wang et al, 2006).
Meta-analytic reviews of studies carried out on Arterial Spin Labelling have compared the two forms of ASL and their applications in modern imaging techniques in adults and children (Miranda, Olofsson & Sidaros, 2006). A review carried out compared PASL and CASL in terms of a myriad of affecting factors (Petersen, Zimine & Golay, 2006). Arterial transit time was identified as one of the quantification errors that occur during the determination of cerebral blood flow using PASL.
Because of interference of readings by this error, modifications on PASL were carried out to obtain methods that were not under the influence of transit time. An example of such methods includes QUIPSS II (quantitative imaging of perfusion using a single subtraction). Other errors identified by the review entailed errors resulting from vascular artifacts, the shape of inversion pulse and its frequency, the partition coefficient of the blood-brain barrier, and errors of equilibrium magnetization in the different regions of the brain (Petersen, Zimine & Golay, 2006).
Amount and extent of perfusion in tumors are important biomarkers in determining the aggressiveness of the particular tumor. ASL has been used in the quantification of perfusion in tumors. Studies have been carried out to compare the use of ASL and other imaging methods in the quantification of tumor perfusion. One such study compared quantification of blood flow in brain tumors using Arterial Spin Labelling and Dynamic Susceptibility weighted Contrast-enhanced MR Imaging (Warmuth, Gunther & Zimmer, 2003).
The authors observed that both methods were highly sensitive in determining blood flow in glioblastomas while low sensitivity was observed in gliomas. The researchers concluded that ASL was a superior method in the determination of Tumour Blood Flow since it made a distinction between low and high-grade gliomas (Warmuth, Gunther & Zimmer, 2003).
For a clear understanding of the working principles behind ASL, researchers have carried out studies in non-human primates using the two-compartment method to simulate cerebral blood flow. In one study carried out, researchers fitted ASL data obtained in monkeys in a two-compartment model under different physiological conditions (Zappe et al, 2007). Furthermore, any expected uncertainties were identified using the Monte Carlo simulations.
The researchers found out that inherent errors were associated with the use of ASL in the determination of cerebral blood flow. In addition, it was demonstrated that ASL is superior in the determination of cerebral blood flow especially in the occipital lobe of the monkey as compared to the use of positron emission tomography (PET). There was a discrepancy in the results obtained when compared to other studies, and this is enough reason to encourage further studies in the use of modified ASL and two-compartment models in the determination of cerebral blood flow in primates and humans alike (Zappe et al, 2007).
Other reviews of ASL have focussed on the applications of ASL in diagnostic purposes for the various conditions and associated sources of error. One review identified transit delay and incorporation of a signal emanating from blood in the intravascular compartment (Wong, Buxton & Frank, 1999). In addition, the authors reviewed current solutions to the problem, which include fine-tuning of CASL and OPASL to eliminate these interferences through modifications such as the (quantitative imaging of perfusion using a single subtraction).
Furthermore, the authors reviewed the use of arterial spin labeling in the measurement of cerebral blood flow under normal and pathologic conditions. The review entailed an assessment of the efficacy of Arterial Spin Labelling, in the determination of cerebral blood flow in stroke patients. The authors identified flaws in the application of ASL in such pathologies since CBF values obtained using ASL were observed to be decreased as a result of elongated transit time. As such, modifications of ASL were identified to be applicable in determining cerebral blood flows in stroke patients (Wong, Buxton & Frank, 1999).
Just like all imaging techniques, ASL has constraints that limit its application in different situations. The diversity in the composition of the voxel tissue affects the quantification of cerebral blood flow when using Arterial Spin Labelling. This occurs as a result of effects of partial volume and errors that occur in the conversion of raw ASL signal to Cerebral Blood Flow units. The study revealed that heterogeneity had a major impact on the results obtained from the cortical regions of the brain but there was little change in the results obtained in the deep white matter (Kim, 1995). The researchers concluded that correction of the result to cater for influences brought about by partial volume effects results in more representative results.
To increase sensitivity and accuracy of determination of cerebral blood flow in normal and pathologic situations, ASL has been modified by coupling it to other advanced techniques. In one instance, researchers modified ASL by reducing the period between the application of the tag image and the time of repletion. The result was increased sensitivity of ASL and increased temporal resolution. By applying a novel method called Turbo Arterial Spin Labelling at 1.5T using a GE Signa LX Echospeed system the researchers were able to delay the transit time to speed up the scan rate and enhance the temporal resolution (Wong, Luh & Liu, 2000).
Even though the temporal resolution was enhanced, there was a minimal decrease in the CNR on each image. The authors identified a lack of compatibility of Turbo ASL with QUIPSS II (quantitative imaging of perfusion using a single subtraction). The two were only compatible when detailed modifications that entailed the use of extra saturation pulse in the tagging region to modulate the temporal duration resulting in nullification of effects due to transit delay from the tag image to the control image (Wong, Luh & Liu, 2000). As such, Turbo Arterial Spin Labelling is meant for qualitative rather than quantitative purposes. Furthermore, Turbo-ASL was shown to be compatible when used in CBF/BOLD (blood oxygenation level-dependent) studies.
Perfusion of the brain varies with the type of activity the brain is undergoing. When the brain is activated, there is increased perfusion. The rate of perfusion affects the results obtained when using imaging techniques such as ASL. This results from the fact that changes in the flow of blood affect parameters such as transit and trailing time of the bolus in ASL and may cause an error in the quantification of cerebral blood flow (Yang et al, 2000).
In one study, researchers set out to determine the absolute effect of the change in the flow characteristics of the blood and its effect on training and transit time. They employed a multislice approach by a combination of Pulsed Arterial Spin Labelling (PASL) and fast spiral scanning for the acquisition of data. It was observed that both the transit and trailing times were significantly reduced when the sensorimotor aspect of the brain was activated. Furthermore, there was an elevation in the cerebral blood flow. The distance between the initialization of the tagging position to the position of image slices (Yang et al, 2000). They concluded that variations in the rate of blood flow during measurements of cerebral perfusion might result in errors in the quantification of CBF during studies of cerebral activation.
Techniques such as finger tapping have been used in modern neuroimaging studies to ascertain the functional activational differences in localization and overlap. Studies have been carried out to assess differences in Arterial Spin Labelling (ASL) and blood oxygenation level-dependent (BOLD) techniques to determine which method is superior in localization of the functional Magnetic Resonance Imaging Signal (fMRI) to the brain tissues.
The researchers compared PICORE Q2TIPS, a form of PASL and BOLD techniques in the localization of the fMRI signal in five volunteers (Luh, Wong, & Hyde, 2000). It was observed that the peak signal in both methods had an overlap of about 40%. In addition, the authors observed lower values of the average T1 values of the voxels activated by ASL when compared to BOLD-activated Voxels.
As such, it was concluded that the BOLD signal receives a major contribution from Cerebrospinal Fluid and such an effect does not occur when ASL is used. Since there was a significant overlap between the two methods; the researchers attributed this overlap to the differences in the regions of the contributing factors (Luh, Wong, & Hyde, 2000). Both methods had contributions from the grey matter and capillaries. As such, the lack of contribution of cerebrospinal fluid to the ASL technique makes it an attractive option in the localization of changes in cerebral neural activities.
Even though ASL has received wide acceptance among the radiological and imaging community, it still has issues with standardization and reproducibility, especially when used in multicentre studies. This lack of reproducibility stems from a myriad of factors that are either intrinsic or extrinsic to the ASL system. Factors such as low Signal to Noise Ratio, equilibration in blood magnetization, and complications in quantification may affect the reproducibility of ASL (Tofts, 2003).
Furthermore, instrumentation errors such as shimming, improper modulation of the RF signal, and human errors such as improper manipulation of the ASL system result in false readings of the cerebral blood flow in the determination of cerebral perfusion in humans. Studies have been carried out to ascertain the reproducibility of ASL (Parkes, Rashid, Chard & Tofts, 2004). This allows for accurate diagnosis, consistent patient follow-up, and seamless interpretation of data on perfusion wherever and whenever with consistent results.
Gevers et al. (2009) carried out a study to assess the reproducibility of CASL and pCASL in multicentre settings and to ascertain whether the obtained results would be extrapolated in other centers as standards of CBF measurements (Tofts, 2003). They did this by subjecting similar individuals to repetitive scanning using the two techniques in various centers. The researchers observed that pCASL had the highest reproducibility in multicentre while CASL had higher reproducibility in the Intra center. In addition, they observed that in the presence of background suppression, the results were more reproducible. They concluded that Intra center reproducibility was more feasible than multicentre one.
Even though most studies carried out so far have focused on the use of arterial spin labeling in the quantification of the flow of blood in the brain, the technique can be employed in the quantification of perfusion in any organ including kidneys, liver, and the bone marrow. Kidneys are the most perfused organs in the body and quantification of blood in these organs is of medical importance since it gives an insight into the origin of a myriad of medical conditions such as hypertension, acute renal failure, and diabetes-associated nephropathy. One study carried out by researchers set out to quantify blood flow into the kidneys of the rat using ASL.
Imaging methods such as the use of gadolinium-diethyltriaminepentaacetic acid (Gd-DTPA) are not feasible in such studies since these tracers are filtered by the renal tubules. The researchers observed regional variations in blood flow with the highest values being recorded in the cortex while the medulla had little perfusion. Since no signal was observed in post mortem rats the observed signal was due to renal perfusion (Rajendran, Yong, Tan, Wang & Chaung, 2005).
Further evidence was obtained by initiating dilatation of the vessels by administration of CO2 and O2 in a ratio of 1:19 to reduce the transit time and elevated perfusion. The authors employed Matlab in the generation of maps of arterial transit time and perfusion regarding stepwise removal of FAIR images. The researchers observed a change in the signal when using FAIR ASL that was attributed to suspected sensitivity of the method to renal perfusion (Kim, 1995). When the results obtained were compared to available literature, the results tallied implying feasibility of this method is the determination of renal perfusion in humans.
Modifications in ASL have resulted in a wide application of this technique. Techniques such as EPISTAR, STAR-HASTE, and FAIR have undergone modification to incorporate more measurements (Robson et al, 2009). In a study combining ASL with cardiac-triggered segmented true-FISP, a sequence was acquired and used to develop two methods of quantification namely STAR-TrueFISP (True Fast Imaging with Steady-state Precession) and FAIR-TrueFISP (FAIR-True Fast Imaging with Steady-state Precession). These two had a very high temporal resolution. The image was acquired through the true-FISP sequencing (Rusinek et al, 2008).
This method, according to the researchers provided very high resolution and had short TR and TE that allows for a rapid imaging process. With such properties, the method can be used in the determination of a variety of vascular diseases such as emboli in the pulmonary branch and renal artery stenosis in both adults and children (Ogawa, Sakurai & Kayama, 1989). The authors observed setbacks in this method, since; a breath-holding delay is required to minimize interference by chest wall components. Other studies have employed the use of True-FISP but at higher t values. One study used this method at 3T to determine blood flow in the brain’s hippocampus region.
This region is implicated in various neurological conditions such as epilepsy, Alzheimer’s, and Huntington’s disease (Alsop, Detre, & Grossman, 2000). The researchers observed that ASL was applicable in the determination of regional cerebral blood flow and as such, it can be used in the monitoring of the progress of temporal lobe neurological conditions such as Alzheimer’s disease and epilepsy (Chen, Mai, Pippa & Edelman, 2001)
References
Adel, J., Sherma, A., Carroll, T., Hage, Z., Miller, J., Walker, M., Hunt, H., et al. (2008). The use of quantitative magnetic resonance perfusion for assessment of CBF in the perioperative management of carotid stenosis: case illustration. The Open Neurosurgery Journal, 1, 1-5.
Alsop, C., Detre, J., & Grossman, M. (2000). Assessment of cerebral blood flow in Alzheimer’s disease by spin-labeled magnetic resonance imaging. Annals of Neurology, 47, 93–100.
Barbier, L., Lamalle, L., & Decorps, M. (2001). Methodology of brain perfusion imaging. Journal of Magnetic Resonance Imaging, 13, 496–520.
Biagi, L., Abbruzzese, A., Bianchi, C., Alsop, C., Del Guerra, A., & Tosetti, M. (2007). Age dependence of cerebral perfusion assessed by magnetic resonance continuous arterial spin labeling. Journal of Magnetic Resonance Imaging, 25(1), 696–702.
Bode, H., & Wais, U. (1988). Age dependence of flow velocities in basal cerebral arteries. Archives of Disease in Childhood, 63, 606–611.
Bokkers, P., Laar, J., & Ven, C. (2008). Arterial spin-labelling MR imaging measurements of timing parameters in patients with a carotid artery occlusion. American Journal Neuroradiology, 29, 1698 –1703.
Buxton, R., Frank, L., Wong, C., Siewert, B., Warach, S., Edelman, R. (1998). A general kinetic model for quantitative perfusion imaging with arterial spin labeling. Journal of Magnetic Resonance Imaging, 40(3), 383-396.
Calamante, F., Thomas, L., Pell, S., Wiersma, J., Turner, R. (1999). Measuring cerebral blood flow using magnetic resonance imaging techniques. Journal of Cerebral Blood Flow and Metabolism, 19(7), 701-735.
Chen, Q., Mai, V., Pippa, S., & Edelman, R. (2001). Dynamic ASL flow imaging with cardiac triggered true FISP acquisition. International Society for Magnetic Resonance in Medicine, 9, 1955-1956.
Chiron, C., Raynaud, C., & Maziere, B. (1992). Changes in regional cerebral blood flow during brain maturation in children and adolescents. Journal Nuclear Medicine, 33, 696 -703.
Chugani, T. (1998). Biological basis of emotions: Brain systems and brain development. Pediatrics, 102, 1225–1229.
Deibler, A., Pollock, J., Kraft, R., Tan, H., Burdette, J., & Maldjian, J. (2008). Arterial spin labeling in routine clinical practice. American Journal of Neuroradiology, 29, 1428-1435.
Detre, J. (2007). Arterial spin-labeled perfusion MRI. Journal of Neurotherapeutics, 4(3), 346–359.
Figueiredo, P., Clare, S., & Jezzard, P. (2005). Quantitative perfusion measurements using pulsed arterial spin labeling: Effects of large region-of-interest analysis. Journal of Magnetic Resonance Imaging, 21, 676–682.
Gevers, S., Osch, J., Hendrikse, J., Bokkers, R., Kies, D., Teeuwisse, W. (2009). Multicentre reproducibility of continuous, pulsed, and pseudo-continuous arterial spin labeling: Can we use general reference values of cerebral blood flow? International Society for Magnetic Resonance in Medicine, 17, 626-630.
Hendrikse, J., Petersen, T., & Laar, J. (2007). Cerebral border zones between distal end branches of intracranial arteries: MR imaging. Radiology, 246, 572–580.
Kim, S. (1995). Quantification of relative cerebral blood flow change by flow sensitive alternating inversion recovery (FAIR) technique: application to functional mapping. Magnetic Resonance Medicine, 34, 293–301.
Luh, W., Wong, E., & Hyde, J. (2000). Arterial spin labeling localizes the fMRI signal too brain tissues better than BOLD. Journal of Neuroimaging, 12(4), 442-456.
MacIntosh, B., Lindsay, A., Kylintireas, I., Kuker, W., Gunther, M., Robson, M., et al. (2008). Multiple inflow pulsed arterial spin-labeling reveals delays in the arterial arrival time in minor stroke and transient ischemic attack. American Journal of Neuroradiology, 31, 1892-1896.
Martin, J., Friston, J., Colebatch, G., & Frackowiak, J. (1991). Decreases in regional cerebral blood flow with normal aging. Journal of Cerebral Blood Flow Metabolism, 11, 684–689.
Miranda, J., Olofsson, K., & Sidaros, K. (2006). Non-invasive measurements of regional cerebral perfusion in preterm and term neonates by magnetic resonance arterial spin labeling. Journal of Paediatric Research, 60(3), 359-363.
Ogawa, A., Sakurai, Y., & Kayama, T. (1989). Regional cerebral blood flow with age: Changes in rCBF in childhood. Neurology Research, 11, 173–176.
Parkes, L., Rashid, W., Chard, T., & Tofts, P. (2004). Normal cerebral perfusion measurements using arterial spin labeling: reproducibility, stability, and age and gender effects. Magnetic Resonance in Medicine, 51(1), 736–743.
Petersen, E., Zimine, I., & Golay, X. (2006). Non-invasive measurement of perfusion: A critical review of arterial spin labeling techniques. The British Journal of Radiology, 79, 688–701.
Rajendran, R., Yong, C., Tan, J., Wang, J., & Chaung, K. (2005). Quantitative mouse renal perfusion imaging using arterial spin labeling, International Society for Magnetic Resonance in Medicine, 16, 456-460.
Robson, M., Madhuranthakam, J., Dai, W., Pedrosa, I., Rofsky, M., & Alsop, C. (2009). Arterial spin labeling. Magnetic Resonance in Medicine, 61, 1374-1375.
Rusinek, H., Brys, M., Switalski, R., Haas, F., Mcgorty, K., Leon, M., et al. (2008). Hippocampal blood flow and vascular reserve: TrueFisp ASL at 3T Proc. International Society for Magnetic Resonance in Medicine, 16, 1918-1919.
Silver, M., & Joseph, I. (1985). Selective spin inversion in nuclear magnetic resonance and coherent optics through an exact solution of the Bloch-Riccati equation. Physical Review, 31, 2753-2755.
Takahashi, T., Shirane, R., & Sato, S. (1999). Developmental changes of cerebral blood flow and oxygen metabolism in children. American Journal of Neuroradiology, 20, 917–922.
Tofts, P. (2003). Quantitative MRI of the brain: measuring changes caused by disease. Chichester: John Wiley & Sons.
Wang, J., Alsop, D., Li, L., Listerud, J., Gonzalez, J., Schnall, M., et al. (2002). Comparison of quantitative perfusion imaging using arterial spin labeling at 1.5 and 4.0 tesla. Magnetic Resonance in Medicine, 48, 242–254.
Wang, J., & Licht, D. (2006). Pediatric perfusion MR imaging using arterial spin labeling. Journal of Neuroimaging Clinics of North America, 16, 149–167.
Wang, J., Licht, D., Jahng, D., Liu, C., Rubin, J., Haselgrove, J., et al. (2003). Pediatric perfusion-imaging using pulsed arterial spin labeling. Journal of Magnetic Resonance Imaging, 18, 404–413.
Warmuth, C., Gunther, M., & Zimmer, C. (2003).Quantification of blood flow in brain tumors: comparison of arterial spin labelling and dynamic susceptibility weighted contrast-enhanced MR imaging. Radiology, 228, 523–532.
Wong, E., Buxton, R., & Frank, L. (1999). Quantitative perfusion imaging using arterial spin labeling. Neuroimaging Clinics of North America, 9(2), 333-340.
Wong, E., Luh, W., & Liu, T. (2000). Turbo ASL: Arterial spin labeling with higher SNR and temporal resolution. Magnetic Resonance in Medicine, 44, 511–515.
Yang, Y., Engelien, W., Xu, S., Gu, H., Silbersweig, D., & Stern, E. (2000). Transit time, trailing time, and cerebral blood flow during brain activation: Measurement using multislice, pulsed spin-labeling perfusion imaging. Magnetic Resonance in Medicine 44, 680–685.
Zappe, A., Reichold, J., Burger, C., Weber, B., Buck, A., Pfeuffer, J., et al. (2007). Quantification of cerebral blood flows in nonhuman primates using arterial spin labeling and a two-compartment model. Magnetic Resonance Imaging, 25, 775–783.
Zhang, W., Silva, A., Williams, D., & Koretsky, A. (1995). NMR measurement of perfusion using arterial spin labeling without saturation of macromolecular spins. Magnetic Resonance in Medicine, 33(3), 370-376.
Zoccoli, G., Lucchi, L., & Andreoli, E. (1996). Arterial spin labelling. Journal of Cerebral Blood Flow Metabolism, 16, 1312–1318.