Abstract
The susceptibility differences at the gas-liquid interface of microbubbles (MBs) allow their use as an intravascular susceptibility contrast agent for in vivo magnetic resonance imaging (MRI). However, the characteristics of MBs are very different from that of the standard Gd-DPTA contrast agent, including size distribution and hemodynamic properties which could influence MRI outcomes. Here we quantitatively investigated the correlation between relative cerebral blood volume (rCBV) derived from Gd-DTPA (rCBVGd) and the MB-induced susceptibility effect
(R2*MB) by conventional dynamic susceptibility contrast MRI (DSC-MRI).
Custom-made MBs had a mean diameter of 0.92 um and were capable of inducing 4.68±3.02% of the maximum signal change. The MB-associated R2*MB was
compared with rCBVGd on 16 rats at 4.7 T MRI. We observed a significant effect of
the time to peak (TTP) on the correlation between R2*MB and rCBVGd, and also found
the noticeable dependence between TTP and maximum signal change (MSC). Our findings suggest that MBs with longer TTPs can be used for estimation of rCBV by DSC-MRI, and emphasize the critical effect of TTP on MB-based contrast MRI.
Keywords: Microbubbles, DSC-MRI, Time To Peak, Cerebral
1. Introduction
Dynamic susceptibility contrast magnetic resonance imaging (DSC-MRI) has long been the standard method for determining cerebral blood volume and perfusion parameters (1-4). The most widely used contrast agents for DSC-MRI are the gadolinium chelates such as Gd-DTPA and Gd-DOTA. After a bolus of the exogenous contrast agent is injected into a vein, its paramagnetic properties will induce strong T2* relaxation, contributing to intensive loss of signal. By tracking the
passage of contrast agent inside the vessels, DSC-MRI can then be used to estimate the relative cerebral blood volume (rCBV) based on the integral of the change of the transverse relaxation rate (R2*) (2, 5).
Gas-filled microbubbles (MBs) were originally developed as a contrast agent for ultrasound imaging because of the difference in acoustic impedance between the contrast agent and surrounding tissues. With diameters in the micrometer range, MBs can traverse the microcirculatory system as blood pool tracers and can be widely applied in the field of ultrasonic medical imaging, including perfusion evaluation in stroke patients (6), assessment of micro circulation volumetric flow (7), and quantification of myocardial blood volume (8). The potential applications of MBs can be further extended to targeted imaging for noninvasive evaluation of tumor angiogenesis (9). In addition, when combined with focused ultrasound (FUS), MBs were shown to temporally and locally disrupt the blood-brain barrier (BBB) thereby enhancing drug delivery into brain tumors (10). MBs also have the potential to serve as drug-carrying vehicles. Given these characteristics, MBs are anticipated to hold great promise for use in diverse medical applications.
MBs are typically stabilized by encapsulation in a thin shell of lipid, protein, or polymer to avoid quick diffusion of gas. In this study, we used custom-made MBs with a lipid shell and a perfluorocarbon gas core, which has the advantage of lower
gas solubility. Recently, MBs were utilized as a novel MR susceptibility contrast agent based on the susceptibility differences between the gas and liquid interfaces (11). MBs have been used as an MR intravascular susceptibility contrast agent for liver imaging in animal studies (12), and Cheung et al. demonstrated the feasibility of applying MBs in rat brains (11). Despite the increasing number of studies on MB-based MRI, there has been limited information on their induced susceptibility effect. Alterations in size, relaxation mechanisms, and chemical characteristics could lead to differences between MB- and Gd-based contrast agents in perfusion imaging. Here we carried out a systematic and quantitative investigation of the correlation of rCBV derived from Gd-DTPA and the induced susceptibility effect from MBs, with the aim of identifying the critical factor that impacts signal changes in MB-based perfusion MRI.
2. Methods and Materials
All MRI experiments were conducted on a 4.7 T animal MRI scanner (Bruker Biospec 47/40) with a 72 mm volume coil used as radio frequency (RF) transmitter and a quadrature surface coil as the receiver. This study was approved by the local institutional animal ethics committee.
Microbubble preparation
The lipid components, 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) and 1,2-distearoyl-sn-glycero-3-phospho-(1'-rac-glycerol) (DSPG), were purchased from Avanti polar lipids (AL, USA). Polyoxyethylene 40 stearate (PEG40S) was obtained from Sigma-Aldrich (MO, USA) and used as an emulsifier to increase the
fluidity of the lipid membrane. To prepare the aqueous lipid solution, the lipid/emulsifier mixture composed of 60 wt% DPPC, 30 wt% DSPG and 10 wt% PEG40S was directly dissolved in filtered (0.2 um) 1% wt glycerol-containing phosphate buffered saline (20 mM, pH7.4) and dispersed by a bath sonicator at 80C° until a transparent and homogeneous solution was obtained. The lipid/emulsifier concentrations were adjusted to 3 mg/mL. The solution was then divided into 1 mL-aliquots in 1.8-mL airtight vials for later degassing and refilling with perfluoropropane (C3F8) at atmospheric pressure. The final formation of MBs was
achieved by agitation with a home-built high speed agitator at 4550 rpm for 45 seconds. To estimate the MB diameter distribution, the size distribution and concentration of MBs in the suspension were measured by the electrical sensing zone (ESZ) system. The concentration was (4.36±0.32)×1010 droplets/mL and the mean
diameter of MBs was 0.92 um (Figure 1a). An optical image of MBs by bright field microscope was presented in Figure 1b to show the various sizes of MBs. The measured volume fraction of MBs was 7%. The prepared MBs have been utilized in the fields beyond MRI for many studies (10, 13). Via intravenous route, these prepared MBs did not induce any adverse effects or animal deaths. Similar lipid formulations have been proved by Food and Drug Administration (FDA) and commercialized for human uses. Thus, we suggested that these lipid-based MBs was
biocompatible and safe for clinical applications. Animal preparation
Dynamic MRI was performed in 16 male Sprague-Dawley rats weighing 250~390 g. A 23-gauge needle connected to a 0.8 m polyethylene-50 tube was inserted into a tail vein for injection of contrast agents. The dead volume in the catheter was about 0.2 mL. A long catheter was used to enable injection outside of the magnet. During imaging, the rats were anesthetized with isoflurane (5% for initial induction and 1.5% for maintenance) via a nose cone with respiratory monitoring. Administration of contrast agents
The MB suspensions were slowly rotated for 2 min until a homogeneous milky-white suspension formed, and 0.2 mL of the suspension was then drawn into a 1-cc syringe. The bolus injection of MB suspensions were administered via the tail vein during dynamic MRI scanning. The catheter was flushed with 0.5 mL 0.9% NaCl solution after injection. To ensure adequate clearance of MBs, the minimum interval before the next injection was 10 min [11]. For comparison, the bolus injection of 0.2 mL Gd-DTPA (Magnevist, Bayer Schering Pharma, Germany) was administered by the same scheme.
MRI acquisition
The rapid acquisition with relaxation enhancement (RARE) sequence was performed on all rats in the axial plane for anatomic images. Scan parameters were
TR=4000 ms, TE=70 ms, field of view (FOV)=29 mm×29 mm, slice thickness=1.5 mm, acquisition matrix=256×128, and echo train length (ETL)=8. DSC-MRI was performed in the axial plane of the anterior commissure by gradient echo echo-planner imaging (GE-EPI). Scan parameters were TR=1000 ms, TE=30 ms, flip angle=90°, FOV=29 mm×29 mm, slice thickness=1 mm, acquisition matrix=96×96, and sampling bandwidth (BW) =200 kHz. Six hundred data points were obtained with a temporal sampling interval of 1 s and total acquisition time of 600 s. To avoid contamination by residual Gd-DTPA, the imaging session with MBs was performed prior to that of Gd-DTPA. Both contrast agents were injected after a period of 2 minutes for acquisition of baseline images.
Data analysis
The dynamic images were first realigned using the SPM 5 software. To assess the brain perfusion of each rat, a ROI was manually drawn along the edge of the brain. To avoid susceptibility artifacts caused by respiratory motion in the trachea (14), the pixels close to the trachea were excluded from the ROI selection. The concentration C(t) during the passage of Gd-DTPA could be expressed as:
* 2 0 S t 1 C t R l n TE S [1]where S(t) is the signal intensity in the tissue at time t, S0 is the average signal
The rCBV maps were obtained by conventional methods by the integration of the area under the concentration time curve of the first pass in each pixel, using the following formula (2, 5):
[2]
Because the heart rate of the rat can be as quick as 300 bpm or more (15), the first passage of concentrated contrast agent was expected to be only a few seconds. To overcome this potential problem, we assumed that the input function was identical for all pixels, as the contrast agent circulated in the blood vessels. Therefore, the whole Ct(t) of Gd-DTPA could be included in the range of integration to calculate the rCBV
(rCBVGd) (16) .
In previous reports using MBs as contrast agent, the R2* was measured for
quantification of the susceptibility effect of MBs. Accordingly, we assessed the MBs by R2* based on a pixel-by-pixel analysis, using the equation 1 (11). S0 was the
average intensity of 50 baseline images. R2*MB was then derived from the average
R2* of 40 consecutive post-injection images at the timing of maximum susceptibility
contrast. For MBs with slow passage through capillary bed, these consecutive data points were located at the plateau region of the R2* time course. In addition, another
value of AreaMB was calculated from the integration of the time course of R2* of
contrast agent with fast bolus passage, as described previously.
Quantitative parameters of time-to-peak (TTP) and maximum signal change (MSC) were extracted from signal change curves for further evaluation. The signal change curve was described by the following formula:
0 f
0
S S t
si gnal change= 100%
S [3]
where Sf(t) was the mean postcontrast signal intensities of ROI varying with time,
using a moving average filter with a span of 5 points of the time course to reduce the noise interference. The MSC was measured as the signal change at the maximum peak, and TTP was defined as the time needed for maximum signal change after injection of MBs.
Statistical analysis
The correlations between the rCBVGd and AreaMB/R2*MB were estimated by using
the Pearson's correlation coefficient r. Then, the Pearson's correlation coefficient r was transformed into the Fisher's Z for a normalizing transformation (17). Differences in correlations between rCBVGd and AreaMB and rCBVGd and R2*MB were assessed
using the paired Student t-test (two-tailed). Due to observation of varied TTPs in our in vivo experiments, we tried to model the relation between the TTP and the transformed correlation between the rCBVGd and R2*MB, and the relation between
coefficient of variation (CV) was used. Values of P < 0.05 were accepted as statistically significant. All data were processed by in-house tool programmed by Matlab (The MathWorks, Natick, MA, USA).
3. Results
We compared the use of MBs and Gd-DTPA by correlating the maps of AreaMB
and R2*MB from MBs trials with the rCBVGd maps from Gd-DTPA trials (Figure 2).
With the P less than 0.05 (P=0.0004), R2*MB exhibited a higher correlation with
rCBVGd when compared with that of AreaMB. Therefore, only R2*MB is used for
further analysis hereafter.
The phenomenon of signal decrease after MB injection was observed in all 16 rats. Typical MRI images of one case with a good correlation between rCBVGd and R2*MB
are shown in Figure 3. The ROI for time course analysis was selected based on the GE-EPI T2*-weighted image (Figure 3a, red dashed line) and excluded voxels near the
trachea (Figure 3a, arrow). rCBVGd and R2*MB maps were constructed (Figure 3b,c)
and T2*-weighted signal time curves during Gd-DTPA and MBs injection were
determined for the whole ROI (Figure 3d). The negative contrast agent Gd-DTPA had an excellent susceptibility effect with a signal change of 80% in this rat. For MBs, however, the induced signal drop was only 6.95%. Although the signal change of MBs was much weaker in comparison to Gd-DPTA, it was significant enough to
estimate brain perfusion. A time-course comparison revealed that the peaks with MBs were delayed and dispersed. The TTP of MBs was 269 sec while that of Gd-DTPA for the same rat was 6 sec (Figure 3d). In addition, the signals did not return to the preinjection baseline after injection of MBs. The R2*MB map was similar to the
rCBVGd map with a correlation coefficient of 0.90 (Figure 3e).
Brain images of a contrary case are shown in Figure 4. In this case, the induced signal change with MBs was 3.35%. The TTP of MBs was 93 sec while that of Gd-DTPA for the same rat was 6 sec (Figure 4d) The Gd-Gd-DTPA provided the advantage of intersubject consistency in terms of TTP. In contrast, the TTPs of MBs were delayed and varied from rat to rat. Furthermore, the correlation coefficient of 0.71 for R2*MB and rCBVGd maps for this case is lower than the previous case (Figure 4e).
Linear regression between the TTP and correlation between the rCBVGd and
R2*MB is shown in Figure 5(a). With the P less than 0.05, there was a significant
relationship between the correlation between rCBVGd and R2*MB and the mean TTP
derived from MBs trials (R2=0.65). The correlation between rCBV
Gd and R2*MB was
an upward trend with increasing TTP. In addition, a positive proportional correlation was found with MSC (R2=0.6). This finding suggested that DSC-MRI trials using
MBs with longer TTP accompanied with enhanced susceptibility effect (Figure 5b). Table 1 shows the MSCs as well as TTPs of Gd-DTPA and MBs. The TTP for
MBs was 121.19±68.23s and the corresponding MSC was 4.68±3.02%. The TTPs of MBs were distributed over a wide range from 51s to 287s, revealing that the dynamic behavior of MBs during MRI scan varied from rat to rat (CV=56.3%). In contrast, the TTPs of Gd-DTPA were consistent in all rats (CV=12.79%). The TTP of MBs suggested that MBs may have been trapped and delayed in the vasculature.
4. Discussion
In this study, MBs were used as an intravascular contrast agent for MRI, and the induced susceptibility effects were quantitatively investigated and correlated to rCBVGd in a normal rat model, for the first time. In a previous pioneering report, the
R2* map of MBs was qualitatively compared with the R2* map of monocrystalline
iron oxide nanoparticle (MION) (11), which is used as a steady-state contrast agent. Due to the superparamagnetism of iron oxide nanoparticles, the steady state R2* of
MION in the blood pool has often been used to monitor the change of rCBV, in both functional and pharmacological MRI animal experiments (18-20). In a clinical setting, the bolus passage of Gd chelates such as Gd-DTPA is a standard method for measuring rCBV. The investigation on the relation between R2*MB and rCBVGd are
worthy of further study.
TTP derived from DSC-MRI is valuable by providing information regarding the hemodynamic alterations. TTP is relatively easy to generate, and quantification of the
TTP is an indirect evaluation of cerebral blood flow (CBF). For tissue with longer TTP, it is significantly associated with perfusion deficit (21). MSC is a mixed measure of contrast dose, rate of injection, blood volume, blood flow, cardiac output as well as other hemodynamic parameters. Although it is not a pure measure of physiology, MSC is found to have a good correlation with rCBV for DSC-MRI (22), which is related with the induced susceptibility effect. However, due to the larger size of MBs, the dynamics of MBs should be different from that of Gd-based contrast agents, and the meaning of TTP and MSC of MBs-based contrast agents have not been fully discovered in previous literature. In this study, adopting TTP as an index in MBs-based perfusion MRI, we demonstrate that it is a potential means for recording the hemodynamic situation whether MBs are trapped in the microvasculature or not. The MSC represents the susceptibility effect, which could be altered by the concentration, size distribution, and trapping effect of MBs.
MBs have been applied in fields beyond ultrasound for years (11, 23-24). Studies also demonstrated the potential capability of MBs to be the MRI-based perfusion contrast agent (11, 25). A promising application of MBs is toward tumor therapy. In the glioma animal model, Ting et al (10) and Fan et al (26) proposed a novel technique of the use of drug-loaded MBs with FUS-induced BBB opening to enhance the local drug delivery and improve the overall survival. The tumor follow-up was
performed via T2-weighted images. If the application of MBs as a contrast agent for perfusion MRI could be applied in the tumor model, it allows to discover comprehensive imaging information, including tumor perfusion and tumor angiogenesis, which is anticipated to benefit the prediction of local tumor control after therapy.
Nanosized Gd chelates are presumably well-mixed in the blood plasma, and a substantial susceptibility contrast is induced as the bolus of contrast agent first passes through the tissue. The area under the R2* time course of the first pass is calculated
as rCBV (2-3). However, this method may be not suitable for estimating rCBV when using MBs as contrast agents. When compared with Gd-DTPA, MBs were delayed and dispersed over time. We therefore inferred that the hemodynamic characteristics of MBs are considerably different from those of Gd-DTPA. In Figure 2, we also demonstrated that R2*MB exhibited a higher capability of estimating rCBV. However,
considerable variations existed in the relation between R2*MB and rCBVGd, which
may hamper the accuracy of rCBV estimation.
Moreover, we observed large variations in TTPs derived from MBs trials in our experiments. For human study, the maximum concentration time and the residue time of MBs were also varied among subjects when using the commercial SonoVueTM (29).
aggregation and trapping of MBs with a wide size distribution, which is also mentioned in the previous report (11). MBs aggregation may accelerate once it has started (25), setting off a chain-reaction that could lead to large differences when the aggregation is initiated at different times and sites within the complex circulatory system. Since the wide range of TTP variation in in vivo studies is unavoidable for MBs, our analysis of the various TTPs has given further valuable information about the passage of MBs in the local tissue vasculature of brain. We found that the correlations between rCBVGd and R2*MB were higher when MBs had longer TTP, and
the signal changes were also increased (Figure 5). Therefore, we speculated that aggregation of MBs could be a possible cause of longer TTP. When the MBs aggregate as clusters, the flow of MBs is expected to be slower. In the capillaries, the clusters of MBs would be more easily trapped and thus remain for a longer time. As more and more gas-filled MBs are retained in the tissue vasculature, a profound susceptibility can thus be induced. Furthermore, at the time when the vasculature is fully loaded with MBs, the susceptibility effect is expected to more closely resemble a well-mixed quasi-steady state for measuring the representative rCBV. Therefore, high correlations between rCBVGd and R2*MB (Pearson’s r=0.8~0.9) were found in those
cases with the longest TTPs. In contrast, cases with short TTPs may have involved fast passage of MBs through the vessels with reduced susceptibility effect. Therefore,
MBs with shorter TTPs would not be suitable for blood volume estimation. If we can tailor the ranges of TTP in MBs, MBs with longer TTP could be a potential contrast agent in perfusion MRI, for the purpose of being dual-modality contrast agents with only one injection of contrast agent.
In the field of ultrasound, the size of MBs is an important parameter that influences the degree of backscatter. In terms of the MR susceptibility effect, R2*
increases approximately with the fifth power of the radius of the MBs (25). The current accepted size for MBs is around 3 um (28), allowing enough backscatter to be induced without filtered occurring in the lung capillary bed. In our study, the mean diameter of MBs was 0.92 um allowing passage through the lung capillary safely. Smaller MBs carry the cost of a lower susceptibility effect. To compensate for this undesired effect, MBs with a higher volume fraction as 7% were used in our study. The signal change for an in vivo MBs experiment was previously found to be about 6% with a 7 T MRI scanner (11). In our experiments, the average signal change was 4.68% at a lower magnetic field of 4.7 T. Thus smaller-sized MBs can be manipulated by using a higher volume fraction to boost the susceptibility effect. The MBs with smaller size were custom made by state-of-the-art manufacture procedures, which was achieved through controlling the formulation and fabricating temperature (29). The advantages of small size are profound. The increased surface area could for example
lead to more binding of surface-modified ligands to MBs, thus increasing their efficiency as drug-carrying vehicles (10, 30). Moreover, small MBs have a high resonant frequency, which allows high frequency ultrasound with stable cavitation for BBB opening (10, 31). However, the size distribution of the fabricated bubbles is still too wide to evaluate the MB size effects. To have a deep understanding of the MBs size effects on the CBV measurement in perfusion MRI, the technique of differential centrifugation is possible to be utilized for further studies (32).
5. Conclusions
In conclusion, we demonstrated the feasibility of MBs as an MR intravascular susceptibility contrast agent in vivo for measuring rCBV. Our results suggested that the observed TTP of MBs is longer than the conventional Gd-DTPA method, and the TTP is a prominent factor for MBs contrasted DSC-MRI. As the TTP of MBs increased, the correlation between R2*MB map and rCBVGd map also increased.
Furthermore, we showed that the susceptibility effect induced by MBs with a higher volume fraction at a lower magnetic field was comparable to previous reports which used MBs with a lower volume fraction at a higher magnetic field. MBs are a promising contrast agent with multiple applications, which are not limited to perfusion imaging. Surface modifications for specificity and affinity could allow MBs to serve as drug-carrying vehicles as well as targeting imaging. The ability of
cavitation effects could also be used to open the BBB to increase the therapeutic effect. Combined with real-time MRI monitoring, MBs could lead to simultaneous treatment and evaluation in the future. Through this systematic study, the characteristics of MBs on DSC-MRI was well demonstrated. We also proposed the essential factors for considering future MRI applications with MBs.
6. Acknowledgements
This work was supported by the National Science Council of Taiwan, grant no. NSC 99-2320-B-007-004-MY3. The authors would also like to acknowledge for the instrument support of Dr. Chen Chang and the Functional and Micro-Magnetic Resonance Imaging Core of Genomics Research Center, Academia Sinica, Taiwan.
References
1. Rosen BR, Belliveau JW, Aronen HJ, Kennedy D, Buchbinder BR, Fischman A, Gruber M, Glas J, Weisskoff RM, Cohen MS, Hochberg FH, Brady TJ. Susceptibility contrast imaging of cerebral blood volume: human experience. Magn Reson Med. 1991; 22: 293-299
2. Rempp KA, Brix G, Wenz F, Becker CR, Guckel F, Lorenz WJ. Quantification of regional cerebral blood flow and volume with dynamic susceptibility contrast-enhanced MR imaging. Radiology. 1994; 193: 637-641. 3. Ostergaard L, Sorensen AG, Kwong KK, Weisskoff RM, Gyldensted C,
Rosen BR. High resolution measurement of cerebral blood flow using intravascular tracer bolus passages .2. Experimental comparison and preliminary results. Magn Reson Med. 1996; 36: 726-736.
4. Calamante F, Gadian DG, Connelly A. Quantification of perfusion using bolus tracking magnetic resonance imaging in stroke: assumptions, limitations, and potential implications for clinical use. Stroke. 2002; 33: 1146-1151.
5. Axel L. Cerebral blood flow determination by rapid-sequence computed tomography: theoretical analysis. Radiology. 1980; 137: 679-686.
6. Meairs S. Contrast-enhanced ultrasound perfusion imaging in acute stroke patients. Eur Neurol. 2008; 59 Suppl 1: 17-26.
using high-resolution, contrast-assisted images. IEEE T Ultrason Ferr. 2008; 55: 74-83.
8. Zwirn G, Beeri R, Gilon D, Friedman Z, Akselrod S. Quantitative evaluation of local myocardial blood volume in contrast echocardiography. Med Image Anal. 2009; 13: 62-79.
9. Ellegala DB, Leong-Poi H, Carpenter JE, Klibanov AL, Kaul S, Shaffrey ME, Sklenar J, Lindner JR. Imaging tumor angiogenesis with contrast ultrasound and microbubbles targeted to alpha(v)beta3. Circulation. 2003; 108: 336-341. 10. Ting CY, Fan CH, Liu HL, Huang CY, Hsieh HY, Yen TC, Wei KC, Yeh CK.
Concurrent blood-brain barrier opening and local drug delivery using drug-carrying microbubbles and focused ultrasound for brain glioma treatment. Biomaterials. 2012; 33: 704-712.
11. Cheung JS, Chow AM, Guo H, Wu EX. Microbubbles as a novel contrast agent for brain MRI. Neuroimage. 2009; 46: 658-664.
12. Wu EX, Wong KK, Huang I, Kim YR, Tang HY, Yang ES, Kwong KK. In vivo study of microbubbles as an MR susceptibility contrast agent. Mag Reson Med. 2004; 52: 445-452.
13. Fan CH, Liu HL, Huang CY, Ma YJ, Yen TC, Yeh CK. Detection of intracerebral hemorrhage and transient blood-supply shortage in focused-ultrasound-induced blood-brain barrier disruption by ultrasound imaging.
Ultrasound Med Biol. 2012; 38: 1372-1382.
14. Yamamoto A, Sato H, Enmi J, Ishida K, Ose T, Kimura A, Fujiwara H, Watabe H, Hayashi T, Iida H. Use of a clinical MRI scanner for preclinical research on rats. Radiol Phys Technol. 2009; 2: 13-21.
15. Wu Y, Wu EX. MR study of postnatal development of myocardial structure and left ventricular function. J Magn Reson Imaging, 2009; 30: 47-53.
16. Wu YL, Chen CC, Wu YC, Chang CH, Wang FN. DSC MRI on Rat Model: Choosing the Integration Inteval for Measuring CBV. Proceedings of the 19th Annual Meeting ISMRM, Stockholm, Sweeden, 2011; 3922.
17. Fisher RA. Frequency distribution of the values of the correlation coefficient in samples from an indefinitely large population. Biometrika. 1914; 10: 507-521.
18. Mandeville JB, Marota JJ, Kosofsky BE, Keltner JR, Weissleder R, Rosen BR, Weisskoff RM. Dynamic functional imaging of relative cerebral blood volume during rat forepaw stimulation. Magn Reson Med. 1998; 39: 615-24. 19. Chen YC, Mandeville JB, Nguyen TV, Talele A, Cavagna F, Jenkins BG.
Improved mapping of pharmacologically induced neuronal activation using the IRON technique with superparamagnetic blood pool agents. J Magn Reson Imaging. 2001; 14: 517-524.
BG, Hui KK, Kwong KK. Inhibition of stimulated dopamine release and hemodynamic response in the brain through electrical stimulation of rat forepaw. Neurosci Lett. 2008; 431: 231-235.
21. Neumann-Haefelin T, Wittsack HJ, Fink GR, Wenserski F, Li TQ, Seitz RJ, Siebler M, Modder U, Freund HJ. Diffusion- and perfusion-weighted MRI: influence of severe carotid artery stenosis on the DWI/PWI mismatch in acute stroke. Stroke. 2000; 31: 1311-7.
22. Cha S, Lu S, Johnson G, Knopp EA. Dynamic susceptibility contrast MR imaging: correlation of signal intensity changes with cerebral blood volume measurements. J Magn Reson Imaging. 2000; 11: 114-9.
23. Willmann JK, Cheng Z, Davis C, Lutz AM, Schipper ML, Nielsen CH, Gambhir SS. Targeted microbubbles for imaging tumor angiogenesis: assessment of whole-body biodistribution with dynamic micro-PET in mice. Radiology. 2008. 249: 212-219.
24. Feshitan JA, Vlachos F, Sirsi SR, Konofagou EE, Borden MA. Theranostic Gd(III)-lipid microbubbles for MRI-guided focused ultrasound surgery. Biomaterials. 2012; 33: 247-255.
25. Wong KK, Huang I, Kim YR, Tang HY, Yang ES, Kwong KK, Wu EX. In vivo study of microbubbles as an MR susceptibility contrast agent. Mag Reson Med. 2004; 52: 445-452.
26. Fan CH, Ting CY, Liu HL, Huang CY, Hsieh HY, Yen TC, Wei KC, Yeh CK. Antiangiogenic-targeting drug-loaded microbubbles combined with focused ultrasound for glioma treatment. Biomaterials. 2013; 34: 2142-2155.
27. Morel DR, Schwieger I, Hohn L, Terrettaz J, Llull JB, Cornioley YA, Schneider M. Human pharmacokinetics and safety evaluation of SonoVue, a new contrast agent for ultrasound imaging. Invest Radiol. 2000; 35: 80-5. 28. Schutt EG, Klein DH, Mattrey RM, Riess JG. Injectable microbubbles as
contrast agents for diagnostic ultrasound imaging: the key role of perfluorochemicals. Angew Chem Int Ed Engl. 2003; 42: 3218-3235.
29. Wang CH, Yeh CK. Controlling the Size Distribution of Lipid-Coated Bubbles via Fluidity Regulation. Ultrasound Med Biol. 2013; 39: 882-892 30. Phillips LC, Klibanov AL, Wamhoff BR, Hossack JA.
Ultrasound-microbubble-mediated drug delivery efficacy and cell viability depend on microbubble radius and ultrasound frequency. IEEE International Ultrasonics Symposium Proceedings, 2010: 1775-8.
31. Wang CH, Huang YF, Yeh CK. Aptamer-conjugated nanobubbles for targeted ultrasound molecular imaging. Langmuir; 2011: 27: 6971-6976.
32. Sirsi S, Feshitan J, Kwan J, Homma S, Borden M. Effect of microbubble size on fundamental mode high frequency ultrasound imaging in mice. Ultrasound Med Biol. 2010; 36: 935-948.
Table 1. Maximum signal change (MSC) and time-to-peak (TTP) for MBs and Gd-DTPA TTP (sec) MSC (%) MBs 121.18±68. 23 4.7±3.0.2 Gd-DTPA 5.6±0.72 77.38±6.61
