3T versus 1.5T phased-array MRI in the presurgical work-up of patients with partial epilepsy of uncertain focus.

PURPOSE: To study 3T compared to 1.5T phased array magnetic resonance imaging (MRI) in the presurgical work-up of patients with epilepsy with complex focus localization. MATERIALS AND METHODS: In all, 37 patients (>10 years) in preoperative work-up for epilepsy surgery were offered 3T in addition to 1.5T MRI if ambiguity existed about the epileptic focus. Scans were randomly reviewed by two observers, blinded for prior imaging, patient-identifying information, and each other's assessments, followed by a consensus meeting. The number of abnormal scans, detected lesions, and interobserver agreement were calculated and compared. The final consensus was compared to original scan reports. RESULTS: One observer identified 22 lesions in both 3 and 1.5T scans, while the second identified more lesions in 1.5T scans (28 vs. 20). 3T MRI had better interobserver agreement. 3T revealed more dysplasias, while 1.5T revealed more tissue loss and mesial temporal sclerosis (MTS). The final consensus yielded 29 lesions, whereas original reports identified only 17 lesions. CONCLUSION: The 3T scans revealed different lesions compared to 1.5T. Patients can benefit most from 3T scans when a dysplasia is suspected. Reevaluation by another experienced neuroradiologist is advised in case of negative or equivocal MRIs.

FID sampling superior to spin-echo sampling for T2*-based quantification of holmium-loaded microspheres: theory and experiment.

This work demonstrates both theoretically and experimentally that multiple gradient-echo sampling of free induction decay (MGEFID) is superior to MGE sampling of spin echo (MGESE) for T2*-based quantification of holmium-loaded microspheres (HoMS). An interleaved sampling strategy was applied in great detail to characterize the MR signal behavior of FID and SE signals of gels and perfused rabbit livers containing HoMS in great detail. Diffusion sensitivity was demonstrated for MGESE sampling, resulting in non-exponential signal decay on both sides of the SE peak and in an underestimation of the HoMS concentration. Other than MGESE sampling, MGEFID sampling was demonstrated to be insensitive to diffusion, to exhibit exponential signal decay, and to allow accurate T2*-based quantification of HoMS. Furthermore, a fit procedure was proposed extending the upper limit of quantifiable R2* relaxation rates to at least 1500 sec(-1). With this post-processing step incorporated, MGEFID was shown to correctly estimate the integral amount of inhomogeneously distributed HoMS in liver tissue, up to a clinically relevant limit. All experimental findings could be explained with the theory of nuclear magnetic resonance (NMR) signal behavior in magnetically inhomogeneous tissues. HoMS were shown to satisfy the static dephasing regime when investigated with MGEFID and to violate the static dephasing conditions for MGESE at longer echo times typically used in SE.

Factors affecting the effectiveness of a projection dephaser in 2D gradient-echo imaging.

Projection dephasers are often used for background suppression and dynamic range improvement in thick-slab 2D imaging in order to promote the visibility of subslice structures, e.g., blood vessels and interventional devices. In this study, we explored the factors that govern the effectiveness of a projection dephaser by simulations and phantom experiments. This was done for the ideal case of a single subslice hyper- or hypointensity against a uniform background in the absence of susceptibility effects. Simulations and experiments revealed a pronounced influence of the slice profile, the nominal flip angle and the TE and TR of the acquisition, the size, intraslice position and MR properties of the subslice structure, and T(1) of the background. The complexity of the ideal case points to the necessity of additional explorations when considering the use of projection dephasers under less ideal conditions, e.g., in the presence of tissue heterogeneities and susceptibility gradients.

Long-term toxicity of holmium-loaded poly(L-lactic acid) microspheres in rats.

The aim of this study was to get insight into the toxic effects of holmium-166-loaded poly(L-lactic acid) microspheres (Ho-PLLA-MS) which have very interesting features for treatment of liver malignancies. Acute, mid- and long-term effects were studied in healthy Wistar rats by evaluating clinical, biochemical and tissue response. Rats were divided into four treatment groups: sham, decayed neutron-irradiated Ho-PLLA-MS, non-irradiated Ho-PLLA-MS and PLLA-MS. After implantation of the microspheres into the liver of the rats, the animals were monitored (body weight, temperature and liver enzymes) for a period of 14-18 months. Some of the rats that received previously neutron-irradiated Ho-PLLA-MS were periodically scanned with magnetic resonance imaging (MRI) to see if holmium was released from the microspheres. After sacrifice, the liver tissue was histologically evaluated. Bone tissue was subjected to neutron-activation analysis in order to examine whether accumulation of released holmium in the bone had occurred. No measurable clinical and biochemical toxic effects were observed in any of the treatment groups. Furthermore, histological analyses of liver tissue samples only showed signs of a slight chronic inflammation and no significant differences in the tissue reaction between rats of the different treatment groups could be observed. The non-irradiated PLLA-MS and Ho-PLLA-MS stayed intact during the study. In contrast, 14 months after administration, the neutron-irradiated Ho-PLLA-MS was not completely spherical anymore, indicating that degradation had started. However, the holmium loading had not been released as was illustrated with MRI and affirmed by neutron-activation analysis of bone tissue. In conclusion, neutron-irradiated Ho-PLLA-MS does not provoke any toxic reaction and can be applied safely in vivo.

White-marker imaging--separating magnetic susceptibility effects from partial volume effects.

By applying dephasing gradients, local magnetic field inhomogeneitiescan selectively visualized with positive contrast, such as those created by magnetically labeled cells. This is known as "white-marker imaging." In white-marker imaging, subvoxel signal variations are also visualized as a result of partial volume (PV) effects and may compromise the identification of magnetic structures (e.g., magnetically-labeled cells). This study presents the theory and proof-of-principle experiments of a strategy to eliminate PV effects during white-marker imaging. The strategy employs the asymmetry of the signal response curves for non-PV effects as a function of externally applied gradients. In the case of PV effects, subtraction of the symmetrical signal responses eliminates their contribution. In vitro experimental images were made using a spherical phantom with cylindrical elements. In vivo images of the brain were obtained at a location that included air cavities (susceptibility effects) and the circle of Willis (PV effect). The results show that PV effects were eliminated in the in vitro experiments and were virtually absent under in vivo conditions.

EEG-fMRI in the preoperative work-up for epilepsy surgery.

Epilepsy surgery requires precise localization of the epileptic source. EEG-correlated functional MRI (EEG-fMRI) is a new technique showing the haemodynamic effects of interictal epileptiform activity. This study assesses its potential added value in the presurgical evaluation of patients with complex source localization. Adult surgical candidates considered ineligible because of an unclear focus and/or presumed multifocality on the basis of EEG underwent EEG-fMRI. Interictal epileptic discharges (IEDs) in the EEG during fMRI were identified by consensus between two observers. Topographically distinct IED sets were analysed separately. Only patients with significant, positive blood oxygen level-dependent (BOLD) responses that were topographically related to the EEG were re-evaluated for surgery. Forty-six IED sets from 29 patients were analysed. In eight patients, at least one BOLD response was significant, positive and topographically related to the IEDs. These patients were rejected for surgery because of an unclear focus (n = 3), presumed multifocality (n = 2) or a combination of both (n = 3). EEG-fMRI improved localization in four out of six unclear foci. In patients with presumed multifocality, EEG-fMRI advocated one of the foci in one patient and confirmed multifocality in four out of five patients. In four patients EEG-fMRI opened new prospects for surgery and in two of these patients intracranial EEG supported the EEG-fMRI results. In these complex cases, EEG-fMRI either improved source localization or corroborated a negative decision regarding surgical candidacy. It is thus a valuable tool in the presurgical evaluation of patients. Guidelines for the use of EEG-fMRI in clinical practice are proposed.

Factors affecting the sensitivity and detection limits of MRI, CT, and SPECT for multimodal diagnostic and therapeutic agents.

Noninvasive imaging techniques like magnetic resonance imaging (MRI), computed tomography (CT) and single photon emission computed tomography (SPECT) play an increasingly important role in the diagnostic workup and treatment of cancerous disease. In this context, a distinct trend can be observed towards the development of contrast agents and radiopharmaceuticals that open up perspectives on a multimodality imaging approach, involving all three aforementioned techniques. To promote insight into the potentialities of such an approach, we prepared an overview of the strengths and limitations of the various imaging techniques, in particular with regard to their capability to quantify the spatial distribution of a multimodal diagnostic agent. To accomplish this task, we used a two-step approach. In the first step, we examined the situation for a particular therapeutic anti-cancer agent with multimodal imaging opportunities, viz. holmium-loaded microspheres (HoMS). Physical phantom experiments were performed to enable a comparative evaluation of the three modalities assuming the use of standard equipment, standard clinical scan protocols, and signal-known-exactly conditions. These phantom data were then analyzed so as to obtain first order estimates of the sensitivity and detection limits of MRI, CT and SPECT for HoMS. In the second step, the results for HoMS were taken as a starting point for a discussion of the factors affecting the sensitivity and detection limits of MRI, CT and SPECT for multimodal agents in general. In this, emphasis was put on the factors that must be taken into account when extrapolating the findings for HoMS to other diagnostic tasks, other contrast agents, other experimental conditions, and other scan protocols.

Lanthanide-loaded liposomes for multimodality imaging and therapy.

UNLABELLED: Many advanced molecular imaging agents are currently being investigated preclinically. Especially, liposomes, have proven to be very promising carrier systems for diagnostic agents for use in single-photon emission computed tomography (SPECT) or magnetic resonance imaging (MRI), as well as for therapeutic agents to treat diseases such as cancer. In this study, nanosized liposomes were designed and labeled with the radionuclides, holmium-166 (both a beta- and gamma-emitter and also highly paramagnetic) or technetium-99m, and coloaded with paramagnetic gadolinium allowing multimodality SPECT and MR imaging and radionuclide therapy with one single agent. METHODS: Diethylenetriaminepentaacetic acid bisoctadecylamide (an amphiphilic molecule with a chelating group suitable for labeling with radionuclides) and gadoliniumacetylacetonate (GdAcAc) (a small lipophilic paramagnetic molecule) were incorporated in liposomes. The liposomes were characterized by measuring their mean size and size distribution, gadolinium content, and radiochemical stability after incubation in human serum at 37 degrees C. The MRI properties (in vitro) were determined by use of relaxivity measurements at 1.5 and 3.0 Tesla in order to evaluate their potency as imaging agents. RESULTS: The liposomes were successfully labeled with holmium-166, resulting in a high labeling efficiency (95% +/- 1%) and radiochemical stability (> 98% after 48 hours of incubation), and coloaded with GdAcAc. Labeling of liposomes with technetium-99m was somewhat less efficient (85% +/- 2%), although their radiochemical stability was sufficient (95% +/- 1% after 6 hours of incubation). MRI measurements showed that the incorporation of GdAcAc had a strong effect on the MRI relaxivity. CONCLUSIONS: The synthesized liposomes allow for multimodality imaging and therapy, which makes these new agents highly attractive for future applications.

Dephased MRI.

In this work gradient dephasing is treated as a mechanism for manipulating contrast in otherwise conventional MR images. The paper provides a theoretical and experimental framework for this approach. It starts from the observation that dephasing gradients invoke a shift in k-space. From this it is inferred that the effects of in-plane and through-plane dephasing can be systematically explored in the context of any given imaging experiment by sampling k-space more widely and densely than dictated by the field of view (FOV) and the spatial resolution of the desired images. The oversampled k-space allows an ensemble of lower-resolution dephased images to be reconstructed in which the degree and direction of dephasing are determined by the off-center position of the reconstruction window. The efficacy of this approach is demonstrated for standard gradient-echo acquisitions in a phantom. The results indicate the potential of the proposed methodology for evaluating 3D image data and optimizing gradient dephasing in applications that rely on the exploitation of partial volume and susceptibility effects (e.g., tracking interventional devices and tracing magnetically labeled substances).

Fully MR-guided hepatic artery catheterization for selective drug delivery: a feasibility study in pigs.

PURPOSE: To demonstrate the feasibility of hepatic catheterization for selective delivery of therapeutic agents using a clinical MRI scanner for real-time image guidance. MATERIALS AND METHODS: Experiments were performed in three domestic pigs (70-80 kg) using a clinical 1.5-T MR scanner. After abdominal three-dimensional contrast-enhanced MR angiography (3D-CE-MRA) was performed, endovascular devices with susceptibility markers were tracked with passive tracking techniques. Catheters were maneuvered into the primary and secondary hepatic arteries. Selective catheterization was verified using selective time-resolved CE angiography. Paramagnetic microspheres were administered to a different region for each liver. The resulting biodistributions were investigated using MR images. RESULTS: Successful selective hepatic catheterization was repeatedly demonstrated using passive tracking techniques. 3D-CE-MRA significantly aided the interventional procedure by showing the vascular anatomy, and maximum-intensity projections (MIPs) were used as roadmaps during the interventions. In all cases, microspheres were successfully delivered to the selected regions. The catheters were visualized at a maximum frame rate of five frames per second, allowing a good depiction of the devices and a reliable catheterization of the hepatic arteries. CONCLUSION: Fully MR-guided real-time navigation of endovascular devices permits complex procedures such as selective intra-arterial delivery of therapeutic agents to parts of the liver.

Spectral characterization of local magnetic field inhomogeneities.

The purpose of this study was the characterization of local magnetic susceptibility deviations by spectral analysis of their induced magnetic field inhomogeneities. Magnetic resonance spectra and related signal decay curves of local susceptibility deviations were simulated for different volume fractions and compositions of the object within the VOI. The size or composition of the object was varied at constant volume fraction, constant object size, or at constant 'magnetic strength' (defined as the product of the volume and the volume susceptibility of the object). Experimental spectra were acquired for individual metal spherical particles and a spherical air cavity. Where possible, spectra were used to characterize objects in terms of volume and composition. By simulations, a numerical relation was determined between the spectral broadening and the object's volume and composition. Comparison of spectra for various spherical objects showed the possibility of characterization with respect to size and composition. Experimental results confirmed the numerical results to a large extent, although the characterization was compromised by background signal decay, low volume fractions and limitations in signal-to-noise. In conclusion, spectral description of the field inhomogeneities related to small objects allows characterization of such objects with respect to size and composition. Practical applicability of the simulation results depends on background signal decay and volume fraction of the object.

Internal radiation therapy of liver tumors: qualitative and quantitative magnetic resonance imaging of the biodistribution of holmium-loaded microspheres in animal models.

In internal radiation therapy of unresectable liver tumors, microspheres containing a radionuclide are injected in the hepatic artery to achieve a preferential deposition of microspheres in the lesions. In this study, MR imaging techniques for qualitative and quantitative assessment of the biodistribution of holmium-loaded microspheres (HoMS) were investigated for their use in selective internal radiation therapy of liver tumors. To achieve this goal, the relaxivity of HoMS was first investigated in gel experiments. The resultant calibration curve was subsequently employed to quantify the biodistribution of HoMS administered to 13 excised rabbit livers and to the livers of 3 live rabbits with an implanted tumor. Finally, the feasibility of MR imaging of the biodistribution during treatment of a large animal was investigated by MR imaging of hepatic administration of HoMS to a live pig. Overall, the study showed that MRI can clearly depict the biodistribution of HoMS, but that quantification by means of the gel calibration curve yields an underestimation that increases for higher amounts of HoMS. The observed underestimation is tentatively attributed to accumulations of HoMS in larger liver vessels. The exploratory quantification experiments suggest the feasibility of MR dosimetry.

Adaptive subtraction as an aid in MR-guided placement of catheters and guidewires.

PURPOSE: To demonstrate the utility of mask subtraction optimization in magnetic resonance (MR)-guided placement of catheters and guidewires. MATERIALS AND METHODS: MR-guided positioning of magnetically prepared catheters and guidewires was done by dynamically imaging a single thick slab at two frames per second. Selective visualization of the prepared parts of the devices was achieved by the use of a conventional baseline subtraction technique and by the use of an adaptive subtraction technique. In the latter, the best reference image is automatically selected from a fixed or a sliding subset of hitherto acquired dynamic images. The efficacy of both approaches was compared by tracking experiments in a flow phantom and in the aortoiliac arteries of a pig. RESULTS: Baseline subtraction produced adequate visualization of paramagnetic markers in the absence of subject motion and for fixed scan conditions. The sensitivity to subject motion and interactive modification of the scan parameters was greatly reduced by using adaptive subtraction. Adaptive subtraction images, other than conventional subtraction images, appeared to be insensitive to slow periodic motion, e.g., respiratory motion, and were only transiently affected by gross subject motion and interactive alterations of the scan parameters. CONCLUSION: Adaptive subtraction is superior to baseline subtraction for guiding the manipulation of catheters and guidewires in the presence of gross and periodic subject motion and whenever scan parameters are modified in the course of a procedure.

Liver tumors: MR imaging of radioactive holmium microspheres--phantom and rabbit study.

PURPOSE: To investigate the use of magnetic resonance (MR) imaging in the administration and biodistribution of holmium-loaded poly(L-lactic acid) microspheres (Ho-PLLA-MS) in liver tumors. MATERIALS AND METHODS: MR imaging measurements were obtained in phantoms, three ex vivo rabbit livers, and four livers in living rabbits. When applicable, measurements were compared with those on scintigraphic images. The transverse relaxivity R2* of the Ho-PLLA-MS was determined in a phantom study. The in vivo animal experiments were performed by using rabbits with an implanted VX2 tumor. Detection of passing Ho-PLLA-MS to estimate lung shunting was performed in a scaled model of the vena cava. RESULTS: In the ex vivo liver experiments, the feasibility of real-time MR imaging during administration of microspheres was demonstrated. The in vivo rabbit experiments demonstrated that MR imaging can depict radioactive, nonradioactive, and decayed Ho-PLLA-MS after treatment for as long as they remain in the body. Furthermore, this study showed the ability of dynamic MR imaging to detect single doses of passing Ho-PLLA-MS. CONCLUSION: Ho-PLLA-MS used for internal radionuclide therapy can be imaged clearly in vivo with MR imaging.

Passive tracking exploiting local signal conservation: the white marker phenomenon.

This article presents a novel approach to passive tracking of paramagnetic markers during endovascular interventions, exploiting positive contrast of the markers to their background, so-called "white marker tracking." The positive contrast results from dephasing of the background signal with a slice gradient, while near the marker the signal is conserved because a dipole field induced by the marker compensates the dephasing gradient. Theoretical investigation shows that a local gradient induced by the local dipole field will nearly always cancel the dephasing gradient somewhere, regardless of marker composition, gradient strength, orientation, and acquisition parameters. The actual appearance of the white marker is determined by the marker strength, echo-time, slice thickness, and gradient strength, as shown both theoretically and experimentally. The novel concept is demonstrated by tracking experiments in a flow phantom and in pig models and is shown to allow reliable and robust depiction of paramagnetic markers with positive contrast and significant suppression of the background signal.