Detection of liver fibrosis with magnetic cross-relaxation

The utility of MRI using magnetization transfer (MT) enhanced pulse sequences to diagnose hepatic cirrhosis in a rat model was investigated. Hepatic T₁ was measured with and without MT off-resonance RF pulses in 17 treated and six control rats. The livers were evaluated histologically, and the hydroxyproline content quantitatively measured. We did not find a statistically significant linear correlation between the MR relaxation times and the degree of tissue injury. However, the MR measurements performed with MT were superior to those without differentiating the treated and control groups. Specifically, the T₁ times were 695 ±76 ms for the treated group, versus 748 ± 61 ms in the controls; P= 0.095. The T_1sat times were also lower in the treated group, with statistical significance: 367 ± 51 ms versus 421 ± 38 ms, P = 0.016. Finally, the change in the relaxation rates (the inverse of the relaxation times) with and without saturation were 1.31 ± 0.22 s^?1 (treated group) versus 1.05 ± 0.12 s^?1 (controls), which differed significantly, P= 0.001.     

Systematic variation of off-resonance prepulses for clinical magnetization transfer contrast imaging at 0.2, 1.5, and 3.0 tesla

OBJECTIVES: The aim of the presented study was to evaluate pulsed magnetization transfer contrast (MTC) effects using saturation pulses of variable off-resonance frequency and radio frequency (RF) amplitude for a variety of tissue types (white and gray matter, liver, kidney, spleen, muscle, and articular cartilage) in human subjects at field strengths of 0.2, 1.5, and 3.0 Tesla. MATERIALS AND METHODS: MTC imaging studies of the head, knee, and abdomen were performed using an adapted multiple MTC (mMTC) module in 3 healthy volunteers for all field strengths. This mMTC pulse module applies a variable Gaussian shaped magnetization transfer (MT) saturation pulse in a proton-density weighted RF-spoiled gradient echo sequence. It allows for both a flexible MT pulse design and performance of consecutive measurements with variation of amplitude and off-resonance frequency, whereas keeping other MT pulse parameters unchanged. Magnetization transfer signal ratio (MTR) maps were calculated on a pixel-by-pixel basis. Additional mMTC imaging measurements were performed using an agar-water phantom. For assessment of undesired direct saturation effects of the MT pulse on the water pool, numerical simulations based on Bloch's equations were performed and analyzed. RESULTS: The results indicate that MTR values for given MT pulses (pulse shape, off-resonance frequency and flip angle) are larger at higher magnetic field strengths. For white matter, gray matter, cartilage, and muscle, an increase of 10% to 30% was found at 3.0 T when compared with 1.5 T. Low magnetic field strength of 0.2 T led to MTR values of one third to half the values at 1.5 T. MTR values for abdominal tissues were partly lower at 3.0 T compared with 1.5 T, which might be related to reduced B1 field strengths at 3.0 T due to dielectric effects. CONCLUSIONS: The increased MT effect at a higher field strength can partly compensate the specific absorption rate related problems in MTC applications. It is shown that for flip angles of 700 degrees to 900 degrees and offset frequencies of 1000 Hz to 1500 Hz, high quality MTR maps could be obtained at an acceptable level of direct saturation for all field strengths. Furthermore, if the better signal-to-noise ratio at higher magnetic fields is taken into account, quality of MTR maps of the head and the knee at 3.0 T was clearly improved compared with lower fields under optimized and comparable conditions.     

Single-breath-hold venous or arterial flow-suppressed pulmonary vascular MR imaging with phased-array coils

A method for acquiring pulmonary vascular magnetic resonance (MR) images with either venous or arterial flow suppression is described. The proposed method only marginally increases the overall imaging time compared with conventional flow-suppression techniques. This enables an acquisition to be completed within a single breath hold with some selectivity as to flow direction. Instead of applying a spatially selective presaturation pulse before each radio-frequency (RF) excitation pulse, the flow presaturation pulse is applied once every 16-20 RF excitation pulses. To avoid image artifacts and to maintain a steady state, each presaturation pulse interval is followed by a normal imaging segment but with data acquisition turned off. Overall imaging time is increased by two TR intervals for each presaturation segment. For a 256 × 128 matrix acquisition, venous flow presaturation increases overall imaging time by approximately 14 TR intervals, while arterial flow suppression increases imaging time by 10 TR intervals.     

Whole-body magnetization transfer contrast imaging

PURPOSE: To demonstrate the feasibility of whole-body magnetization transfer (MT) contrast imaging. MATERIALS AND METHODS: Whole-body MT imaging was performed on eight healthy volunteers and five patients (mean age=40.5+/-17.8 years) with diagnoses of dermatomyositis (N=1), B-symptoms with suspicion of paraneoplastic disease (N=1), metastatic malignant melanoma (N=1), and multiple sclerosis (MS) (N=2). Measurements were carried out on a 1.5-Tesla whole-body MR scanner capable of parallel signal reception. A three-dimensional (3D) gradient-echo sequence (TR=17 msec, TE=4.8 msec, flip angle=10 degrees) was applied in combination with a Gaussian off-resonance MT preparation pulse acting at an off-resonance of 1.500 Hz with a 500 degrees effective flip angle. Whole-body images were constructed from five different body regions. RESULTS: In all subjects, whole-body MT contrast images were obtained within less than 20 minutes of measuring time. The images showed sufficient diagnostic image quality to assess the patients' pathologies. The MT ratios (MTRs, in percent units) for the volunteers were as follows: white matter (WM) 51.1+/-1.0, gray matter (GM) 42.2+/-1.3, skeletal muscle (mean value of four muscle groups) 50.3+/-2.1, liver 39.4+/-3.2, spleen 31.8+/-2.6, renal cortex 30.4+/-1.9, and renal medulla 25.6+/-1.3. The MTRs for the pathologies were as follows: skeletal muscle in dermatomyositis approximately 30, metastases in malignant melanoma 30.7-36.0, uterus myoma 49.3, and MS lesions 30-40. CONCLUSION: Our preliminary data indicate that MT contrast in whole-body MRI is feasible, and may be useful for rapid whole-body assessment of diseases that exhibit high contrast in MT imaging, such as MS and muscular disorders.     

Magnetization transfer contrast MR in lesions of the head and neck.

PURPOSETo compare lesion-to-background contrast with and without magnetization transfer (MT) in lesions of the head and neck.METHODSTwenty lesions (16 malignant, 4 benign) were evaluated in 17 patients (11 men, 6 women; mean age, 58 years; age range, 39-76 years). In 13 patients, MR imaging was performed at 0.1 T with continuous-wave, off-resonance MT; in 4 patients, MR imaging was performed at 1.5 T with on-resonance, binomial MT prepulses. Fifteen sequences were conducted before the administration of gadopentetate dimeglumine; 13 were conducted after the administration of that contrast material. The ratio of signal intensity with the MT pulses (Ms) to signal intensity without the MT pulses (Mo) was calculated, as were the lesion-to-background contrast and the contrast-to-noise ratios.RESULTSMs/Mo showed both wide variability and considerable overlap among different lesion types. Images from MT sequences showed better contrast than those from non-MT sequences in 23 of 28 lesions (12 of 15 before and 11 of 13 after the administration of contrast material). The mean contrast improvement percentages (+/- standard deviation) were 165.5% (+/- 58%) on unenhanced images and 186.6% (+/- 84.8%) on contrast-enhanced images. The mean improvements in contrast-to-noise ratios were 156% (+/- 60%) on unenhanced images and 171.6% (+/- 98.1%) on contrast-enhanced images.CONCLUSIONMT improved contrast between nodes or tumors showing an MT effect and background tissue (usually fat) not showing an MT effect. MT also improved contrast between contrast-enhanced neoplastic lesions and background tissue that showed an MT effect.     

Hepatic tumors: Magnetization transfer MR imaging with gadolinium enhancement

Thirty patients with 15 hepatocellular carcinomas, 10 metastases, four hemangiomas, and one cholangiocarcinoma underwent magnetic resonance imaging at 1.5 T with T1-weighted, T2- weighted spin-echo (SE) images, gradient-echo (GRE) magnetization transfer (MT) images, and gadolinium-enhanced T1-weighted SE and MT- GRE images. The MT effect and lesion-liver contrast-to-noise ratio (C/N) were calculated and visual assessment (qualitative analysis) performed for unenhanced and enhanced MT-GRE images and enhanced Tl-weighted SE images. The C/N values for hepatic adenocarcinomas (seven metastases and one cholangiocarcinoma) and hemangiomas were larger for enhanced MT-GRE images (adenocarcinoma, 8.4 ± 2.3 [P < 0.01); hemangioma, 24 ± 2.1 [P < 0.05]) than for enhanced GRE images (5.0 ± 1.9 and 18 ± 2.7, respectively). These enhancing tumors had the highest scores in the qualitative analysis. Enhanced MT-GRE images showed no advantage for depiction of hepatocellular carcinomas relative to the other images.     

Assessment of magnetization transfer effects in myocardial tissue using balanced steady‐state free precession (bSSFP) cine MRI

Magnetization transfer imaging (MTI) by means of MRI exploits the mobility of water molecules in tissue and offers an alternative contrast mechanism beyond the more commonly used mechanisms based on relaxation times. A cardiac MTI method was implemented on a commercially available 1.5 T MR imager. It is based on the acquisition of two sets of cardiac‐triggered cine balanced steady‐state free precession (bSSFP) images with different levels of RF power deposition. Reduction of RF power was achieved by lengthening the RF excitation pulses of a cine bSSFP sequence from 0.24 ms to 1.7 ms, while keeping the flip angle constant. Normal volunteers and patients with acute myocardial infarcts were imaged in short and long axis views. Normal myocardium showed an MT ratio (MTR) of 33.0 ± 3.3%. In acute myocardial infarct, MTR was reduced to 24.5 ± 9.2% (P < 0.04), most likely caused by an increase in water content due to edema. The method thus allows detection of acute myocardial infarct without the administration of contrast agents. Magn Reson Med, 2009. © 2009 Wiley‐Liss, Inc.     

NMR properties of human median nerve at 3 T: Proton density,T1, T2, and magnetization transfer

Purpose

To measure the proton density (PD), the T1 and T2 relaxation time, and magnetization transfer (MT) effects in human median nerve at 3 T and to compare them with the corresponding values in muscle.

Materials and Methods

Measurements of the T1 and T2 relaxation time were performed with an inversion recovery and a Carr‐Purcell‐Meiboom‐Gill (CPMG) imaging sequence, respectively. The MT ratio was measured by acquiring two sets of 3D spoiled gradient‐echo images, with and without a Gaussian saturation pulse.

Results

The median nerve T1 was 1410 ± 70 msec. The T2 decay consisted of two components, with average T2 values of 26 ± 2 msec and 96 ± 3 msec and normalized amplitudes of 78 ± 4% and 22 ± 4%, respectively. The dominant component is likely to reflect myelin water and connective tissue, and the less abundant component originates possibly from intra‐axonal water protons. The value of proton density of MRI‐visible protons in median nerve was 81 ± 3% that of muscle. The MT ratio in median nerve (40.3 ± 2.0%) was smaller than in muscle (45.4 ± 0.5%).

Conclusion

MRI‐relevant properties, such as PD, T1 and T2 relaxation time, and MT ratio were measured in human median nerve at 3 T and were in many respects similar to those of muscle. J. Magn. Reson. Imaging 2009;29:982–986. © 2009 Wiley‐Liss, Inc.     

Off-resonance metabolite magnetization transfer measurements on rat brain in situ.

Off-resonance metabolite magnetization transfer (MT) experiments were performed on rat brain in vivo and post mortem, with short (18 msec) and long (144 msec) echo-time 1H nuclear magnetic resonance (NMR) spectroscopy. In vivo and post mortem, the methyl protons of total creatine and all protons from glutamate/glutamine showed a strong MT effect on off-resonance saturation, as well as the methyl protons from lactate post mortem. Other resonances, like that of A-acetyl aspartate, showed a much smaller, but detectable, MT effect. The results obtained were confirmed by combining off-resonance saturation with two-dimensional correlation spectroscopy. Three water suppression techniques, i.e., presaturation, chemical shift-selective (CHESS), and selective water eliminated Fourier transform (WEFT) were evaluated for their ability to generate an MT effect, to assess their possible influence on metabolite quantification. Presaturation and selective WEFT led to alterations of the total creatine, lactate, and N-acetyl aspartate resonance intensities, while CHESS had no effect. Finally, it was shown that water protons play an important role in the generation of the observed metabolite MT effects.     

Spoiled gradient-echo as an arterial spin tagging technique for quick evaluation of local perfusion

PURPOSE: To introduce a simple gradient-echo arterial spin tagging (GREAST) technique available for most commercial magnetic resonance (MR) systems, for a quick evaluation of tissue perfusion. MATERIALS AND METHODS: The GREAST technique uses a combination of a short TR spoiled gradient-echo (SPGR) sequence with a selective presaturation radio frequency (RF) pulse that allows acquiring each tagged and control image within 10-20 seconds. The phantom and human studies were performed for evaluating the feasibility in measurement of local perfusion and the efficacy in alleviation of the asymmetric magnetization transfer (MT) and slice profile effects. RESULTS: Results show a good linear relationship between the signal attenuation caused by the presaturation pulse and flow rates in the phantom experiment and effective alleviation of the asymmetric MT and slice profile effects for various orientations of imaging slices. Human studies showed good perfusion results in brain imaging. Perfusion imaging on the liver and kidney were also conducted. The results could be significantly improved by effectively lessening motion-related artifacts. CONCLUSION: The GREAST technique is simple, easy to use, and applicable to examine local perfusion of the brain and other organs in the trunk.     

Suppression of effects of gradient imperfections on imaging with alternate ascending/descending directional navigation

Alternate ascending/descending directional navigation (ALADDIN) is a new imaging technique that provides interslice perfusion‐weighted and magnetization transfer (MT) asymmetry images. In this article, we investigated the effects of gradient imperfections on ALADDIN MT asymmetry (MTA) signals. Subtraction artifacts increasing with readout offsets were detectable in ALADDIN MTA images from an agarose phantom but not from a water phantom. Slice‐select offsets had no significant effect on the artifacts in MTA. The artifacts were suppressed by averaging signals over the readout gradient polarities independent of scan parameters. All these results suggested that the subtraction artifacts were induced by readout eddy currents. With suppression of the artifacts, ALADDIN signals in human brain and skeletal muscle varied less with scan conditions. Percent signal changes of MTA in human skeletal muscle (0.51 ± 0.11%, N = 3) were about 30% of those in white matter. The new averaging scheme will allow for more accurate MTA imaging with ALADDIN, especially at off‐center positions. Magn Reson Med, 2012. © 2012 Wiley Periodicals, Inc.     

Image quality using dynamic MR imaging of the temporomandibular joint with true-FISP sequence.

We evaluated the quality of dynamic magnetic resonance (MR) imaging of the temporomandibular joint (TMJ) in 20 normal volunteers with 40 TMJs. To confirm TMJ, we obtained static proton density weighted images (PDWI) before performing dynamic MR imaging with true-fast imaging in a steady-state precession (true-FISP) sequence. Four sequences of the first 10 volunteers were examined to determine the optimal sequence. The 4 sequences included the integrated parallel acquisition technique (iPAT) and/or fat saturation technique. The optimal sequence was then determined and performed in all 20 volunteers. The quality of imaging was evaluated, especially with respect to the conspicuity of the articular disk, mandibular condyle, articular eminence and lateral pterygoid muscle. One of 3 confidence levels was assigned for this evaluation. Neither iPAT nor fat saturation obtained the best quality imaging. Detection rates in the 20 volunteers were 83% for the articular disk, 95% for the mandibular condyle, 96% for the articular eminence and 7.5% for the lateral pterygoid muscle. We recommend dynamic MR imaging of the TMJ with the true-FISP sequence using neither iPAT nor fat saturation. Nevertheless, dynamic MR imaging was inferior to static imaging in detecting the articular disk and still requires improvement.     

Treatable chondral injuries in the knee: frequency of associated focal subchondral edema

OBJECTIVE: In the knee, chondral flaps and fractures are radiographically occult articular cartilage injuries that can mimic meniscal tears clinically; once correctly diagnosed, these injuries can be treated surgically. We investigated an associated MR imaging finding--focal subchondral bone edema--in a series of surgically proven lesions. MATERIALS AND METHODS: Two musculoskeletal radiologists retrospectively reviewed the MR studies of 18 knees with arthroscopically proven treatable cartilage infractions, noting articular surface defects and associated subchondral bone edema; subchondral edema was defined as focal regions of high signal intensity in the bone immediately underlying an articular surface defect on a T2-weighted or short inversion time inversion recovery (STIR) image. RESULTS: The first observer saw focal subchondral edema deep relative to a cartilage surface defect in 15 (83%) of the 18 cases; in two additional cases a surface defect was seen without underlying edema. The second observer identified 13 knees (72%) with surface defects and associated subchondral edema and three with chondral surface defects and no associated edema. Subchondral edema was seen more frequently on fat-suppressed images and on STIR images than non-fat-suppressed images. CONCLUSION: Focal subchondral edema is commonly visible on MR images of treatable, traumatic cartilage defects in the knee; this MR finding may prove to be an important clue to assist in the detection of these traumatic chondral lesions.     

Pulsed holmium:yttrium-aluminum-garnet (Ho:YAG) laser ablation of fibrocartilage and articular cartilage

A new, near-infrared, pulsed holmium laser (wavelength, 2.1 microns; pulse duration, 400 microseconds) was used to ablate bovine articular cartilage and meniscal fibrocartilage. Microscopic examination revealed zones of thermal damage extending 550 microns from ablation sites. Ablation rates were measured with a mass loss technique. Above threshold, mass removal rates were proportional to laser radiant exposure. Threshold radiant exposure for ablation was 50 J/cm2 for articular cartilage and 11 J/cm2 for meniscal fibrocartilage. Because the holmium laser can precisely and rapidly resect cartilaginous tissues with only moderate necrosis, function in a saline environment in direct contact with tissue, and be transmitted through conventional optical fibers, it has the potential to become a useful tool for the precise arthroscopic removal of intraarticular tissue.     

Fast fat suppression RF pulse train with insensitivity to B1 inhomogeneity for body imaging

In higher‐field magnetic resonance imaging scanners, a spectrally selective fat saturation radiofrequency (RF) pulse does not work well because B₁ inhomogeneity increases. An adiabatic 180° pulse is used to improve nonuniform fat suppression, but requires inversion recovery time. Therefore, a new RF pulse that achieves flip angles near 90° and is B₁ insensitive has been developed. The pulse consists of three sinc‐shaped RF pulses with different flip angles and with different time intervals between each RF pulse. Using the Bloch equations, we analyzed the optimal combination of flip angles. Experimental results demonstrated that M_z was maintained at less than 0.05 M₀ for a B₁ inhomogeneity of ±35%. The optimal net flip angles was adjusted to 95° by varying the time interval between RF pulses. The pulse duration was 77 ms, which is less than half of the 170‐ms inversion recovery time required for the adiabatic pulse. We demonstrated excellent fat suppression for body imaging. Magn Reson Med, 2012. © 2011 Wiley Periodicals, Inc.     

Spinal epidural synovial sarcoma: a case of homogeneous enhancing large paravertebral mass on MR imaging

We report the MR imaging findings in a 44-year-old man with a low-grade synovial sarcoma. There was a right-sided epidural and paravertebral mass, widening of the ipsilateral neural foramen at the L4-L5 level, and focal erosion of the right superior articular process of the L5 vertebra. The mass was relatively homogeneous, hyperintense to muscle and isointense to fat on T2-weighted images, and isointense to muscle on T1-weighted images, and it demonstrated moderate homogeneous enhancement.     

Meniscal tears: MR and arthrographic findings after arthroscopic repair

Magnetic resonance (MR) imaging was performed on 29 previously repaired menisci and one conservatively treated meniscus (total, 30 menisci). Intermediate- and T1-weighted MR sequences revealed persistent signal intensity extending to an articular surface (grade 3 signal intensity) in 27 of the 30 menisci. On T2-weighted images, seven of the 30 menisci were found to contain unequivocally higher signal intensity, defined by a full-thickness defect (grade 3 signal intensity involving two articular surfaces) increasing in signal intensity to a level equivalent to that of joint fluid. The MR imaging and arthrographic appearances of 23 of the 30 menisci were compared. Arthrographic examination revealed partial or complete healing in 13 menisci and tears in 10. The presence of grade 3 signal intensity on intermediate- and T1-weighted MR images did not reliably predict a tear seen at arthrography. Unequivocally higher signal intensity on T2-weighted images is a useful sign in the prediction of a persistent meniscal tear (sensitivity, 60%; specificity, 92%; P less than .02). Since presence of grade 3 signal intensity on intermediate- and T1-weighted images does not reliably predict a tear and unequivocal T2 increase in intensity has a sensitivity of only 60%, arthrography should be considered for assessment of the symptomatic, previously repaired meniscus.     

磁化传递对比成像在大鼠肝硬化的初步应用

目的 探讨磁化传递对比（MTC）成像在大鼠肝硬化的初步应用价值。材料与方法 二乙基亚硝胺诱导的单纯性肝硬化Wistar大鼠6只,5只正常大鼠作为对照组;全部大鼠均行附加磁化传递（MT）脉冲前后SE序列T1WI及FSE序列T2WI扫描。测量感兴趣区平均信号强度、背景噪声,计算信噪比及磁化传递率（MTR）。结果 6只大鼠病理证实均为单纯性肝硬化。电镜下见肝硬化细胞间隔增宽,胶原纤维明显增多。T1WI上正常肝实质附加MT脉冲后信噪比（SNR）低于无MT脉冲图像（P=0．002）;肝硬化附加MT后SNR明显低于无MT时（P=0．000）;肝硬化的MTR显著高于正常肝实质（P=0．000）。T2WI上正常肝实质附加MT后SNR低于无MT时（P=0．021）;肝硬化附加MT后SNR显著低于无MT时（P=0．000）;肝硬化的MTR较正常肝实质明显增高（P=0．002）。结论 MTC成像能间接反映肝硬化组织中胶原成分变化,为肝硬化的诊断和分期提供依据,可作为评价肝硬化的有效手段。     

MR imaging of the wrist: comparison between 1.5- and 3-T MR imaging--preliminary experience

Institutional review board approval and informed consent were obtained from 25 healthy volunteers and 15 consecutive patients with chronic wrist pain or suspected carpal mass, and 1.5- and 3-T magnetic resonance (MR) imaging of the wrist was prospectively performed with comparable sequence parameters and surface coils of the same geometric design. Imaging protocols at both field strengths included a T1-weighted spin-echo sequence, two intermediate-weighted fast SE sequences with different echo times and with and without fat saturation, and a three-dimensional fast field-echo sequence. The contrast-to-noise ratio (CNR) between muscle and bone and between bone and cartilage was calculated for both field strengths. The visibility of various anatomic structures, including the triangular fibrocartilage complex, carpal ligaments, nerves, and cartilage, was analyzed with a four-point grading scale. CNRs were significantly higher on 3-T MR images than on 1.5-T MR images (P < .001; analysis of variance) for all sequences. Visibility of the triangular fibrocartilage complex and intercarpal ligaments and cartilage was significantly better on 3-T MR images than on 1.5-T MR images (paired sign test).     

1.5-T MR of the knee: the correlations with arthroscopy

The authors report their experience of knee articular pathologies studied with MR imaging. Forty patients (36 with and 4 without articular trauma) were examined with both MR imaging and arthroscopy. Arthroscopic findings were correlated with MR results in order to evaluate the accuracy of the latter in detecting meniscal tears. MR images were obtained with a 1.5 T unit and a transmit-receiver surface coil. Spin-echo sequences and, occasionally, fast-scan sequences were used both in the coronal and in the sagittal planes. High resolution MR imaging clearly depicted knee anatomy. As far as meniscal lesions were concerned, the authors employed a grading system to classify intrameniscal signals, as proposed by Lotysch. The preliminary results are encouraging and suggest the use of MR imaging as the first imaging modality for use in knee pathology before arthroscopy is performed.