Effect of field strengths on magnetic resonance angiography: comparison of an ultrasmall superparamagnetic iron oxide blood-pool contrast agent and gadopentetate dimeglumine in rabbits at 1.5 and 3.0 tesla

OBJECTIVES: We sought to compare the intravascular enhancement of an ultrasmall superparamagnetic iron oxide (USPIO) blood-pool contrast agent to gadopentetate dimeglumine for contrast-enhanced magnetic resonance angiography (CE-MRA) at field strengths of 1.5 and 3.0 T in rabbits. MATERIALS AND METHODS: CE-MRA at 1.5 and 3.0 T was performed at several time points (50 seconds and 5, 10, 20, and 30 minutes) after the manual intravenous injection of 40 micromol Fe/kg body weight of an USPIO (SH U 555 C; Schering AG, Berlin, Germany) and 100 micromol/kg body weight gadopentetate dimeglumine (Magnevist; Schering AG, Berlin, Germany). MRA was performed with comparable acquisition parameters at both field strengths (Turbo-gradient sequence; 1.5 T: TR/TE/alpha: 5.5/1.7 milliseconds/40 degrees ; 3.0 T: TR/TE/alpha: 5.1/1.8 milliseconds/40 degrees ) on clinical imaging systems (both: Gyroscan Intera, Philips Medical Systems, Best, The Netherlands). At either field strength, 6 rabbits were studied with both contrast agents (n = 24 in total). Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated from signal intensity measurements in the abdominal aorta. RESULTS: Compared with 1.5 T, the SNR and CNR of gadopentetate dimeglumine significantly increased at 3.0 T by a factor of 2.2 and 2.3, respectively (P or= 0.05). At both field strength and either time point, CNR and SNR of SH U 555 C were significantly higher compared with gadopentetate dimeglumine at 3.0 T (P 相似文献   

3.0 Tesla magnetic resonance angiography of endovascular aortic stent grafts: phantom measurements in comparison with 1.5 Tesla

RATIONALE AND OBJECTIVES: This study evaluated different stent grafts by 3 T magnetic resonance angiography (MRA) with respect to lumen visibility, susceptibility-induced signal loss, and type of stent artifacts compared with 1.5 T MRA in a phantom model. METHODS: Six different stent-grafts (tube: n = 3, bifurcated: n = 3) were evaluated by 3 T and 1.5 T MRA using a tube phantom. MRA was performed using T1-weighted sequences at both systems with comparable parameters (3T: TR 5.4/TE 2.0/FA 30 degrees, 1.5 T: TR 6.2/TE 2.2/FA 30 degrees). A blind study of the image quality, including artifacts, was performed by 3 radiologists. Furthermore, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) values were calculated. Statistical analysis was performed with Student's t test (P < 0.05). RESULTS: One Elgiloy stent graft showed almost a complete intraluminal signal loss at 1.5 and 3 T. All other models could be evaluated by both systems by MRA, resulting in a favorable lumen visibility (score: 1) for prostheses made of nitinol. Scores for overall image quality and artifacts were the same for both MR systems. SNR and CNR values of the stented part of the vessel phantom increased from 320 +/- 33 to 618 +/- 40 and from 306 +/- 34 to 596 +/- 40 at 3 T when compared with 1.5 T, resulting in a significant signal gain of 93% at the higher field strength. CONCLUSIONS: 3 Tesla MRA of aortic stent grafts in a phantom model demonstrates an increase in SNR and CNR when compared with 1.5 T. However, the magnitude of imaging artifacts as well as coherent intraluminal signal loss within the stent does not increase equally in both MR systems.     

Time-resolved contrast-enhanced three-dimensional pulmonary MR-angiography: 1.0 M gadobutrol vs. 0.5 M gadopentetate dimeglumine

PURPOSE: To compare contrast characteristics and image quality of 1.0 M gadobutrol with 0.5 M Gd-DTPA for time-resolved three-dimensional pulmonary magnetic resonance angiography (MRA). MATERIALS AND METHODS: Thirty-one patients and five healthy volunteers were examined with a contrast-enhanced time-resolved pulmonary MRA protocol (fast low-angle shot [FLASH] three-dimensional, TR/TE = 2.2/1.0 msec, flip angle: 25 degrees, scan time per three-dimensional data set = 5.6 seconds). Patients were randomized to receive either 0.1 mmol/kg body weight (bw) or 0.2 mmol/kg bw gadobutrol, or 0.2 mmol/kg bw Gd-DTPA. Volunteers were examined three times, twice with 0.2 mmol/kg bw gadobutrol using two different flip angles and once with 0.2 mmol/kg bw Gd-DTPA. All contrast injections were performed at a rate of 5 mL/second. Image analysis included signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) measurements in lung arteries and veins, as well as a subjective analysis of image quality. RESULTS: In patients, significantly higher SNR and CNR were observed with Gd-DTPA compared to both doses of gadobutrol (SNR: 35-42 vs.17-25; CNR 33-39 vs. 16-23; P < or = 0.05). No relevant differences were observed between 0.1 mmol/kg bw and 0.2 mmol/kg bw gadobutrol. In volunteers, gadobutrol and Gd-DTPA achieved similar SNR and CNR. A significantly higher SNR and CNR was observed for gadobutrol-enhanced MRA with an increased flip angle of 40 degrees. Image quality was rated equal for both contrast agents. CONCLUSION: No relevant advantages of 1.0 M gadobutrol over 0.5 M Gd-DTPA were observed for time-resolved pulmonary MRA in this study. Potential explanations are T2/T2*-effects caused by the high intravascular concentration when using high injection rates.     

Low flip angle spin-echo MR imaging to obtain better Gd-DTPA enhanced imaging with ECG gating]

ECG-gated spin-echo imaging (ECG-SE) can reduce physiological motion artifacts. However, ECG-SE does not provide strong T1-weighted images because repetition time (TR) depends on heart rate (HR). We investigated the usefulness of low flip angle spin-echo imaging (LFSE) in obtaining more T1-dependent contrast with ECG gating. in computer simulation, the predicted image contrast and signal-to-noise ratio (SNR) obtained for each flip angle (0-180 degrees) and each TR (300 msec-1200 msec) were compared with those obtained by conventional T1-weighted spin-echo imaging (CSE: TR = 500 msec, TE = 20 msec). In clinical evaluation, tissue contrast [contrast index (CI): (SI of lesion-SI of muscle)2*100/SI of muscle] obtained by CSE and LFSE were compared in 17 patients. At a TR of 1,000 msec, T1-dependent contrast increased with decreasing flip angle and that at 38 degrees was identical to that with T1-weighted spin-echo. SNR increased with the flip angle until 100 degrees, and that at 53 degrees was identical to that with T1-weighted spin-echo. CI on LFSE (74.0 +/- 52.0) was significantly higher than CI on CSE (40.9 +/- 35.9). ECG-gated LFSE imaging provides better T1-dependent contrast than conventional ECG-SE. This method was especially useful for Gd-DTPA enhanced MR imaging.     

Time-reduced T2-weighted celebral MRI with optimized flip angles

This study was set up to see whether lowering the flip angle in proton density- and T2-weighted double-spin echo sequences allows for shortening of repetition time (TR) and imaging time without significant change of image quality. Ten patients with celebral white matter lesions were investigated with an 1.5 T MR scanner using a conventional long- TR double-spin echo sequence (TR = 2500 ms, TE = 15 and 70 ms) and reduced-TR double-spin echo sequences (TR = 1900 ms, TE = 15 and 70 ms) at flip angles of 90°, 80°, 70°, 60°, and 50°. Lowering the flip angle resulted in less T1-contrast and a relative increase of T2-contrast. At a flip angle of 70°, contrast-to noise ratios (NNRs) between lesions and brain, as well as image artifacts of the reduced-TR sequence (CNR: 22.4) were similar to the conventional long-TR sequence (CNR:21.1), while imaging time was shortened by about 25%. Offprint requests to: Peter Schubeus     

Magnetic resonance imaging of lung tissue: influence of body positioning, breathing and oxygen inhalation on signal decay using multi-echo gradient-echo sequences

PURPOSE: To assess susceptibility related signal decay in lung tissue and to measure the influence of body positioning, together with inspiration and expiration, as well as oxygen inhalation. T2* maps and line shape maps of lung parenchyma were derived from datasets acquired at 0.2 T and compared with findings at 1.5 T. The line shape maps allow for a visualization of the intravoxel frequency distribution of lung parenchyma. MATERIALS AND METHODS: A multiecho spoiled gradient-echo sequence with 16 echoes was implemented both on a 0.2 T [repetition time (TR) = 100 milliseconds, echo time (TE)1 = 2.15 milliseconds, DeltaTE = 2.94 milliseconds, flip angle 30 degrees] and on a 1.5 T magnetic resonance scanner (TR = 100 milliseconds, TE1 = 1.25 milliseconds, DeltaTE = 1.65 milliseconds, flip angle 30 degrees). Sagittal datasets were recorded in 8 healthy volunteers at 0.2 T in supine position under maximal expiration and inspiration and during oxygen breathing. Additional measurements were performed after 20 minutes inside the scanner in supine position and after prone repositioning. In 2 volunteers, further datasets were acquired at 1.5 T. Color-encoded T2* maps and full-width-at-half-maximum (FWHM) maps of the frequency distribution were computed on a pixel-by-pixel basis. T2* maps were generated by mono-exponential fitting and, additionally, with an extended nonexponential fitting approach. The FWHM maps were calculated with a model-free approach using a discrete Fast Fourier Transformation. RESULTS: A notably slower T2* decay was found at 0.2 T (T2*: 5.9-11.8 milliseconds) when compared with 1.5 T (T2*: 1.0-1.4 milliseconds), allowing for the measurement of up to 6 to 8 gradient echoes above the noise level. The T2* maps and the FWHM maps computed from the datasets acquired at 0.2 T allowed regional comparison of the derived parameters. If volunteers were positioned in supine position, expiration resulted in a T2* of 10.9 +/- 1.0 milliseconds and a FWHM of 47.1 +/- 4.0 Hz in the dorsal lung. Significant changes (P < 0.05) were found, eg, in the ventral lung in expiration (T2*: 7.5 +/- 0.8, FWHM: 76.7 +/- 11.2) versus dorsal lung in expiration, in the dorsal lung in inspiration (T2*: 8.4 +/- 1.0, FWHM: 67.8 +/- 12.5) versus dorsal lung in expiration, in the dorsal lung during oxygen breathing (T2*: 8.7 +/- 1.1, FWHM: 52.2 +/- 5.2) versus dorsal lung while breathing room air, and in the dorsal lung in prone position (T2*: 8.5 +/- 0.6, FWHM: 67.0 +/- 9.2) versus dorsal lung in supine position. CONCLUSION: The proposed method allows for the computation of color-encoded T2* maps and FWHM maps of lung parenchyma in good image quality using datasets acquired at 0.2 T. The technique is robust and sensitive to physiological changes of lung magnetic resonance properties, eg, due to the type of body positioning or oxygen breathing.     

MR imaging of the liver with Gd-BOPTA: quantitative analysis of T1-weighted images at two different doses.

This study evaluates the efficacy of gadobentate-dimeglumine (Gd-BOPTA) for enhancement of liver signal-to-noise ratio (SNR) and lesion-liver contrast-to-noise ratio (CNR) on T1-weighted spin-echo (SE) and gradient-recalled-echo (GRE) images at two different doses. Fifty patients with known or suspected liver lesions were examined at 1.5 T. T1-weighted SE (TR/TE 300/12 msec) and GRE images (TR/TE80/4.2 msec/flip angle 80 degrees) were obtained before and at 40-80 minutes and 90-120 minutes after administration of 0.05 or 0.1 mmol/kg Gd-BOPTA. Quantitative measurements of tissue signal intensity were performed at each dose. Liver showed significant enhancement after Gd-BOPTA on T1-weighted SE and GRE images (0.05 mmol: P < 0.05; 0.1 mmol: P < 0.001). The dose of 0.1 mmol/kg provided higher liver SNR than 0.05 mmol/kg. Mean liver SNR was higher on GRE than SE images (P < 0.0001). Lesion-liver CNR significantly increased on GRE images after 0.1 mmol (P < 0.05). There was a trend toward superiority of 0.1 mmol over 0.05 mmol/kg. GRE images were superior to SE images for pre- and post Gd-BOPTA lesion-liver CNR (P < 0.05). Our study suggests that Gd-BOPTA provides prolonged enhancement of liver SNR and CNR, that a dose of 0.1 mmol/Kg appears to be superior than 0.05 mmol/Kg, and that GRE techniques should be used in preference over SE techniques.     

Myocardial first pass perfusion imaging with gadobutrol: impact of parallel imaging algorithms on image quality and signal behavior

OBJECTIVE: To implement parallel imaging algorithms in fast gradient recalled echo sequences for myocardial perfusion imaging and evaluate image quality, signal-to-noise ratio (SNR), contrast-enhancement ratio (CER), and semiquantitative perfusion parameters. MATERIALS AND METHODS: In 20 volunteers, myocardial perfusion imaging with gadobutrol was performed at rest using an accelerated TurboFLASH sequence (TR 2.3 milliseconds, TE 0.93 milliseconds, flip angle [FA] 15 degrees) with GRAPPA, R=2. A nonaccelerated TurboFLASH sequence with similar scan parameters served as standard of reference. Artifacts were assessed qualitatively. SNR, CER, and CNR were calculated and semiquantitative perfusion parameters were determined from fitted SI-time curves. RESULTS: Phantom measurements yielded significant higher SNR for nonaccelerated images (P<0.001). CER was equal; differences in CNR were statistically nonsignificant. The evaluation of semiquantitative perfusion parameters yielded significantly higher peak signal intensities in nonaccelerated images (P<0.001). Differences in maximum upslope were statistically nonsignificant. A qualitative examination of all images for artifacts by 2 board-certified radiologists yielded a significant reduction in dark rim artifacts with GRAPPA, R=2 (P<0.001). CONCLUSIONS: The application of GRAPPA with an acceleration factor of R=2 leads to a significant reduction of dark rim artifacts in fast gradient recalled echo sequences.     

Optimization of sequence parameters in fast MR imaging of the brain with FLASH

Owing to the intrinsically complex behavior of the signal intensity of fast gradient-refocusing MR sequences, agreement as to the clinically most useful sequence parameters has not yet been reached. This study evaluates the FLASH (fast low-angle shot) sequence for gray-white matter differentiation on normal volunteers at 1.5 T. The FLASH gradient-echo sequence is essentially T1-dependent. For very fast imaging and T1 weighting, the following parameters yield the best results: a flip angle of 30-50 degrees with TR = 20 and TE = 10. To replace T1-weighted SE by the faster FLASH sequence, the best results are achieved by a flip angle of 70-120 degrees with TR = 150-300 and TE = 10 (or shorter, if possible). The most valuable proton-density aspect is achieved by a flip angle of 30 degrees with TR = 300 and TE = 16.     

Parallel acquisition techniques in cardiac cine magnetic resonance imaging using TrueFISP sequences: comparison of image quality and artifacts

PURPOSE: To compare image quality, artifacts, and signal-to-noise ratio (SNR) in cardiac cine TrueFISP magnetic resonance imaging (MRI) with and without parallel acquisition techniques (PAT). MATERIALS AND METHODS: MRI was performed in 16 subjects with a TrueFISP sequence (1.5 T; Magnetom Sonata, Siemens): TR, 3.0 msec; TE, 1.5 msec; flip angle (FA), 60 degrees. Three axes were scanned without PAT (no PAT) and using the generalized autocalibrating partially parallel acquisition (GRAPPA) and modified sensitivity encoding (mSENSE) reconstruction algorithms with an autocalibration mode to reduce scan time. A conventional spine array and a body flex array were used. Artifacts, image noise, and overall image quality were classified on a 4-point scale by an observer blinded to the implemented technique; for quantitative comparison, SNR was measured. RESULTS: With a PAT factor of two, acquisition time could be reduced by 39%. No PAT did not show artifacts, and GRAPPA revealed fewer artifacts than mSENSE. PAT provided inferior-quality scores concerning image noise and overall image quality. In quantitative measurements, GRAPPA and mSENSE (20.1 +/- 6.2 and 15.6 +/- 6.2, respectively) yielded lower SNR than no PAT (30.6 +/- 20.1; P < 0.05) and P < 0.001). CONCLUSION: Time savings in PAT are accompanied by artifacts and an increase in image noise. The GRAPPA algorithm was superior to mSENSE concerning image quality, noise, and SNR.     

Renal magnetic resonance angiography at 3.0 Tesla using a 32-element phased-array coil system and parallel imaging in 2 directions

PURPOSE: The aim of the present study was to assess the feasibility of renal magnetic resonance angiography at 3.0 T using a phased-array coil system with 32-coil elements. Specifically, high parallel imaging factors were used for an increased spatial resolution and anatomic coverage of the whole abdomen. MATERIALS AND METHODS: Signal-to-noise values and the g-factor distribution of the 32 element coil were examined in phantom studies for the magnetic resonance angiography (MRA) sequence. Eleven volunteers (6 men, median age of 30.0 years) were examined on a 3.0-T MR scanner (Magnetom Trio, Siemens Medical Solutions, Malvern, PA) using a 32-element phased-array coil (prototype from In vivo Corp.). Contrast-enhanced 3D-MRA (TR 2.95 milliseconds, TE 1.12 milliseconds, flip angle 25-30 degrees , bandwidth 650 Hz/pixel) was acquired with integrated generalized autocalibrating partially parallel acquisition (GRAPPA), in both phase- and slice-encoding direction. Images were assessed by 2 independent observers with regard to image quality, noise and presence of artifacts. RESULTS: Signal-to-noise levels of 22.2 +/- 22.0 and 57.9 +/- 49.0 were measured with (GRAPPAx6) and without parallel-imaging, respectively. The mean g-factor of the 32-element coil for GRAPPA with an acceleration of 3 and 2 in the phase-encoding and slice-encoding direction, respectively, was 1.61. High image quality was found in 9 of 11 volunteers (2.6 +/- 0.8) with good overall interobserver agreement (k = 0.87). Relatively low image quality with higher noise levels were encountered in 2 volunteers. CONCLUSION: MRA at 3.0 T using a 32-element phased-array coil is feasible in healthy volunteers. High diagnostic image quality and extended anatomic coverage could be achieved with application of high parallel imaging factors.     

Spontaneous acute dissection of the internal carotid artery: high-resolution magnetic resonance imaging at 3.0 tesla with a dedicated surface coil

PURPOSE: Magnetic Resonance Imaging (MRI) has become the method of choice in the evaluation of patients with suspected cervical artery dissection (CAD). However, reliable identification of acute CAD might be impaired by the limited spatial resolution of standard 1.5 T MRI. In this preliminary study, we implemented a multicontrast high-resolution noninvasive vessel wall imaging approach at 3.0 T in patients with spontaneous CAD. METHODS AND MATERIALS: Ten patients with CAD of the internal carotid artery (ICA) were included in the study. 3.0 T MRI (Gyroscan Intera, Philips) was acquired using a dedicated phased-array coil. MRI-protocol consisted of: (1) bright blood 3D inflow MRA (TR/TE/FA = 25 milliseconds/3.1 millisecond/16 degrees , 120 slices, reconstructed voxel size 0.3 x 0.3 x 0.8 mm); (2) black blood cardiac-gated water-selective T1w 3D spoiled GE (TR/TE/FA = 31 milliseconds/7.7 milliseconds/15 degrees , 36 slices, 0.3 x 0.3 x 1.0 mm); and (3) black blood cardiac triggered fat suppressed T2w TSE (TR/TE/ETL = 3 heart beats/44 milliseconds/7, 18 slices, 0.3 x 0.3 x 2 mm). Three observers in consensus performed image analysis. Special attention was paid to the integrity of the luminal and adventitial vessel boundary and the presence of a communicating intimal tear or flap. RESULTS: 3.0 T MRI provided excellent delineation of vessel lumen and vessel wall as a result of the nearly complete suppression of arterial blood signal. An intramural hematoma could be identified in all patients, confined between the luminal and adventitial vessel boundary. In no patient a communicating intimal tear could be identified. Clear distinction between intramural hematoma and thrombus was possible. CONCLUSION: High-resolution vessel wall imaging in patients with acute CAD is feasible. The increased signal-to-noise ratio at 3.0 T can be invested to obtain a higher spatial resolution, permitting depiction of intimal and adventitial vessel wall boundary and the intramural hematoma in the diseased vessel segment. The morphologic information that is gained is helpful in the understanding of the underlying pathomechanismen of CAD.     

Detection of asymptomatic cerebral microbleeds: a comparative study at 1.5 and 3.0 T

RATIONALE AND OBJECTIVES: The magnitude of iron-induced susceptibility changes in gradient echo T2*-weighted magnet resonance imaging (T2* MRI) increases with the field strength and should increase the sensitivity for detection of cerebral microbleeds (CMBs) at 3.0 T. To test these hypotheses, we prospectively examined individuals with documented CMBs at 1.5 and 3.0 T. MATERIALS AND METHODS: Five hundred fifty elderly individuals, who participated in an interdisciplinary study of healthy aging, were examined at 3.0 T using T2* MRI sequences (repetition time [TR]/echo time [TE]/flip angle [FA] = 573 ms/16 ms/18 degrees ). Individuals positive for CMBs were asked to undergo an additional examination at 1.5 T (TR/TE/FA = 663 ms/23 ms/18 degrees ). Images were analyzed independently by two observers. CMBs were counted throughout the brain and were qualitatively analyzed comparing the degree of visible hypointensity on a 5-point scale from 1 (complete signal loss) to 5 (no detection) for both field strengths. Contrast-to-noise ratio of CMBs to surrounding brain tissue was calculated. RESULTS: At 3.0 T, CMBs were detected in 45 of 550 individuals; 25 agreed to an additional examination at 1.5 T. In this group (n = 25), a total of 53 CMBs were detected at 3.0 T, compared to 41 CMBs at 1.5 T. The mean contrast-to-noise ratio of CMBs was significantly increased at 3.0 T compared to 1.5 T (27.4 +/- 8.2 vs. 17.4 +/- 8.0; p < .001). On qualitative analysis, visibility of CMBs was ranked significantly higher at 3.0 T (1.3 +/- 0.4 vs. 2.9 +/- 1.1; p < .001). CONCLUSION: Evidence of past microbleeds may even be found in neurologically normal elderly individuals by MRI. Detection rate and visibility of CMBs benefit from the higher field strength, resulting in a significantly improved depiction of iron-containing brain structures (CMBs) at 3.0 T with potential clinical relevance.     

Three-dimensional breathhold SSFP coronary MRA: a comparison between 1.5T and 3.0T

PURPOSE: To assess the feasibility of three-dimensional breathhold coronary magnetic resonance angiography (MRA) at 3.0T using the steady-state free precession (SSFP) sequence, and quantify the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) gains of coronary MRA from 1.5T to 3.0T using whole-body and phased-array cardiac coils as the signal receiver. MATERIALS AND METHODS: Eight healthy volunteers were scanned on 1.5T and 3.0T whole-body systems using the SSFP sequence. Numerical simulations were performed for the SSFP sequence to optimize the flip angle and predict signal enhancement from 1.5T to 3.0T. Coronary artery images were acquired with the whole-body coil in transmit-receive mode or transmit-only with phased-array cardiac coil receivers. RESULTS: In vivo studies of the same volunteer group at both field strengths showed increases of 87% in SNR and 83% in CNR from 1.5T to 3.0T using a whole-body coil as the signal receiver. The corresponding increases using phased-array receivers were 53% in SNR and 92% in CNR. However, image quality at 3.0T was more variable than 1.5T, with increased susceptibility artifacts and local brightening as the result of increased B(0) and B(1) inhomogeneities. CONCLUSION: Coronary MRA at 3.0T using a three-dimensional breathhold SSFP sequence is feasible. Improved SNR at 3.0T warrants the use of coronary MRA with faster acquisition and/or improved spatial resolution. Further investigations are required to improve the consistency of image quality and signal uniformity at 3.0T.     

High-resolution magnetic resonance imaging (MRI) at 3.0 Tesla in the short-term follow-up of patients with proven cervical artery dissection

PURPOSE: For the imaging evaluation of patients with suspected cervical artery dissection (CAD) in the last decade, magnetic resonance imaging (MRI) has become the first line imaging modality. However, CAD is a highly dynamic process with rapid changes over time. Aim of this study was to assess the short-term morphologic changes in patients with proven CAD by MRI within 2 weeks after the initial diagnosis using a multicontrast high-resolution noninvasive vessel wall imaging approach at 3.0 T. MATERIALS AND METHODS: Eighty-two patients with clinically suspected CAD were examined using a 3.0 T system (Gyroscan Intera, Philips). Imaging protocol consisted of 3-dimensional inflow MRA (repetition time [TR]/echo time [TE]/flip angle [FA] = 25 milliseconds/3.1 milliseconds/16 degrees, reconstructed voxel size 0.3 x 0.3 x 0.8 mm), black blood T1w 3-dimensional spoiled gradient echo (TR/TE/FA = 31 milliseconds/7.7 milliseconds/15 degrees, 0.3 x 0.3 x 1.0 mm), and fat suppressed T2w turbo spin echo (TSE) (TR/TE/echo train length = 3 heart beats/44 milliseconds/7, 0.3 x 0.3 x 2 mm). Three observers in consensus performed image analysis. Images were assessed with regard to presence and size of intramural hematoma, degree of stenosis, presence of intraluminal thrombus, development of pseudoaneurysm, and incidence of additional dissections. In 29 patients (35%) a dissection had initially been proven by direct visualization of an intramural hematoma. Twenty-one patients (72%; 7 male, 14 female; mean age 41.5 years) were available for follow-up studies leading to a total of 24 diseased cervical arteries being reevaluated 2 weeks later for prospective follow-up. RESULTS: Mean interval between initial study and follow-up was 14.2 days (range 7-30 days). Eighteen patients had presented with an acute CAD in 1 artery, 3 patients with an acute CAD in 2 arteries. At follow-up, degree of stenosis had increased in 2 arteries, remained unchanged in 13, and decreased in 5 arteries. Four initially occluded arteries were recanalized at follow-up. In 3 arteries a pseudoaneurysm had been visible in the initial study and remained unchanged at follow-up; in 1 artery a new pseudoaneurysm was observed. In 3 arteries, new dissections were identified during follow-up. CONCLUSION: High-resolution MRI of acute CAD at 3.0 T permits a refined cross-sectional and longitudinal analysis of the morphologic features of CAD. The increased signal-to-noise ratio at 3.0 T allows for a high spatial resolution permitting detailed analysis of the diseased vessel segment. An unequivocal distinction between intramural hematoma and thrombus was possible. Information could be gained with regard to recanalization, degree of stenosis, formation of pseudoaneurysm, and appearance of new dissections making short-term follow-up in pts with acute CAD recommendable. Further studies are needed to assess the relationship between short-term results and definite outcome.     

High-resolution MRI of the triangular fibrocartilage complex (TFCC) at 3T: comparison of surface coil and volume coil

PURPOSE: To evaluate high-resolution MRI of the triangular fibrocartilage complex (TFCC) at 3T using a surface coil (SC) or volume coil (VC). MATERIALS AND METHODS: MRI was obtained from nine volunteers in the supine position with a 3-inch SC and in prone position with a transmit-receiver wrist VC at 3 T. Coronal two-dimensional-gradient echo (2D-GRE) images (TR/TE/FA = 500 msec/15 msec/40 degrees , 1 mm slice-thickness, 60 mm field of view [FOV], 192 x 256 matrix) and coronal 3D-GRE images (TR/TE/FA = 33 msec/15 msec/10 degrees , 0.8 mm slice-thickness, 80 mm FOV, 256 x 256 matrix) were used. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the TFCC and surrounding structures were measured. For qualitative measurement, visualization of TFCC and intercarpal ligaments was graded. RESULTS: SNR of TFCC, cartilage, and bone marrow on 2D-GRE with SC/VC was as follows: 5.3/5.3 (TFCC), 16.5/14.4 (cartilage), and 3.61/3.96 (bone marrow). 3D-GRE showed similar SNR. Cartilage-TFCC/cartilage-bone marrow CNR were 11.1/12.8 (SC-2D-GRE), 8.8/10.5 (VC-2D-GRE), 14.1/15.5 (SC-3D-GRE), and 11.9/15.0 (VC-3D-GRE). Quantitative values were not significantly different between SC and VC. Visualization of TFCC and intercarpal ligament with SC was superior to that with VC. All structures show higher scores with 3D-GRE imaging compared to 2D-GRE imaging. CONCLUSION: SC may provide superior qualitative and quantitative results and can be an alternative in case of difficulty in prone position at 3T.     

High-resolution MRI of the labyrinth: optimization of scan parameters with 3D-FSE

The aim of our study was to optimize the parameters of high-resolution MRI of the labyrinth with a 3D-FSE sequence. We investigated TR, TE, Matrix, FOV, and coil selection in terms of CNR (contrast-to-noise ratio) and SNR (signal-to-noise ratio) by comparing axial images and/or three-dimensional images. The optimal 3D-FSE sequence parameters were as follows: 1.5 Tesla MR unit (Signa LX, GE Medical Systems), 3D-FSE sequence, dual 3-inch surface coil, acquisition time=12.08 min., TR=5000 msec, TE=300 msec, 3 NEX, FOV=12 cm, matrix=256 x 256, slice thickness=0.5 mm/0.0 sp, echo train=64, bandwidth=+/-31.5 kHz. High-resolution MRI of the labyrinth using the optimized 3D-FSE sequence parameters permits visualization of important anatomic details (such as scala tympani and scala vestibuli), making it possible to determine inner ear anomalies and the patency of cochlear turns. To obtain excellent heavily T2-weighted axial and three-dimensional images in the labyrinth, high CNR, SNR, and spatial resolution are significant factors at the present time. Furthermore, it is important not only to optimize the scan parameters of 3D-FSE but also to select an appropriate coil for high-resolution MRI of the labyrinth.     

Steady-state free precession sequences in myocardial first-pass perfusion MR imaging: comparison with TurboFLASH imaging

The aim of this study was to compare the image quality of a saturation-recovery gradient-recalled echo (GRE; TurboFLASH) and a saturation-recovery SSFP (SR-TrueFISP) sequence for myocardial first-pass perfusion MRI. Eight patients with chronic myocardial infarction and 8 volunteers were examined with a TurboFLASH (TR 2.1 ms, TE 1 ms, FA 8°) and a SR-TrueFISP sequence (TR 2.1 ms, TE 0.9 ms, FA, 50°) on a 1.5 T scanner. During injection of 0.05 mmol/kg BW Gd-DTPA at 4 ml/s, three short axis slices (8 mm) of the left ventricle (LV) were simultaneously scanned during breath-hold. Maximum signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR) between infarcted and normal myocardium, and percentage signal intensity change (PSIC) were measured within the LV lumen and in four regions of the LV myocardium for the three slices separately. For the LV lumen, SR-TrueFISP was superior in SNR and PSIC (factor 3.2 and 1.6, respectively). Mean maximum SNR, PSIC, and CNR during peak enhancement in the LV myocardium were higher for SR-TrueFISP compared with TurboFLASH (factor 2.4, 1.25, and 1.24, respectively). The SNR was higher in the septal portion of the ventricle than in anterior/posterior and lateral regions. The SR-TrueFISP provides higher SNR and improves image quality compared with TurboFLASH in first-pass myocardial perfusion MRI.     

Free-breathing renal magnetic resonance angiography with steady-state free-precession and slab-selective spin inversion combined with radial k-space sampling and water-selective excitation.

The impact of radial k-space sampling and water-selective excitation on a novel navigator-gated cardiac-triggered slab-selective inversion prepared 3D steady-state free-precession (SSFP) renal MR angiography (MRA) sequence was investigated. Renal MRA was performed on a 1.5-T MR system using three inversion prepared SSFP approaches: Cartesian (TR/TE: 5.7/2.8 ms, FA: 85 degrees), radial (TR/TE: 5.5/2.7 ms, FA: 85 degrees) SSFP, and radial SSFP combined with water-selective excitation (TR/TE: 9.9/4.9 ms, FA: 85 degrees). Radial data acquisition lead to significantly reduced motion artifacts (P < 0.05). SNR and CNR were best using Cartesian SSFP (P < 0.05). Vessel sharpness and vessel length were comparable in all sequences. The addition of a water-selective excitation could not improve image quality. In conclusion, radial k-space sampling reduces motion artifacts significantly in slab-selective inversion prepared renal MRA, while SNR and CNR are decreased. The addition of water-selective excitation could not improve the lower CNR in radial scanning.     

3.0 Tesla high spatial resolution contrast-enhanced magnetic resonance angiography (CE-MRA) of the pulmonary circulation: initial experience with a 32-channel phased array coil using a high relaxivity contrast agent

PURPOSE: To evaluate the technical feasibility of high spatial resolution contrast-enhanced magnetic resonance angiography (CE-MRA) with highly accelerated parallel acquisition at 3.0 T using a 32-channel phased array coil, and a high relaxivity contrast agent. MATERIALS AND METHODS: Ten adult healthy volunteers (5 men, 5 women, aged 21-66 years) underwent high spatial resolution CE-MRA of the pulmonary circulation. Imaging was performed at 3 T using a 32-channel phase array coil. After intravenous injection of 1 mL of gadobenate dimeglumine (Gd-BOPTA) at 1.5 mL/s, a timing bolus was used to measure the transit time from the arm vein to the main pulmonary artery. Subsequently following intravenous injection of 0.1 mmol/kg of Gd-BOPTA at the same rate, isotropic high spatial resolution data sets (1 x 1 x 1 mm3) CE-MRA of the entire pulmonary circulation were acquired using a fast gradient-recalled echo sequence (TR/TE 3/1.2 milliseconds, FA 18 degrees) and highly accelerated parallel acquisition (GRAPPA x 6) during a 20-second breath hold. The presence of artifact, noise, and image quality of the pulmonary arterial segments were evaluated independently by 2 radiologists. Phantom measurements were performed to assess the signal-to-noise ratio (SNR). Statistical analysis of data was performed by using Wilcoxon rank sum test and 2-sample Student t test. The interobserver variability was tested by kappa coefficient. RESULTS: All studies were of diagnostic quality as determined by both observers. The pulmonary arteries were routinely identified up to fifth-order branches, with definition in the diagnostic range and excellent interobserver agreement (kappa = 0.84, 95% confidence interval 0.77-0.90). Phantom measurements showed significantly lower SNR (P < 0.01) using GRAPPA (17.3 +/- 18.8) compared with measurements without parallel acquisition (58 +/- 49.4). CONCLUSION: The described 3 T CE-MRA protocol in addition to high T1 relaxivity of Gd-BOPTA provides sufficient SNR to support highly accelerated parallel acquisition (GRAPPA x 6), resulting in acquisition of isotopic (1 x 1 x 1 mm3) voxels over the entire pulmonary circulation in 20 seconds.