Coronary sinus flow measurement by means of velocity-encoded cine MR imaging: validation by using flow probes in dogs.

PURPOSE: To validate coronary sinus flow measurements for quantification of global left ventricular (LV) perfusion by means of velocity-encoded cine (VEC) magnetic resonance (MR) imaging and flow probes. MATERIALS AND METHODS: Measurements of coronary sinus flow were performed in seven dogs by using VEC MR imaging at baseline, single coronary arterial stenosis, dipyridamole stress, and reactive hyperemia. These measurements were compared with flow probe measurements of coronary blood flow (CBF) in the left anterior descending coronary (LAD) and circumflex (CFX) arteries (CBF(LAD+CFX)) and coronary sinus. LV blood perfusion was calculated in milliliters per minute per gram from coronary sinus flow, and LV mass was obtained by using VEC and cine MR imaging. LV mass was validated at autopsy. RESULTS: CBF(LAD+CFX) and coronary sinus flow at VEC MR imaging showed close correlation (r = 0.98, P: <.001). The difference between CBF(LAD+CFX) and MR coronary sinus flow was 3.1 mL/min +/- 8.5 (SD). LV mass at cine MR imaging was not significantly different from that at autopsy (73.2 g +/- 12.8 vs 69. 4 g +/- 12.8). At baseline, myocardial perfusion was 0.40 mL/min/g +/- 0.09 at VEC MR imaging, and CBF(LAD+CFX) was 0.44 mL/min/g +/- 0. 08 (not significant). Reactive hyperemia resulted in 2.7- and 2. 3-fold increases in coronary sinus flow at VEC MR imaging and flow probe CBF(LAD+CFX), respectively. CONCLUSION: VEC MR imaging has the potential to measure coronary sinus flow during different physiologic conditions and can serve as a noninvasive modality to quantify global LV perfusion in patients.

Single-shot fluid attenuated inversion recovery (FLAIR) magnetic resonance imaging of the bladder.

The purpose of this study was to reduce artifacts and increase imaging speed in fluid-attenuated inversion recovery (FLAIR) imaging of the urinary bladder. An existing half-Fourier, single-shot fast spin-echo imaging sequence was modified to allow presaturation with a non-slice-selective inversion recovery pulse (NSI SSFLAIR). Four independent, blinded readers rated severity of bladder artifacts and image quality in six normal male volunteers. NSI SSFLAIR effectively suppressed bladder urine signal in all six cases using a TI of 2900-3100 msec. Although NSI SSFLAIR images were noisier than standard fast spin-echo images, imaging time was only 10 seconds per slice location. Furthermore, perceived image sharpness was only minimally reduced, and conspicuity of the seminal vesicles and peripheral zone of the prostate were nearly equivalent. NSI SSFLAIR provides rapid T2-weighted imaging of the bladder wall and perivesicular tissues with nearly complete negation of signal from urine in the bladder.

Fast 3D cardiac cine MR imaging.

We describe a technique for three-dimensional cine MR imaging. By using short repetition times (TR) and interleaved slice encoding, volumetric cine data can be acquired throughout the cardiac cycle with a temporal resolution of approximately 80 msec. A T1-shortening agent is used to produce contrast between blood and myocardium. A comparison between the acquisition times of this and several other two-dimensional techniques is presented.

Quantification of coronary artery volume flow rate using fast velocity-encoded cine MR imaging.

OBJECTIVE: Breath-hold velocity-encoded cine (VENC) MR imaging has been proposed as a method for measuring coronary blood flow. However, most studies have measured velocity rather than volume flow rate in the coronary arteries. The purpose of this study was to measure volume flow rate in the coronary artery of dogs using high-speed gradients and to compare MR flow measurements with those obtained with a sonographic flowmeter. MATERIALS AND METHODS: Fast VENC MR images were obtained with a high-speed-gradient 1.5-T MR system in seven anesthetized dogs before and after administration of dipyridamole. Images were acquired on double oblique planes perpendicular to the left anterior descending arteries with a slice thickness of 5 mm, a field of view of 20 x 10 cm, a velocity window of +/- 1 m/sec, an average imaging time of 21 sec, a TR/TE of 11/5, and a temporal resolution of 44 msec. RESULTS: Coronary flow measured with VENC MR imaging correlated well with flow measured by the flowmeter (r = .95, slope = 0.97, n = 88). Interobserver variability in measuring coronary flow volume was 8%. CONCLUSION: Fast VENC MR imaging with high-speed gradients can provide accurate quantification of volume flow rate in coronary arteries.

Measurement of coronary blood flow velocity during handgrip exercise using breath-hold velocity encoded cine magnetic resonance imaging.

Coronary blood flow velocity was measured during handgrip exercise using breath-hold velocity encoded cine magnetic resonance imaging. Peak diastolic coronary flow velocity in the left anterior descending artery was 20.6 +/- 9.3 cm/s (mean +/- SD) at baseline and increased significantly to 31.1 +/- 16.4 cm/s after exercise (50.7 +/- 31.3% increase, p <0.01).

Evaluation of the iliac arteries: comparison of two-dimensional time of flight magnetic resonance angiography with cardiac compensated fast gradient recalled echo and contrast-enhanced three-dimensional time of flight magnetic resonance angiography.

We compared dynamic contrast-enhanced three-dimensional time of flight (3DTOF) magnetic resonance angiography (MRA) with two-dimensional time of flight (2DTOF) MRA with cardiac compensated fast gradient recalled echo (C-MON) and conventional angiography (CA) when it was available. C-MON re-orders the normal data acquisition to minimize ghosting artifacts generated by pulsatile flow. The initial phase of the study involved optimization of parameters and comparison C-MON with no C = MON in eight patients and volunteers. The final phase of the study involved 53 patients who were imaged with contrast-enhanced 3DTOF MRA and 2DTOF MRA with C-MON. Thirty of these patients also had CA. In the initial phase, 2DTOF MRA with C-MON was found to be equal (n = 3) or superior (n = 5) to 2DTOF without C-MON. In the final phase, the agreement among all imaging modalities varied from substantial to almost perfect (Cohen's kappa = .6-.83). The lowest agreement was using 2DTOF to evaluate the external iliac segments. The among suggested treatments varied from substantial to almost perfect for all imaging modalities (Cohen's kappa = .73-93). The diagnostic efficacies of 2DTOF with C-MON and contrast-enhanced 3DTOF were high overall, with the lowest value being a specificity of 63% for one reader in the evaluation of an external iliac segment using 2DTOF. In summary, 2DTOF with C-MON helped to eliminate artifacts due to pulsatility in the iliac arterial segments. In our experience, both dynamic contrast-enhanced 3DTOF MRA and 2DTOF MRA with C-MON performed well in the evaluation of the iliac arteries. Both studies have high interobeserver agreement and high diagnostic efficacy. Contrast-enhanced 3DTOF MRA should be reserved for situations in which the iliac vessels are extremely tortuous or occluded or the external iliac segments are poorly seen.

Motion suppression in MR imaging of the liver: comparison of respiratory-triggered and nontriggered fast spin-echo sequences.

OBJECTIVE: Our purpose was to compare the effectiveness of a respiratory-triggered fast spin-echo (RTFSE) pulse sequence and a nontriggered fast spin-echo pulse sequence for imaging liver masses. MATERIALS AND METHODS: Forty-one patients with suspected liver masses were imaged at 1.5 T with a fast spin-echo sequence and an RTFSE sequence designed to trigger data acquisition at end expiration. All other imaging parameters were identical. MR images were compared qualitatively for lesion detection and conspicuity, anatomic sharpness, vascular definition, phase artifacts, and overall image quality. Quantitative analysis was performed to obtain lesion-liver contrast and contrast-to-noise ratio (CNR) measurements of all liver masses. RESULTS: Thirty-three patients had liver masses. The RTFSE images showed superior anatomic sharpness in 83% of examinations and superior overall image quality in 85% of examinations. Lesion detection and conspicuity were superior for the RTFSE images in 53% of examinations and were similar for the two techniques in 38%. In 10 patients we detected more lesions on RTFSE images, and in one patient we detected more lesions on fast spin-echo images. In the remaining patients the number of lesions detected on RTFSE images was identical to the number detected on fast spin-echo images. Respiratory artifacts were less noticeable on the RTFSE images in 66% of examinations and on the fast spin-echo images in 14%. Quantitative analysis showed a 29% increase in mean relative lesion-liver contrast and a 34% increase in mean relative CNR for the RTFSE images. Mean lesion-liver contrast and CNR for the RTFSE images (152.6 +/- 100.9, 14.2 +/- 9.3) were superior to corresponding values for the fast spin-echo images (123.4 +/- 88.0, 10.9 +/- 7.4) (p < .0001). CONCLUSION: Compared with nontriggered fast spin-echo MR images, the RTFSE MR images were superior for our evaluation of liver masses. By acquiring data during a period of reduced respiratory motion, the RTFSE sequence produced images with sharper anatomic detail, equal or less phase ghosting, and measurable improvement in the lesion-liver contrast and CNR.

Dynamic contrast-enhanced breath-hold MR imaging of thoracic malignancy using cardiac compensation.

The purpose of this paper was to evaluate the use of dynamic gadopentetate dimeglumine-enhanced, breath-hold spoiled gradient-recalled (SPGR) MR imaging with cardiac compensation (CMON) compared to spin-echo MR imaging in patients with thoracic malignancy. We retrospectively reviewed MR images from 29 patients with thoracic tumors. MR imaging included axial electrocardiogram (ECG)-gated T1-weighted, fast spin echo (FSE) T2-weighted, and contrast-enhanced breath-hold fast multiplanar SPGR imaging with CMON, which selects the phase-encoding gradient based on the phase within the cardiac cycle. Images were reviewed for lung masses, mediastinal or hilar tumor, disease of the pleura, chest wall, and bones and vascular compression or occlusion. Contrast-enhanced fast multiplanar SPGR imaging with CMON produces images of the chest that are free of respiratory artifact and have diminished vascular pulsation artifact. ECG-gated T1-weighted images were preferred for depicting mediastinal and hilar tumor. The gadopentetate dimeglumine-enhanced fast multiplanar SPGR images were useful for depicting chest wall tumor, vascular compression or thrombosis, osseous metastases, and in distinguishing a central tumor mass from peripheral lung consolidation. Pleural tumor was depicted best on the FSE T2-weighted images and the contrast-enhanced SPGR images. As an adjunct to spin echo T1-weighted and T2-weighted imaging, contrast-enhanced fast multiplanar SPGR imaging with CMON is useful in the evaluation of thoracic malignancy.

Breath-hold MR measurements of blood flow velocity in internal mammary arteries and coronary artery bypass grafts.

Breath-hold velocity-encoded cine MR (VENC-MR) imaging is a feasible method for measuring phasic blood flow velocity in small vessels that move during respiration. The purposes of the current study are to compare breathhold VENC-MR measurements of flow velocities in the internal mammary arteries (IMA) with nonbreath-hold measurements and to characterize the systolic and diastolic flow velocity curves in a cardiac cycle in native IMA and IMA grafts. Flow velocity in 30 native IMA and 8 IMA grafts were evaluated with a breath-hold VENC-MR sequence with K-space segmentation and view-sharing reconstruction (TR/TE = 16/9 msec, VENC = 100 cm/s). In 10 native IMA, nonbreath-hold VENC-MR images were acquired as well for comparison. Breath-hold VENC-MR imaging showed significantly higher systolic and diastolic peak velocities in native IMA (43.1 cm/second +/- 15.0 and 10.0 cm/second +/- 4.8), in comparison to those of nonbreath-hold VENC-MR imaging (27.6 cm/second +/- 10.2 and 7.3 cm/second +/- 3.9, P < .05). The diastolic/systolic peak velocity ratio in the IMA grafts (.88 +/- .41) was significantly higher than that in native IMA (.24 +/- .08, P < .01). Interobserver variability in the flow velocity measurement was less than 4%. Breath-hold VENC-MR imaging demonstrated higher peak flow velocity in the IMA than nonbreath-hold VENC-MR imaging. This technique is a rapid and effective method for the noninvasive assessment of blood flow velocity in IMA grafts.

Improved reproducibility in measuring LV volumes and mass using multicoil breath-hold cine MR imaging.

There is a generally recognized need for improvement in quality of fast cardiac MR images. Consequently, breath-hold cine MR images were obtained with multiple surface coils connected to phased array receivers, and C/N, intra-observer and inter-observer variabilities for LV volumes and mass were evaluated. Two sets of short-axis images of the LV, one with multiple surface coils and another with a body coil, were acquired in eight subjects with a fast cine MR sequence using k-space segmentation (TR/TE = 7/2.2 msec, temporal resolution = 56 msec). C/N with multicoil imaging was 32.2 +/- 7.6 (mean +/- SD), significantly higher than that with a body coil (11.0 +/- 3.3, P < .01). The mean percentage differences in intra-observer and inter-observer measurements with multicoil imaging were significantly better than those with a body coil. In conclusion, multicoil imaging provides significant gain in C/N on breath-hold cine MRI of the heart. In addition, intra-observer and inter-observer reproducibilities are improved with multicoil imaging.

Evaluation of thoracic aortic dissection using breath-holding cine MRI.

OBJECTIVE: Our goal was to determine if breath-hold cine MRI in transaxial planes can be used for the evaluation of thoracic aortic dissection instead of conventional cine MRI since rapid imaging is required in this clinical setting. MATERIALS AND METHODS: Twelve patients with thoracic aortic dissection were imaged using a 1.5 T imager. Breath-hold images were acquired with fast cine MR sequence (TR/TE = 9/2.8, 20 degrees flip angle) using segmented k-space data acquisition. Conventional non-breath-hold cine MR images (TR/TE = 22/7.5, 35 degrees flip angle, 2 averages) were taken with flow and respiratory compensation. RESULTS: Sharpness of edges of the vessels on fast cine MR images was better than that on conventional cine MR images in 34 (57%) of 60 images. Inhomogeneous blood signal in aortic lumen due to motion artifacts was found in 2 (3%) of fast cine MR images and in 15 (25%) of conventional cine MR images. The contrast-to-noise ratios of fast cine MR images were significantly better than those of conventional cine MR images (26.4 +/- 9.1 vs. 18.5 +/- 10.1; p < 0.05) when the region of interest for noise was placed to include ghosting artifacts. CONCLUSION: Breath-hold cine MRI is a rapid technique that gives high quality images of thoracic aortic dissection and can provide a diagnosis in < 10 min of imaging time.

Breath-hold MR cine angiography of coronary arteries in healthy volunteers: value of multiangle oblique imaging planes.

OBJECTIVE: Breath-hold MR cine angiography was used to depict the coronary arteries in healthy volunteers. Multiangle oblique imaging planes were evaluated for feasibility in showing continuous segments of the proximal and middle portions of the left anterior descending and right coronary arteries. SUBJECTS AND METHODS: Eighteen healthy subjects were examined with a 1.5-T MR imager. Fat-suppressed fast gradient-echo images (TR = 9.8 msec, TE = 3.5 msec) were acquired with a 13-cm receive surface coil. A segmented k-space data acquisition was used to obtain images of the coronary arteries at several phases of the cardiac cycle within a single breath-hold. Multiangle double oblique images that were tangential and sequential to the epicardial surface of the left ventricle were used to show the left anterior descending artery, and oblique coronal images were used to show the right coronary artery. Images of consecutive slice locations were shown in a cine format, and the length of each major coronary artery that was continuously visualized was measured. RESULTS: The left main coronary artery, proximal left anterior descending artery, and right coronary artery were demonstrated in all subjects. The mid and distal portions of the left anterior descending artery and diagonal branches were visualized best on multiangle oblique imaging planes. Continuous segments (> 6 cm) of the left anterior descending artery and right coronary artery were imaged in 14 subjects (78%) and 12 subjects (67%), respectively. Cine display was useful for showing the continuity of the coronary arterial segments and also for distinguishing arteries from veins. CONCLUSION: Double oblique imaging planes were useful in showing long segments of left anterior descending and right coronary arteries on coronary MR angiograms. Further work is necessary to improve detection of the left circumflex artery.

Fast spin-echo MR imaging of the abdomen: contrast optimization and artifact reduction.

The effects of various fast spin-echo (FSE) magnetic resonance (MR) imaging parameters and artifact reduction techniques on FSE image contrast and quality were studied. The authors performed 139 abdominal MR examinations, comparing standard FSE images (echo train length [ETL] = 8, echo space [E-space] = 17 msec, bandwidth = +/- 16-kHz) with FSE images with an ETL of 16 (n = 22) or FSE images with a +/- 32-kHz bandwidth and an E-space of 11-14 msec (n = 22). FSE artifact reduction techniques were evaluated with spectral fat saturation (n = 40) or with a new flow compensation FSE sequence (n = 55). Images of liver lesions were reviewed qualitatively and with contrast-to-noise ratio (C/N) measurements. Decreasing the time of echo train sampling produced superior image quality, with increased anatomic sharpness, less image artifact, and improved liver-lesion C/N. Images obtained with an ETL of 16 showed more image blurring and a 23% decrease in relative contrast and 28% decrease in relative C/N for liver tumors. Increasing the bandwidth reduced E-space, producing a 12% decrease in background noise. Artifact reduction with fat saturation or flow compensation produced images with less ghosting artifact and superior overall image quality, with 39% and 20% increases in liver-tumor C/N, respectively. FSE image quality and contrast in the depiction of hepatic disease can be optimized with careful selection of imaging parameters and the use of artifact reduction techniques.

Single breath-hold pulmonary magnetic resonance angiography. Optimization and comparison of three imaging strategies.

RATIONALE AND OBJECTIVES: Ultrafast gradient-recalled-echo techniques for obtaining high-quality pulmonary magnetic resonance angiograms within a single breath-hold were optimized. METHODS: Fourteen subjects were imaged with both the body coil and a phased-array surface coil, using three gradient-recalled-echo pulse sequences: 1) two-dimensional sequential; 2) two-dimensional interleaved; and 3) volumetric acquisitions. Image quality was assessed with varied flip angle, receiver bandwidth, slice thickness/number, and matrix size. Cardiac compensation diminished ghost artifacts in the interleaved sequence. Individual sagittal sections and maximum intensity projections were reviewed. RESULTS: Pulmonary magnetic resonance angiograms acquired with volumetric and two-dimensional interleaved gradient-recalled-echo pulse sequences benefit greatest from intravenous gadolinium and result in greater pulmonary arterial visualization than traditional time-of-flight techniques. Phased-array coils result in improved vessel detection. CONCLUSIONS: High-quality breath-held pulmonary magnetic resonance angiography can be obtained with an intravenous contrast-enhanced gradient-recalled-echo acquisition; however, image quality is dependent on the pulse sequence.

Evaluation of malignant biliary obstruction: efficacy of fast multiplanar spoiled gradient-recalled MR imaging vs spin-echo MR imaging, CT, and cholangiography.

OBJECTIVE: Although CT and cholangiography have proven value in the detection of biliary obstruction, determining the extent of biliary tumors and imaging small pancreatic or ampullar tumors remain problematic. We hypothesized that the superior contrast resolution of MR, coupled with contrast-enhanced breath-hold imaging, might increase the sensitivity for tumor detection and improve the depiction of the point of obstruction in patients with malignant biliary disease. SUBJECTS AND METHODS: Twenty-one MRI studies were performed prospectively in patients with malignant biliary obstruction by obtaining breath-hold contrast-enhanced fast multiplanar spoiled gradient-recalled (FMPSPGR) images at 0 and 10 min, conventional spin-echo T1-weighted images, and fast spin-echo T2-weighted images. Findings on MR images were correlated with findings on CT scans (15 cases) and/or cholangiograms (14 cases) by two observers. All MR images, CT scans, and cholangiograms were reviewed to evaluate tumor detection, visualization of dilated bile ducts, and conspicuity of the obstructing tumor. A four-point scale (1 = excellent tumor depiction and conspicuity, 4 = tumor not detected) was used for evaluation. Contrast-to-noise ratios for tumor and bile were calculated for the three MR pulse sequences. RESULTS: The contrast-enhanced FMPSPGR images and CT scans provided excellent depiction of the dilated biliary tree in 95% and 93% of examinations, respectively, with both techniques superior to fast spin-echo and T1-weighted images (p < .005). Tumor detection was best with the immediate FMPSPGR MR images (20/21), compared with fast spin-echo MR images (16/21) (p = .04), T1-weighted MR images (16/21) (p = .04), CT scans (12/15) (p > .05), and cholangiograms (13/14) (p > .05). Of 13 examinations showing proximal biliary obstruction, the mean score for tumor conspicuity was best with the immediate enhanced FMPSPGR MR images (1.38 +/- .65), compared with T1-weighted MR images (2.38 +/- 1.3) and fast spin-echo MR images (2.08 +/- 1.0) (p < .05), but it was not different from the delayed FMPSPGR MR images (1.75 +/- 1.1) or CT scans (1.9 +/- 0.99) (p > .05). For five of six cholangiocarcinomas, the immediate and delayed enhanced FMPSPGR MR images showed excellent tumor conspicuity owing to their enhancement with gadopentetate dimeglumine. Data for contrast-to-noise ratios of tumor showed that the immediate FMPSPGR MR images (15.8 +/- 10.2) were superior to T1-weighted images (6.3 +/- 3.5, p < .01), but were not different from fast spin-echo images (13.5 +/- 6.7) or delayed FMPSGR images (11.5 +/- 8.9). For eight examinations in patients with distal biliary obstruction, the mean score for tumor conspicuity was greater with the immediate FMPSPGR MR images (1.38 +/- 0.52), compared with fast spin-echo images (3.25 +/- 0.71, p < .005), T1-weighted images (2.63 +/- 1.06, p < .05), and delayed FMPSPGR MR images (2.60 +/- 1.5, p < .05), but was similar to that with CT scans (1.40 +/- 0.89, p > .05). Data for contrast-to-noise ratios of tumor showed an advantage for the immediate FMPSPGR MR images (12.0 +/- 7.7) over T1-weighted images (4.0 +/- 2.6, p < .01) and delayed FMPSPGR images (4.3 +/- 2.6, p < .025), but no difference from fast spin echo images (6.6 +/- 8.8, p = .05). CONCLUSION: Contrast-enhanced FMPSPGR MR imaging is sensitive for the detection of tumors causing biliary obstruction. For proximal obstruction, it may be particularly effective for detecting and defining tumor extent of hilar cholangiocarcinomas because of their enhancement with gadopentetate dimeglumine. For cases of distal obstruction, this technique showed improved tumor detection and conspicuity compared with T1- and fast spin-echo T2-weighted images, but showed no advantage over CT.

Shunt flow measurement and evaluation of valve oscillation with a spin-echo phase-contrast MR sequence.

PURPOSE: To present a spin-echo phase-contrast (SEPC) magnetic resonance pulse sequence designed to measure the very slow flow in ventricular shunt tubing. MATERIALS AND METHODS: A flow phantom constructed of shunt tubing and incorporating no valve or a high-, medium-, or low-pressure valve was connected to a flow pump. Flow rates were 0.05-1.00 mL/min (72-1,440 mL/d). Flow measurement was performed with the thin-section SEPC sequence. RESULTS: The flow rates measured with SEPC imaging correlated closely with the pump flow rate for the entire physiologic spectrum of shunt flow rates. This was true for all valves, resulting in overall R2s of .974 at 4 cm/sec and .980 at 2 cm/sec. Shunt flow was pulsatile with valves in place. There was a linear relationship between flow rate and the frequency of valve opening and closing. CONCLUSION: The SEPC technique is an accurate and noninvasive method of measuring shunt flow.

Evaluation of left ventricular volume and mass with breath-hold cine MR imaging.

Left ventricular (LV) volumes and mass were evaluated in 10 healthy volunteers with breath-hold cine magnetic resonance (MR) imaging. The results were compared with those obtained with conventional cine MR imaging. The breath-hold studies showed no ghosting artifact, and cardiac edges were clearly identified because of the reduced blurring. Measurements of LV end-diastolic volume (LVEDV), LV end-systolic volume (LVESV), and LV mass obtained with breath-hold cine MR imaging showed close correlation with those obtained with conventional cine MR imaging (r = .98, .97, and .99, respectively). The interobserver variabilities for LVEDV, LVESV, and LV mass determined with breath-hold cine MR imaging (4.0%, 8.0%, and 3.7%, respectively) were equal to or less than those determined with conventional cine MR imaging (4.0%, 8.6%, and 5.0%, respectively). The authors conclude that breath-hold cine MR imaging is highly useful because an accurate assessment of cardiac function is obtained in less than 5 minutes.

Fast multiplanar spoiled gradient-recalled imaging of the liver: pulse sequence optimization and comparison with spin-echo MR imaging.

OBJECTIVE: The purpose of this study was to optimize a new rapid-acquisition MR pulse sequence, called fast multiplanar spoiled gradient-recalled (FMPSPGR) imaging, for breath-hold imaging of the liver and to compare unenhanced and contrast-enhanced FMPSPGR with standard spin-echo imaging in detecting liver tumors. MATERIALS AND METHODS: The pulse sequence was optimized at 1.5 T with a healthy volunteer. Various scanning parameters were evaluated, and liver-spleen signal difference/noise measurements were used to estimate lesion contrast-to-noise ratios. We examined 24 patients with hepatic masses using the optimized sequence with spin-echo T1-weighted and T2-weighted imaging as well as unenhanced and gadopentetate dimeglumine-enhanced FMPSPGR imaging. The contrast-to-noise ratio for the hepatic tumors was determined for each sequence. Three radiologists who did not know the biopsy or test results reviewed all images for lesion conspicuity, lesion tissue specificity, and overall image quality. RESULTS: A comparison of unenhanced FMPSPGR images with spin-echo T1-weighted images showed a 40% improvement in mean contrast-to-noise ratio and a 70% improvement in liver signal-to-noise ratio for the FMPSPGR images. A comparison of gadopentetate dimeglumine-enhanced FMPSPGR images with spin-echo T1- and T2-weighted images showed a superior contrast-to-noise ratio for the enhanced FMPSPGR images in 17 (68%) of 25 hepatic lesions, which included all hepatic cysts (n = 3) and all hepatomas (n = 6), and in six of 12 patients with other liver tumors. The results of contrast-to-noise ratio for four patients with hemangiomas were mixed. For the remaining eight lesions, the contrast-to-noise ratio for spin-echo T1- and T2-weighted images predominated in three and five cases, respectively. Contrast-enhanced FMPSPGR images revealed a 40% and 300% increase in contrast-to-noise ratio compared with T2- and T1-weighted images, respectively. All three radiologists preferred the contrast-enhanced FMPSPGR images for overall image quality. For lesion conspicuity and specificity, however, the three radiologists differed, with a preference for the FMPSPGR images in 52%, 80%, and 40% of cases for lesion conspicuity and in 68%, 40%, and 60% of cases for lesion specificity. CONCLUSION: FMPSPGR is a new, ultrafast MR sequence that provides T1-weighted images of the liver during suspended respiration. Contrast-to-noise ratio and liver signal-to-noise ratio are significantly improved over those on conventional spin-echo T1-weighted images. The combination of breath-hold FMPSPGR with gadopentetate dimeglumine is an excellent technique that can be used to rapidly evaluate the liver with superior overall image quality. Contrast-to-noise ratios are generally superior to T2-weighted spin-echo images, making this technique a useful adjunct to conventional spin-echo MR imaging.

Minimizing TE in moment-nulled or flow-encoded two- and three-dimensional gradient-echo imaging.

A method for minimizing field-echo delay in moment-nulled gradient-echo imaging is presented. Even though ramps are accounted for, the analysis yields simple closed-form solutions. The method is then generalized to the section-select waveform for three-dimensional volume imaging and to flow encoding for phase-contrast imaging. Three strategies for first-moment selection in phase-contrast imaging are discussed, including a new strategy that always yields the minimum echo delay. Trapezoidal and triangular gradient lobe shapes are analyzed.

Quantitation of the susceptibility difference between trabecular bone and bone marrow: experimental studies.

In this study we quantify the effects of different relaxation mechanisms on the signal intensity in gradient-echo images of tissue such as bone marrow in the presence of trabecular bone. The susceptibility difference between trabecular bone and soft tissue produces distortions in the magnetic lines of force which induce strong inhomogeneities in the static magnetic field. Diffusion of tissue protons in such magnetic field gradients produce a shortening of the transverse relaxation time T2, while the dephasing of the transverse magnetization due to susceptibility differences produces a shortening of the apparent relaxation time T2* as demonstrated in gradient-echo images. We have used specimens of dried human vertebrae with different bone densities immersed in either saline to simulate tissue water or an emulsion of oil and water to simulate bone marrow to quantify these relaxation mechanisms in vitro. We have measured the MR relaxation times T1, T2, and T2* of protons within the trabecular spaces and correlated their variations with trabecular bone density. We have found that in vitro, at 1.5 T, the relaxation times T1 and T2 do not show significant variations with bone density and there are no significant contributions to the transverse relaxation rate due to the diffusion of tissue water in the magnetic field gradients. However, the relaxation rate, 1/T2*, of saline in the presence of trabecular bone increases at a rate of 0.2 s-1/mg/cc due to the dephasing of the transverse magnetization in the magnetic field inhomogeneities. Similar bone density-related T2* variations were observed for fat protons within the trabeculae where the chemical-shift-induced modulations of signal intensity in an oil-water emulsion have been separated from the susceptibility-induced relaxation effects. In addition, we have verified these effects in vivo and quantified in vivo variations in fat and water relaxation rates of bone marrow in the epiphysis and diaphysis in the appendicular skeleton of normal volunteers and found that both fat and water T2* are shorter in the epiphysis compared to the diaphysis, which correlates well with previous observations.