No abstract available

[en] In 132 patients Gd-DTPA enhanced and nonenhanced FLASH sequences were compared with enhanced and nonenhanced spin-echo sequences (SES). Maximal contrast between neoplasm and muscle was obtained with enhanced FLASH-90, but these images were 60% inferior to those from a T2-weighted spin-echo sequence. Highest contrast between sclerotic and fibrotic tumors and bone marrow or fat was found in the nonenhanced FLASH-90, while in lytic lesions contrast was highest in FLASH-10, and was 64% of that in T1-weighted spin-echo sequences. Edema in muscle was reliably visualized by enhanced FLASH-90 only, while edema in fat was mostly missed by all FLASH sequences

[en] Ultrafast MR imaging techniques acquire whole-body images in as little as 1/25th of a second. The instant scan technique (Advanced NMR Systems) requires only a single excitation. Thus MR images have a repetition time that is infinitely long and an echo time that can be varied from 30 to 600 msec. Water or fat protons can be imaged independently. This clinical study was undertaken to investigate the utility of this technique in five healthy volunteers and 16 patients. The latter were selected on the basis of known liver disease. Images showed no degradation secondary to physiologic motion. Lesions as small as 5 mm were detected with high lesion-liver contrast. Due to much longer T2 times, the contrast-to-noise ratio between fluid-containing structures such as the gallbladder, bladder, and renal cysts was even greater. The authors results show that this technique offers considerable promise for motion-free abdominal MR imaging

[en] Spectral lines from metabolites of interest, such as lactate, may be obscured because of the signal from a variety of protonated metabolites within the volume of interest (VOI). To eliminate interfering peaks, the authors combined localized proton spectroscopy in vivo with a method for volume-localized lactate editing that uses zero-quantum coherence created in a stimulated-echo (STEAM) pulse sequence with a 1.5-T clinical MR unit. H-1 spectra of the human brain in vivo were obtained in five normal volunteers and seven patients. All spectra were obtained for a VOI of 3x3x3 cm. In normal volunteers, neither the baseline spectrum nor the lactate editing spectrum exhibited a lactate peak. In patients with cerebral infarction ( n = 4) or brain tumors (n = 3), complex spectra were found before lactate editing due to interfering peaks in four of seven patients. With the lactate editing method, interfering lipid peaks could be removed from the lactate resonances in all four patients. The authors conclude that volume-localized lactate editing with zero-quantum coherence is useful in eliminating interfering peaks that may obscure the spectral line of lactate

[en] Conventional T1-weighted spin-echo (SE) (260/14[repetition time (TR) msec/echo time (TE) msec]) and T2-weighted SE (2,350/60-180) and phase-contrast (PC) T2-dependent methods (2,350/30 or 2,350/60) were compared at 0.6 T. The contrast-to-noise ratio magnitudes on SE260/14(-12.4+-6.7) and PC2,350/60(10.8+-4.2) images were significantly higher than on the SE2,350/60 and 2,350/180(7.8+-3.6 and 7.9+-4.5) or the PC2,350/30(4.6+-2.9)(P < .05, analysis of variance). Seventy-eight of 88 confirmed lesions were detected on the SE260/14 images and 81 of 88 on the PC2,350/60 images. The T1-weighted SE and the PC2,350/60 sequences demonstrated more (P < .05) metastases than any of the conventional T2-weighted SE sequences (TE 60:71 of 88; TE 120:69 of 88; TE 180: 65 of 88). This analysis indicates that for detection of hepatic metastases at mid field strengths the T1-weighted short TR, short TE (SE 260/14) and the T2-weighted PC (2,350/60) sequences offer comparable performance

[en] This exhibit illustrates the spectrum of morphologic and signal intensity features of a variety of hepatic neoplasms seen with MR imaging. The cases have been selected from our experience with 442 patients, evaluated with 0.6-T imager and a wide range of pulse sequences. Known or suspected primary and secondary liver neoplasms included 330 metastases, 14 hepatomas, three hepatic adenomas, two cases of focal nodular hyperplasia, 20 cases of hepatic cysts, 37 hemangionas and 11 miscellaneous conditions; 25 subjects were healthy. In virtually all cases correlative contrast agent-enhanced CT scans were obtained; selected ones are shown. Pathologic or historical confirmation of etiology was available in most patients, and gross anatomic correlation was available for the 18 primary liver neoplasms that were resected

[en] To compare the diagnostic performance of high-field strength MR imaging and CT in the detection of hepatic metastases, the authors have studied the receiver operating characteristics (ROCs) of 700 images. Images were obtained from 45 patients with proved primary cancer. The presence or absence of lesions was based on intraoperative sonographic and intraoperative or pathologic findings. MR images were obtained at 1.5 T

[en] To determine the diagnostic specificity of scar tissue associated with primary liver tumors (PLT), the authors compared MR tissue characteristics and morphologic features with gross and microscopic pathologic findings in 38 patients with PLT. Central scars were detected in three of 14 hemangiomas, three of 18 hepatocellular carcinomas, one of four hepatic adenomas, and one of two focal nodular hyperplasias. The scar shape on MR images correlated well with the morphologic scar formation on the cut pathologic specimen. Histopathologic analysis shows that scars composed of dense collagenous tissue (n = 4) appeared hypointense on T1-weighted (spin-echo repetition time = 260 msec, echo time = 14 msec) [SE 260/14] and T2-weighted (SE 2,350/60, 120, 180) images. However, scars composed of loose fibrous tissue with greater cellular content, vascular structures, edema, hemorrhage and/or necrosis appeared isointense (n = 3) or hyperintense (n = 1) to tumor on T2-weighted images. This study shows no significant tissue specificity for PLT based on the MR appearance of central scars

[en] SPECT has been used in a comparative study with planar bone scintigraphy in 54 patients. Spatial resolution of the method is sufficient to localise the vertebral bodies, the spinous processes, the intervertebral, costotransverse and costovertebral joints. In all patients it was possible to relate the areas of increased uptake to specific anatomical sites of the spine known to be affected in the different conditions. A detailed localisation is rarely possible using planar scintigraphy alone, due to the complex osseous anatomy of the spine. In some patients lesions could be seen only with SPECT. SPECT is an invaluable supplement to planar scintigraphy of the spine. (orig)

[en] A retrospective analysis of signal intensity (SI), lesion margin, and internal architecture (descriptors) was performed in 133 patients with cancer (102), hemangiomas (24), and cysts (seven). More than 392 individual lesions were imaged with multiple T1-weighted (T1W) and T2-weighted (T2W) pulse sequences. The authors conclude (1) 96% of all lesions have an SI less than liver on T1W; on T2W, over 89% have a higher SI; (2) very sharply marginated lesions are more typical of benign lesions, and indistinct margins characterize cancer (P < .005); (3) 92% of patients with cancer have one or more cancer-specific descriptors, including amorphous (47%), target (25%), halo (16%), and chameleon (12%); and (4) the lightbulb and doughnut signs favor benign and malignant disease, respectively, but are less specific