Sequences and techniques in spinal MR imaging

The optimal protocol in spinal MR imaging is not evident. Sagittal T2 weighted FSE, sagittal T1 weighted SE and axial T2 weighted FSE sequences are widely accepted for imaging patients with sciatica and/or lumbar pain. Because of the limited amount of CSF compared to the lumbar spine, the choice of sequences is much more complex in the study of the cervical spine. Sagittal T2 FSE, sagittal T1 SE and axial 2D GE images are suggested in routine cervical spine imaging. To assess the bone marrow, a STIR sequence can be added to this protocol on both lumbar and cervical spine examinations. The 2D GE produces an acceptable image quality to differentiate between the disc and bony protrusions. The use of FLAIR for imaging spinal cord lesions remains controversial in the literature.     

Fast spin echo vs conventional spin echo in cervical spine imaging

The major attraction of fast-spin-echo (FSE) imaging is reduced acquisition time; however, careful review of the literature reveals many weaknesses: phase-encoded blurring, truncation artefact, bright fat signal, reduced magnetic susceptibility and increased motion artefact. Our aim was a prospective, blinded comparison of FSE and conventional spin echo (CSE) in the cervical spine. Both sequences were performed in 43 patients (19 males and 24 females; mean age 45 years, range 15-66 years). Twenty-eight patients were studied at 1.5 T and 15 at 0.5 T. Typical sequence parameters were: at 1.5 T, TR/TE 2000/90 CSE and 3000/120 FSE, and at 0.5 T, 2200/80 CSE and 2800/120 FSE. Time saved on the FSE was used to increase the matrix and the number of acquisitions. Two neuroradiologists evaluated the images for pathology, artefacts, disc signal intensity, thecal sac compression and image quality. Ten patients had cord lesions; 2 (20 %) were missed on CSE. In 4 of 10 patients with moderate/severe thecal sac compression, the degree of stenosis was apparently exaggerated on CSE. The mean degree of confidence for the CSE sequences was 1.8 and for the FSE 1.1, where 1 is optimal. For cervical spine imaging, FSE should be preferred to CSE.     

Postoperative sella: evaluation with fast spin echo T2-weighted high-resolution imaging

OBJECTIVE: The purpose of this study was to investigate the magnetic resonance features of the postoperative sella with fast spin echo (FSE) T2-weighted high-resolution imaging and to evaluate the benefits of the sequence using a follow-up magnetic resonance imaging protocol after transsphenoidal surgery. METHODS: Coronal spin echo (SE) T1-weighted and FSE T2-weighted images were prospectively obtained in 24 patients after surgery for pituitary adenomas. We observed the signals and the contour of normal structures, fluid collection, implanted materials, and mass lesions, including granulation tissue and adenoma. RESULTS: The pituitary gland was delineated in 51 of 59 FSE T2-weighted images, 90.2% of which presented clear boundaries. Whereas the gland was detected in 49 of 58 SE T1-weighted images, only 20.4% showed the boundaries. A mass lesion was identified in each of 12 patients with good resolution on FSE T2-weighted images. SE T1-weighted images detected mass lesions in 7 of 12 patients without distinctive boundaries. Contrast enhancement had little advantage in clarifying the boundaries between normal and abnormal structures. For the detection of mass lesions in the sella, the kappa values for interobserver agreement were 0.8 for FSE T2-weighted images and 0.25 for SE T1-weighted images. CONCLUSION: FSE T2-weighted imaging is a reliable method with which to assess the sella with sufficient resolution after transsphenoidal surgery. The combination of unenhanced SE T1-weighted and FSE T2-weighted images may reduce the use of contrast material after pituitary surgery.     

Hyperparathyroidism--comparison of flash imaging with spin echo MR imaging

MR images of the neck were prospectively studied in 19 patients with hyperparathyroidism. Fast low angle shot (FLASH) sequence was performed in addition to T1- and T2-weighted spin echo (SE) sequences. FLASH images were obtained with 320/12/20 degrees (TR/TE/flip angle) using presaturation technique. TE of 12 ms was chosen to eliminate high signal of fat tissue. In the evaluation of detectability, a combination of T1-weighted SE and FLASH images (T1WI + FLASH) was compared with a combination of T1- and T2-weighted SE images (T1WI + T2WI). MR imaging correctly depicted 20 of 30 abnormal glands on both T1WI + FLASH and T1WI + T2WI. FLASH imaging effectively eliminated high signal of fat tissue. Nineteen abnormal glands demonstrated higher signal than surrounding tissues on FLASH images, whereas 12 glands were high-intense on T2-weighted SE images. We conclude that FLASH imaging provides improved tissue contrast and anatomic delineation and, thus, may replace T2-weighted SE imaging in the neck.     

Comparison of fat-saturation fast spin echo versus conventional spin-echo MRI in the detection of rotator cuff pathology

The purpose of this study was to compare the diagnostic performance of fat-saturation fast-spin-echo (FSE) T2-weighted (T2W) sequences with conventional spin-echo (CSE) T2W sequences in the detection of rotator cuff pathology using surgery as the reference standard. Oblique coronal dual-echo CSE and FSE T2W images with fat saturation from 50 surgically confirmed MR shoulder examinations were acquired on a 1.5-T MR scanner. Blinded MR readers retrospectively analyzed each imaging sequence separately and ultimately correlated both sequences together with findings at surgery. FSE was 100% sensitive and 94% specific in detection of full-thickness tears (n = 19) and 73% sensitive and 97% specific in the detection of partial-thickness rotator cuff tears (n = 13). There was no statistically significant difference in the performance of FSE with fat saturation compared with CSE. The two discrepancies between imaging sequences related to the extent of partial-thickness tears. Our findings suggest that fat-saturation FSE imaging can effectively replace CSE imaging in the evaluation of rotator cuff pathology.     

Hepatic T2-weighted MRI: a prospective comparison of sequences, including breath-hold, half-Fourier turbo spin echo (HASTE)

The purpose of this study was to quantitatively compare the hepatic contrast characteristics of conventional spin-echo (CSE) and fast spin-echo (FSE) sequences with breath-hold T2-weighted images acquired with half-Fourier turbo spin echo (HASTE). Forty-five patients were examined with a phased-array surface coil. Nineteen patients had focal hepatic lesions, including eight malignant tumors, 10 cavernous hemangiomas, and one hepatic adenoma. Twenty-six patients had no focal hepatic lesions. T2-weighted images with comparable TE were acquired with CSE, FSE, and HASTE pulse sequences. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) for liver, spleen, and lesions were measured. FSE demonstrated significantly better quantitative performance than CSE for liver-spleen CNR (P .0084). No statistically significant difference was demonstrated between FSE and CSE for liver or spleen SNR. FSE demonstrated clear scan time and resolution advantages over CSE. HASTE performed significantly poorer than CSE and FSE for liver-spleen CNR (P < .0001), liver SNR (P = .0002 for CSE and P < .0001 for FSE), and spleen SNR (P < .0001). Optimized FSE images with a short echo train length performed comparably to CSE images of equivalent TE. Liver-lesion CNR was suppressed on HASTE images, suggesting that long echo train length FSE sequences could diminish solid lesion detection compared to CSE and short echo train length FSE.     

Four new mutations in the apolipoprotein B gene causing hypobetalipoproteinemia, including two different frameshift mutations that yield truncated apolipoprotein B proteins of identical length

Magnetic resonance imaging is frequently complicated by the presence of motion and susceptibility gradients. Also, some biologic tissues have short T2s. These problems are particularly troublesome in fast spin-echo (FSE) imaging, in which T2 decay and motion between echoes result in image blurring and ghost artifacts. The authors reduced TE in conventional spin-echo (SE) imaging to 5 msec and echo spacing (E-space) in FSE imaging to 6 msec. All magnetic gradients (except readout) were kept at a maximum, with data sampling as fast as 125 kHz and only ramp waveforms used. Truncated sinc radio-frequency pulses and asymmetric echo sampling were also used in SE imaging. Short TE (5.8 msec) SE images of the upper abdomen were compared with conventional SE images (TE = 11 msec). Also, FSE images with short E-space were compared with conventional FSE images in multiple body sites. Short TE significantly improved the liver-spleen contrast-to-total noise ratio (C/N) (7.9 vs 4.1, n = 9, P < .01) on T1-weighted SE images, reduced the intensity of ghost artifacts (by 34%, P < .02), and increased the number of available imaging planes by 30%. It also improved delineation of cranial nerves and reduced susceptibility artifacts. On short E-space FSE images, spine, lung, upper abdomen, and musculoskeletal tissues appeared crisper and measured spleen-liver C/N increased significantly (6.9 vs 4.0, n = 12, P < .01). The delineation of tissues with short T2 (eg, cartilage) and motion artifact suppression were also improved. Short TE methods can improve image quality in both SE and FSE imaging and merit further clinical evaluation.     

Study of susceptibility-induced artefacts in GRASE with different echo train length

The aim of this study was to evaluate the sensitivity of gradient-and-spin-echo (GRASE) sequences to susceptibility effects. GRASE sequences with 21 and 33 echoes per echo train were compared with a T2-weighted FSE sequence with an echo train length of 5 by means of MRI in phantoms, volunteers (n = 10), and patients (n = 19) with old hemorrhagic brain lesions. All experiments were performed on a 1.0-T clinical MR system (Impact Expert, Siemens AG, Erlangen, Germany) with constant imaging parameters. Contrast-to-noise ratios (CNRs) of tubes doped with iron oxides at different concentrations, of brain areas with physiological iron deposition (red nucleus, substantia nigra), and of areas of old brain hemorrhage were calculated for FSE and GRASE pulse sequences. Areas of old brain hemorrhage were also qualitatively analyzed for the degree of visible susceptibility effects by blinded reading. The CNR of iron oxide tubes and iron-containing brain areas decreased with increasing echo trains of GRASE sequences. The CNR of GRASE sequences decreased when compared with CNR of their FSE counterparts (GRASE 21 echo trains 23.8 +/- 0.8, FSE 5 echo trains 26.7 +/- 0.9; p 相似文献   

Effects of feeding on protein turnover in healthy children and in children with cystic fibrosis

In 24 patients presenting with 55 renal lesions (mean size, 20.8 mm), single-breath-hold (SBH) fast spin-echo (FSE) techniques allowing T1 and T2 images to be produced within 20 and 23 sec, respectively, were compared with routine non-breath-hold (NBH) spin-echo (SE) T1 and NBH-FSE T2 sequences. Contrast-to-noise ratios (CNRs) measured from SBH-FSE T1 images were an average of 97% higher than their NBH counterparts (P = .0001) and allowed an improved lesion conspicuity in 80% of the cases (P = 0.0001). For T2 imaging, SBH-FSE and NBH-FSE sequences were not statistically different with respect to lesion conspicuity (P = .55) and CNR values (P = .19). This was observed despite a 35% average decrease in CNR of SBH-FSE compared to NBH-FSE images. By reducing respiratory motion artifacts while preserving SE-like image contrast, SBH-FSE techniques have the potential to replace routine NBH sequences for an optimal diagnosis of renal masses.     

Single shot fast spin echo (SSFSE)

Single Shot Scan methods, which acquire all the necessary data for reconstruction with one excitation, are very useful methods to minimize motion artifacts. Single Shot Scan can be categorized in 2 groups, one is EPI method using gradient by Echo data acquisition, and another is SSFSE method using RF. SSFSE method is Single Shot FSE method using 0.5 NEX. SSFSE produces images with less motion artifacts and heavy PD and T2 images with high contrast. New RF reduces Echo Space and minimizes image quality degradation caused by Single Shot. SSFSE is useful for long T2 like MRCP. Long TE mode can be used with very long ETE even with 0.5 NEX. New applications development for Single Shot using RF is awaited in the future.     

Chronic pulmonary histoplasmosis in Brazil: report of two cases with cavitation diagnosed by transthoracic needle biopsy

OBJECTIVE: To compare fat-suppressed fast spin-echo (FSE) T2-weighted images with gradient-recalled echo (GRE) T2*-weighted images in the evaluation of anteroinferior labral tears. DESIGN: MR images were retrospectively reviewed by two radiologists masked to the history and arthroscopic findings. They separately interpreted the anteroinferior labrum as torn or intact, first on one pulse sequence and then, 4 weeks later, on the other sequence. The MR interpretations were correlated with the arthroscopic findings. PATIENTS: Nine patients with anteroinferior labral tears, and nine similarly-aged patients with normal, labra were studied. RESULTS AND CONCLUSIONS: Observer 1 had a sensitivity of 0.56 on the GRE images and 0.67 on the FSE images (P > 0.5), with a specificity of 1.0 for both sequences. Observer 2 had a sensitivity of 0.78 and a specificity of 0.89 for both sequences. In this small study there is no significant difference between GRE and fat-suppressed FSE images in their ability to diagnose anteroinferior labral tears. When evaluating the labrum with conventional MRI, axial fat-suppressed FSE T2-weighted images can be used in place of GRE images without a loss of accuracy.     

Fat-suppressing STIR sequences with and without contrast media in the MRT of ENT tumors

Fat-suppressed STIR (short TI inversion recovery) sequences were compared to plain and contrast-enhanced T1-weighted SE sequences of head and neck tumors. 19 patients underwent MR imaging on a 0.5 Telsa system (T5-II, Philips). STIR imaging parameters: TR/TE = 1000/20 ms, inversion pulse 100 ms. All films were read by four radiologists. The image quality was graded: score from 0 to 5, by means that grade 5 = optimal quality. Sensitivity was 89% in STIR, 96% in SE sequences. Tumor delineation was graded good in the enhanced T1-weighted and enhanced fat suppression images. The unenhanced imaging was superior in STIR (STIR/T1 = 2.8/2.43). The tumor contrast was best in contrast enhanced and plain STIR sequences (STIR contrast = 3.41), and in the contrast enhanced T1-weighted SE (3.33). STIR almost equaled T1 post-contrast in respect of tumour conspicuity, but the sensitivity was lower. STIR can be a supplement to SE, but cannot substitute T1 postcontrast. The combined use is expected to have the highest assessment value.     

MRI of lumbar spondylosis: a comparison of sagittal T2 weighted and three sequence examinations

The aim of this study was to determine whether a single T2 weighted sagittal sequence could replace the conventional three sequence examination of the lumbar spine. The T2 weighted sagittal image of 79 lumbar spine MRI examinations were retrospectively reported by three radiologists. Features relating to degenerative disease were recorded and an assessment made of whether further sequences were likely to add information. On a separate occasion the T1 weighted and T2 weighted sagittal and T2 weighted axial sequences were reported blind in relation to the initial assessment. Areas of disagreement were resolved by consensus opinion. The T2 weighted sequence was compared with the three sequences, taking the three sequence examination as the standard. Disc protrusions were diagnosed from the T2 weighted sagittal images with a sensitivity of 38% and a specificity of 97%. 22 discs reported as a disc bulge on the T2 weighted sequence were re-classified as a disc protrusion on axial images because of their focal nature. Central canal stenosis was diagnosed on the T2 weighted sagittal sequence with a sensitivity of 60% and a specificity of 95%. After assessing the T2 weighted sequence, it was thought unlikely that further sequences would add extra information in 60% of cases (48/79). However, further information was obtained in 21% of these cases (10/48) when all the sequences were assessed. The extra information gained by using all three sequences was considered to be of greater benefit than the time saved by using a single T2 sagittal sequence. Other diagnoses where the additional sequences proved helpful are discussed.     

MR imaging of the anterior cruciate ligament: frequency of discordant findings on sagittal-oblique images and correlation with arthroscopic findings

OBJECTIVE: The purposes of this study were to determine the frequency of discordant MR findings of the anterior cruciate ligament (ACL) using sagittal-oblique images and the arthroscopic integrity of the ACL in knees with a discordant MR appearance. MATERIALS AND METHODS: The frequency of discordant MR features was determined by independently reviewing the sagittal-oblique images of 53 consecutive patients undergoing subsequent arthroscopic surgery. The first 25 consecutive knees were examined using a conventional, dual-echo, T2-weighted spin-echo (SE) sequence; the subsequent 28 consecutive knees were examined using both a T1-weighted SE(T1) and a fat-suppressed, fast spin-echo (FSE) T2-weighted sequence. Two examiners who were unaware of patient identification and arthroscopic results evaluated only the sagittal images that included the ACL. Each reviewer graded the ACL as torn or intact. The MR appearance was considered discordant when one MR sequence showed disrupted or poorly seen ACL fibers and the other sequence showed intact ACL fibers. The arthroscopic integrity of the ACL was determined for the 20 knees with a discordant MR appearance. RESULTS: A discordant MR appearance of the ACL was seen in 20 of the 53 knees (38%), including 10 knees evaluated using conventional T2-weighted sequences (40%) and 10 knees using T1/FSE T2 (36%) sequences. Arthroscopic examination of these knees showed intact ACL fibers in all 20 knees; the ACL was normal in 18 knees and a partial tear was questioned in two knees. CONCLUSION: A discordant appearance of the ACL was frequently encountered (38%) using either conventional T2-weighted sequences or T1 and FSE MR sequences in the sagittal-oblique plane. Intact ACL fibers were found during arthroscopy in all 20 knees with a discordant MR appearance.     

Contrast-enhanced MR "magnetization transfer technique". Improved tumor contrast, delineation and visibility of intracranial malignant gliomas and metastases in radiosurgical treatment planning

AIMS: To improve tumor conspicuity and delineation on contrast-enhanced T1-weighted MR images with and without magnetization transfer (MT) contrast as a strategy to improve the macroscopic boost volume definition in the planning process of radiosurgery in patients with high grade gliomas or metastatic brain lesions. PATIENTS AND METHODS: Thirty-two patients (mean age 47 years) with histologically proven or suspected high grade glioma (n = 12) or metastatic brain lesions (n = 20) were prospectively examined by MR imaging. After the administration of gadolinium dimeglumine (0.1 mmol/kg body weight) the lesions were imaged with a T1-weighted MT-fast low angle shot (FLASH) pulse sequence and with a conventional T1-weighted SE sequence without MT saturation. RESULTS: The mean CNR of enhancing lesions on T1-weighted MT-FLASH was 15 +/- 5 compared to 11 +/- 4 on SE images, representing a significant (p < .01) improvement. The mean tumor diameter of malignant gliomas was significantly (p < .01) larger measured on T1-weighted MT-FLASH images compared to those obtained from T1-weighted SE images and were comparable for metastatic lesions. Lesion conspicuity and delineation were improved in 50% of patients with high grade gliomas and in 35% of patients with brain metastases. Lesion conspicuity was markedly improved in the posterior fossa. Additional contrast enhancing lesions were detected in 10% of patients with metastases on MT-FLASH images. CONCLUSIONS: It is concluded that contrast-enhanced MT-FLASH images may improve lesion detection and delineation in the planning process of radiosurgery in patients with intracranial high grade gliomas or metastases or even alter the treatment approach.     

Studies on utility of MR T2-weighted images using multishot echo-planar imaging for hepatic mass lesions

MR T2-weighted images using multishot echo-planar imaging (EPI) and fast spin-echo (FSE) sequences were obtained in 22 patients with hepatic masses. Multishot EPI sequences included eight-shot breath-hold EPI and 16-shot EPI without breath-hold, while FSE sequences included nonfat-suppressed respiratory-triggered FSE, fat-suppressed respiratory-triggered FSE, and nonfat-suppressed breath-hold FSE. Signal-to-noise ratio, contrast-to-noise ratio and artifacts were compared between EPI and FSE images of 47 hepatic masses. In evaluating solid tumors, EPI provided image quality equal or superior to that of FSE, whereas in the evaluation of nonsolid tumors FSE showed better image quality than EPI. In conclusion, it was demonstrated that in the evaluation of hepatic solid tumors T2-weighted eight-shot breath-hold EPI can replace both nonfat-suppressed respiratory-triggered FSE and breath-hold FSE, and it was suggested that eight-shot breath-hold EPI can replace fat-suppressed respiratory-triggered FSE to reduce patient discomfort and increase examination throughput.     

A comparison of magnetic resonance and computed tomographic image quality after the implantation of tantalum and titanium spinal instrumentation

STUDY DESIGN: Tantalum- and titanium-based lumbar interbody fusion devices were implanted into two fresh human cadavers, and magnetic resonance and computed tomographic imaging were performed to evaluate adjacent spinal structures and the amount of metallic artifact. OBJECTIVE: The objective of this study was to prospectively compare the preliminary results of magnetic resonance imaging and computed tomography scanning image quality after the implantation of both titanium and tantalum spinal implants. SUMMARY OF BACKGROUND DATA: The availability of tantalum and titanium spinal implants brings theoretical magnetic resonance imaging compatibility along with several other desirable characteristics. The magnetic resonance imaging and computed tomographic imaging of tantalum spinal instrumentation has never been studied previously or compared with titanium instrumentation. METHODS: Titanium and tantalum spinal implants produced for anterior spinal fusion were each placed at two levels in the lumbar spine of two fresh cadaver specimens. Sequential spin echo T1-weighted and T2-weighted magnetic resonance imaging studies and computed tomographic scans were obtained. The resulting images were then graded to describe and compare the behavior of tantalum metal in magnetic resonance imaging and computed tomographic studies. RESULTS: Good T1 and T2 images were obtained that allowed visualization of the neural structures with minimal artifact. The optimal T1 images for tantalum metal were similar in quality to the optimal T1 parameters for titanium metal. T2 images for both tantalum and titanium metal were obtained with similar results for both metals. Gradient echo magnetic resonance imaging scans of both were poorly imaged with a large amount of artifact. Computed tomographic studies of tantalum implants produced a large amount of metal artifact when compared with computed tomographic studies of titanium implants. CONCLUSIONS: High-quality magnetic resonance imaging studies can be obtained after the implantation of both titanium and tantalum spinal instrumentation. Both of the metals produce similar images on magnetic resonance imaging studies with comparable amounts of metallic artifact. High-quality computed tomographic scans of titanium implants can be obtained with minimal distortion secondary to artifact. However, computed tomographic scanning is not the imaging modality of choice for the tantalum spinal implants because of the large amounts of artifact.     

Three-dimensional echocardiographic planimetry of maximal regurgitant orifice area in myxomatous mitral regurgitation: intraoperative comparison with proximal flow convergence

The purpose of this study was to evaluate the usefulness of multishot echo-planar imaging in detecting liver tumors in comparison with respiratory triggered T2-weighted fast-spin-echo (FSE) imaging. Thirty-two patients with 70 focal liver lesions were imaged using a 1.5-T high speed MR imager. Eight-shot echo-planar images covering the whole liver were acquired during a single breath-hold period. FSE images were acquired with respiratory triggering in approximately 4 minutes. Lesion detectability and image quality of the two pulse sequences were analyzed qualitatively. Quantitative analysis was performed by means of signal-to-noise and tumor-liver contrast-to-noise analysis. Lesion detectability was comparable in both solid (86.3% vs 90.2%: .3 < P < .5) and nonsolid lesions (89.5% vs 100%: .3 < P < .5) between echo-planar and FSE images. Echo-planar imaging provided significantly reduced image artifact, better lesion conspicuity, and anatomic detail compared with FSE imaging. The signal-to-noise and contrast-to-noise ratios of echo-planar images were significantly higher than those of FSE images. Breath-hold eight-shot echo-planar imaging can be an alternative to T2-weighted FSE imaging because it can provide comparable image quality in a substantially decreased acquisition time.     

MR diagnosis of labral tears of the shoulder: value of T2*-weighted gradient-recalled echo images made in external rotation

OBJECTIVE: Because several studies have shown that conventional MR imaging can fail to diagnose a significant percentage of labral tears, some authors have proposed obtaining T2*-weighted gradient-recalled echo images with the humerus in external rotation. The purpose of our study was to determine whether the diagnostic accuracy of detecting anteroinferior labral tears by MR imaging would be improved by adding a T2*-weighted gradient-recalled echo sequence with the humerus in external rotation. MATERIALS AND METHODS: The study included 24 patients for whom axial MR images of the shoulder were obtained with the humerus in both the neutral position and external rotation. Two observers interpreted the images made in the neutral position and then noted any change in their interpretations after viewing findings on the external-rotation images. MR results were correlated with surgical findings. At surgery, 14 anteroinferior labra were found to be torn and 10 were found to be intact. RESULTS: Both observers identified one patient for whom a surgically proved labral tear was seen only on the external-rotation images. The sensitivity increased from 0.43 to 0.50 (p = .35) for observer 1 and from 0.36 to 0.43 (p = .35) for observer 2. The specificity of 0.90 for both observers remained unchanged. The accuracy improved from 0.62 to 0.67 for observer 1 and from 0.55 to 0.62 for observer 2. CONCLUSION: The addition of external-rotation T2*-weighted gradient-recalled echo images to the MR examination for tears of the anteroinferior labrum leads to a small but statistically insignificant increase in diagnostic sensitivity. We conclude that the small increase in sensitivity does not justify the routine use of this sequence.     

Hepatic lesion detection at MR imaging: a comparative study with four sequences

PURPOSE: To compare results with the following magnetic resonance (MR) imaging sequences in the detection of focal hepatic lesions: fast spin-echo (SE) before and after administration of superparamagnetic iron oxide (SPIO) particles, fat-suppressed T2-weighted SE, and dynamic gadolinium-enhanced fast low-angle shot (FLASH). MATERIALS AND METHODS: In 26 patients with known malignancy and documented focal liver lesions, 1.0-T MR imaging was performed prior to hepatic resection. All images were reviewed independently by four blinded observers. Sensitivity was calculated for each sequence and for each observer by means of alternative-free-response receiver operating characteristic (ROC) methods. RESULTS: The mean area under the alternative-free-response ROC curve with SPIO-enhanced fast SE (0.85) was significantly greater (P < .02) than that with all other sequences. Detection was significantly improved with SPIO-enhanced fast SE compared with dynamic gadolinium-enhanced FLASH (which was the best of the other three sequences) for all lesions (P < .002) and for malignant lesions (P < .0002). CONCLUSION: Findings with the SPIO-enhanced fast SE sequence improved detection of focal liver lesions and had the highest diagnostic accuracy.