Neuroradiology/Head and Neck Imaging Original Research

Transcription

1 Neuroradiology/Head and Neck Imaging Original Research Sayah et al. Use of 3D T2-Weighted SPCE MRI for Evaluation of Lumbar Spine Neuroradiology/Head and Neck Imaging Original Research nousheh Sayah 1 nn K. Jay Jacob S. Toaff Erini V. Makariou Frank erkowitz Sayah, Jay K, Toaff JS, Makariou EV, erkowitz F Keywords: 3D imaging, MRI, spine, spondylosis DOI: /JR K. Jay, J. S. Toaff, and E. V. Makariou contributed equally to this work. Received October 22, 2015; accepted after revision March 21, ased on a presentation at the merican Society of Spine Radiology 2012 annual meeting, Miami, FL. Supported by grant UL1RR from the National Center for Research Resources, National Institutes of Health, through the Clinical and Translational Science wards Program, a trademark of the Department of Health and Human Services, as part of the Road Map Initiative, Re-Engineering the Clinical Research Enterprise. 1 ll authors: Division of Neuroradiology, Department of Radiology, Medstar Georgetown University Hospital, Ground Fl, CG 201, 3800 Reservoir Rd, NW, Washington, DC ddress correspondence to. Sayah (anousheh.sayah@gunet.georgetown.edu). This article is available for credit. JR 2016; 207: X/16/ merican Roentgen Ray Society Effectiveness of a Rapid Lumbar Spine MRI Protocol Using 3D T2-Weighted SPCE Imaging Versus a Standard Protocol for Evaluation of Degenerative Changes of the Lumbar Spine OJECTIVE. Reducing lumbar spine MRI scanning time while retaining diagnostic accuracy can benefit patients and reduce health care costs. This study compares the effectiveness of a rapid lumbar MRI protocol using 3D T2-weighted sampling perfection with application-optimized contrast with different flip-angle evolutions (SPCE) sequences with a standard MRI protocol for evaluation of lumbar spondylosis. MTERILS ND METHODS. Two hundred fifty consecutive unenhanced lumbar MRI examinations performed at 1.5 T were retrospectively reviewed. Full, rapid, and complete versions of each examination were interpreted for spondylotic changes at each lumbar level, including herniations and neural compromise. The full examination consisted of sagittal T1-weighted, T2-weighted turbo spin-echo (TSE), and STIR sequences; and axial T1- and T2-weighted TSE sequences (time, 18 minutes 40 seconds). The rapid examination consisted of sagittal T1- and T2-weighted SPCE sequences, with axial SPCE reformations (time, 8 minutes 46 seconds). The complete examination consisted of the full examination plus the T2-weighted SPCE sequence. Sensitivities and specificities of the full and rapid examinations were calculated using the complete study as the reference standard. RESULTS. The rapid and full studies had sensitivities of 76.0% and 69.3%, with specificities of 97.2% and 97.9%, respectively, for all degenerative processes. Rapid and full sensitivities were 68.7% and 66.3% for disk herniation, 85.2% and 81.5% for canal compromise, 82.9% and 69.1% for lateral recess compromise, and 76.9% and 69.7% for foraminal compromise, respectively. CONCLUSION. Isotropic SPCE T2-weighted imaging provides high-quality imaging of lumbar spondylosis, with multiplanar reformatting capability. Our SPCE-based rapid protocol had sensitivities and specificities for herniations and neural compromise comparable to those of the protocol without SPCE. This protocol fits within a 15-minute slot, potentially reducing costs and discomfort for a large subgroup of patients. L ower back pain affects about half of the U.S. adult population annually, with about two-thirds of adults affected at some point in their lifetime [1]. The total associated annual costs, including medical costs and lost wages, exceed over $100 billion a year [2], with direct medical costs exceeding $86 billion in 2005 [3]. Lumbar spine MRI is often performed for lower back pain, with or without radiculopathy, to evaluate for a potential source, and its use is steadily rising. Over a 12-year interval ( ) the number of lumbar spine MRI examinations performed among Medicare recipients increased by more than 300% [4]. This trend may continue if we consider the increasing age of the U.S. population and the prevalence of degenerative spine conditions. The typical unenhanced lumbar spine MRI protocol usually includes sagittal T1-weighted, sagittal T2-weighted, sagittal STIR, axial T2-weighted, and axial T1-weighted images. This protocol takes approximately 20 minutes on a typical highfield MRI system. In view of the high number of lumbar spine MRI examinations performed nationwide, the total scanning times for these studies are staggering. In addition, patients must lie flat and supine and remain completely motionless during the scanning, a task that often is very difficult for patients with back pain. Equally problematic is the anxiety and claustrophobia that many patients experience while in the bore of the magnet, which often intensifies with long scanning times. 614 JR:207, September 2016

2 Use of 3D T2-Weighted SPCE MRI for Evaluation of Lumbar Spine earing in mind the cost, time, and discomfort associated with lumbar spine MRI, reducing MRI scanning time while retaining diagnostic accuracy should benefit patients and reduce health care costs. Thus, we propose a shorter MRI examination protocol using 3D T2-weighted sampling perfection with application-optimized contrast with different flip-angle evolutions (SPCE) imaging for the routine evaluation of lower back pain. Lighvani and Melhem [5] concluded that a highresolution 3D T2-weighted SPCE sequence is feasible for routine clinical imaging of the spine. Meindl et al. [6] described excellent 3D T2-weighted SPCE imaging of the cervical spine with clinically acceptable imaging times and superior delineation of neural structures over conventional T2-weighted imaging. Lee et al. [7] showed comparable interpretations of 3D T2-weighted lumbar spine imaging against conventional 2D T2-weighted turbo spin-echo (TSE) imaging. The idea of a rapid lumbar MRI for spondylosis is not new. Robertson et al. [8] used a two-sequence protocol with sagittal and axial T2-weighted imaging and found adequate accuracy compared with the standard protocol. Our purpose was to evaluate the diagnostic effectiveness of a rapid lumbar MRI protocol consisting of only sagittal T1-weighted imaging and sagittal 3D T2-weighted SPCE imaging with axial reformats. We calculated the sensitivities and specificities of this rapid examination as well as those of the same study without SPCE imaging (full examination), using our reference standard version of the study a version that contains all sequences (complete examination) for the evaluation of lumbar spondylosis. Our hypothesis was that the shortened 3D-based protocol will be of similar diagnostic accuracy for diagnosing spondylotic changes, TLE 1: Imaging Parameters Parameter Sagittal T1-Weighted Imaging including disk herniations and neural compromise, compared with the conventional series commonly used in practice. Materials and Methods Method Design and MRI This is an institutional review board approved HIP-compliant retrospective review of 250 consecutive unenhanced lumbar spine MRI studies performed at 1.5 T (Magnetom vanto, Siemens Healthcare) at our institution. Studies were performed from January 2010 through September 2011 and included adult patients between the ages of 18 and 100 years. Studies consisted of the following sequences: sagittal T1-weighted, sagittal T2-weighted, sagittal STIR, axial T1-weighted, axial T2-weighted, and sagittal 3D T2-weighted SPCE sequences. T1-weighted, T2-weighted, and STIR sequences were all obtained with turbo or fast parameters (i.e., TSE). conventional spine surface coil was used in all cases. Each study was converted and anonymized using an eight-digit alphanumeric identification code into three separate examinations: a full examination consisting of sagittal T1-weighted, T2-weighted, and STIR images and axial T1- and T2-weighted images (18 minutes 40 seconds); a rapid examination consisting of sagittal T1-weighted images and T2-weighted SPCE images, with axial SPCE reformations (8 minutes 46 seconds); and a complete examination containing all sequences. SPCE was set to a resolution of 0.8 mm to keep an isotropic acquisition. Separate axial reformats of 2.5-mm intervals were created by the MRI technologists using the sagittal SPCE dataset to take full advantage of the 3D capability. Parameters for the various sequences are listed in Table 1. T1-weighted images were included in the rapid examination to evaluate for fatty lesions and endplate marrow changes and as an additional evaluation of foraminal stenosis. Sagittal T2-Weighted Imaging Sagittal STIR Sagittal SPCE Evaluation for Spondylosis and Neural Compromise Five fellowship-trained neuroradiologists each independently reviewed 50 studies for spondylotic changes. separate set of 50 studies was assigned to each reader with a main goal of eliminating interobserver bias from the study. Rapid examinations were reviewed first, followed by full examinations a minimum of 2 weeks later and complete examinations 1 year later, to minimize recall bias. The anonymized examinations were reviewed at a multimonitor clinical PCS. t each disk space between L1 and S1, any right-sided, central, and left-sided disk herniations were reported as present or absent. Herniations included protrusions and extrusions but were not noted by herniation type or size. Disk bulges were not included. In addition, at these levels, neural compromise was evaluated at the central spinal canal, bilateral lateral recesses, and bilateral neural foramina using a point value scale of 0 3 (0, normal; 1, mild; 2, moderate; and 3, severe). Non-disk-related causes for neural compromise, such as osteophyte encroachment, were not specifically noted. reas evaluated are depicted in Figure 1. binary point value system was then used to analyze the data collected from the 250 MRI examinations, in all three versions, at the following sites: right disk, central disk, left disk, canal, right lateral recess, left lateral recess, right foramen, and left foramen. Disk herniations were given a value of 1 if present and 0 if absent. Canal, lateral recess, and neural foraminal narrowing were given a value of 0 if normal or mild and 1 if moderate or severe. This was done to simplify the analysis and to group the likely more clinically significant lesions (i.e., moderate and severe) from likely asymptomatic lesions (i.e., normal or mild). Discrepant areas were then highlighted if values differed between rapid and complete examinations and between full and complete examinations. xial T1-Weighted Imaging xial T2-Weighted Imaging TR/TE 514/ / / / / /124 Flip angle ( ) Partition thickness (mm) FOV (mm) Matrix size Interslice gap (mm) N Receiver bandwidth No. of signals averaged cquisition time 3 min 33 s 3 min 51 s 2 min 51 s 4 min 32 s 4 min 8 s 4 min 24 s Note SPCE = sampling perfection with application-optimized contrast with different flip-angle evolutions, N = not applicable. JR:207, September

3 Sayah et al. Subjective Image nalysis ll discrepant regions were reevaluated on the original study (with all sequences) to determine the subjective cause of the difference in reader interpretation. This analysis was performed by the readers on their own studies, choosing from a list of causes for discrepancy, including a write-in option. Statistical nalysis Five neuroradiologists read a full, rapid, and complete version of 50 separate lumbar spine MRI studies, for a total of 250 examinations. Five levels were assessed on each MR image and each was treated independently, resulting in 1250 total levels. Eight anatomic locations were evaluated at each level and also were treated independently, resulting in total of 10,000 locations. Statistical analysis was performed using the complete examination as the reference standard to which the rapid and full studies were compared. Sensitivity, specificity, and discrepancy rates of the rapid and full studies were calculated for total number of locations and for each location, using the binary data. symptotic standard errors and 95% confidence limits for sensitivity, specificity, and discrepancy rates were obtained using binomial distribution. nalysis was performed using SS software (version 9.1.3, SS Institute). Results For the 1250 total lumbar disk levels evaluated for degenerative pathologic abnormalities, the rapid study had overall sensitivity of 76.0% and specificity of 97.2%, with a discrepancy rate of 4.24%. The full study had Fig year-old woman with low back pain. MR images show anatomic areas of spine that were evaluated., Red outlined areas show lateral recess narrowing. Green outlined area shows central spinal canal stenosis., Yellow outlined area shows central disk herniation. Purple outlined areas show right and left disk herniations. C and D, On axial (C) and sagittal (D) images, blue outlined areas show neural foraminal narrowing. overall sensitivity of 69.3% and specificity of 97.9%, with a discrepancy rate of 4.0%. The sensitivities, specificities, and discrepancy rates, respectively, of the rapid study for the various locations are as follows: 68.7%, 95.8%, and 5.9% for disk herniations; 85.2%, 98.8%, and 1.8% for canal; 82.9%, 98.0%, and 2.9% for lateral recess; and 76.9%, 97.4%, and 4.3% for foramina. The sensitivities, specificities, and discrepancy rates, respectively, of the full study for the various locations are as follows: 66.3%, 97.3%, and 4.8% for disk herniations; 81.5%, 98.5%, and 2.2% for canal; 69.1%, 98.5%, and 3.3% for lateral recess; and 69.7%, 97.7%, and 4.6% for foramina. Data including 95% CIs are shown in Figure 2. Subjective analysis of the discrepant levels established that the most frequent reason for discrepancy was intraobserver variation between the rapid and full studies, which was reported in 245 instances. The next most common reason was the higher resolution of SPCE, in which the abnormality was better or only identified on SPCE versus conventional T2-weighted images (Fig. 3), reported in 150 instances. In some cases, discrepant levels showed no significant abnormality on the SPCE images for example, a disk herniation on conventional imaging was deemed to be bulge on the isotropic data, or lateral recess narrowing was less prominent on highresolution SPCE imaging (Fig. 4). Inadequate coverage of the lateral spinal regions on the SPCE images was reported to result in missed lesions on the rapid examinations in 10 instances. pparent higher tissue contrast on conventional T2-weighted images and motion on the SPCE images were each reported as causes for discrepancies in four instances. Other reported reasons for discrepancy totaled eight instances. Discussion In this study, we evaluated the effectiveness of a rapid unenhanced lumbar spine MRI protocol using 3D T2-weighted SPCE imaging as compared with a conventional examination. We observed slightly higher sensitivities and comparable specificities of this rapid study over the examination without 3D imaging. The isotropic T2-weighted SPCE sequence is a variant of TSE but with variable flip-an- C D 616 JR:207, September 2016

4 Use of 3D T2-Weighted SPCE MRI for Evaluation of Lumbar Spine Sensitivity (%) Overall Disk Herniation Full Foraminal Rapid gle radiofrequency pulses less than 180 along the echo train. The nonselective low-flip-angle pulses allow less radiofrequency energy deposition, longer echo trains, shorter echo spacing, higher turbo factors, and rapid filling of the k-space. The result is a relatively high signal-to-noise ratio, high spatial resolution, and short imaging time secondary to parallel acquisition. SPCE is proprietary to Siemens Healthcare, with similar techniques on other platforms, including Cube (GE Healthcare) and Vista (Philips Healthcare). efore SPCE imaging, the existing spinecho and TSE schemes were not clinically efficient at creating 3D T2-weighted images in a timely manner. SPCE was originally developed and used for brain imaging by Mugler et al. [9 11]. Lichy et al. [12] first described the clinical use of SPCE in pelvic, lumbar spine, cervical spine, and extremity imaging and found that the contrast-tonoise and signal-to-noise ratios of SPCE are of comparable diagnostic value to those of conventional 2D T2-weighted TSE imaging. Since then, the clinical use of 3D T2-weighted SPCE has been successfully applied to multiple diagnostic issues, including prostate cancer imaging and MRCP [13, 14]. Rodegerdts et al. [15] described visualization of the whole spine using this technique, with high-resolution detail, ability to create curved planar reformations, and retrospective angulation. SPCE has been reported for high-resolution evaluation of the small craniocervical ligaments at the base of the skull in various planes [16] and the detection of aqueductal stenosis in the brain [17]. Lateral Recess Canal Overall Fig. 2 Graphs of sensitivity and specificity of full versus rapid examinations for evaluating various spinal abnormalities., ars show sensitivities, and lines and whiskers show associated 95% CIs., ars shows specificities and lines and whiskers show associated 95% CIs. Specificity (%) The SPCE sequence lacks significant CSF flow artifacts and has adequate tissue characterization. In addition, multiplanar reformats can be produced after the patient leaves the scanner, saving time and decreasing the need for repeat examinations in cases of missed pathologic abnormalities. This can be extremely helpful with spinal scoliosis (Fig. 5) and postoperative change where anatomy is distorted. Gerigk et al. [18] recently described the use of SPCE in the postsurgical cervical spine with diagnostic caliber images, despite the presence of metallic hardware. We designed the study such that no study was read by different readers; each reader was assigned a separate set of studies to evaluate. This was done to eliminate interobserver bias from the study, which is a significant phenomenon among MRI spine reads in daily practice. We thought any other design would introduce significant interreader variability that would considerably affect the data analysis. Limitations of this type of study design include decreased statistical power. Our subjective analysis showed that the most common cause for discrepancy between rapid and full studies was intrareader variability. This intrareader variability is ubiquitous throughout all of radiology and is extremely difficult to manage. We did not quantitatively evaluate our cases for intrareader variability, and no guidelines were used to grade mild, moderate, and severe variability. Of note, rant-zawadzki et al. [19] reported 86% intrareader agreement for disk abnormalities on lumbar spine MRI. The use of a standardized pictorial method of grading spondylosis, such Disk Herniation Full Foraminal Rapid Lateral Recess Canal as that used by Lee et al. [20], may help with reader variability, although it is likely more useful for interreader variability. In addition, we did not distinguish the subtypes of disk herniations because our goal was to evaluate the degree of neural compromise. This study also focused solely on degenerative changes of the lumbar spine. The 3D SPCE sequence is of great use for spondylotic changes, but its major disadvantage is that it is less accurate for lesions in the spinal cord or soft tissues. nother disadvantage is a slight relative decrease in its ability to evaluate the spinal canal in the presence of spinal hardware. potential focus for future studies would be optimizing the SPCE sequence to better detect cord and soft-tissue abnormalities; we did not tackle this in the current study. Our studies were unenhanced, so inflammatory, infectious, or neoplastic processes could be missed by this proposed protocol. We suggest that gadolinium-enhanced T1-weighted imaging be added to the rapid protocol if there is suspicion for these entities. In addition, STIR imaging was not included in our rapid examination, and we argue that it adds little new diagnostic information in lumbar spondylosis. The lack of axial T1-weighted imaging did not affect our ability to diagnose spondylosis; it is often used for a second look at the foramina or for nonspondylotic items, including fatty or hemorrhagic lesions. We think that the robust information obtained from the SPCE sequence trumps that gained from axial T1-weighted imaging or sagittal STIR sequences for spondylotic disease. In a handful of the studies included in our statistical analysis, the lateral paraspinal re- JR:207, September

5 Sayah et al. gions were sometimes not imaged using the SPCE sequence (4% of our studies), which limited the evaluation of far-lateral herniations. This was mainly seen in patients with lumbar scoliosis or relatively large body habitus and could be overcome by extending SPCE coverage to include more of the lateral regions or by imaging in the coronal plane. Lumbar lordosis, however, often causes coronal plane imaging to take longer than sagittal imaging. We have since increased this coverage in our larger patients or those with scoliosis with mild increases in acquisition time of about seconds. lso, our axial SPCE reformats were created 2.5 mm thick to take advantage of the high-resolution data, whereas the conventional axial T2-weighted images were created with 4-mm-thick reformats. The argument can be made that acquiring the axial T2-weighted images at 2.5 mm would give the same information as the reformats, but this would come with increased scanning time. Of note, the time for lumbar spine 3D T2-weighted SPCE imaging performed at 3 T is approximately 5 minutes 30 seconds, which is slightly longer than the time needed at 1.5 T. The use of SPCE imaging for the spine has been extremely useful in our practice for Fig year-old woman with lower back pain and right lower extremity radiculopathy. and, xial T2-weighted () and corresponding axial T2-weighted sampling perfection with applicationoptimized contrast with different flip-angle evolutions (SPCE) reformat () at L4 5 disk space both show right paracentral disk herniation. C and D, xial T2-weighted (C) and axial T2-weighted SPCE reformat (D) obtained one slice below level in and show second central disk herniation on SPCE reformat (D) that is not seen on conventional axial T2-weighted image (C), likely secondary to partial volume averaging. its high-resolution imaging of pathologic abnormalities and multiplanar reformatting capabilities. The resolution afforded by SPCE allows better evaluation of nerve roots, small facet joint cysts, bone spurs, and disk fragments that cause nerve root impingement at all lumbar levels, some of which are not detected on the conventional T2-weighted images. Disk bulges and herniations appeared clearer on the SPCE series because the edges of the disk were more distinct on the high-resolution images, especially on sagittal views. We also believe that the lateral recesses are better discerned on SPCE images, and it may be that lateral recess stenosis is, in fact, overcalled on conventional image series. Interestingly, despite the longer acquisition time of SPCE, there were no significant motion artifacts in our cases. The standard spine coil afforded excellent images; we did not combine the standard and the phasedarray coils in any cases. The ability to reformat the data into various planes (in a matter of seconds) improves evaluation and detection of abnormalities in scoliotic spines and exaggerated lumbar lordoses. We also use SPCE regularly for cervical and thoracic spondylosis and find it extremely useful for detailed evaluation of epidural extent of neoplastic disease. We urge radiologists to work and become familiar with the SPCE sequence and test its strong ability to diagnose spondylosis. The implications for patients are multifold. Lying flat in the MRI bore often exacerbates pain and anxiety in patients with back pain. The reduction in scanning time afforded by using SPCE in lieu of other conventional sequences in this rapid examination (> 50% in most cases) decreases time in the machine and thus lessens discomfort. The spatial detail of C D 618 JR:207, September 2016

6 Use of 3D T2-Weighted SPCE MRI for Evaluation of Lumbar Spine SPCE may increase radiologist confidence and decrease interpretation time. The reduced time increases throughput of patients, translating to a reduction in cost. Interestingly, reduction of cost is becoming increasingly important in our current situation as reimbursement in the U.S. moves from a fee-for-service model to a capitated model. The overall implications include more-efficient patient care and the ability to meet increasing demands by the aging population. The 3D acquisition allows more-detailed evaluation of the spine Fig year-old woman with peripheral neuropathy in bilateral lower extremities. and, Narrowing of spinal canal and bilateral lateral recesses on conventional axial T2-weighted image () may be graded as more severe on conventional imaging than appears on higher resolution axial T2-weighted sampling perfection with application-optimized contrast with different flip-angle evolutions reformat (), where space provided to cauda equina is more clearly depicted. with multiplanar analysis and higher resolution. lthough we did not specifically evaluate the appearance of metallic artifacts on the SPCE sequences, our impression is that the susceptibility artifacts are less than those on T2-weighted TSE images. Future directions of our research include evaluation of a similar rapid examination for the evaluation of cervical spondylosis. In addition, further investigation of 3D T2-weighted SPCE images in comparison with conventional 2D T2-weighted spin-echo or TSE imaging for the detection of spinal cord lesions would also be of great clinical interest. Conclusion The rapid lumbar MRI protocol with 3D T2-weighted imaging has comparable sensitivities and specificities in diagnosing herniations and neural compromise compared with the conventional examination. We conclude that the use of sagittal T2-weighted SPCE imaging is robust and could decrease scanning times and result in similar accuracies in the Fig year-old woman with scoliosis and lower back pain. C, Coronal reformats of sagittal T2-weighted sampling perfection with application-optimized contrast with different flip-angle evolutions (SPCE) () show significant thoracolumbar scoliosis. xial T2-weighted image though L3 4 disk level () is distorted anatomically because of marked curvature. T2-weighted SPCE data were used to create axial T2-weighted reformats (C) aligned to disk space to better visualize spinal anatomy. C JR:207, September

7 Sayah et al. evaluation of lumbar spondylosis and neural compromise. ecause this rapid protocol can be completed in 15 minutes, it can reduce cost and discomfort for a large subgroup of patients. cknowledgment We thank Sameer Desale, iostatistician, in the iostatistics and ioinformatics Department at the Medstar Health Research Institute. References 1. Deyo R, Mirza SK, Martin I. ack pain prevalence and visit rates: estimates from U.S. national surveys, Spine 2006; 31: Katz JN. Lumbar disc disorders and low-back pain: socioeconomic factors and consequences. J one Joint Surg m 2006; 88(suppl 2): Martin I, Deyo R, Mirza SK, et al. Expenditures and health status among adults with back and neck problems. JM 2008; 299: Deyo R, Mirza SK, Turner J, Martin I. Overtreating chronic back pain: time to back off? J m oard Fam Med 2009; 22: Lighvani, Melhem ER. dvances in high-field MR imaging of the spine. ppl Radiol 2009; 38: Meindl T, Wirth S, Weckbach S, Dietrich O, Reiser M, Schoenberg SO. Magnetic resonance imaging of the cervical spine: comparison of 2D T2-weighted turbo spin echo, 2D T2*weighted gradient-recalled echo and 3D T2-weighted variable flip-angle turbo spin echo sequences. Eur Radiol 2009; 19: Lee S, Jee WH, Jung JY, Lee SY, Ryu KS, Ha KY. MRI of the lumbar spine: comparison of 3D isotropic turbo spin-echo SPCE sequence versus conventional 2D sequences at 3.0 T. cta Radiol 2015; 56: Robertson WD, Jarvik JG, Tsuruda JS, Koepsell TD, Maravilla KR. The comparison of a rapid screening MR protocol with a conventional MR protocol for lumbar spondylosis. JR 1996; 166: Mugler JP 3rd, ao S, Mulkern RV, et al. Optimized single-slab three-dimensional spin-echo MR imaging of the brain. Radiology 2000; 216: Mugler JP, Meyer H, Kiefer. Practical implementation of optimized tissue-specific prescribed signal evolutions for improved turbo-spin-echo imaging. In: Ehman RL, ed. Proceedings of the International Society for Magnetic Resonance in Medicine 11 th scientific meeting & exhibition. Toronto, ON, Canada: International Society for Magnetic Resonance in Medicine, 2003: Mugler JP, Menzel MI, Horger W. High-resolution, multi-contrast 3D imaging of the brain in 15 minutes. In: Kucharyczyk W, ed. Proceedings of the International Society for Magnetic Resonance in Medicine 13 th scientific meeting & exhibition. Miami each, FL: International Society for Magnetic Resonance in Medicine, 2005: Lichy MP, Wietek M, Mugler JP 3rd, et al. Magnetic resonance imaging of the body trunk using a single-slab, 3-dimensional, T2-weighted turbospin-echo sequence with high sampling efficiency (SPCE) for high spatial resolution imaging: initial clinical experiences. Invest Radiol 2005; 40: Rosenkrantz, Neil J, Kong X, et al. Prostate cancer: comparison of 3D T2-weighted with conventional 2D T2-weighted imaging for image quality and tumor detection. JR 2010; 194: Morita S, Ueno E, Masukawa, et al. Comparison of SPCE and 3D TSE MRCP at 1.5T focusing on difference in echo spacing. Magn Reson Med Sci 2009; 8: Rodegerdts E, oss, Riemarzik K, et al. 3D imaging of the whole spine at 3T compared to 1.5T: initial experiences. cta Radiol 2006; 47: aumert, Wortler K, Steffinger D, Schmidt GP, Reiser MF, aur-melnyk. ssessment of the internal craniocervical ligaments with a new magnetic resonance imaging sequence: three-dimensional turbo spin echo with variable flip-angle distribution (SPCE). Magn Reson Imaging 2009; 27: lgin O, Turkbey. Evaluation of aqueductal stenosis by 3D sampling perfection with applicationoptimized contrasts using different flip angle evolutions sequence: preliminary results with 3T MR imaging. JNR 2012; 33: Gerigk L, ostel T, Hegewald, et al. Dynamic magnetic resonance imaging of the cervical spine with high-resolution 3-dimensional T2-imaging. Clin Neuroradiol 2012; 22: rant-zawadzki MN, Jensen MC, Obuchowski N, Ross JS, Modic MT. Interobserver and intraobserver variability in interpretation of lumbar disc abnormalities: a comparison of two nomenclatures. Spine 1995; 20: ; discussion, Lee S, Lee JW, Yeom JS, et al. practical MRI grading system for lumbar foraminal stenosis. JR 2010; 194: FOR YOUR INFORMTION This article is available for CME and Self-ssessment (S-CME) credit that satisfies Part II requirements for maintenance of certification (MOC). To access the examination for this article, follow the prompts associated with the online version of the article. 620 JR:207, September 2016