Advanced magnetic resonance imaging (MRI) techniques of the spine and spinal cord in children and adults

Advanced magnetic resonance imaging (MRI) techniques of the spine and spinal cord in children and... In this article, we illustrate the main advanced magnetic resonance imaging (MRI) techniques used for imaging of the spine and spinal cord in children and adults. This work focuses on daily clinical practice and aims to address the most common questions and needs of radiologists. We will also provide tips to solve common problems with which we were confronted. The main clinical indications for each MR technique, possible pitfalls and the challenges faced in spine imaging because of anatomical and physical constraints will be discussed. The major advanced MRI techniques dealt with in this article are CSF, (cerebrosopinal fluid) flow, diffusion, diffusion tensor imaging (DTI), MRA, dynamic contrast-enhanced T1-weighted perfusion, MR angiography, susceptibility-weighted imaging (SWI), functional imaging (fMRI) and spectroscopy. Teaching Points � DWI is essential to diagnose cord ischaemia in the acute stage. � MRA is useful to guide surgical planning or endovascular embolisation of AVMs. � Three Tesla is superior to 1.5 T for spine MR angiography and spectroscopy. � Advanced sequences should only be used together with conventional morphological sequences. . . . . Keywords Spinal cord Perfusion Spectroscopy Magnetic resonance angiography Diffusion tensor imaging Introduction These techniques are feasible at 1.5 and 3 T with a clear advantage of 3 T for magnetic resonance angiography (MRA) Advanced MRI techniques applied to the spinal cord have and spectroscopy. remained difficult to put into practice until recently. Until Characterisation of intramedullary lesions is challenging now, advanced imaging techniques of the spine have relied with conventional sequences, and, on numerous occasions, it on significant contributions from MR physicists to apply them is difficult to identify the origin of the lesion, distinguish be- to clinical routine. tween inflammation and ischaemia, and correctly date an isch- aemic lesion as acute or hyperacute. Conversely, advanced sequences allow a much better depiction of the anatomy, such as diffusion tensor imaging (DTI) for pre-surgical planning of spinal tumours or MRA to accurately localise the shunt and * M. I. Vargas nidus of an AVM. maria.i.vargas@hcuge.ch In this article, we illustrate the main advanced techniques, such as cerebrospinal fluid (CSF) flow, diffusion and diffusion Division of Neuroradiology, DISIM, Geneva University Hospitals tensor imaging (DTI), dynamic magnetic resonance and Faculty of Medicine, Rue Gabrielle-Perret-Gentil 4, 1211, Geneva 14, Switzerland angiograpy (MRA), dynamic contrast-enhanced T1-weighted perfusion, susceptibility-weighted imaging (SWI), functional Division of Radiology, DISIM, Geneva University Hospitals, Geneva, Switzerland imaging and spectroscopy. The main technical challenges faced in spine imaging and the clinical applications of these Division of Neuroradiology, Strasbourg University Hospitals, Strasbourg, France techniques in children and adults are also discussed. 550 Insights Imaging (2018) 9:549–557 Advantages and disadvantages of high field the phase encoding gradient is applied in the slice encoding strength direction. These phase-contrast sequences require ECG trig- gering since the blood and CSF flow velocities vary during the The use of high field strength on MRI in brain imaging has cardiac cycle. allowed an increased signal-to-noise ratio (SNR) and de- Flow measurement is performed in the sagittal plane to creased acquisition time. However, the situation is not the visualise in-plane CSF flow. In this case, care has to be taken same with advanced sequences in spine and spinal cord imag- if the phase encoding direction is chosen since cardiac and ing. The main challenges faced in spine imaging relate to the respiratory motion are highly deleterious to image quality small dimensions of the spinal cord (approximately 12 to and flow quantification. 14 mm in diameter), inhomogeneity of B0 on T1 imaging, The velocity encoding gradient in the sequence should artefacts due to the CSF flow, which are more significant on be set between 10 and 20 cm/s and be increased if aliasing the dorsal spine, and also artefacts due to breathing, patient artefacts are observed. Bunck et al. [5]reported10cm/sin motion and swallowing [1]. Magnetic susceptibility artefacts volunteers and 20 cm/s in patients where a specific pathol- are also important because of interfaces between different ogy could lead to increased velocity due to narrowing of structures such as bone, lungs and fat. the CSF canal. Sometimes, high field strength may be a limitation be- Flow can be measured directly in the three spatial direc- cause of the greater effect of susceptibility artefacts (going tions to improve sensitivity to both the in- and through-plane linearly with the field strength). The distortions of the im- velocities and also to determine the flow direction. This se- ages are thus important particularly in the spine where the quence, which provides data in a 3D volume, is promising but proximity with the lungs creates high susceptibility arte- has the disadvantage of long acquisition times (12–14 min facts. On the other hand, high field strength in the DTI of versus 3–5 min for the 2D sequence [5]). the cervical spine causes geometric distortion. This has been partially solved by the new sequence types [readout segmentation of long variable echo train (RESOLVE) Three-dimensional MR T2 high resolution available on Siemens systems or zoomed EPI available on all systems, for example]. Three-dimensional MR T2 high resolution is an isotropic se- For other sequences such as CSF flow and perfusion imag- quence also known as the 3D CISS (constructive interference ing, 1.5- or 3-T MR can be used with the same final result. in steady state), 3D True-FISP (fast imaging with steady-state Additionally, high field strength offers a real added value for precession) or FIESTA (fast imaging employing steady-state MRA and spectroscopy; it produces higher SNRs and faster acquisition) sequence (Fig. 2). acquisition minimising motion artefacts. This sequence allows good depiction of smaller structures because of high spatial resolution, high T2 contrast and its isotropic properties, which permit visualisation in three planes Advanced sequences without distortion. The main clinical indications are: CSF flow – thin septa in post-traumatic cysts The application of this sequence in spinal cord imaging is for – walls of arachnoid cysts or arachnoid webs (Figs. 1 and 2) depicting cystic lesions, such as arachnoid or leptomeningeal – inner structure of cystic tumours cysts (Fig. 1), the latter often resulting from haematomas after – normal and dilated vessels on the surface of the spinal trauma, which breakdown into haemosiderin and its deriva- cord in dural fistulas and AVMs tives and may cause arachnoiditis [2, 3]. CSF flow techniques – post-traumatic pseudomeningoceles are usually coupled with T2 high-resolution sequences, which – dural breach and CSF leak in hypotension of CSF serve the double purpose of helping to better depict lesions – Tarlov cysts and also to generate an anatomical mask. Phase contrast MRI is a unique approach to measuring flow A larger FOV can be obtained in the coronal and sagittal in vivo. It relies on the principle that motion in a voxel induces planes, but axial images are also possible. additional dephasing of the signal. With an appropriate se- Other 3D T2 sequences such as space, CUBE or VISTA, quence providing phase images in addition to magnitude im- which are spin echo sequences, are less sensitive to flow ages, it is possible to measure the flow velocity in a voxel [4]. and susceptibility artefacts [6], allowing clear depiction of A phase encoding gradient can be applied in one of the three the nerve root sheaths. The most common pitfall of these directions of space to detect flow in that specific direction. If sequences is the Gibbs artefact mimicking superficial the flow going through the imaging plane needs to measured, siderosis [6](Fig. 3). Insights Imaging (2018) 9:549–557 551 Fig. 1 Sagittal (a)and axial (b) T2WI shows hyperintensity of the cord (arrow), which is deformed (scalpel sign), suggesting an arachnoid web. Sagittal 3D T2WI (b) better demonstrates the web (arrow). Sagittal flow sequences (d,e) show decreased CSF flow posteriorly to the cord due to the arachnoid web Dynamic MRA thrombus in the vessel lumen, particularly in a small artery such as the AKA. On the other hand, dynamic MRA can be Clinical applications include investigation of vascular very useful in showing a dissection or a partially malformations such as dural fistulas (Fig. 4) and arteriove- thrombosed aneurysm of the aorta. However, it plays no nous malformations and identification of the exact location role in the workup of cavernomas as they are angiograph- of the artery of Adamkiewicz (AKA) for presurgical plan- ically silent. CT is an excellent technique for imaging of ning for tumours to facilitate endovascular treatment. In large vessels, but it does not allow visualisation of spinal case of ischaemia, it is very difficult to visualise the cord ischaemia. Fig. 2 CT scan (a,b,c)performed in a 61-year-old females complaining of back pain after a fall shows several paraspinal lesions (arrows), which are difficult to characterise by this technique. A CISS MR sequence (d) clearly shows these to be extradural cysts. Note that there is no enhancement on T1 FS Gd (e,f,g) 552 Insights Imaging (2018) 9:549–557 Fig. 3 FSE T2 shows an enlarged ependymal canal at the C5 level (white arrow in a). Note that the same abnormality appears larger on the SPACE T2 sequence and also the false image of superficial siderosis on the cervical spinal cord (black arrows in b) Fig. 4 Sagittal T2WI (a) demonstrates multiple dilated vessels with corkscrew appearance surrounding the spinal cord (a); 3D MRA reformat nicely illustrates the arteriovenous shunt at the level of T6 on the right (arrow in b) confirmed by DSA (c) Insights Imaging (2018) 9:549–557 553 The main challenges of dynamic MRA are the small size of 18] because of the lower intracellular water content. This is vascular structures in the spine and the dynamic aspect of this the main reason why b500 or b900 is used in spinal imaging sequence. These are also the main reasons why high field and not b1000. strength is useful because of its higher SNR. DWI and DTI are challenging techniques in spinal imaging There are two techniques for performing dynamic MRA: for several reasons, including the small size of the cord rela- dynamic sequences with a temporal resolution of approxi- tive to the brain and respiratory and cardiac motion artefacts. mately 1 min (3 phases: arterial, venous and delayed, and an Therefore, spine diffusion imaging requires high spatial reso- additional later acquisition at high spatial resolution) [7]and lution, which should be combined with distortion reduction 4D imaging [8]. techniques and homogeneous fat saturation. These goals are The first relies on a 3D gradient echo T1-weighted se- difficult to achieve with the broadly used single-shot spin echo quence with a field of view comprising the descending aorta EPI diffusion sequence, especially when image acquisition is as well as the spine in the sagittal plane. The angiography in the sagittal plane, which is preferred for the evaluation of technique uses a noncontrast image that is subsequently the spine. Specific aspects are (1) fat saturation, (2) imaging subtracted from the arterial phase image. An important aspect distortion and (3) b-values and directions. is the timing of the imaging. The arterial bolus remains in the arteries for a short period of time; thus, imaging should be (1) Conventional fat saturation with spectral selection of the done at a precise time to eliminate venous contamination. fat peak based on CHESS (chemical shift selective) has This timing depends on the injection rate and also on the the advantage of being fast but often delivers poor results cardiovascular status of the patient. in spine imaging. In dorsal areas, an inversion recovery The utilisation of blood pool agents or doubly concentrated technique, such as STIR (short tau inversion recovery), is contrast media may nevertheless facilitate image acquisition more robust in eliminating the signal from fat and im- and subsequent analysis. proving image quality. However, this causes a reduction The other technique for MRA relies on a 4D sequence also in the signal due to the inversion pulse. A compromise is known as time-resolved angiography with stochastic trajecto- to use SPAIR (spectral attenuated inversion recovery) ries (TWIST) [9] or 4D time-resolved MR angiography with a preparation, which shows relatively robust saturation keyhole (4D-TRAK) with a spatial resolution of 1 mm and provided the shim box is placed correctly in the area of temporal resolution of approximately 1.3 s. interest avoiding the lungs. Diffusion and diffusion tensor imaging A recently available method to suppress the fat signal is the Dixon technique [19]. It relies on the principle that water and fat do not precess at the exact same frequency and that they Acute ischaemia is one of the main clinical indications for DWI, is seen as high signal on trace images and decreased ADC can be either in or out of phase after the preselected time to without enhancement (Fig. 5), which only appears in the sub- echo. This technique provides very homogeneous fat- acute phase [10]. The main causes in adults are atherosclerosis, saturated images on large fields of view. cardiac surgery and minimally invasive procedures, compres- sion of the radicular artery by a disc [11]and minortraumato (2) Diffusion imaging uses a single-shot EPI sequence. the cervical spine in the setting of degenerative changes. Throughout the long echo train, phase errors will ac- In children, minor trauma is a cause of ischaemia related to cumulate, resulting in spatial mismatch in the recon- fibrocartilage emboli [12] (Fig. 5) and also arterial spasm. structed image. The longer the echo train and the Other causes include traction for scoliosis after orthopaedic higher the resolution, the more pronounced the distor- surgery [13], complications of cardiac surgery, sickle cell tions will be. Distortions will also be enhanced be- anaemia and umbilical artery catheter in the neonate. cause of susceptibility differences between different DWI is also used to differentiate between spondylodiscitis spinal tissues (bone, intervertebral discs, cerebrospi- and inflammatory degenerative changes [14]. FA and ADC nal fluid, etc.). values may be used to predict gain of function in patients with cervical spondylotic myelopathy after decompressive surgery Reducing the readout bandwidth minimises this distortion. [15]. To achieve this, parallel imaging can be used together with a DTI tractography is used for pre-surgical planning of rectangular field of view. Another option is to choose a trans- tumours [16](Fig. 6) as the generated cartography is the verse orientation with an isotropic voxel resolution. Another only method allowing the neurosurgeon to visualise the alternative to this problem is segmentation of the EPI readout tracts in vivo [17]. in either the phase or readout direction. A more detailed ex- Unlike in the brain, diffusion of water molecules in the planation of distortion reduction in spine diffusion imaging can be found in: [20–22]. spinal cord occurs mainly in the cranio-caudal direction [8, 554 Insights Imaging (2018) 9:549–557 Fig. 5 A 14-year-old male who suffered minor trauma. Spine MR performed 24 h later shows increased signal intensity on T2WI (a,d)(arrows) and DWI (e) reflecting ischaemia. Note an acute wedge fracture of T12 (*) (3) b = 500 s/mm is often chosen at 1.5 T since it produces a generate DTI parameters such as FA or MD is six, a more sufficient SNR to allow satisfactory interpretation of the reasonable value would be around 20. images without being too low, and thus too sensitive, to perfusion effects [23]. This value can be increased at 3 T Dynamic contrast-enhanced T1-weighted perfusion because of the higher SNR. Dynamic contrast enhancement (DCE) is a technique that al- For DTI, the optimal number of diffusion directions varies lows dynamic visualisation of contrast behaviour in tissues. It depending on the authors [24], a higher number being often is the technique of choice to assess microvascularisation, in preferred. While the minimum number of directions to particular in the context of tumour growth, because it provides Fig. 6 Patient with multiple myeloma and several vertebral fractures, medullaris (b,c,f,g). On tractography (d,e), destruction of the fibers of treated by vertebroplasty. No enlargement or enhancement of the conus the conus medullaris can be seen; this translates into a secondary lesion medullaris is visible (a). Spine MR performed approximately 7 months with rapid growth later shows hyperintensity and nodular enhancement of the conus Insights Imaging (2018) 9:549–557 555 information about the tumour vasculature and the effects of Susceptibility weighted imaging (SWI) treatment (Fig. 7). This technique is used in brain imaging for tumour characterisation and for distinguishing between SWI is a sequence based on the magnetic susceptibility differ- radionecrosis and true tumour progression. In spine imaging, ences between tissues. Reconstructions of magnitude and DCEisusedtocharacterisetumours andtoevaluate phase images are possible. The acquisition of this sequence extradural spinal metastases and their vascularisation [25], does not need the administration of contrast medium. which in turn helps in the selection of patients amenable to SWI is principally used at the level of the brain to detect endovascular treatment. micro-haemorrhages, calcifications, iron and deoxy-Hb. The goal of DCE is to quantify tissue permeability with the Concerning the spine, few articles exist concerning the use use of specific models such as the Tofts model or equivalent of this sequence at 1.5 T. In our opinion, its use is possible [26]. This two-compartment model provides physiologically but has not been extended to daily clinical practice because of relevant parameters such as the K [volume transfer con- limitations in spatial resolution and multiple artefacts due to trans stant between blood plasma and extravascular extracellular phase-encoding directions, bone-tissue interfaces, flow and space (EES)], K (rate constant between EES and blood plas- increased noise. This sequence is particularly sensitive to sub- ep ma) and V (volume of EES per unit volume of tissue, i.e., the tle changes of the local magnetic susceptibility variances, de- volume fraction of the EES). creased signal-to-noise ratios, etc. [30]. To generate these parameters, imaging should be per- The possibly clinical indications are visualisation of normal formed at relatively high temporal resolution (between 2 to venous anatomy [30], identification of haemorrhage, evalua- 15 s) and over 5 to 10 min post administration of contrast tion of the efficiency of treatment of spinal arteriovenous media. Using a 3D-T1 spoiled gradient recalled echo sequence malformations and evaluation of changes in venous oxygena- to dynamically image the contrast arrival and washout is rec- tion [31] with SW phase imaging. ommended [27]. This sequence is very sensitive to T1 varia- tions and is fast enough to produce a suitable temporal reso- Spectroscopy lution while maintaining a sufficient SNR. For the modelling, it is necessary to convert the signal intensity curve into a Gd Spectroscopy shows the concentration of normal metabolites concentration curve, which can only be done with knowledge in a specific anatomic location and changes in those metabo- of the T1 values before contrast injection. Usually, the two flip lites in case of pathology. Few works have shown the feasi- angle method is chosen because it is fast and reliable. This bility of spectroscopy in spinal cord imaging. Spectroscopy technique has been more widely used in the brain but has also has, nevertheless, been used to characterise and differenti- shown promising results in spine imaging, such as preclinical ate tumours [32] from inflammatory pathologies, in amyo- research in spinal cord injury assessment [28, 29]. trophic lateral sclerosis or in the follow-up of cervical Fig. 7 A patient with spinal cord glioblastoma. a Sagittal T1 post- curve) in the mass but also in the posterior paraspinal soft tissues gadolinium MR image shows an enhancing intramedullary mass. b reflecting contrast extravasation due to postoperative changes. This case Corrected Vp (volumetric plasma volume) parameter of T1 perfusion is illustrates that T1 perfusion, particularly corrected Vp, allows detection of increased in the mass. c and d Increased K and AUC (area under the true tumour hypervascularisation trans 556 Insights Imaging (2018) 9:549–557 creativecommons.org/licenses/by/4.0/), which permits unrestricted use, spondylotic myelopathy [33]. Furthermore, Holly et al. distribution, and reproduction in any medium, provided you give appro- [33] showed that spectroscopy may be of value in priate credit to the original author(s) and the source, provide a link to the predicting neurological outcome in patients with cervical Creative Commons license, and indicate if changes were made. spondylotic myelopathy after surgery. Spectroscopy of the spinal cord presents a real challenge because of the small dimensions of the cord, flow artefacts, motion artefacts during the cardiac and respiratory cycles, the References deep anatomical location [34] and B0 inhomogeneity, which is particularly deleterious to spectroscopy of the spine. For this 1. Vargas MI, Delavelle J, Kohler R, Becker CD, Lovblad K (2009) reason, B0 shimming is essential. Saturation bands and pulse Brain and spine MRI artifacts at 3Tesla. J Neuroradiol 36:74–81 2. Caremel R, Hamel O, Gerardin E et al (2013) Post-traumatic syrin- triggering [32] are used to reduce CSF flow artefacts. gomyelia: what should know the urologist? Prog Urol 23:8–14 The sequence used is PRESS (point-resolved spectrosco- 3. Fehlings MG, Austin JW (2011) Posttraumatic syringomyelia. J py), which produces better results with a long TE of around Neurosurg Spine 14:570–572 discussion 572 135 or 280 ms than with short TE below 40 ms. Currently, this 4. Levy LM, Di Chiro G (1990) MR phase imaging and cerebrospinal technique is used exclusively in research. fluid flow in the head and spine. Neuroradiology 32:399–406 5. Bunck AC, Kroger JR, Juttner A et al (2011) Magnetic resonance 4D flow characteristics of cerebrospinal fluid at the craniocervical junction and the cervical spinal canal. Eur Radiol 21:1788–1796 Functional imaging (fMRI) 6. Dietemann JL, Bogorin A, Abu Eid M et al (2012) Tips and traps in neurological imaging: imaging the perimedullary spaces. Diagn Interv Imaging 93:985–992 fMRI provides information concerning spinal cord motor 7. Vargas MI, Nguyen D, Viallon M et al (2010) Dynamic MR angi- function. The sequence used is essentially the same as that ography (MRA) of spinal vascular diseases at 3T. Eur Radiol 20: on the brain; however, modifications are necessary because 2491–2495 of the size of the spinal cord. Spinal fMRI illustrates neuronal 8. Wheeler-Kingshott CA, Hickman SJ, Parker GJ et al (2002) function indirectly by changes in blood flow and blood oxy- Investigating cervical spinal cord structure using axial diffusion tensor imaging. NeuroImage 16:93–102 gen levels (potential clinical indication of fMRI) [35]. 9. Amarouche M, Hart JL, Siddiqui A, Hampton T, Walsh DC (2015) This tool is only used for research purposes. Time-resolved contrast-enhanced MR angiography of spinal vascu- The potential clinical applications are: to determine pre- lar malformations. AJNR Am J Neuroradiol 36:417–422 served motor function in patients with spinal injury, to plan 10. Vargas MI, Gariani J, Sztajzel R et al (2015) Spinal cord ischemia: treatment or to evaluate the treatment response of tumours to practical imaging tips, pearls, and pitfalls. AJNR Am J Neuroradiol 36:825–830 understand the physiopathology of cervical headaches and to 11. Hirsch E, Vautravers P, Dietemann JL, Vetter D, Jesel M (1986) monitor functional changes in diseases such as multiple scle- Acute lumbar spinal cord disease caused by lumbar disk hernia. rosis and amyotrophic lateral sclerosis [35]. Presse Med 15:843–844 12. Reisner A, Gary MF, Chern JJ, Grattan-Smith JD (2013) Spinal Funding The authors state that this work has not received any funding. cord infarction following minor trauma in children: fibrocartilaginous embolism as a putative cause. J Neurosurg Pediatr 11:445–450 Compliance with ethical standards 13. Lewis SJ, Gray R, Holmes LM et al (2011) Neurophysiological changes in deformity correction of adolescent idiopathic scoliosis Guarantor The scientific guarantor of this publication is Maria Isabel with intraoperative skull-femoral traction. Spine (Phila Pa 1976) 36: Vargas. 1627–1638 14. Patel KB, Poplawski MM, Pawha PS, Naidich TP, Tanenbaum LN Conflict of interest The authors of this manuscript declare no relation- (2014) Diffusion-weighted MRI Bclaw sign^ improves differentia- ships with any companies, whose products or services may be related to tion of infectious from degenerative modic type 1 signal changes of the subject matter of the article. the spine. AJNR Am J Neuroradiol 35:1647–1652 15. Tessitore E, Broc N, Mekideche A et al (2016) A modern multidis- Statistics and biometry No complex statistical methods were necessary ciplinary approach to patients suffering from cervical spondylotic for this paper. myelopathy. J Neurosurg Sci. https://www.ncbi.nlm.nih.gov/ pubmed/27175620 Informed consent Written informed consent was waived by the 16. Vargas MI, Delavelle J, Jlassi H et al (2008) Clinical applications of Institutional Review Board. diffusion tensor tractography of the spinal cord. Neuroradiology 50: 25–29 Ethical approval Institutional Review Board approval was obtained. 17. Kurzbuch AR, Rilliet B, Vargas MI, Boex C, Tessitore E (2010) Coincidence of cervical spondylotic myelopathy and intramedullary ependymoma: a potential diagnostic pitfall. J Methodology Review article Neurosurg Spine 12:249–252 18. Basser PJ, Mattiello J, LeBihan D (1994) MR diffusion tensor Open Access This article is distributed under the terms of the Creative spectroscopy and imaging. Biophys J 66:259–267 Commons Attribution 4.0 International License (http:// Insights Imaging (2018) 9:549–557 557 19. Dixon WT (1984) Simple proton spectroscopic imaging. Radiology 29. Tatar I, Chou PC, Desouki MM, El Sayed H, Bilgen M (2009) Evaluating regional blood spinal cord barrier dysfunction following 153:189–194 20. Le Bihan D, Poupon C, Amadon A, Lethimonnier F (2006) spinal cord injury using longitudinal dynamic contrast-enhanced Artifacts and pitfalls in diffusion MRI. J Magn Reson Imaging MRI. BMC Med Imaging 9:10 24:478–488 30. Ishizaka K, Kudo K, Fujima N et al (2010) Detection of normal 21. Bammer R, Augustin M, Prokesch RW, Stollberger R, Fazekas F spinal veins by using susceptibility-weighted imaging. J Magn (2002) Diffusion-weighted imaging of the spinal cord: interleaved Reson Imaging 31:32–38 echo-planar imaging is superior to fast spin-echo. J Magn Reson 31. Fujima N, Kudo K, Terae S et al (2010) Spinal arteriovenous mal- Imaging 15:364–373 formation: evaluation of change in venous oxygenation with 22. Andre JB, Bammer R (2010) Advanced diffusion-weighted mag- susceptibility-weighted MR imaging after treatment. Radiology netic resonance imaging techniques of the human spinal cord. Top 254:891–899 Magn Reson Imaging 21:367–378 32. Hock A, Henning A, Boesiger P, Kollias SS (2013) (1)H-MR spec- 23. Tanenbaum LN (2013) Clinical applications of diffusion imaging in troscopy in the human spinal cord. AJNR Am J Neuroradiol 34: the spine. Magn Reson Imaging Clin N Am 21:299–320 1682–1689 24. Jones DK, Horsfield MA, Simmons A (1999) Optimal strategies for 33. Holly LT, Ellingson BM, Salamon N (2015) Metabolic imaging measuring diffusion in anisotropic systems by magnetic resonance using proton magnetic spectroscopy as a predictor of outcome fol- imaging. Magn Reson Med 42:515–525 lowing surgery for cervical spondylotic myelopathy. J Spinal 25. Mazura JC, Karimi S, Pauliah M et al (2014) Dynamic contrast- Disord Tech. https://doi.org/10.1097/BSD.0000000000000248 enhanced magnetic resonance perfusion compared with digital sub- 34. Carew JD, Nair G, Pineda-Alonso N, Usher S, Hu X, Benatar M traction angiography for the evaluation of extradural spinal metas- (2011) Magnetic resonance spectroscopy of the cervical cord in tases: a pilot study. Spine (Phila Pa 1976) 39:E950–E954 amyotrophic lateral sclerosis. Amyotroph Lateral Scler 12:185–191 26. Tofts PS, Brix G, Buckley DL et al (1999) Estimating kinetic pa- 35. Kornelsen J, Mackey S (2007) Potential clinical applications for rameters from dynamic contrast-enhanced T(1)-weighted MRI of a spinal functional MRI. Curr Pain Headache Rep 11:165–170 diffusable tracer: standardized quantities and symbols. J Magn Reson Imaging 10:223–232 27. Leach MO, Morgan B, Tofts PS et al (2012) Imaging vascular function for early stage clinical trials using dynamic contrast- Publisher’sNote enhanced magnetic resonance imaging. Eur Radiol 22:1451–1464 28. Bilgen M, Narayana PA (2001) A pharmacokinetic model for quan- Springer Nature remains neutral with regard to jurisdictional claims in titative evaluation of spinal cord injury with dynamic contrast- published maps and institutional affiliations. enhanced magnetic resonance imaging. Magn Reson Med 46: 1099–1106 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Insights into Imaging Springer Journals

Advanced magnetic resonance imaging (MRI) techniques of the spine and spinal cord in children and adults

Free
9 pages

Loading next page...
 
/lp/springer_journal/advanced-magnetic-resonance-imaging-mri-techniques-of-the-spine-and-TA4wq94WNe
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Medicine & Public Health; Imaging / Radiology; Diagnostic Radiology; Interventional Radiology; Neuroradiology; Ultrasound; Internal Medicine
eISSN
1869-4101
D.O.I.
10.1007/s13244-018-0626-1
Publisher site
See Article on Publisher Site

Abstract

In this article, we illustrate the main advanced magnetic resonance imaging (MRI) techniques used for imaging of the spine and spinal cord in children and adults. This work focuses on daily clinical practice and aims to address the most common questions and needs of radiologists. We will also provide tips to solve common problems with which we were confronted. The main clinical indications for each MR technique, possible pitfalls and the challenges faced in spine imaging because of anatomical and physical constraints will be discussed. The major advanced MRI techniques dealt with in this article are CSF, (cerebrosopinal fluid) flow, diffusion, diffusion tensor imaging (DTI), MRA, dynamic contrast-enhanced T1-weighted perfusion, MR angiography, susceptibility-weighted imaging (SWI), functional imaging (fMRI) and spectroscopy. Teaching Points � DWI is essential to diagnose cord ischaemia in the acute stage. � MRA is useful to guide surgical planning or endovascular embolisation of AVMs. � Three Tesla is superior to 1.5 T for spine MR angiography and spectroscopy. � Advanced sequences should only be used together with conventional morphological sequences. . . . . Keywords Spinal cord Perfusion Spectroscopy Magnetic resonance angiography Diffusion tensor imaging Introduction These techniques are feasible at 1.5 and 3 T with a clear advantage of 3 T for magnetic resonance angiography (MRA) Advanced MRI techniques applied to the spinal cord have and spectroscopy. remained difficult to put into practice until recently. Until Characterisation of intramedullary lesions is challenging now, advanced imaging techniques of the spine have relied with conventional sequences, and, on numerous occasions, it on significant contributions from MR physicists to apply them is difficult to identify the origin of the lesion, distinguish be- to clinical routine. tween inflammation and ischaemia, and correctly date an isch- aemic lesion as acute or hyperacute. Conversely, advanced sequences allow a much better depiction of the anatomy, such as diffusion tensor imaging (DTI) for pre-surgical planning of spinal tumours or MRA to accurately localise the shunt and * M. I. Vargas nidus of an AVM. maria.i.vargas@hcuge.ch In this article, we illustrate the main advanced techniques, such as cerebrospinal fluid (CSF) flow, diffusion and diffusion Division of Neuroradiology, DISIM, Geneva University Hospitals tensor imaging (DTI), dynamic magnetic resonance and Faculty of Medicine, Rue Gabrielle-Perret-Gentil 4, 1211, Geneva 14, Switzerland angiograpy (MRA), dynamic contrast-enhanced T1-weighted perfusion, susceptibility-weighted imaging (SWI), functional Division of Radiology, DISIM, Geneva University Hospitals, Geneva, Switzerland imaging and spectroscopy. The main technical challenges faced in spine imaging and the clinical applications of these Division of Neuroradiology, Strasbourg University Hospitals, Strasbourg, France techniques in children and adults are also discussed. 550 Insights Imaging (2018) 9:549–557 Advantages and disadvantages of high field the phase encoding gradient is applied in the slice encoding strength direction. These phase-contrast sequences require ECG trig- gering since the blood and CSF flow velocities vary during the The use of high field strength on MRI in brain imaging has cardiac cycle. allowed an increased signal-to-noise ratio (SNR) and de- Flow measurement is performed in the sagittal plane to creased acquisition time. However, the situation is not the visualise in-plane CSF flow. In this case, care has to be taken same with advanced sequences in spine and spinal cord imag- if the phase encoding direction is chosen since cardiac and ing. The main challenges faced in spine imaging relate to the respiratory motion are highly deleterious to image quality small dimensions of the spinal cord (approximately 12 to and flow quantification. 14 mm in diameter), inhomogeneity of B0 on T1 imaging, The velocity encoding gradient in the sequence should artefacts due to the CSF flow, which are more significant on be set between 10 and 20 cm/s and be increased if aliasing the dorsal spine, and also artefacts due to breathing, patient artefacts are observed. Bunck et al. [5]reported10cm/sin motion and swallowing [1]. Magnetic susceptibility artefacts volunteers and 20 cm/s in patients where a specific pathol- are also important because of interfaces between different ogy could lead to increased velocity due to narrowing of structures such as bone, lungs and fat. the CSF canal. Sometimes, high field strength may be a limitation be- Flow can be measured directly in the three spatial direc- cause of the greater effect of susceptibility artefacts (going tions to improve sensitivity to both the in- and through-plane linearly with the field strength). The distortions of the im- velocities and also to determine the flow direction. This se- ages are thus important particularly in the spine where the quence, which provides data in a 3D volume, is promising but proximity with the lungs creates high susceptibility arte- has the disadvantage of long acquisition times (12–14 min facts. On the other hand, high field strength in the DTI of versus 3–5 min for the 2D sequence [5]). the cervical spine causes geometric distortion. This has been partially solved by the new sequence types [readout segmentation of long variable echo train (RESOLVE) Three-dimensional MR T2 high resolution available on Siemens systems or zoomed EPI available on all systems, for example]. Three-dimensional MR T2 high resolution is an isotropic se- For other sequences such as CSF flow and perfusion imag- quence also known as the 3D CISS (constructive interference ing, 1.5- or 3-T MR can be used with the same final result. in steady state), 3D True-FISP (fast imaging with steady-state Additionally, high field strength offers a real added value for precession) or FIESTA (fast imaging employing steady-state MRA and spectroscopy; it produces higher SNRs and faster acquisition) sequence (Fig. 2). acquisition minimising motion artefacts. This sequence allows good depiction of smaller structures because of high spatial resolution, high T2 contrast and its isotropic properties, which permit visualisation in three planes Advanced sequences without distortion. The main clinical indications are: CSF flow – thin septa in post-traumatic cysts The application of this sequence in spinal cord imaging is for – walls of arachnoid cysts or arachnoid webs (Figs. 1 and 2) depicting cystic lesions, such as arachnoid or leptomeningeal – inner structure of cystic tumours cysts (Fig. 1), the latter often resulting from haematomas after – normal and dilated vessels on the surface of the spinal trauma, which breakdown into haemosiderin and its deriva- cord in dural fistulas and AVMs tives and may cause arachnoiditis [2, 3]. CSF flow techniques – post-traumatic pseudomeningoceles are usually coupled with T2 high-resolution sequences, which – dural breach and CSF leak in hypotension of CSF serve the double purpose of helping to better depict lesions – Tarlov cysts and also to generate an anatomical mask. Phase contrast MRI is a unique approach to measuring flow A larger FOV can be obtained in the coronal and sagittal in vivo. It relies on the principle that motion in a voxel induces planes, but axial images are also possible. additional dephasing of the signal. With an appropriate se- Other 3D T2 sequences such as space, CUBE or VISTA, quence providing phase images in addition to magnitude im- which are spin echo sequences, are less sensitive to flow ages, it is possible to measure the flow velocity in a voxel [4]. and susceptibility artefacts [6], allowing clear depiction of A phase encoding gradient can be applied in one of the three the nerve root sheaths. The most common pitfall of these directions of space to detect flow in that specific direction. If sequences is the Gibbs artefact mimicking superficial the flow going through the imaging plane needs to measured, siderosis [6](Fig. 3). Insights Imaging (2018) 9:549–557 551 Fig. 1 Sagittal (a)and axial (b) T2WI shows hyperintensity of the cord (arrow), which is deformed (scalpel sign), suggesting an arachnoid web. Sagittal 3D T2WI (b) better demonstrates the web (arrow). Sagittal flow sequences (d,e) show decreased CSF flow posteriorly to the cord due to the arachnoid web Dynamic MRA thrombus in the vessel lumen, particularly in a small artery such as the AKA. On the other hand, dynamic MRA can be Clinical applications include investigation of vascular very useful in showing a dissection or a partially malformations such as dural fistulas (Fig. 4) and arteriove- thrombosed aneurysm of the aorta. However, it plays no nous malformations and identification of the exact location role in the workup of cavernomas as they are angiograph- of the artery of Adamkiewicz (AKA) for presurgical plan- ically silent. CT is an excellent technique for imaging of ning for tumours to facilitate endovascular treatment. In large vessels, but it does not allow visualisation of spinal case of ischaemia, it is very difficult to visualise the cord ischaemia. Fig. 2 CT scan (a,b,c)performed in a 61-year-old females complaining of back pain after a fall shows several paraspinal lesions (arrows), which are difficult to characterise by this technique. A CISS MR sequence (d) clearly shows these to be extradural cysts. Note that there is no enhancement on T1 FS Gd (e,f,g) 552 Insights Imaging (2018) 9:549–557 Fig. 3 FSE T2 shows an enlarged ependymal canal at the C5 level (white arrow in a). Note that the same abnormality appears larger on the SPACE T2 sequence and also the false image of superficial siderosis on the cervical spinal cord (black arrows in b) Fig. 4 Sagittal T2WI (a) demonstrates multiple dilated vessels with corkscrew appearance surrounding the spinal cord (a); 3D MRA reformat nicely illustrates the arteriovenous shunt at the level of T6 on the right (arrow in b) confirmed by DSA (c) Insights Imaging (2018) 9:549–557 553 The main challenges of dynamic MRA are the small size of 18] because of the lower intracellular water content. This is vascular structures in the spine and the dynamic aspect of this the main reason why b500 or b900 is used in spinal imaging sequence. These are also the main reasons why high field and not b1000. strength is useful because of its higher SNR. DWI and DTI are challenging techniques in spinal imaging There are two techniques for performing dynamic MRA: for several reasons, including the small size of the cord rela- dynamic sequences with a temporal resolution of approxi- tive to the brain and respiratory and cardiac motion artefacts. mately 1 min (3 phases: arterial, venous and delayed, and an Therefore, spine diffusion imaging requires high spatial reso- additional later acquisition at high spatial resolution) [7]and lution, which should be combined with distortion reduction 4D imaging [8]. techniques and homogeneous fat saturation. These goals are The first relies on a 3D gradient echo T1-weighted se- difficult to achieve with the broadly used single-shot spin echo quence with a field of view comprising the descending aorta EPI diffusion sequence, especially when image acquisition is as well as the spine in the sagittal plane. The angiography in the sagittal plane, which is preferred for the evaluation of technique uses a noncontrast image that is subsequently the spine. Specific aspects are (1) fat saturation, (2) imaging subtracted from the arterial phase image. An important aspect distortion and (3) b-values and directions. is the timing of the imaging. The arterial bolus remains in the arteries for a short period of time; thus, imaging should be (1) Conventional fat saturation with spectral selection of the done at a precise time to eliminate venous contamination. fat peak based on CHESS (chemical shift selective) has This timing depends on the injection rate and also on the the advantage of being fast but often delivers poor results cardiovascular status of the patient. in spine imaging. In dorsal areas, an inversion recovery The utilisation of blood pool agents or doubly concentrated technique, such as STIR (short tau inversion recovery), is contrast media may nevertheless facilitate image acquisition more robust in eliminating the signal from fat and im- and subsequent analysis. proving image quality. However, this causes a reduction The other technique for MRA relies on a 4D sequence also in the signal due to the inversion pulse. A compromise is known as time-resolved angiography with stochastic trajecto- to use SPAIR (spectral attenuated inversion recovery) ries (TWIST) [9] or 4D time-resolved MR angiography with a preparation, which shows relatively robust saturation keyhole (4D-TRAK) with a spatial resolution of 1 mm and provided the shim box is placed correctly in the area of temporal resolution of approximately 1.3 s. interest avoiding the lungs. Diffusion and diffusion tensor imaging A recently available method to suppress the fat signal is the Dixon technique [19]. It relies on the principle that water and fat do not precess at the exact same frequency and that they Acute ischaemia is one of the main clinical indications for DWI, is seen as high signal on trace images and decreased ADC can be either in or out of phase after the preselected time to without enhancement (Fig. 5), which only appears in the sub- echo. This technique provides very homogeneous fat- acute phase [10]. The main causes in adults are atherosclerosis, saturated images on large fields of view. cardiac surgery and minimally invasive procedures, compres- sion of the radicular artery by a disc [11]and minortraumato (2) Diffusion imaging uses a single-shot EPI sequence. the cervical spine in the setting of degenerative changes. Throughout the long echo train, phase errors will ac- In children, minor trauma is a cause of ischaemia related to cumulate, resulting in spatial mismatch in the recon- fibrocartilage emboli [12] (Fig. 5) and also arterial spasm. structed image. The longer the echo train and the Other causes include traction for scoliosis after orthopaedic higher the resolution, the more pronounced the distor- surgery [13], complications of cardiac surgery, sickle cell tions will be. Distortions will also be enhanced be- anaemia and umbilical artery catheter in the neonate. cause of susceptibility differences between different DWI is also used to differentiate between spondylodiscitis spinal tissues (bone, intervertebral discs, cerebrospi- and inflammatory degenerative changes [14]. FA and ADC nal fluid, etc.). values may be used to predict gain of function in patients with cervical spondylotic myelopathy after decompressive surgery Reducing the readout bandwidth minimises this distortion. [15]. To achieve this, parallel imaging can be used together with a DTI tractography is used for pre-surgical planning of rectangular field of view. Another option is to choose a trans- tumours [16](Fig. 6) as the generated cartography is the verse orientation with an isotropic voxel resolution. Another only method allowing the neurosurgeon to visualise the alternative to this problem is segmentation of the EPI readout tracts in vivo [17]. in either the phase or readout direction. A more detailed ex- Unlike in the brain, diffusion of water molecules in the planation of distortion reduction in spine diffusion imaging can be found in: [20–22]. spinal cord occurs mainly in the cranio-caudal direction [8, 554 Insights Imaging (2018) 9:549–557 Fig. 5 A 14-year-old male who suffered minor trauma. Spine MR performed 24 h later shows increased signal intensity on T2WI (a,d)(arrows) and DWI (e) reflecting ischaemia. Note an acute wedge fracture of T12 (*) (3) b = 500 s/mm is often chosen at 1.5 T since it produces a generate DTI parameters such as FA or MD is six, a more sufficient SNR to allow satisfactory interpretation of the reasonable value would be around 20. images without being too low, and thus too sensitive, to perfusion effects [23]. This value can be increased at 3 T Dynamic contrast-enhanced T1-weighted perfusion because of the higher SNR. Dynamic contrast enhancement (DCE) is a technique that al- For DTI, the optimal number of diffusion directions varies lows dynamic visualisation of contrast behaviour in tissues. It depending on the authors [24], a higher number being often is the technique of choice to assess microvascularisation, in preferred. While the minimum number of directions to particular in the context of tumour growth, because it provides Fig. 6 Patient with multiple myeloma and several vertebral fractures, medullaris (b,c,f,g). On tractography (d,e), destruction of the fibers of treated by vertebroplasty. No enlargement or enhancement of the conus the conus medullaris can be seen; this translates into a secondary lesion medullaris is visible (a). Spine MR performed approximately 7 months with rapid growth later shows hyperintensity and nodular enhancement of the conus Insights Imaging (2018) 9:549–557 555 information about the tumour vasculature and the effects of Susceptibility weighted imaging (SWI) treatment (Fig. 7). This technique is used in brain imaging for tumour characterisation and for distinguishing between SWI is a sequence based on the magnetic susceptibility differ- radionecrosis and true tumour progression. In spine imaging, ences between tissues. Reconstructions of magnitude and DCEisusedtocharacterisetumours andtoevaluate phase images are possible. The acquisition of this sequence extradural spinal metastases and their vascularisation [25], does not need the administration of contrast medium. which in turn helps in the selection of patients amenable to SWI is principally used at the level of the brain to detect endovascular treatment. micro-haemorrhages, calcifications, iron and deoxy-Hb. The goal of DCE is to quantify tissue permeability with the Concerning the spine, few articles exist concerning the use use of specific models such as the Tofts model or equivalent of this sequence at 1.5 T. In our opinion, its use is possible [26]. This two-compartment model provides physiologically but has not been extended to daily clinical practice because of relevant parameters such as the K [volume transfer con- limitations in spatial resolution and multiple artefacts due to trans stant between blood plasma and extravascular extracellular phase-encoding directions, bone-tissue interfaces, flow and space (EES)], K (rate constant between EES and blood plas- increased noise. This sequence is particularly sensitive to sub- ep ma) and V (volume of EES per unit volume of tissue, i.e., the tle changes of the local magnetic susceptibility variances, de- volume fraction of the EES). creased signal-to-noise ratios, etc. [30]. To generate these parameters, imaging should be per- The possibly clinical indications are visualisation of normal formed at relatively high temporal resolution (between 2 to venous anatomy [30], identification of haemorrhage, evalua- 15 s) and over 5 to 10 min post administration of contrast tion of the efficiency of treatment of spinal arteriovenous media. Using a 3D-T1 spoiled gradient recalled echo sequence malformations and evaluation of changes in venous oxygena- to dynamically image the contrast arrival and washout is rec- tion [31] with SW phase imaging. ommended [27]. This sequence is very sensitive to T1 varia- tions and is fast enough to produce a suitable temporal reso- Spectroscopy lution while maintaining a sufficient SNR. For the modelling, it is necessary to convert the signal intensity curve into a Gd Spectroscopy shows the concentration of normal metabolites concentration curve, which can only be done with knowledge in a specific anatomic location and changes in those metabo- of the T1 values before contrast injection. Usually, the two flip lites in case of pathology. Few works have shown the feasi- angle method is chosen because it is fast and reliable. This bility of spectroscopy in spinal cord imaging. Spectroscopy technique has been more widely used in the brain but has also has, nevertheless, been used to characterise and differenti- shown promising results in spine imaging, such as preclinical ate tumours [32] from inflammatory pathologies, in amyo- research in spinal cord injury assessment [28, 29]. trophic lateral sclerosis or in the follow-up of cervical Fig. 7 A patient with spinal cord glioblastoma. a Sagittal T1 post- curve) in the mass but also in the posterior paraspinal soft tissues gadolinium MR image shows an enhancing intramedullary mass. b reflecting contrast extravasation due to postoperative changes. This case Corrected Vp (volumetric plasma volume) parameter of T1 perfusion is illustrates that T1 perfusion, particularly corrected Vp, allows detection of increased in the mass. c and d Increased K and AUC (area under the true tumour hypervascularisation trans 556 Insights Imaging (2018) 9:549–557 creativecommons.org/licenses/by/4.0/), which permits unrestricted use, spondylotic myelopathy [33]. Furthermore, Holly et al. distribution, and reproduction in any medium, provided you give appro- [33] showed that spectroscopy may be of value in priate credit to the original author(s) and the source, provide a link to the predicting neurological outcome in patients with cervical Creative Commons license, and indicate if changes were made. spondylotic myelopathy after surgery. Spectroscopy of the spinal cord presents a real challenge because of the small dimensions of the cord, flow artefacts, motion artefacts during the cardiac and respiratory cycles, the References deep anatomical location [34] and B0 inhomogeneity, which is particularly deleterious to spectroscopy of the spine. For this 1. Vargas MI, Delavelle J, Kohler R, Becker CD, Lovblad K (2009) reason, B0 shimming is essential. Saturation bands and pulse Brain and spine MRI artifacts at 3Tesla. J Neuroradiol 36:74–81 2. Caremel R, Hamel O, Gerardin E et al (2013) Post-traumatic syrin- triggering [32] are used to reduce CSF flow artefacts. gomyelia: what should know the urologist? Prog Urol 23:8–14 The sequence used is PRESS (point-resolved spectrosco- 3. Fehlings MG, Austin JW (2011) Posttraumatic syringomyelia. J py), which produces better results with a long TE of around Neurosurg Spine 14:570–572 discussion 572 135 or 280 ms than with short TE below 40 ms. Currently, this 4. Levy LM, Di Chiro G (1990) MR phase imaging and cerebrospinal technique is used exclusively in research. fluid flow in the head and spine. Neuroradiology 32:399–406 5. Bunck AC, Kroger JR, Juttner A et al (2011) Magnetic resonance 4D flow characteristics of cerebrospinal fluid at the craniocervical junction and the cervical spinal canal. Eur Radiol 21:1788–1796 Functional imaging (fMRI) 6. Dietemann JL, Bogorin A, Abu Eid M et al (2012) Tips and traps in neurological imaging: imaging the perimedullary spaces. Diagn Interv Imaging 93:985–992 fMRI provides information concerning spinal cord motor 7. Vargas MI, Nguyen D, Viallon M et al (2010) Dynamic MR angi- function. The sequence used is essentially the same as that ography (MRA) of spinal vascular diseases at 3T. Eur Radiol 20: on the brain; however, modifications are necessary because 2491–2495 of the size of the spinal cord. Spinal fMRI illustrates neuronal 8. Wheeler-Kingshott CA, Hickman SJ, Parker GJ et al (2002) function indirectly by changes in blood flow and blood oxy- Investigating cervical spinal cord structure using axial diffusion tensor imaging. NeuroImage 16:93–102 gen levels (potential clinical indication of fMRI) [35]. 9. Amarouche M, Hart JL, Siddiqui A, Hampton T, Walsh DC (2015) This tool is only used for research purposes. Time-resolved contrast-enhanced MR angiography of spinal vascu- The potential clinical applications are: to determine pre- lar malformations. AJNR Am J Neuroradiol 36:417–422 served motor function in patients with spinal injury, to plan 10. Vargas MI, Gariani J, Sztajzel R et al (2015) Spinal cord ischemia: treatment or to evaluate the treatment response of tumours to practical imaging tips, pearls, and pitfalls. AJNR Am J Neuroradiol 36:825–830 understand the physiopathology of cervical headaches and to 11. Hirsch E, Vautravers P, Dietemann JL, Vetter D, Jesel M (1986) monitor functional changes in diseases such as multiple scle- Acute lumbar spinal cord disease caused by lumbar disk hernia. rosis and amyotrophic lateral sclerosis [35]. Presse Med 15:843–844 12. Reisner A, Gary MF, Chern JJ, Grattan-Smith JD (2013) Spinal Funding The authors state that this work has not received any funding. cord infarction following minor trauma in children: fibrocartilaginous embolism as a putative cause. J Neurosurg Pediatr 11:445–450 Compliance with ethical standards 13. Lewis SJ, Gray R, Holmes LM et al (2011) Neurophysiological changes in deformity correction of adolescent idiopathic scoliosis Guarantor The scientific guarantor of this publication is Maria Isabel with intraoperative skull-femoral traction. Spine (Phila Pa 1976) 36: Vargas. 1627–1638 14. Patel KB, Poplawski MM, Pawha PS, Naidich TP, Tanenbaum LN Conflict of interest The authors of this manuscript declare no relation- (2014) Diffusion-weighted MRI Bclaw sign^ improves differentia- ships with any companies, whose products or services may be related to tion of infectious from degenerative modic type 1 signal changes of the subject matter of the article. the spine. AJNR Am J Neuroradiol 35:1647–1652 15. Tessitore E, Broc N, Mekideche A et al (2016) A modern multidis- Statistics and biometry No complex statistical methods were necessary ciplinary approach to patients suffering from cervical spondylotic for this paper. myelopathy. J Neurosurg Sci. https://www.ncbi.nlm.nih.gov/ pubmed/27175620 Informed consent Written informed consent was waived by the 16. Vargas MI, Delavelle J, Jlassi H et al (2008) Clinical applications of Institutional Review Board. diffusion tensor tractography of the spinal cord. Neuroradiology 50: 25–29 Ethical approval Institutional Review Board approval was obtained. 17. Kurzbuch AR, Rilliet B, Vargas MI, Boex C, Tessitore E (2010) Coincidence of cervical spondylotic myelopathy and intramedullary ependymoma: a potential diagnostic pitfall. J Methodology Review article Neurosurg Spine 12:249–252 18. Basser PJ, Mattiello J, LeBihan D (1994) MR diffusion tensor Open Access This article is distributed under the terms of the Creative spectroscopy and imaging. Biophys J 66:259–267 Commons Attribution 4.0 International License (http:// Insights Imaging (2018) 9:549–557 557 19. Dixon WT (1984) Simple proton spectroscopic imaging. Radiology 29. Tatar I, Chou PC, Desouki MM, El Sayed H, Bilgen M (2009) Evaluating regional blood spinal cord barrier dysfunction following 153:189–194 20. Le Bihan D, Poupon C, Amadon A, Lethimonnier F (2006) spinal cord injury using longitudinal dynamic contrast-enhanced Artifacts and pitfalls in diffusion MRI. J Magn Reson Imaging MRI. BMC Med Imaging 9:10 24:478–488 30. Ishizaka K, Kudo K, Fujima N et al (2010) Detection of normal 21. Bammer R, Augustin M, Prokesch RW, Stollberger R, Fazekas F spinal veins by using susceptibility-weighted imaging. J Magn (2002) Diffusion-weighted imaging of the spinal cord: interleaved Reson Imaging 31:32–38 echo-planar imaging is superior to fast spin-echo. J Magn Reson 31. Fujima N, Kudo K, Terae S et al (2010) Spinal arteriovenous mal- Imaging 15:364–373 formation: evaluation of change in venous oxygenation with 22. Andre JB, Bammer R (2010) Advanced diffusion-weighted mag- susceptibility-weighted MR imaging after treatment. Radiology netic resonance imaging techniques of the human spinal cord. Top 254:891–899 Magn Reson Imaging 21:367–378 32. Hock A, Henning A, Boesiger P, Kollias SS (2013) (1)H-MR spec- 23. Tanenbaum LN (2013) Clinical applications of diffusion imaging in troscopy in the human spinal cord. AJNR Am J Neuroradiol 34: the spine. Magn Reson Imaging Clin N Am 21:299–320 1682–1689 24. Jones DK, Horsfield MA, Simmons A (1999) Optimal strategies for 33. Holly LT, Ellingson BM, Salamon N (2015) Metabolic imaging measuring diffusion in anisotropic systems by magnetic resonance using proton magnetic spectroscopy as a predictor of outcome fol- imaging. Magn Reson Med 42:515–525 lowing surgery for cervical spondylotic myelopathy. J Spinal 25. Mazura JC, Karimi S, Pauliah M et al (2014) Dynamic contrast- Disord Tech. https://doi.org/10.1097/BSD.0000000000000248 enhanced magnetic resonance perfusion compared with digital sub- 34. Carew JD, Nair G, Pineda-Alonso N, Usher S, Hu X, Benatar M traction angiography for the evaluation of extradural spinal metas- (2011) Magnetic resonance spectroscopy of the cervical cord in tases: a pilot study. Spine (Phila Pa 1976) 39:E950–E954 amyotrophic lateral sclerosis. Amyotroph Lateral Scler 12:185–191 26. Tofts PS, Brix G, Buckley DL et al (1999) Estimating kinetic pa- 35. Kornelsen J, Mackey S (2007) Potential clinical applications for rameters from dynamic contrast-enhanced T(1)-weighted MRI of a spinal functional MRI. Curr Pain Headache Rep 11:165–170 diffusable tracer: standardized quantities and symbols. J Magn Reson Imaging 10:223–232 27. Leach MO, Morgan B, Tofts PS et al (2012) Imaging vascular function for early stage clinical trials using dynamic contrast- Publisher’sNote enhanced magnetic resonance imaging. Eur Radiol 22:1451–1464 28. Bilgen M, Narayana PA (2001) A pharmacokinetic model for quan- Springer Nature remains neutral with regard to jurisdictional claims in titative evaluation of spinal cord injury with dynamic contrast- published maps and institutional affiliations. enhanced magnetic resonance imaging. Magn Reson Med 46: 1099–1106

Journal

Insights into ImagingSpringer Journals

Published: Jun 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off