3t magnetic resonance neurography: preliminary experience · neuroradiology state of the art 3t...

Neuroradiology

State of the art

3T Magnetic Resonance Neurography: preliminary experienceMartín Aguilar, Natalia Caneo, Laura Falcón, Cecilia Rollán, Hernán Chaves

ResumenPropósito. Presentar las ventajas de la secuenciaIDEAL (Iterative Decomposition of water and fat withEcho Asymmetry and Least-squares estimation) en laevaluación de los nervios periféricos y plexos braquial(PB) y lumbosacro (PLS) para el diagnóstico de lasneuropatías por atrapamiento o compresión, las neuro-patías sin atrapamiento y las condiciones subyacentes.La secuencia IDEAL proporciona 4 tipos de imágenesa partir de una sola adquisición, permitiendo la supre-sión uniforme de la grasa y el agua, y la obtención deimágenes en fase o fuera de fase de agua, grasa o de lacombinación de éstas.Materiales y Métodos. Estudio retrospectivo de enerode 2011 a junio de 2011. Se realizaron 11 neurografíascon secuencia IDEAL en resonancia magnética 3T (HDX3T, GE Healthcare, USA), con bobinas phased array de 8canales en planos coronal y sagital, cortes de 1,2 - 0 mm,y plano axial 3D spoiled gradient recalled (SPGR) T1, cor-tes 1 - 0 mm sin y con gadolinio. El campo de visión(FOV) fue variable según el nervio o plexo a explorar. Resultados. Se encontraron 2 schwannomas (plexobraquial, nervio ciático), 1 neuritis (inflamación porcompresión del nervio mediano), 2 casos con neurofi-bromas múltiples (uno en plexo lumbosacro y ciático yotro en plexo braquial), 3 neuromas postraumáticos(ciático poplíteo externo -CPE-), una avulsión conpseudomeningocele (plexo braquial) y 2 casos sin alte-raciones (plexo lumbosacro y ciático poplíteo externo).Conclusión. En esta presentación preliminar, la neuro-grafía por resonancia magnética (RM) mostró unaexcelente diferenciación entre el nervio y las estructu-ras circundantes y permitió reconocer la estructura delnervio y su señal, determinar variantes anatómicas ycausas de neuropatías, así como también se pudieronevaluar los cambios de denervación en los gruposmusculares involucrados.Palabras clave. Neurografía. Nervios periféricos.Plexo. Neuropatía. Resonancia magnética.

Abstract3T MRI neurography: preliminary experience.Purpose. Demonstrate the advantages of the IDEAL(Iterative Decomposition of water and fat with EchoAsymmetry and Least-squares estimation) sequence in theevaluation of peripheral nerves, brachial plexus and lumbarplexus, for the diagnosis of compression or entrapment neu-ropathies, non-entrapment neuropathies, and the underly-ing conditions.The IDEAL sequence provides 4 types of images from a sin-gle acquisition, allowing uniform fat or water suppressionand in phase/out of phase images of water, fat or a combina-tion of both.Materials and Methods. This is a retrospective study,from January 2011 to June 2011. Eleven neurographies wereperformed on 3T MRI (HDX 3T, GE Healthcare, USA),with 8-channel phased array coils on sagittal and coronalplanes, with 1.2-0 mm slices with no gap, axial 3D spoiledgradient recalled (SPGR) T , with 1-0-mm slice thicknesswith and without gadolinium injection and variable field ofview (FOV) according to the nerve or plexus to explore.Results. We found 2 schwannomas (brachial plexus andsciatic nerve), 1 neuritis (compression to median nerve), 2cases of multiple neurofibromas (lumbosacral plexus, sciaticnerve, brachial plexus), 3 traumatic neuromas (peronealnerve) and 1 pseudomeningocele avulsion (brachial plexus),and 2 with no structural alterations (lumbosacral plexusand peroneal nerve).Conclusion. In this preliminary experience, the use of high-resolution sequences in magnetic resonance imaging neu-rography studies provided excellent signal homogeneity,improving the recognition of the nerve structure and signal,the identification of anatomical variations, and causes ofneuropathy, as well as the characterization of denervationchanges of the affected muscle groups.Keywords. Neurography. Peripheral nerves. Plexus.Neuropathy. Magnetic resonance.

INTRODUCTION

Peripheral nerve disease or peripheral neuropathy(PN) affects all age groups, being a major cause ofmorbidity (1).

The standard classification of peripheral neuropa-

thies divides them into two large groups: those causedby entrapment or compression (cPN) and those inwhich there is no entrapment (ncPN) (1, 2).

The etiologies of peripheral neuropathy are diver-se and include anatomical or extrinsic compression,traumatic injury (penetrating injury or injury from

Servicio de Diagnóstico por Imágenes, FLENI, CABA, Argentina.Correspondencia: Dr. Martín Aguilar – [email protected]

Recibido: enero 2012; aceptado: julio 2012Received: january 2012; accepted: july 2012©SAR

RAR - Volumen 76 - Número 4 - 2012 Página 1

3T Magnetic Resonance Neurography

stretch and/or friction), inflammatory-infectious pro-cesses, tumor invasion, metabolic, ischemic or geneticcauses and physical-chemical damage (thermal orradiation injuries) (1).

The study of PNs has traditionally been within thefield of neurophysiology with the use of electromyo-graphy (EMG), providing functional and qualitative-quantitative data on the conduction properties of theaffected nerve (velocity, latency, etc.), serving as a sup-plement to clinical examination in the delimitation ofthe affected territory, and thus allowing the identifica-tion of the injured nerve location (3).

The advent of the magnetic resonance (MRI) neu-rography in 1992 (4) made a major contribution to thenon-invasive diagnostic study of peripheral neuropa-thies. Magnetic resonance neurography complementsthe EMG evaluations providing further anatomicaldata and allowing characterization of the distributionof muscle involvement, based on signal changes andthe trophism of tissue (denervation edema, fattyreplacement, atrophy). Thus, it is possible to estimatethe time of injury evolution and in many cases to cla-rify the underlying pathology.

As a result of the growing use of MRI in periphe-ral neuropathies, this technique has become a widelyused diagnostic method.

The standard imaging protocol for the evaluationof PN includes T1- and T2-weighted spin-echo imagesand T1-weighted images with intravenous gadoli-nium injection, using variable section thickness. Thethree anatomical planes are used, as well as the obli-que plane with reduced slice thickness to increaseresolution at specific locations. In these studies, T1-and T2-weighted images are often complementedwith fat saturation images, usually weighted on T2

(T2 Fat Sat, STIR) which, due to the abundant fat inthe perineural tissue, provide improved sensitivity forthe detection of signal abnormalities and higher speci-ficity for the chronological characterization of suchabnormalities in the studied structures (5).

With the appearance of new high-resolution neu-rography imaging (3D T2 SPACE - SiemenesHealthcare, IDEAL - GE Healthcare, T2 CUBE - GEHealthcare, etc) for MRI assessment of PNs, furtherdiagnostic data are available and, often, explorationbecomes simpler.

The aim of this study is to report our preliminaryexperience in the use of high-resolution neurographysequences for the study of PN in high field MRI (3T),to discuss the advantages and disadvantages of themethod and illustrate its applications in the evalua-tion of PN, considering some basic concepts of thispathology.

MATERIALS AND METHODS

A retrospective study was performed from January toJune 2011. Eleven patients with a clinical and/or elec-tromyographic diagnosis of peripheral neuropathyand an age range between 3 and 61 years (mean age:25.9 years) were examined (Table 1).Eleven neurographies were performed on high fieldMRI (HDX 3 Tesla, GE Healthcare, USA) using highresolution neurographic sequences called IDEAL bythe manufacturer (Iterative Decomposition of waterand fat with Echo Asymmetry and Least-squares esti-mation - GE Healthcare, USA) on coronal and sagittalplanes with 1.2-0-mm slices (Table 2). Axial T1-weigh-ted 3D SPGR (spoiled gradient recalled) images were

Página 2 RAR - Volumen 76 - Número 4 - 2012

a b

Fig 1: Thermal injury (secondary to laser treatment for varicose veins) of the common peroneal nerve. STIR axial image. (a) Increased nerve sig-nal (arrow) and hyperintensity of perineural tissue; (b) IDEAL T2 Fat Sat images allow for the detection of signal abnormalities in the affectednerve and perineural tissues with improved definition of nerve architecture (please note the fascicular pattern of the nerve).

Martín Aguilar et al.

also acquired with 1-0-mm slice thickness, before andafter gadolinium injection. Eight-channel phasedarray coils appropriate for the anatomical region ofinterest were used. The field of view (FOV) usedvaried according to the explored nerve or plexus.Clinical data (history, duration of the condition,physical examination, etc) were taken into account, aswell as data provided by the EMG when it was perfor-med prior to the MRI exam.

RESULTS

In 2 of the 11 patients studied (Table 1) no structu-ral causes of PN were identified on high-resolutionneurographies. Of those cases in which pathology was

documented by MRI, only 3 had a history of trauma (2traffic accidents and 1 secondary to laser treatment forvaricose veins in the lower limb) (Fig. 1). We alsofound: 2 schwannomas (brachial plexus -BP- and scia-tic nerve) (Fig. 2); an inflammatory process (neuritisdue to compression to median nerve) (Fig. 3); 2 casesof multiple neurofibromas (one involving the lumbo-sacral plexus -LSP- and the sciatic nerve, and the otherin the BP) (Fig. 4), of which only one had a history ofneurofibromatosis at the time of examination; 1 post-traumatic neuroma (external popliteal sciatic nerve-EPSN-) and 1 avulsion with associated pseudomenin-gocele (BP) (Fig. 5).

In cases of cPN with trauma history, signal chan-ges in the affected nerve and surrounding tissueswere mainly observed, with hyperintense signal on


Gender Age Clinical EMG Neurography Diagnosishistory findings

Patient 1 F 14 ys-o Chronic low back - Complex regional pain radiating to neuropathic painleft lower limb

Patient 2 F 13 ys-o Pain and functional Myelinic neuropathy - Regional impairment in the with axonal neuropathic painterritory of the right component of the common peroneal nerve common peroneal nerve

Patient 3 F 38 ys-o Carpal Tunnel Edema of the median Carpal Tunnel Syndrome nerve in carpal tunnel Syndrome

Patient 4 M 35 ys-o Progressive functional Axonal neurogenic Increased signal - Entrapment neuritisimpairment of the injury and denervation thickening of the of the commonfoot and associated in the L4-L5 territory. common peroneal peroneal nerveatrophy nerve

Patient 5 M 15 ys-o Trauma after fall from Hypersignal of the right Traumatic paralysismotorcycle popliteal nerve and of the common

muscle atrophy peroneal nerve

Patient 6 M 35 ys-o Fall from motorcycle Pseudomeningocele Post-traumaticwith fracture after left C7-D1 avulsion of left TIdislocation of the left avulsion and secondary root with sternoclavicular joint to lower trunk brachial pseudomeningocele

plexus avulsion

Patient 7 F 61 ys-o Right common peroneal Peroneal nerve injury Peroneal nerve nerve injury secondary neuromato laser treatment for varicose veins

Patient 8 F 46 ys-o Pain in the neck and Right supraclavicular Brachial plexusright arm Schwannoma Schwannoma

Patient 9 M 7 ys-o Non-traumatic acute Image in common Neuromaonset right common peroneal nerve - peroneal nerve palsy right peroneal nerve

Patient 10 M 3 ys-o Neurofibromatosis Expansive lesions in Neurofibromatosistype I and giant lumbosacral plexus, type Ineurofibroma in right gluteal region and lower limb lower limbs

Patient 11 M 31 ys-o Operated neck Bilateral foraminal MultipleSchwannoma neurofibromas Schwannomatosis

C4-C5, C5-C6 and C6-C7

Table 1: Demographics.


T2-weighted pulse sequences. A case of loss of nervecontinuity and associated pseudomeningocele wasalso observed.

Non-traumatic cPN demonstrated a frank associa-tion between signal changes and adjacent anatomicalstructures in the upper limb (carpal tunnel) and lowerlimb (EPSN - fibular head), as well as tumor compres-sion in cases of neurofibromatosis and schwannomas.

Those pathological cases with no compressioninvolvement proved to be morphologically healthy,but with signal changes.

DISCUSSION

Peripheral neuropathies result from an isolated ora group of damaged peripheral nerves with orwithout involvement of a nerve plexus. They affect allage groups but, depending on which nerve is invol-ved, there may be some sex-specific prevalence (forexample, there is a relationship between the femalegender and the carpal tunnel syndrome). In addition,PNs are a major cause of morbidity with a significanteconomic and occupational impact. Thus, in the gene-ral population, only entrapment neuropathies (themost frequent) generate approximately 100,000 surgi-cal procedures per year in USA and Europe (6).

Peripheral neuropathies may be classified into twolarge groups: those caused by entrapment or compression(cPN) and those in which there is no entrapment (ncPN).

Regarding etiology, in the case of cPN, nervesdamage is caused by physical forces from both anato-

mical and external structures (1, 2). Such forces, as descri-bed by H. Seddon in 1943 (7), cause different grades ofnerve damage, called neuropraxia, axonotmesis andneurotmesis, each one with different functional prog-noses related to the nerve capacity of regeneration.

Neuropraxia is the lesser degree of severity in caseof peripheral nerve injury, as there is nerve injury butno discontinuity of the sheath and axon. It is generallya temporary disorder with regeneration ad integrum.

Axonotmesis is a more severe nerve injury withdisruption of the axon but preservation of the encapsu-lating connective tissue of the nerve (endoneurium,perineurium and epineurium), with capacity for rege-neration being variable.


Fig.2. Schwannoma at the right supraclavicu-lar triangle: (a) and (b) IDEAL images, T1-weighted (a) and T2-weighted (b) coronal pla-nes: a nodular lesion can be seen (short arrows)with a hypointense signal on T1 and a hyperin-tense signal on T2 in relation to structures ofthe right brachial plexus (long arrows in a); (c)Multiplanar axial and (d) coronal obliquereconstructions of IDEAL T2 Fat Sat images(short arrows). Reconstructions allow to openout the brachial plexus (long arrows in d) andto show the relationship between the tumor andthe surrounding nerves.

a b

c d

FOV: 16 - 35 cm

Slice thickness / spacing between slices: 1 - 1.2 mm / 0 - 0.2 mm

Frequency: 320

NEX: 3

TE: 90 ms (T2) / 9.1 ms (T1)

TR: 7160 ms (T2) / 575 ms (T1)

Echo Train: 20 (T2) / 3 (T1)

Matrix: 320 x 256

Table 2: T1- and T2-weighted IDEAL sequence acquisi-tion parameters.


Finally, the most severe form of entrapment injuryof the peripheral nerve is neurotmesis, in which boththe axon and perineural sheaths are damaged. Theinjured nerve looses its capacity for regeneration andthis may result in aberrant regeneration processes (asin case of neuromas) (7).

As mentioned above, these three degrees of nerveinjuries imply different functional prognoses. Surgicalrepair is performed in cases of neurotmesis and severeaxonotmesis (8).

In patients in whom damage is caused by anatomi-cal compression of the nerve (i.e.: nerve entrapmentsyndromes), pain is determined both by the pressureexerted on the nerve (a conflict between the containerand its content) and the duration of such pressure.

In this type of mechanism, the pressure exertedaffects nerve irrigation, first impairing venous bloodflow and higher pressures are required to impair arte-rial blood flow in nerve structures. In case of compres-sion pressures above 80 mmHg, ischemic damage


Fig. 3. Entrapment neuropathy of the median nerve. IDEAL T2 Fat Satimage showing diffuse thickening of the median nerve (arrowhead) withcentral hyperintensity (neuritis) and thickening of carpal tunnel ten-dons with synovitis (long arrows).

Fig. 4. Neurofibromatosis type I: diffuse involvement of the right thighin a 3-year child. IDEAL T1 image, axial plane with IV contrast.

Fig. 5: Traumatic left brachial plexus lowertrunk avulsion. T2 IDEAL Fat Sat coronal planeimages. Hyperintensity of the left brachial plexus(arrows) and scalene muscles (*), consistent withdenervation edema. In (b), avulsion and retrac-tion of the left brachial plexus lower trunk(arrow). Image (c) shows a lesion consistent withpseudomeningocele in the left C7-D1 neuralforamen (arrow). (d) Axial and (e) sagittal mul-tiplanar reconstructions showing traumaticpseudomeningocelein the left neural foramen(arrows).

a b c

d e


becomes irreversible. In addition, if compression beco-mes chronic (>6 months), the aforementioned ischemicdamage is associated with reactive fibrosis (7).

Non-entrapment peripheral neuropathies includethose caused by inflammatory and or infectious condi-tions, metabolic and ischemic disorders and physical-chemical damage (thermal or ionizing radiation inju-ries). These processes may affect both the cytoarchitec-ture of the axon and the constitution of the myelin she-ath at molecular level (dysmyelinating disorders) andstructural level (demyelinating disorders), resulting innerve conduction abnormalities of PNs (9, 10).

Tumor involvement in PNs may be related totumors of the nerve itself (i.e.: schwannomas) or fromtumors of adjacent structures that provoke nerve com-pression and/or infiltration (11, 12).

The study of PNs has traditionally been within thefield of neurophysiology, with EMG being the stan-dard diagnostic test. This test evaluates and recordsthe spontaneous or induced electrical activity of amuscle related to its innervating nerve, providing dataon the conduction properties of such nerve. Even ifelectromyographic testing may be performed with sur-face electrodes, it is more often performed percutane-ously, with needle electrodes, for a higher diagnosticefficacy (2).

With the advent of neurographic sequences in 1992(4), the diagnostic efficacy of MRI for PN has improved,providing further anatomical details and making pos-sible to trace the course of the studied nerve. Thus,MRI neurography contributes to reveal signs of anato-mic conflict in cPN and to rule out such conflicts inncPN, providing a better assessment of the morpho-logy and signal of the nerve of interest. Due to theseadvantages, and its noninvasive nature, MRI neuro-graphy has become a currently validated method forthe study of PN and, often, the first diagnostic test,making possible to complement the functional andqualitative-quantitative data provided by EMG withanatomical data.

As a first step in the evaluation of PN, the normalappearance of the nerve should be taken into accountto identify changes in morphology that may be asso-ciated with signal abnormalities; the presence of anato-mical variants should also be detected.

Consequently, normal peripheral nerves have afascicular appearance, uniform caliber and well-defi-ned borders, with a straight course without sharpangulations (characteristics that are absent in a largenumber of peripheral neuropathies). Normal periphe-ral nerves signal in MRI is isointense to skeletal mus-cles or minimally hyperintense, both on T1- and T2-weighted images, in relation to fat content. In addition,normal peripheral nerves demonstrate an adequatecleavage plane in relation to perineural tissue, with thelatter showing a homogeneous signal.

These characteristics are absent in PN. The nervelooses its normal fascicular pattern and straight appe-arance, with nodular or fusiform irregularities in its

caliber, and even loss of nerve continuity in extremecases. Sharp angulations or deviations may also beobserved in the course of the affected nerve (as seen incPN) (2, 8, 13, 14).

Regarding MRI images, both in cPN and in ncPN,there is an abnormal signal intensity of the injurednerve, with T2 hyperintensity and signal abnormalitiesof perineural structures with loss of their interface. Fordetecting these abnormalities and due to of the fat con-tent of the nerve, the use of fat saturation imaging (T2Fat Sat, STIR, SPIR, etc) becomes highly useful, provi-ding increased sensitivity (13, 14).

Depending on the etiology of nerve injury (e.g.,infections or tumors), findings derived from the admi-nistration of paramagnetic contrast medium should beconsidered, which become evident on T1-weightedsequences (to which fat saturation pulses may beadded) (5).

Changes in the perineural tissue signal are oftenassociated with changes in the abnormal nerve. Thistissue loses its signal homogeneity and there is peri-neural fat stratification, poor fat saturation in T2 pul-ses with fat saturation and linear tracts of low signalintensity corresponding to fibrosis (6, 8, 13).

It is important to highlight the presence of signalchanges in muscles innervated by the affectednerve(s).

At an early stage, and due to an early denervationedema, muscle structures show a variable hypointen-sity on T1 and hyperintensity on T2, with no decreasein signal intensity on T2-weighted fat saturation pulsesequences. These signal changes are not usually asso-ciated with significant changes in muscle volume (15).

Approximately one month after nerve injury, if theharmful agent persists, signs of fatty replacement willbe seen in the affected muscle. In this setting, signalintensity gradually increases on T1-weighted imagesand decreases in fat saturation sequences. In this sub-acute stage, a slight decrease in muscle volume may beobserved; therefore, it may be useful to compare itwith the contralateral muscle (16-18).

In chronic stages (> 12 months), progressive repla-cement of muscle fibers becomes evident, and therefo-re the muscle becomes markedly hyperintense on T1-and T2-weighted images and hypointense on fat satu-ration images. Even if muscle has a tendency toatrophy with a decrease in volume, in rare cases, anenlargement may be observed due to conspicuousfatty infiltration (16-18).

Regarding the study of PN by MRI, traditionalimage acquisition protocols used standard T1- and T2-weighted images, as well as fat saturation sequences,usually weighted on T2 (T2 Fat Sat, STIR, etc.), withadministration of intravenous paramagnetic contrastdepending on clinical suspicion.

In our practice, as from January 2011, we have gra-dually added to 3T MRI neurographies (HDX 3T, GEHealthcare, USA), high-resolution multiple contrastsequences. This recently developed sequence, called



IDEAL (Iterative decomposition of water and fat withecho asymmetry and least-squares estimation - GEHealthcare, USA) (19) by the manufacturer, allows toobtain with a single acquisition T1- or T2-weightedimages and combine them with fat-only or water-onlyimages, or both, as well as with in-phase or out-of-phase images.

These imaging sequences are derived from theprinciples first described by Dixon, based on decom-posing fat proton signals according to their precessionfrequency (a widely used principle for in-phase or out-of-phase imaging) (5).

This type of imaging allows for an adequate explo-ration and its advantages include optimization of thesignal-to-noise ratio and a more uniform fat saturationthan that achieved with traditional fat saturationsequences (T2 Fat Sat, STIR, etc), reduced presence ofartifacts (because of less field inhomogeneity), and anadequate visualization of the structures being imaged,mainly in challenging anatomies (plexus, nerve canals,etc.). Furthermore, it is important to highlight the pos-sibility offered by these sequences and 3D T1 SPGRand T2 CUBE images for multiplanar and/or curvedreformatted images using the workstation.

Probably, the main advantage of this type of ima-ging sequence is the optimization of the signal-to-noise ratio, as it allows for improved signal homoge-neity. This development becomes even more evident infat saturation sequences, where there is a significantdecrease in the incidence of artifacts.

Regarding the different saturation options ofIDEAL sequences, we would like to highlight the use-fulness of water-saturation and out-of-phase images.Both sequences provide a better delineation of thenerve structure, allowing for best resolution of theneural structure and surrounding tissue.

Anyway, at least some standard imaging sequen-ces should be used, particularly because of the diffe-rent contrasts between both types of images. Then, wethink it would be wise to have a training time, duringwhich both standard and neurographic sequencesshould be used (20, 21).

In addition, clinical and EMG data should be avai-lable to optimize acquisition parameters and improveimaging accordingly. This would avoid to explore vastregions, which affects the length of the examinationand, secondarily, its quality.

In our patients, high-resolution neurographiesdemonstrated pathology in most cases and ruled it outin patients with normal imaging findings in whomstructural pathology was suspected based on clinicaland / or EMG data.

Regarding imaging findings, the morphology orsignal changes of the different nerves that we observedin our group of patients, were consistent with thosereported in the literature (1,8,9,13,14,19,21).

For the characterization of associated muscleabnormalities, traditional imaging sequences with awide FOV are a useful supplement to neurographic

sequences for the study of muscle groups innervatedby the nerve of interest.

A limitation of the study was the small sample sizeand the lack of a study comparing standard imagingsequences with high-resolution neurographic imaging.We think that further studies with larger sample sizesare needed to provide further data on this describedtechnique.

CONCLUSION

It is important to be aware of the normal characte-ristics of peripheral nerves on MRI and their appearan-ce in pathological conditions, considering the changesin signal and volume of muscle structures as a result oftheinnervating nerve injury.

Although our study did not compare the benefit ofthese imaging sequences with those of traditional ima-ging, we may affirm that in this group of patients, theuse of high resolution imaging for MRI neurographyallowed to detect abnormalities in most cases and torule them out in 2 cases with clinical suspicion of asso-ciated structural pathology. Hence, high resolutionneurographic sequences allowed for an adequatevisualization of the nerve and the perineural tissue,optimizing their evaluation, mainly by improving sig-nal homogeneity.

References

1. Chhabra A, Lee PP, Bizzell C, Soldatos T. 3 Tesla MR neuro-graphy--technique, interpretation, and pitfalls. SkeletalRadiol 2011; 40:1249-60.

2. Hoff JJ, Kliot M, Slimp J, Haynor DR. What´s new in MRI ofperipheral nerve entrapment? Neurosurg Clin N Am, 2008;19:583-95.

3. Leiguarda R, Nogues M. Enfermedades de los nervios peri-féricos y plexos. En: Leiguarda R, ed. Neurología. BuenosAires: El Ateneo; 2005:391-426.

4. Howe F, Filler AG, Bell BA, Griffiths JR. Magnetic resonanceneurography. Magn Reson Med 1992; 28:328-38.

5. Maravilla KR, Bowen BC. Imaging of the peripheral nervoussystem: evaluation of peripheral neuropathy and plexo-pathy. AJNR Am J Neuroradiol 1998; 19:1011-23.

6. Siemionow M, Brzezicki G. Chapter 8: Current techniquesand concepts in peripheral nerve repair. Int Rev Neurobiol2009; 87:141-72.

7. Seddon H. Three types of nerve injury. Brain 1943; 66:237-88.8. Chhabra A, Williams EH, Wang KC, Dellon AL, Carrino JA.

MR neurography of neuromas related to nerve injury andentrapment with surgical correlation. AJNR Am JNeuroradiol 2010; 31:1363-8.

9. Chhabra A, Faridian-Aragh N, Chalian M, et al. High-reso-lution 3-T MR neurography of peroneal neuropathy. SkeletalRadiol 2012; 41:257-71.

10. Plewnia C, Wallace C, Zochodne D. Traumatic sciatic neuro-pathy: a novel cause, local experience, and a review of theliterature. J Trauma 1999; 47:986-91.

11. Gupta G, Mammis A, Maniker A. Malignant peripheralnerve sheath tumors. Neurosurg Clin N Am 2008; 19:533-43.



12. Rodero L, Canga A, Figols J, Berciano J, Combarros O.Buttock mass and malignant sciatic nerve tumor. Neurología2004; 19:27-31.

13. Chhabra A, Andreisek G, Soldatos T, et al. MR neurography:past, present, and future. AJR Am J Roentgenol 2011;197:583-91.

14. Kim S, Choi JY, Huh YM, et al. Role of magnetic resonanceimaging in entrapment and compressive neuropathy- what,where, and how to see the peripherals nerves on the muscu-loskeletal magnetic resonance image: part I. Overview andlower extremity. Eur Radiol 2007; 17:139-49.

15. McMahon CJ, Wu JS, Eisenberg RL. Muscle edema. AJR AmJ Roentgenol 2010; 194:W284-92.

16. May DA, Disler DG, Jones EA, Balkissoon AA, Manaster BJ.Abnormal signal intensity in skeletal muscle at MR imaging:patterns, pearls, and pitfalls. Radiographics 2000; 20:S295-315.

17. Yamabe E, Nakamura T, Oshio K, Kikuchi Y, Ikegami H,

Toyama Y. Peripheral nerve injury: diagnosis with MR ima-ging of denervated skeletal muscle -- experimental study inrats. Radiology 2008; 247:409-17.

18. Fleckenstein JL, Watumull D, Conner KE, et al. Denervatedhuman skeletal muscle: MR imaging evaluation. Radiology1993; 187:213-8.

19. Kim S, Choi JY, Huh YM, et al. Role of magnetic resonanceimaging in entrapment and compressive neuropathy-- what,where, and how to see the peripherals nerves on the muscu-loskeletal magnetic resonance image: part II. Upper extre-mity. Eur Radiol 2007; 17:509-22.

20. Costa DN, Pedrosa I, McKenzie C, Reeder SB, Rofsky NM.Body MRI using IDEAL. AJR Am J Roentgenol 2008;190:1076-84.

21. Thawait SK, Wang K, Subhawong TK, et al. Peripheral nervesurgery: the role of high-resolution MR neurography. AJNRAm J Neuroradiol 2012; 33:203-10.


The authors declare no conflicts of interest.

3t magnetic resonance neurography: preliminary experience · neuroradiology state of the art 3t...

Documents