Differentiation of Osteophytes and Disc Herniations in ...clinical-mri.com/wp-content/uploads/2017/...Disc.1.pdf · osteophytes, and/or posterior disc herniations of the cervical

ORIGINAL ARTICLE

Differentiation of Osteophytes and Disc Herniations in SpinalRadiculopathy Using Susceptibility-Weighted Magnetic

Resonance Imaging
Yvonne Yi-Na Bender, MD,* Gerd Diederichs, MD,* Thula Cannon Walter, MD,* Moritz Wagner, MD,*
Thomas Liebig, MD, PhD,† Marcus Rickert, MD,‡ Kay-Geert Hermann, MD,*Bernd Hamm, MD, PhD,* and Marcus Richard Makowski, MD, PhD*

Objective: The aim of this study was to evaluate the diagnostic performance ofsusceptibility-weighted magnetic resonance imaging (SW-MRI) for the differen-tiation of osteophytes and disc herniations of the spine compared with that of con-ventional spine MR sequences and radiography.Materials and Methods: This study was approved by the local ethics reviewboard; written consent was obtained from all subjects. Eighty-one patients withsuspected radiculopathy of the spine were included prospectively. Radiography,T1/T2, and SW-MRI of the cervical/lumbar spine were performed. As referencestandard, 93 osteophytes (n = 48 patients) were identified on radiographs in combi-nation with conventional T1/T2 images. One hundred fourteen posterior disc herni-ations (n = 60 patients) were identified on T1/T2 in combination with radiographyexcluding osteophytes. For this study, 2 observers independently assessed the pres-ence of osteophytes and disc herniations on T1/T2 and SW-MRI, with radiographsexcluded from the analysis. In a subgroup of patients (n = 19), additional com-puted tomography images were evaluated. Sensitivity, specificity, and interobserveragreement were calculated.Results: Most osteophytes (n = 92 of 93) and disc herniations (n = 113 of 114)could be identified and differentiated on SW-MRImagnitude/phase images, if ra-diographs were excluded from analysis. Susceptibility-weighted magnetic reso-nance imaging achieved a sensitivity of 98.9% and specificity of 99.1% for theidentification of osteophytes. Conventional T1/T2 spine MR sequences achieveda sensitivity and specificity of 68.6% and 86.5%, respectively, if radiographswereexcluded from analysis. Regarding the size of osteophytes, SW-MRI showeda strong correlation with computed tomography (R2 = 0.96) and radiography(R2 = 0.95). In addition, SW-MRI achieved a higher interobserver agreementcompared with conventional MR.Conclusions: Susceptibility-weighted magnetic resonance imaging enables thereliable differentiation of osteophytes and disc herniations in patients with spinalradiculopathy with a higher sensitivity and specificity compared with conven-tional T1/T2 MR sequences.

Key Words: magnetic resonance imaging, susceptibility-weighted MRI,spinal radiculopathy, osteophytes, disc herniations

(Invest Radiol 2017;52: 75–80)

W ith increasing age, a relatively large proportion of the populationshows radiological signs of spinal degeneration, usually located

in the cervical or lumbar spine.1 If surgical intervention is requireddue to radiculopathy or myelopathy, cross-sectional imaging is impor-tant to localize the site and degree of spine disease and to differentiatebetween disc herniations, nerval compression due to osteophytes, and

Received for publication March 21, 2016; and accepted for publication, after revision,June 17, 2016.

From the Departments of *Radiology, and †Neuroradiology, Charité–UniversityMed-icine Berlin, Berlin; and ‡Spine Department, Orthopedic University HospitalFriedrichsheim, Frankfurt am Main, Germany.

Conflicts of interest and sources of funding: none declared.Correspondence to: Yvonne Yi-Na Bender, MD, Department of Radiology,

Charité–University Medicine Berlin, Charitéplatz 1, 10117 Berlin, Germany.E-mail: [email protected].

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.ISSN: 0020-9996/17/5202–0075DOI: 10.1097/RLI.0000000000000314

Investigative Radiology • Volume 52, Number 2, February 2017

Copyright © 2016 Wolters Kluwer H

other abnormalties.2,3 Commonly used imaging modalities includemagnetic resonance imaging (MRI), radiography, and computed to-mography (CT). Radiography is a widely available and inexpensivemethod to visualize osteophytes, facet and uncovertebral arthrosis,and narrowing of disc space. Functional radiographs can be useful forthe 2-dimensional assessment of the spine, including instability. A pre-cise 3-dimensional (3D) evaluation of osseous changes can be achievedby using CT. This technique is, however, associated with radiation3 anda limited diagnostic performance for the identification of disc hernia-tions due to a relatively low soft tissue contrast.4 Therefore, it is not rou-tinely performed in all patients.

Magnetic resonance imaging provides a great range of infor-mation with a focus on soft tissue structures and has developed intothe method of choice to assess the involvement of spinal nerves anddamage to the spinal cord.5 However, a major limitation of spineMRI is that osseous degenerative changes such as osteophytes can-not reliably be distinguished from disc herniations, as both struc-tures can appear hypointense on T1- and T2-weighted images.Because of this limitation, additional radiography and CT are oftenacquired for preoperative planning.

The development of susceptibility-weightedMRI (SW-MRI) en-ables the MR-based differentiation of calcified structures from soft tis-sues based on their magnetic susceptibility. The potential of SW-MRIhas so far mainly been investigated for the identification of calcifica-tions in the brain.6–15

The aim of this study was to evaluate the diagnostic perfor-mance of SW-MRI for the differentiation of osteophytes and discherniations of the spine compared with that of conventional MR se-quences and radiography.

MATERIALS AND METHODS

Study PopulationThis study was prospectively approved by the local ethics re-

view board. Written consent was obtained from all subjects beforethey underwent the study protocol. In all investigated 81 subjectswith suspected radiculopathy of the cervical or lumbar spine, a clin-ical MRI scan was required. As part of the study protocol, an addi-tional sequence (susceptibility-weighted imaging) was performed.Overall, we investigated 14 subjects without degenerative changesof the spine and a total of 67 patients with degenerative changes,osteophytes, and/or posterior disc herniations of the cervical/lumbar spine prospectively using a standardized imaging protocol.The 14 subjects without osteophytes and/or posterior disc hernia-tions presented with clinical symptoms, suggesting radiculopathyof the spine. The scan, however, did not show degenerative or in-flammatory changes of the spine. The study population included41 male (50.6%) and 40 female (49.4%) patients who visited our in-stitution and underwent MRI of the spine between July 2013 andFebruary 2016. Sixty-seven patients were diagnosed with degenerativechanges of the cervical/lumbar spine (36 male patients, mean age59.4 ± 17.0 years; 31 female patients, mean age 55.0 ± 17.5 years,

www.investigativeradiology.com 75

ealth, Inc. All rights reserved.

mailto:[email protected]

www.investigativeradiology.com

Bender et al Investigative Radiology • Volume 52, Number 2, February 2017

P > 0.05) based on the detection of osteophytes and posterior disc her-niations on radiographs in combination with conventional spine MRIscans in at least 1 vertebral segment. Fourteen patients (6 male patients,mean age 45.0 ± 18.3 years; 8 female patients, mean age 43.6 ± 16.7years) without degenerative changes on radiographs and conventionalspine MRI scans were used as a reference group. Susceptibility-weighted magnetic resonance imaging, conventional cervical/lumbarspine T1/T2 MR sequences, and radiography were performed in allpatients. For a subgroup of patients (n = 19 of 67), additional CTscans of the cervical/lumbar spine were available.

Imaging ProtocolMagnetic resonance imaging was performed on a Siemens 1.5 T

scanner (Avanto; Siemens Medical Solutions, Erlangen, Germany) in astandardized supine position with a standard neck coil for the cervicaland a standard body coil for the lumbar spine. The following spineMRI protocol, which is routinely used in our department for the diagno-sis of cervical/lumbosacral spine lesions, was applied: sagittal T1 turbospin echo (TSE), T2 TSE, and axial T2 TSE (lumbar) or T2 MEDIC(cervical). Cervical spine [T1 TSE: field of view 240 � 240 mm2, ma-trix 448 � 448, TR/TE = 803/21 milliseconds, 150-degree flip angle,and 3-mm slice thickness; T2 TSE: field of view 240 � 240 mm2, ma-trix 448 � 448, TR/TE = 2800/77 milliseconds, 150-degree flip an-gle, and 3-mm slice thickness], lumbar spine [T1 TSE: field of view280 � 280 mm2, matrix 448 � 448, TR/TE = 726/13 milliseconds,150-degree flip angle, and 3-mm slice thickness; T2 TSE: field ofview 280� 280 mm2, matrix 448� 448, TR/TE = 3300/88 millisec-onds, 150-degree flip angle, and 3-mm slice thickness]). In addition,a 3D fast low-angle gradient-echo sequence (SW-MRI) was per-formed. A standard ventral saturator was used to minimize artifacts.The SW-MRI magnitude image derives from a velocity-compensated3D-gradient-recalled echo (GRE) sequence, which is part of theSW-MRI. This sequence is comparable to standard GRE sequencesfor the detection of T2* time-shortening lesions. In addition to thevelocity-compensated 3D-GRE sequence, SW-MRI includes phase in-formation.16,17 Imaging parameters of the SW-MRI sequence, whichwas automatically aligned to the sagittal T1/T2-weighted sequences, in-cluded the following: TR/TE = 49/14 milliseconds, 15-degree flip an-gle, and 3-mm slice thickness. Field of view and matrix were adaptedto the respective T1 and T2 TSE sequences. Acquisition time forSW-MRI was 5 minutes and 11 seconds in average. After the acquisi-tion, SW-MRI magnitude and phase images were reconstructed.12

Imaging AnalysisAll images were analyzed using PACS workstations (Centricity

Radiology RA1000; GE Healthcare, Little Chalfont, United Kingdom).As reference standard, osteophytes were identified on radiographs incombination with conventional T1/T2 spine MRI scans. Osteophyteswere visually detected on sagittal radiographs and sagittal conventionalT1/T2 spine MRI scans based on their cortical and medullary continuitywith the vertebral body and based on their specific shape. On conven-tional T1/T2-weighted spineMRI scans, osteophytes display accordinglyan isointense signal to the vertebra. Off-center marginal osteophytes ofthe end plates were excluded from the analysis as their detection is lim-ited on radiographs. On SW-MRI scans, osteophytes were identifiedbased on the following properties: hyperintense on inverse SW-MRImagnitude images with a hyperintense surface on SW-MRI phase im-ages. Sizes of osteophytes were measured on SW-MRI magnitude, T1-and T2-weighted MRI scans, and radiographs to assess intermodalitycorrelations. Regarding posterior disc herniations, the combination ofT1/T2-weighted MRI scans with radiographs excluding osteophytesserved as reference standard. Posterior disc herniations were identifiedon MRI scans as herniations in the spinal canal with contiguous connec-tion to the intervertebral disc. The herniation showed a hypointense

76 www.investigativeradiology.com


signal to the adjacent disc on T1-weighted and an isointense to hyperin-tense signal on T2-weighted images, depending on the degree of disc hy-dration. Radiographswere additionally reviewed to exclude that posteriorosteophytes were present at that location.

Disc herniations were identified on inverse SW-MRI magnitudeimages as hypointense or without signal on SW-MRI phase images. Forthis study, all MRI scans were reviewed independently by 2 experiencedradiologists with 4 and 8 years of diagnostic experience inmusculoskel-etal MRI. Both readers were blinded to radiographs. Osteophytes werevisually detected based on their specific shape and signal intensity. Themaximal extend of the osteophyte was measured. No grading systemwas used.

Computed TomographyFor a subgroup of patients (n = 19 of 67) additional CT scans of

the cervical/lumbar spine were available. Correlations of size measure-ments between SW-MRI, conventional radiography, and CTwere eval-uated. Diameters of spondylophytes were measured manually onsusceptibility-weighted magnitude images and radiographs to assessintermodality correlations. Clinical standard CT scans were performedby using a 64-section Aquillion system (Toshiba Medical Systems,Otawara, Japan).

Contrast-to-Noise MeasurementsThe signal(osteophytes), signal(intervertebral disc), and the signal(noise)

was assessed by manually drawing a region of interest in intervertebraldiscs, osteophytes, and noise in standard T2-weighted and inversesusceptibility-weighted magnitude images. Noise refers to the meansignal in a region of air dorsal to the spine. Contrast-to-noise ratio(CNR) was calculated by subtracting the signal measured in the osteo-phyte by the signal measured in the intervertebral disc and dividing thedifference by the signal measured in the noise.

Statistical AnalysisVariables are reported as mean ± standard deviation. Sensitivities

and specificities of SW-MRI comparedwith that of T1/T2-weightedMRIscans in combination with radiographs were computed. To display thespread of data and the limits of agreement, interobserver/intraobserveragreement was assessed using Bland-Altman plots. The relationship be-tween size measurements on SW-MRI, T1- and T2-weightedMRI scans,and radiographs was determined using linear regression. A P < 0.05 wasconsidered statistically significant.

RESULTS

Sensitivity and Specificity for the Detectionof Osteophytes

As reference standard, 93 osteophytes (in 48 patients) were iden-tified on conventional T1/T2MR in combination with radiographs con-firming osteophytes. If radiographs were excluded from the analysis,the majority (n = 92 of 93) of osteophytes could be identified on in-verse SW-MRI magnitude images with a hyperintense surface onsusceptibility-weighted phase images (Figs. 1, 2). Susceptibility-weighted magnetic resonance imaging achieved a sensitivity of98.9% and a specificity of 99.1% for the identification of osteophyteswith a 95% confidence interval of 93.4% to 99.9% and 94.4% to99.9%, respectively (Figs. 1B3-B4, Figs. 2A3-A4). If radiographs wereexcluded from the analysis, it was in some cases difficult to distin-guish osteophytes from disc herniations and surrounding tissues(eg, ligaments) on standard T1/T2-weighted MR sequences. There-fore, only 69 lesions (72.6%) could be delineated (Figs. 1B1-B2,Figs. 2A1-A2). Standard spine MR sequences achieved a sensitivityof 68.6% and a specificity of 86.5% for the identification of

© 2016 Wolters Kluwer Health, Inc. All rights reserved.



FIGURE 1. SW-MRI for the evaluation of dorsal osteophytes and disc herniations in the lumbar spine in a 61-year-old male patient with radiculopathy. A,Dorsal disc herniation (arrows) on conventional sagittal T1/T2 MR (A1, A2), susceptibility-weighted magnitude/phase images (A3, A4), andCT/radiographs (A5, A6). The dorsal disc herniation can be clearly identified on susceptibility-weighted magnitude images (A3, A4) as a hypointenseherniation. Sagittal CT/radiographs excluded osteophytes (A5, A6). Because of the low signal of the disc herniation on conventional T1/T2-weightedimages (A1, A2), it remains challenging to determine, whether a dorsal osteophyte or dorsal disc herniation is present as both appear with a hypointensesignal. B, Dorsal osteophytes (arrows) on sagittal T1/T2 conventionalMR sequences (B1, B2), susceptibility-weightedmagnitude and phase images (B3,B4), and CT/radiographs (B5, B6). The osteophyte can be clearly identified on susceptibility-weighted magnitude images (B3, B4) as a hyperintenseosseous structure. CT/radiography confirmed the presence of an osteophyte (B5, B6). On conventional T1/T2-weighted images, the dorsal osteophyte(B1, B2) cannot be differentiated from the dorsal disc herniation (A1, A2).

Investigative Radiology • Volume 52, Number 2, February 2017 SW-MRI for Spinal Radiculopathy

osteophytes with a 95% confidence interval of 56.2% to 78.9% and79.9% to 91.3%, respectively.

Sensitivity and Specificity for the Detection of PosteriorDisc Herniations

As reference standard, 114 posterior disc herniations (in60 patients) were identified on conventional T1/T2 MR in combina-tion with radiographs excluding osteophytes. If radiographs wereexcluded from the analysis, the majority (n = 113 of 114) of dis-cal herniations were identified on SW-MRI magnitude images(Figs. 1A3-A4, Figs. 2B3-B4). Susceptibility-weighted magneticresonance imaging achieved a sensitivity of 99.1% and a specificityof 98.9% for the identification of disc herniations with a 95% confi-dence interval of 94.4% to 99.9% and 93.4% to 99.9%, respectively.If radiographs were excluded from the analysis, conventional T1/T2spine MR sequences achieved a sensitivity of 86.5% and a specific-ity of 68.6% for the identification of dorsal disc herniations with a95% confidence interval of 79.9% to 91.3% and 56.2% to 78.9%, re-spectively (Figs. 1A1-A2, Figs. 2B1-B2).

Assessment of Size of OsteophytesSize measurements revealed a close correlation for osteophytes

between SW-MRI and radiographs (R2 = 0.95, P < 0.05). There was nosignificant difference of size measurements between SW-MRI and radio-graphs (6.3 ± 4.0 mm vs 6.4 ± 4.2 mm). Conventional T2-weighted MRspine sequences showed a moderate correlation (R2 = 0.82, P < 0.05)



compared with radiography-based size measurements of osteophytes.Conventional T1-weighted MRI scans showed a lower moderate correla-tion (R2 = 0.62, P < 0.05) compared with radiography-based size mea-surements of osteophytes. In T2-weighted sequences, the size ofcalcifications was underestimated compared with radiographs (5.8 ±4.0 mm vs 7.2 ± 4.5 mm). In 19 of 81 patients, a total of 33 osteophyteswere detected on CT images. All of these could be identified onsusceptibility-weightedmagnitude images. Sizemeasurements revealed aclose correlation for osteophytes between CT and SW-MRI (R2 = 0.96,P < 0.05) as well as for CTand radiographs (R2 = 0.92, P < 0.05). Therewas no significant difference of size measurements between SW-MRIand CT (6.0 ± 3.7 vs 6.2 ± 3.8 mm) as well as radiographs and CT(6.7 ± 3.9 vs 6.8 ± 3.8 mm).

Signal(osteophytes)-to-Signal(intervertebral disc) Ratio (CNR)On T2-weighted images, a CNR of 35.5 ± 17.1 was mea-

sured. On inverse susceptibility-weighted magnitude, a CNR of50.4 ± 0.20.8 was measured, demonstrating the significantlyhigher (P < 0.05) difference in signal or contrast for osteophytescompared with intervertebral discs on inverse susceptibility-weightedmagnitude images.

Interobserver CorrelationBetween both readers, there was a close correlation of interob-

server agreement regarding size measurements (R2 = 0.97) in SW-MRIwith only a minimum difference in mean (0.01) (Fig. 3, A–B). The




FIGURE 2. SW-MRI for the evaluation of dorsal osteophytes and disc herniations in the cervical spine in a 72-year-old (A) and 59-year-old (B)male patientwith radiculopathy. A, Dorsal osteophyte (arrows) on sagittal native T1/T2 conventional MR sequences (A1, A2), susceptibility-weightedmagnitude/phase images (A3, A4), and radiographs (A5). The osteophyte can be clearly identified on susceptibility-weighted magnitude/phase images(A3, A4) as a hyperintense osseous structure. Radiography confirmed the presence of an osteophyte (A5). On conventional T1/T2-weighted images(A1, A2), it is challenging to differentiate whether a dorsal osteophyte or dorsal disc herniation is present as both appear with a hypointense signal.B, Dorsal disc herniation (arrows) on sagittal T1/T2 conventionalMR sequences (B1, B2) and susceptibility-weightedmagnitude/phase images (B3, B4).The dorsal disc herniation can be clearly identified on susceptibility-weighted images (B3, B4) as a hypointense herniation. Radiography (B5)excluded osteophytes. On conventional T1/T2-weighted images (A1, A2), the dorsal osteophyte cannot reliably be differentiated from the dorsaldisc herniation (B1, B2).


95% confidence intervals ranged from −1.26 to 1.24 in Bland-Altmanplots. Interobserver correlation for size measurements in conventionalT2-weighted MR sequences showed a moderate correlation of measure-ments (R2 = 0.94) with a small difference in mean (0.2) (Fig. 3, C−D). InBland-Altman plots, 95% confidence intervals ranged from −3.54to 3.26.

DISCUSSIONThis study demonstrated that SW-MRI enables the differentia-

tion of osteophytes and posterior disc herniations in patients with spinalradiculopathy with a higher sensitivity and specificity compared withconventional T1/T2-weighted MR sequences.

Current Clinical Imaging Assessment of Osteophytesand Posterior Disc Herniations

Radiography is the most frequently used technique for the detec-tion of osteophytes and osseous changes of the degenerative spine.Computed tomography imaging of the spine provides more detailed



information about the dimension of osseous changes but is associatedwith a significant radiation dose. Magnetic resonance imaging is usu-ally performed as the first-line imaging study in patients with progres-sive neurologic symptoms and especially if plain radiography filmsare negative. With its superior soft tissue resolution, MRI provides ex-cellent anatomic detail.18 Changes in signal intensity, disc structure,disc height, and distinction between nucleus and annulus are usuallyevaluated on T2-weighted sagittal images. Guidelines for a more con-sistent description of disc herniations were released.18–21 Especially,the differentiation of dorsal osteophytes and disc herniations can bechallenging on conventional T1/T2 MR sequences, as both structurescan appear hypointense on T1- and T2-weightedMRI scans. Therefore,radiography and CTare currently used as additional imaging modalitiesfor the visualization of these changes.

Susceptibility-Weighted MRICurrently, the interpretation of conventional MRI of the spine is

based on the magnitude information of the image. Because of phase




FIGURE 3. Interobserver correlation and agreement of size measurements of osteophytes. A and B, Size measurement showed a strong correlation andinterobserver agreement for SW-MRI. C and D, For conventional T2 images, a moderate correlation with a wider 95% confidence interval wasmeasured. Bland-Altman plots: centerline, mean absolute difference central line; upper and lower line, limits of agreement (95% confidence intervals).

Investigative Radiology • Volume 52, Number 2, February 2017 SW-MRI for Spinal Radiculopathy

artifacts, it has been challenging in the past to extract useful phase infor-mation for further tissue characterization. Susceptibility-weighted im-aging is based on a high-resolution 3D gradient-echo sequence. Itmakes use of magnitude and filtered phase information separatelyand in combination with each other during image acquisition to createnew sources of contrast.11–13,22 Diamagnetic (eg, calcified lesions),paramagnetic (eg, hemosiderin), and ferromagnetic (eg, iron) com-pounds interact with the local magnetic field and therefore alter thephase of local tissues.22,23 Calcified structures appear with a hyperin-tense signal on inverse magnitude images and with a strong hyperintensesignal on SW-MRI filtered phase images. Phase artifacts, for example, ontissue interfaces, appear dark on magnitude as well as on SW-MRI phaseimages.15,22,24 The main challenge for MRI with standard T1/T2 se-quences is the differentiation between calcified spondylophytes and otherlesions that cause a signal loss in T1/T2 sequences. The most importantinformation for the differentiation of these lesions can be derived fromthe phase image. Phase images are, however, more prone to artifacts,for example, minor motion artifacts. In these cases, it can be difficultto clearly differentiate small spondylophytes from tissue artifacts.Susceptibility-weighted magnetic resonance imaging sequences havebeen previously used mainly for brain imaging to visualize venous vas-culature11,12,22,25 and to distinguish between intracerebral hemorrhage,calcifications, and changes in iron content in, for example, stroke, neu-rodegenerative disorders, trauma, and tumors.6,12,22,23,26–29 In addition,it was shown that SW-MRI can be useful for the identification of plaquecalcifications in atherosclerosis30 and the distinction between calcifica-tions and hemorrhagic lesions in patients with prostatic cancer.10

SW-MRI for the Differentiation of Osteophytes andPosterior Disc Herniations of the Spine

In this study, SW-MRI enabled the reliable identification ofosteophytes and disc herniations in the cervical and lumbar spinewith a higher sensitivity and specificity compared with conventionalT1/T2 MR sequences. In addition, the interobserver correlation and



agreement for the size measurement of osteophytes was higher forSW-MRI compared with that for conventional T1/T2 MR. Thiscould be explained by 2 features of SW-MRI. First, SW-MRI allowsthe differentiation of diamagnetic calcified lesions from surroundingstructures, based on the magnitude and phase information. Second,the contrast (CNR) between osteophytes and disc herniations wassignificantly higher on SW-MRI compared with that on conven-tional MR sequences.

In a clinical setting, the analysis of the additional sequences can bebased on a visual assessment of the additional magnitude and phase im-ages and therefore is not time-consuming. The analysis directly adds tothe clinical and surgical decision tree as it provides the differentiation be-tween spondylophytes and disc herniations without the need for furtherradiographs or CT imaging. Especially for preoperative planning, it is im-portant to determine whether discectomy is sufficient or if osteophyteshave to be removed additionally for successful treatment. Based on thecause of spinal radiculopathy, surgery can be performed with an anterior,posterior, or even combined approach. A prospective study demonstratedthat surgical decompression arrests progression and improves neurologi-cal outcomes, functional status, and quality of life; therefore, surgery isincreasingly recommended as the standard treatment for patients withdegenerative myelopathy.31

In addition, SW-MRI can provide important information aboutthe dimension and the precise spatial localization of osteophytes, whichcan be challenging based on 2-dimensional radiography. Therefore, itcan help to plan the dimension of anterior decompression preoperatively,for example, if a resection of spondylophytes cranially or caudally is suf-ficient enough or if even a corpectomy as a wide decompression needsto be taken into account. Because SW-MRI shows a good performancein detecting these changes, it should also be a future goal to reduce theuse of ionizing radiation, especially in young patient collectives.

In future studies, the applicability of SW-MRI for the assessmentof other types of calcifying musculoskeletal diseases such as calcifyingbone lesions, bone tumors, and calcifying soft tissue tumors shouldbe examined.





LimitationsOne limitation of the current study was that CTwas not available

in all patients, as it was not clinically required and the additional radia-tion dose was therefore not justified. As the diagnosis of off-center mar-ginal osteophytes of the end plates is limited on radiographs, this type ofosteophyte was excluded from our analysis.

No correlation analysis of SW-MRI findings with symptoms ofpatients was performed. Future studies are needed for further evaluationof the potential of SW-MRI in a larger patient collective. Susceptibility-weighted magnetic resonance imaging artifacts from local field inho-mogeneity due to metallic implants are seen as concentric rings ofhyperintensity and hypointensity and result in limited assessment ofthe affected region.

CONCLUSIONSSusceptibility-weighted magnetic resonance imaging enables

the differentiation of osteophytes and disc herniations in patientswith spinal radiculopathy with a higher sensitivity and specificitycompared with conventional T1/T2 MR sequences. This is of rele-vance for routine MRI of the spine, as an additional SW-MRI se-quence may avoid further workup with CT imaging or radiographyfor assessing osseous structures.

REFERENCES

1. Meyer F, Börm W, Thomé C. Degenerative cervical spinal stenosis: current strat-egies in diagnosis and treatment. Dtsch Arztebl Int. 2008;105:366–372.

2. Fu MC, Webb ML, Buerba RA, et al. Comparison of agreement of cervical spinedegenerative pathology findings in magnetic resonance imaging studies. Spine J.2016;16:42–48.

3. Green C, Butler J, Eustace S, et al. Imagingmodalities for cervical spondylotic ste-nosis and myelopathy. Adv Orthop. 2012;2012:908324.

4. Yi JS, Cha JG, Han JK, et al. Imaging of herniated discs of the cervical spine:inter-modality differences between 64-slice multidetector CT and 1.5-T MRI.Korean J Radiol. 2015;16:881–888.

5. Morishita Y, Naito M, Hymanson H, et al. The relationship between the cervicalspinal canal diameter and the pathological changes in the cervical spine.Eur Spine J. 2009;18:877–883.

6. ChenW, ZhuW, Kovanlikaya I, et al. Intracranial calcifications and hemorrhages:characterization with quantitative susceptibility mapping. Radiology. 2014;270:496–505.

7. Zhu WZ, Qi JP, Zhan CJ, et al. Magnetic resonance susceptibility weighted imag-ing in detecting intracranial calcification and hemorrhage. Chin Med J (Engl).2008;121:2021–2025.

8. Wisnieff C, Ramanan S, Olesik J, et al. Quantitative susceptibility mapping(QSM) of white matter multiple sclerosis lesions: interpreting positive susceptibil-ity and the presence of iron. Magn Reson Med. 2015;74:564–570.

9. Wang Y, Liu T. Quantitative susceptibility mapping (QSM): decoding MRI datafor a tissue magnetic biomarker. Magn Reson Med. 2015;73:82–101.

10. Bai Y, Wang MY, Han YH, et al. Susceptibility weighted imaging: a new tool inthe diagnosis of prostate cancer and detection of prostatic calcification. PLoSOne. 2013;8:e53237.



11. Mittal S, Wu Z, Neelavalli J, et al. Susceptibility-weighted imaging: technical as-pects and clinical applications, part 2.AJNRAm J Neuroradiol. 2009;30:232–252.

12. Haacke EM, Mittal S, Wu Z, et al. Susceptibility-weighted imaging: technical as-pects and clinical applications, part 1. AJNR Am J Neuroradiol. 2009;30:19–30.

13. Haacke EM, Xu Y, Cheng YC, et al. Susceptibility weighted imaging (SWI).Magn Reson Med. 2004;52:612–618.

14. Zulfiqar M, Dumrongpisutikul N, Intrapiromkul J, et al. Detection of intratumoralcalcification in oligodendrogliomas by susceptibility-weighted MR imaging.AJNR Am J Neuroradiol. 2012;33:858–864.

15. Wu Z, Mittal S, Kish K, et al. Identification of calcification with MRI usingsusceptibility-weighted imaging: a case study. J Magn Reson Imaging. 2009;29:177–182.

16. Chavhan GB, Babyn PS, Thomas B, et al. Principles, techniques, and applicationsof T2*-based MR imaging and its special applications. Radiographics. 2009;29:1433–1449.

17. Cheng AL, Batool S, McCreary CR, et al. Susceptibility-weighted imaging ismore reliable than T2*-weighted gradient-recalled echo MRI for detectingmicrobleeds. Stroke. 2013;44:2782–2786.

18. Wilmink JT. MR imaging of the spine: trauma and degenerative disease.Eur Radiol. 1999;9:1259–1266.

19. Maataoui A, Vogl TJ, KhanMF.Magnetic resonance imaging-based interpretationof degenerative changes in the lower lumbar segments and therapeutic conse-quences.World J Radiol. 2015;7:194–197.

20. Pfirrmann CW, Metzdorf A, Zanetti M, et al. Magnetic resonance classifica-tion of lumbar intervertebral disc degeneration. Spine (Phila Pa 1976). 2001;26:1873–1878.

21. Lurie JD, DomanDM, Spratt KF, et al. Magnetic resonance imaging interpretationin patients with symptomatic lumbar spine disc herniations: comparison of clini-cian and radiologist readings. Spine (Phila Pa 1976). 2009;34:701–705.

22. Nörenberg D, Ebersberger HU, Walter T, et al. Diagnosis of calcific tendonitis ofthe rotator cuff by using susceptibility-weighted MR imaging. Radiology. 2016;278:475–484.

23. Tong KA, Ashwal S, Obenaus A, et al. Susceptibility-weightedMR imaging: a re-view of clinical applications in children. AJNR Am J Neuroradiol. 2008;29:9–17.

24. Ogg RJ, Langston JW, Haacke EM, et al. The correlation between phase shifts ingradient-echo MR images and regional brain iron concentration.Magn Reson Im-aging. 1999;17:1141–1148.

25. Sehgal V, Delproposto Z, Haacke EM, et al. Clinical applications of neuroimagingwith susceptibility-weighted imaging. J Magn Reson Imaging. 2005;22:439–450.

26. Gao T, Wang Y, Zhang Z. Silent cerebral microbleeds on susceptibility-weightedimaging of patients with ischemic stroke and leukoaraiosis. Neurol Res. 2008;30:272–276.

27. Nandigam RN, Viswanathan A, Delgado P, et al. MR imaging detection of cere-bral microbleeds: effect of susceptibility-weighted imaging, section thickness,and field strength. AJNR Am J Neuroradiol. 2009;30:338–343.

28. Santhosh K, Kesavadas C, Thomas B, et al. Susceptibility weighted imaging: anew tool in magnetic resonance imaging of stroke. Clin Radiol. 2009;64:74–83.

29. Tong KA, Ashwal S, Holshouser BA, et al. Hemorrhagic shearing lesions in chil-dren and adolescents with posttraumatic diffuse axonal injury: improved detectionand initial results. Radiology. 2003;227:332–339.

30. Yang Q, Liu J, Li K, et al. Visualization of carotid plaque calcification—a novelapproach using susceptibility weighted MR imaging. J Cardiovasc Magn Reson.2010;12(Suppl 1).

31. Fehlings MG, Wilson JR, Kopjar B, et al. Efficacy and safety of surgical decom-pression in patients with cervical spondylotic myelopathy: results of the AOSpineNorth America prospective multi-center study. J Bone Joint Surg Am. 2013;95:1651–1658.




Differentiation of Osteophytes and Disc Herniations in ...clinical-mri.com/wp-content/uploads/2017/...Disc.1.pdf · osteophytes, and/or posterior disc herniations of the cervical

Documents