This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chan, J H M Articles by Wong, K P C Search for Related Content PubMed PubMed Citation Articles by Chan, J H M Articles by Wong, K P C

Acute vertebral body compression fractures: discrimination between benign and malignant causes using apparent diffusion coefficients

Departments of ¹ Diagnostic Radiology and ⁴ Orthopedic Surgery, Tuen Mun Hospital, Hong Kong, ² Department of Diagnostic Radiology, Singapore General Hospital, Singapore and Departments of ³ Diagnostic Radiology and ⁵ Orthopedic Surgery, North District Hospital, Hong Kong

Correspondence: Wilfred C G Peh, Department of Diagnostic Radiology, Singapore General Hospital, Outram Road, Singapore 169608

	Abstract

Top Abstract Introduction Materials and methods Results Discussion References

 
Diffusion weighted MRI was performed on patients with acute vertebral body compression. The usefulness of the apparent diffusion coefficient (ADC) in differentiating between benign and malignant fractures was evaluated. A total of 49 acute vertebral body compression fractures were found in 32 patients. 25 fractures in 18 patients were due to osteoporosis, 18 fractures in 12 patients were histologically proven to be due to malignancy, and 6 fractures in 2 patients were due to tuberculosis. Signal intensities on T₁ weighted, short tau inversion recovery (STIR) and diffusion weighted images were compared. ADC values of normal and abnormal vertebral bodies were calculated. Except for two patients with sclerotic metastases, benign acute vertebral fractures were hypointense and malignant acute vertebral fractures were hyperintense with respect to normal bone marrow on diffusion weighted images. Mean combined ADCs (ADC_cmb; average of the combined ADCs in the x, y and z diffusion directions) were 0.23x10^-3 mm² s^-1 in normal vertebrae, 0.82x10^-3 mm² s^-1 in malignant acute vertebral fractures and 1.94x10^-3 mm² s^-1 in benign acute vertebral fractures. The differences between ADC_cmb values were statistically significant (p<0.001). The ADC is useful in differentiating benign from malignant acute vertebral body compression fractures, but there may be overlapping ADC values between malignant fractures and tuberculous spondylitis.

	Introduction

Top Abstract Introduction Materials and methods Results Discussion References

 
The spine is the most common location of bony metastases in patients with an underlying malignancy [1]. Although previous reports have claimed that T₁ and T₂ weighted MRI is the examination of choice for differentiating benign from malignant vertebral fractures [2–5], discrimination between acute benign and malignant vertebral fractures may sometimes be difficult. Baur et al [6] reported their preliminary results using MR diffusion weighted imaging (DWI) in discriminating between benign and malignant acute vertebral body compression fractures. Similar experiments were performed by Castillo et al [7] and Spuentrup et al [8]. Castillo et al [7] found that DWI offered no additional value in making a diagnosis, while Spuentrup et al [8] found that DWI provided excellent distinction between benign and malignant fractures. The rationale for using DWI is that differences between benign and malignant fractures are mainly due to cellularity and free water content. As DWI is highly sensitive to cellularity and free water molecule mobility, DWI should be useful in differentiating between vertebral body compression fractures caused by tumour infiltration (malignant) and trauma (benign).

Both Baur et al [6] and Castillo et al [7] utilized the steady-state free precession (SSFP) diffusion weighted (DW) pulse sequence with a b-factor of 165 s mm^-2 [6, 7]. As the b-factor is rather low for diffusion weighting to be dominant on the MR images, we feel that the signal hyperintensity within malignant vertebral body compression fractures may possibly be due to the T2 "shine through" effect [9]. The SSFP DW pulse sequence also suffers from a severe trade-off in that it does not allow accurate determination of apparent diffusion coefficients (ADCs). This is because the signal attenuation factor, and hence the resulting MR signal, depends on both the diffusion process as well as T1 and T2 relaxation [9]. The present study aims to perform MRI on patients with acute vertebral body compression fractures with the single shot echo planar DW pulse sequence and to evaluate the usefulness of ADC values in differentiating between vertebral body compression fractures caused by either tumour infiltration (malignant) or solely trauma (benign).

	Materials and methods

Top Abstract Introduction Materials and methods Results Discussion References

 
32 consecutive symptomatic patients with acute vertebral body compression fractures of less than 14 days duration and detected by radiography were referred for MRI of the spine during a 4-month period from 1 April 2000 to 31 July 2000. Inclusion criteria of patients with acute fracture were: (i) definite history of acute pain onset; and (ii) physical examination showing local tenderness that corresponded to the site of radiographic abnormalities. Informed consent for the study was obtained from all patients. 23 patients had a solitary vertebral body fracture. A total of 49 vertebral body fractures were found in the 32 patients. 25 vertebral fractures were found in 18 patients (7 men and 11 women; mean age 63.8 years) who did not undergo bone biopsy. Biopsy was not performed in these patients because of obvious history of trauma, no known primary malignancy, and the absence of posterior element involvement and epidural soft tissue mass on MR images of the vertebrae. Following a 3-month follow-up by physical examination (12 patients) or MR examination (6 patients), all these patients were presumed to have benign vertebral fractures either due to trauma or osteoporosis.

14 patients (8 men and 6 women; mean age 64.2 years) had a known primary malignancy and all these patients underwent bone biopsy. For those patients with multiple pathological fractures, bone biopsy was performed at the sites judged as most likely to be abnormal. Histopathology confirmed that 18 fractures in 12 patients were due to malignancy. Histopathological diagnoses were pulmonary carcinoma (n=7), breast carcinoma (n=4), lymphoma (n=3), prostatic carcinoma (n=3) and multiple myeloma (n=1). Two male patients (mean age 68.4 years) had two and four vertebral body fractures, respectively, which were initially thought to be malignant compression fractures but were later found on histopathologic examination to be due to TB spondylitis.

MRI was performed using a 1.5 T whole body superconducting MR imager (Signa Horizon, Echospeed, Software version 5.8.1; General Electric Medical Systems, Milwaukee, WI) equipped with high speed gradients. The maximal gradient strength was 23 mT m^-1 and the slew rate was 120 T m^-1 s^-1. A phased array cervico–thoraco–lumbar spine coil was used for T₁ and T₂ weighted imaging, and a linearly polarized body coil was used for DWI. Following T₁ weighted fast gradient echo (TR 10 ms; TE 4 ms; flip angle 30°) coronal localizer images, sagittal T₁ weighted spin echo (TR 500 ms; TE 9 ms) and sagittal fast short tau inversion recovery (STIR) (TR 6000 ms; TE 85 ms; TI 150 ms; echo train length 8) images were acquired in all patients.

The MR pulse sequence used for DWI was a single shot echo planar pulse sequence with the following imaging parameters: TR 8000 ms; TE 92 ms; receiver bandwidth ±133 kHz; field of view 35 cmx35 cm; matrix size 128x128; slice thickness 8 mm; interslice gap 2 mm; and number of excitations 1. Data acquisition was repeated with four different b-factors (200 s mm^-2, 500 s mm^-2, 800 s mm^-2, 1000 s mm^-2) while all other imaging parameters were kept constant. For each b-value, data acquisition was also repeated with the diffusion-sensitising gradients placed along the three orthogonal (x, y and z) directions. Only sagittal DW images were acquired.

The signal intensities of the corresponding T₁ weighted images, STIR images and DW images with b=1000 s mm^-2 were noted and compared. Only DW images with b=1000 s mm^-2 were used for qualitative analysis, since these DW images possess sufficient diffusion weighting [9, 10]. The vertebral bodies were considered abnormal based on both the radiographic and MRI findings. The radiographs (both frontal and lateral views), sagittal T₁ weighted images and STIR MR images of the spine were reviewed by two experienced radiologists who were blinded to clinical information. The following vertebral features were assessed: (1) break in continuity of bone and loss of normal morphology on radiography; (2) abnormal bone marrow signal intensity; (3) posterior element involvement; and (4) epidural mass lesion on MR images. Decision was made by consensus.

The ADC values of the normal and abnormal vertebral bodies in all patients were calculated using the formula:

where ADC_(x,y,z) is the ADC value of a voxel at location (x,y,z), S_1(x,y,z) is the signal intensity of a voxel at location (x,y,z) of a DW image acquired with b₁, and S₂(x,y,z) is the signal intensity of the same voxel of another DW image acquired with b₂. Linear regression algorithms (i.e. least-square fitting of logarithm of the signal intensity log(S) vs four different b-values) were used to compute more accurate ADCs. Quantitative evaluation of vertebral body signal intensity was measured by placing a region of interest (ROI) cursor in the centre of each abnormal vertebral body (Figure 1) and measuring the signal intensity in each pixel within the ROI. This procedure was repeated for all the images acquired with different b-factors.

View larger version (48K): [in this window] [in a new window]   Figure 1. 68-year-old man with malignant acute vertebral compression fracture. Sagittal diffusion weighted MR image with a b-factor of 200 s mm-2 showing location of region of interest cursor in the fractured lumbar vertebral body.

	Results

Top Abstract Introduction Materials and methods Results Discussion References

 
8% of vertebral fractures were found in the cervical spine 38% in the thoracic spine and 54% in the lumbar spine. All but two malignant vertebral body compression fractures were hypointense on T₁ weighted spin echo images and hyperintense on fast STIR images with respect to the normal bone marrow. The remaining two vertebral body fractures (prostatic carcinoma) showed sclerotic features and were markedly hypointense on both T₁ weighted spin echo and fast STIR images. Of the 18 malignant fractures, 7 fractures were associated with a soft tissue mass in the epidural space and 3 fractures showed posterior element involvement. 6 (33%) malignant vertebral fractures displayed similar features to benign vertebral body compression fractures on conventional MR images (Figures 2 and 3) but could be definitely differentiated using their ADC values.

View larger version (170K): [in this window] [in a new window]   Figure 2. 65-year-old man with malignant vertebral body compression fracture due to metastasis from pulmonary carcinoma. (a) Sagittal T1 weighted spin echo MR image (TR=500 ms, TE=8 ms) showing L2 vertebral body signal hypointensity with respect to the normal bone marrow. (b)Sagittal fast STIR MR image (TR=3220 ms, TE=65 ms, TI=150 ms, echo train length=8) showing L2 vertebral body signal hyperintensity with respect to the normal bone marrow. Mild vertebral retropulsion is present.

View larger version (163K): [in this window] [in a new window]   Figure 3. 43-year-old man with traumatic vertebral body compression fracture. (a) Sagittal T1 weighted spin echo MR image (TR=500 ms, TE=8 ms) showing L1 vertebral body signal hypointensity with respect to the normal bone marrow. (b) Sagittal fast STIR MR image (TR=3220 ms, TE=65 ms, TI=150 ms, echotrain length=8) showing L1 vertebral body signal hyperintensity with respect to the normal bone marrow.

For malignant vertebral body compression fractures, 12 fractures had complete marrow replacement and 6 fractures showed incomplete marrow replacement on conventional MRI. On the other hand, most (n=18) benign vertebral fractures displayed incomplete marrow replacement and 7 showed complete marrow replacement on conventional MRI. Benign acute vertebral body compression fractures were hypointense and malignant acute vertebral body compression fractures were hyperintense on DW (b=1000 s mm^-2) images with respect to normal bone marrow (Figure 4). For the two patients with sclerotic metastases, the vertebral fractures were hypointense on T₁ weighted, STIR and DW images with respect to normal bone marrow, and had an insignificantly low ADC (i.e. an ADC value close to zero, indicating virtually no diffusion).

View larger version (68K): [in this window] [in a new window]   Figure 4. (a) 60-year-old man with benign acute vertebral body compression fractures. Sagittal MR diffusion weighted image (TR/TE 8000/92) with b-factor of1000 s mm-2 showing signal hypointensity in the fractured vertebral body (arrow) of the lumbar spine. (b) 71-year-old woman with malignant (pulmonary carcinoma metastasis) acute vertebral bodycompression fractures. Sagittal MR diffusion weighted image (TR/TE 8000/92) with b-factor of 1000 s mm-2 showing signal hyperintensity in the fractured vertebral bodies of the lumbar spine (arrow).

View larger version (148K): [in this window] [in a new window]   Figure 5. 67-year-old man with TB spondylitis. (a) Sagittal T1 weighted spin echo MR image (TR=450 ms, TE=8 ms) showing L2 and L3 vertebral body signal hypointensity with respect to the normal bone marrow. (b) Sagittal fast STIR MR image (TR=3300 ms, TE=65 ms, TI=150 ms, echo train length=8) showing L2 and L3 vertebral body signal hyperintensity with respect to the normal bone marrow.

View larger version (9K): [in this window] [in a new window]   Figure 6. Graph showing the relative apparent diffusion coefficients (ADCs) of normal vertebral bodies, malignant acute vertebral bodycompression fractures, TB spondylitis and benign acute vertebral bodycompression fractures. There is overlap between the ADC values of the malignant and TB spondylitis vertebral body compression fractures.

	Discussion

Top Abstract Introduction Materials and methods Results Discussion References

The rapidity of the EPI pulse sequence also enables data acquisition with many different b-values within a reasonably short time, thus enabling accurate calculation of ADC. This is particularly important if diffusion imaging in the x, y and z directions is needed. This ultrafast imaging technique has broadened the clinical utility of DWI to other parts of the body, such as the spine. The main trade-off of the EPI pulse sequence is that it is very sensitive to magnetic susceptibility effects, resulting in geometric distortion artefacts that tend to be more severe with increasing b-values. The image distortion artefact may cause significant errors in the computation of the ADC. The geometric distortion can be reduced by using high receiver bandwidth, using a large field of view, acquiring thick slices and using a radiofrequency (RF) coil of cylindrical geometry, such as a body coil.

The signal-to-noise ratio could be improved by using a phased array RF coil with multi-channel RF receivers. Unfortunately, our MR system does not allow the single shot EPI DW pulse sequence to be executed with phased array RF coils. In our study, the image quality and anatomical details were not vitally important since our results were entirely based upon measurement of ADC values. To reduce chemical shift artefacts, the EPI pulse sequence was defaulted with a fat suppression (spectral saturation) technique. On all DW images, fat marrow signal was nulled. Normal vertebral bodies were thus hypointense on both DW images and STIR images because the fat marrow signal was suppressed and there was a negligible amount of free water content in the interstitial space. Thus, there were a minimal amount of mobile protons available for generating the MR signal. The calculated ADCs of the normal vertebral bodies were very low but they failed to show signal hyperintensity on DW images owing to lack of mobile protons in the vertebral bodies. In other words, there was minimal diffusion in normal vertebral bodies. Identical signal hypointensity was observed from skull bones when a DW EPI pulse sequence was used. We retrospectively calculated the ADCs of the skull bone from 10 examinations of brain scans, and the mean ADC was found to be (0.28±0.05)x10^-3 mm² s^-1. Ward et al [14] reported that normal red and yellow bone marrow showed minimal diffusion, with a very small mean ADC.

In malignant acute vertebral fractures, there is replacement of the fat marrow cells in the vertebral bodies by an accumulation of tumour cells. The diffusion of water molecules inside the tumour cells is relatively restricted, resulting in signal hyperintensity on DW images. The ADCs of the malignant vertebral body fractures were larger than those of the normal vertebral bodies owing to the presence of mobile protons in the tumour tissue. Benign acute vertebral fractures have an increased amount of free water content in the interstitial space, thus increasing water mobility. Thus the diffusion process is markedly increased, resulting in signal hypointensity on DW images. The ADCs of the benign acute vertebral body fractures are much larger than those of pathological vertebral body fractures. In summary, the ADC value indicates the amount of free water content in the interstitial space of the vertebral bodies, being maximal in benign acute vertebral body fractures and minimal in normal vertebral bodies. It should be pointed out that two cases with vertebral body sclerotic metastases from prostatic carcinoma, like normal vertebral bodies, were hypointense on both DW images and STIR images with respect to normal bone marrow, and had a very low ADC value owing to the lack of free water content. In this situation, the signal characteristics on T₁ and T₂ weighted MR images are relied upon.

The signal intensity of DW images is governed by two terms, namely the diffusion term (i.e. water diffusibility) and the T2 relaxation term (i.e. the intrinsic T2 property). The resultant contrast of the DW images is thus a mixture of diffusion contrast and T2 contrast. A tissue with a long T2 relaxation time may appear hyperintense on DW images because of the T2 shine through effect [15, 16]. Baur et al [6] reported that malignant vertebral fractures were hyperintense on all MR images acquired with (i.e. diffusion weighted) and without (i.e. T₂ weighted) the diffusion-sensitizing gradient. It was very likely that the signal hyperintensity on DW images was due to T2 shine through effect since these investigators used an extremely low b-factor. To make the signal intensity dependent on the diffusion term, it is necessary to use a sufficiently large b-factor. Otherwise, ADC maps must be reconstructed, since the ADC maps are free of the T2 shine through effect. With the high b-value of 1000 s mm^-2 used in our study, the shine through effect was reduced. It was interesting to note that, except for the two patients with sclerotic metastases, the DW images were able to distinguish benign from malignant acute vertebral body fractures.

In our study, the actual ADC values were used directly as markers for achieving the differential diagnosis, thus eliminating other sources of error apart from the T2 shine through effect. One limitation of the study was that it failed to differentiate tuberculous spondylitis from malignant vertebral fractures, since the ADCs of the vertebral fractures with tuberculous spondylitis are close to those of malignant acute vertebral fractures (Figure 6). In addition, the exact age of fractures may not be accurately known in routine clinical MRI, and a longitudinal study of DWI, ADC maps and ADC values in patients with benign and malignant compression fractures of known ages would be useful. Theoretically, a healing benign compression fracture may, over time, pass through ADC values that would be within the range of malignant compression fractures, en route to producing ADC values typical of normal bone marrow. There may therefore be a potential for confusion when using ADC values in benign compression fractures during the non-acute or acute-on-chronic stage.

Diffusion in red and yellow bone marrow is anisotropic [14] and should be completely described by a diffusion tensor that requires measurements with at least seven different diffusion directions. A combined ADC, which is the average of the individual ADCs in three orthogonal axes, has been shown to be adequate in clinical applications [17–19]. From our clinical results, we conclude that apparent diffusion coefficients are a useful and reliable supplementary tool in differentiating benign from malignant acute vertebral body compression fractures.

Received for publication May 24, 2001. Revision received October 16, 2001. Accepted for publication November 29, 2001.

	References

Top Abstract Introduction Materials and methods Results Discussion References

 

1. Ratanatharathorn V, Powers WE. Epidural spinal cord compression for metastatic tumor: diagnosis and guidelines for management. Cancer Treat Rev 1991;18:55–71.[Medline]
2. Moulopoulos LA, Yoshimitsu K, Johnston DA, Leeds NE, Libshitz HI. MR prediction of benign and malignant vertebral compression fractures. J Magn Reson Imaging 1996;6:667–74.[Medline]
3. Baker LL, Goodman SB, Perkash I, Lane B, Enzmann DR. Benign versus pathologic compression fractures of vertebral bodies: assessment with conventional spin-echo, chemical-shift, and STIR MR imaging. Radiology 1990;174:495–502.[Abstract/Free Full Text]
4. Rupp RE, Ebraheim NA, Coombs RJ. Magnetic resonance imaging differentiation of compression spine fractures or vertebral lesions caused by osteoporosis or tumor. Spine 1995;20:2499–504.[Medline]
5. Shih TT, Huang KM, Li YW. Solitary vertebral collapse: distinction between benign and malignant causes using MR patterns. J Magn Reson Imaging 1999;9:635–42.[Medline]
6. Baur A, Stäbler A, Brüning R, Bartl R, Krödel A, Reiser M, et al. Diffusion-weighted MR imaging of bone marrow: differentiation of benign versus pathologic compression fractures. Radiology 1998;207:349–56.[Abstract/Free Full Text]
7. Castillo M, Arbelaez A, Smith JK, Fisher LL. Diffusion-weighted MR imaging offers no advantage over routine noncontrast MR imaging in the detection of vertebral metastases. Am J Neuroradiol 2000;21:948–53.[Abstract/Free Full Text]
8. Spuentrup E, Buecker A, Adam G, van Vaals JJ, Guenther RW. Diffusion-weighted MR imaging for differentiation of benign fracture edema and tumor infiltration of the vertebral body. AJR 2001;176:351–8.[Abstract/Free Full Text]
9. Le Bihan DJ. Differentiation of benign versus pathologic compression fractures with diffusion-weighted MR imaging: a closer step toward the "holy grail" of tissue characterization. Radiology 1998;207:305–7.[Free Full Text]
10. Kim YJ, Chang KH, Song IC, Kim HD, Seong SO, Kim YH, Han MH. Brain abscess and necrotic or cystic brain tumor: discrimination with signal intensity on diffusion-weighted MR imaging. AJR 1998;171:1487–90.[Abstract/Free Full Text]
11. Stejskal EO, Tanner JE. Spin diffusion measurements: spin echoes in the presence of time-dependent field gradient. J Chem Phys 1965;42:288–92.
12. Kerttula LI, Jauhiainen JP, Tervonen O, Suramo IJ, Koivula A, Oikarinen JT. Apparent diffusion coefficient in thoracolumbar intervertebral discs of healthy young volunteers. J Magn Reson Imaging 2000;12:255–60.[Medline]
13. Tanner SF, Ramenghi LA, Ridgway JP, Berry E, Saysell MA, Martinez D, et al. Quantitative comparison of intrabrain diffusion in adults and preterm and term neonates and infants. AJR 2000;174:1643–9.[Abstract/Free Full Text]
14. Ward R, Caruthers S, Yablon C, Blake M, DiMasi M, Eustace S. Analysis of diffusion changes in posttraumatic bone marrow using navigator-corrected diffusion gradients. AJR 2000;174:731–4.[Abstract/Free Full Text]
15. Provenzale JM, Engelter ST, Petrella JR, Smith JS, MacFall JR. Use of MR exponential diffusion-weighted images to eradicate T₂ "shine-through" effect. AJR 1999;172:537–9.[Free Full Text]
16. Burdette JH, Elster AD, Ricci PE. Acute cerebral infarction: quantification of spin-density and T2 shine-through phenomena on diffusion-weighted MR images. Radiology 1999;212:333–9.[Abstract/Free Full Text]
17. Nusbaum AO, Lu D, Tang CY, Atlas SW. Quantitative diffusion measurements in focal multiple sclerosis lesions: correlations with appearance on T1-weighted MR images. AJR 2000;175:821–5.[Abstract/Free Full Text]
18. Moseley ME, Cohen Y, Kucharczyk J, Mintorovitch J, Asgaris HS, Wendland MF, et al. Diffusion-weighted MR imaging of anisotropic water diffusion in cat central nervous system. Radiology 1990;176:439–45.[Abstract/Free Full Text]
19. Mukherjee P, Bahn MM, McKinstry RC, Shimony JS, Cull TS, Akbudak E, et al. Differences between gray matter and white matter water diffusion in stroke: diffusion-tensor MR imaging in 12 patients. Radiology 2000;215:211–20.[Abstract/Free Full Text]

This article has been cited by other articles:

				Z. Bozgeyik, S. Ozgocmen, and E. Kocakoc Role of Diffusion-Weighted MRI in the Detection of Early Active Sacroiliitis Am. J. Roentgenol., October 1, 2008; 191(4): 980 - 986. [Abstract] [Full Text] [PDF]

				H.G. Hatipoglu, A. Selvi, D. Ciliz, and E. Yuksel Quantitative and Diffusion MR Imaging as a New Method To Assess Osteoporosis AJNR Am. J. Neuroradiol., November 1, 2007; 28(10): 1934 - 1937. [Abstract] [Full Text] [PDF]

				R. Dammers, H. W C Bijvoet, M. J Driesse, and C. C J Avezaat Occurrence of malignant vertebral fractures in an emergency room setting Emerg. Med. J., October 1, 2007; 24(10): 707 - 709. [Abstract] [Full Text] [PDF]

				A.J. Nemeth, J.W. Henson, M.E. Mullins, R.G. Gonzalez, and P.W. Schaefer Improved Detection of Skull Metastasis with Diffusion-Weighted MR Imaging AJNR Am. J. Neuroradiol., June 1, 2007; 28(6): 1088 - 1092. [Abstract] [Full Text] [PDF]

				D.-M. Koh and D. J. Collins Diffusion-Weighted MRI in the Body: Applications and Challenges in Oncology Am. J. Roentgenol., June 1, 2007; 188(6): 1622 - 1635. [Abstract] [Full Text] [PDF]

				A.Y. Oner, T. Tali, F. Celikyay, A. Celik, and P. Le Roux Diffusion-Weighted Imaging of the Spine with a Non-Carr-Purcell-Meiboom-Gill Single-Shot Fast Spin-Echo Sequence: Initial Experience AJNR Am. J. Neuroradiol., March 1, 2007; 28(3): 575 - 580. [Abstract] [Full Text] [PDF]

				D.M. Yousem, K. Gad, and R.P. Tufano Resectability Issues with Head and Neck Cancer AJNR Am. J. Neuroradiol., November 1, 2006; 27(10): 2024 - 2036. [Abstract] [Full Text] [PDF]

				A. P. White,, B. K. Kwon,, D. M. Lindskog,, G. E. Friedlaender,, and J. N. Grauer, Metastatic disease of the spine. J. Am. Acad. Ortho. Surg., October 1, 2006; 14(10): 587 - 598. [Abstract] [Full Text] [PDF]

				M. Maeda, H. Sakuma, S. E. Maier, and K. Takeda Quantitative Assessment of Diffusion Abnormalities in Benign and Malignant Vertebral Compression Fractures by Line Scan Diffusion-Weighted Imaging Am. J. Roentgenol., November 1, 2003; 181(5): 1203 - 1209. [Abstract] [Full Text] [PDF]

				M. Castillo Diffusion-Weighted Imaging of the Spine: Is It Reliable? AJNR Am. J. Neuroradiol., June 1, 2003; 24(6): 1251 - 1253. [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chan, J H M Articles by Wong, K P C Search for Related Content PubMed PubMed Citation Articles by Chan, J H M Articles by Wong, K P C