| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Mie University School of Medicine, 2-174 Edobashi,
Tsu, Mie 514-8507, Japan.
2 Department of Radiology, Brigham and Women's Hospital, Boston, MA 02115.
Received January 17, 2003;
accepted after revision May 14, 2003.
Address correspondence to M. Maeda
(mmaeda{at}clin.medic.mie-u.ac.jp).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Line scan diffusion-weighted imaging was prospectively performed in 17 patients with 20 acute vertebral compression fractures caused by osteoporosis or trauma, in 12 patients with 16 vertebral compression fractures caused by malignant tumors, and in 35 patients with 47 metastatic vertebrae without collapse. Images were obtained at b values of 5 and 1,000 sec/mm2. The apparent diffusion coefficient (ADC) was measured in vertebral compression fractures and metastatic vertebrae without collapse.
RESULTS. The ADC (mean ± SD) was 1.21 ± 0.17 x 10–3 mm2/sec in benign compression fractures, 0.92 ± 0.20 x 10–3 mm2/sec in malignant compression fractures, and 0.83 ± 0.17 x 10–3 mm2/sec in metastatic vertebral lesions without collapse. The ADC was significantly higher in benign compression fractures than in malignant compression fractures (p < 0.01), although the two types showed considerable overlap.
CONCLUSION. Although the quantitative assessment of vertebral diffusion provides additional information concerning compressed vertebrae, the benign and malignant compression fracture ADC values overlap considerably. Therefore, even a quantitative vertebral diffusion assessment may not always permit a clear distinction between benign and malignant compression fractures.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Subsequently, a small number of investigators have quantified the diffusion in abnormal vertebrae in terms of the apparent diffusion coefficient (ADC) value and have concluded that quantitative assessment is more useful than qualitative assessment in differentiating benign vertebral fractures from malignant lesions [5, 6]. Although their data suggested that the quantification of diffusion was more precise in the characterization of collapsed vertebrae and was therefore more useful in differentiating benign and malignant compression fractures, their methods had limitations that may have affected the validity of their findings. The first is related to the low b value (250 sec/mm2) used in diffusion-weighted imaging [5], and the second is caused by the use of the echoplanar sequence [6], which is easily affected by susceptibility artifacts.
Line scan diffusion-weighted imaging is an advantageous method for spine imaging because it is inherently insensitive to motion artifacts and susceptibility artifacts [7]. We performed line scan diffusion-weighted imaging with a higher b value of 1,000 sec/mm2 to permit the quantification of diffusion as well as to obtain high-quality diffusion-weighted images. The objective of this study was to evaluate the usefulness of the ADC value of the bone marrow in differentiating benign and malignant vertebral compression fractures.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The diagnoses were essentially confirmed in 16 of 17 patients with benign compression fracture by clinical follow-up after more than 6 months. In these patients, back pain resolved or decreased gradually 2–7 months after the initial MRI. In one patient, the clinical history ruled out malignancy as the cause of the fracture, which occurred during a motor crash. Surgery was performed in this patient because the vertebral fracture resulted in compression of the spinal canal. In nine patients, follow-up MRIs were performed 2–14 months after the initial MRI to rule out neoplasms. Four patients received steroid treatment at the time of onset for hypereosinophilic syndrome, retroperitoneal fibrosis, Churg-Strauss syndrome, and pemphigoid. Six patients had a clinical history of malignant tumors, including breast cancer, rectal cancer, cholangiocarcinoma, hepatocellular carcinoma, transitional cell carcinoma, and malignant lymphoma.
The second group comprised 12 patients (mean age, 61.9 years) with 16 vertebral compression fractures caused by malignant tumors. None of these patients had undergone chemotherapy or radiation therapy for malignant bone lesions before the MRI study. The malignant vertebral tumors included one primary malignant tumor (sarcoma not otherwise specified) and 15 metastatic malignant tumors from breast cancer (n = 3), lung cancer (n = 3), prostate cancer (n = 3), multiple myeloma (n = 3), renal cell carcinoma (n = 1), stomach cancer (n = 1), and malignant melanoma (n = 1). In nine of the 12 patients, the diagnosis of malignant fracture was established on the basis of the deteriorated clinical course and the findings of follow-up MRI and CT. A malignant nature of involvement of posterior elements and paravertebral soft-tissue masses was rated as an unequivocal imaging finding of malignancy [1–3]. In one patient (with sarcoma not otherwise specified), the primary tumor was not detected, and open biopsy was subsequently performed. Autopsy was performed in one patient with lung cancer. In another patient with lung cancer, a pathologic diagnosis was obtained when the patient underwent emergency surgery because the vertebral lesion had progressed, resulting in severe compression of the spinal canal.
The third group comprised 35 patients (mean age, 62.2 years) with 47 metastatic vertebrae without compression fractures. The diagnoses were confirmed at pathologic examination of specimens obtained by CT-guided needle biopsy in eight patients, at surgery in five patients, and at autopsy in three patients. In the remaining 19 patients, the clinical diagnosis was established by unequivocal MRI and CT findings of metastatic disease of the spine with primary neoplasms, as described in the second group. The primary cancers included lung cancer (n = 18), breast cancer (n = 10), prostate cancer (n = 6), stomach cancer (n = 4), esophageal cancer (n = 2), multiple myeloma (n = 2), renal cell carcinoma (n = 2), Paget's cancer (n = 1), malignant melanoma (n = 1), and thyroid cancer (n = 1). No patients with metastatic vertebral lesions had received radiation therapy or chemotherapy for the bone lesions before the MRI study.
MRI Sequence
The line scan diffusion-weighted imaging method described previously
[7] was performed using a
spinal array surface coil on a 1.5-T MRI system (Signa, General Electric
Medical Systems, Milwaukee, WI). Neither cardiac gating nor respiratory
triggering was used. Images were acquired with a rectangular field of view (32
x 16 cm or 32 x 24 cm), a matrix size of 128 x 128 columns,
and a bandwidth of 7.81–10.4 kHz. The effective section thickness was
set at 5 mm, with an intersection gap of 1 mm. Diffusion was assessed in three
orthogonal directions, in each of which two images were obtained, one with a
low diffusion weighting (b factor) of 5 sec/mm2 and the other with
a high b factor of 1,000 sec/mm2. Other parameters were as follows:
TR of 2,093–3,498 msec, TE of 70.5 msec, and 1 excitation. The scan time
per slice was 26–37 seconds, and totals of three or five slices were
obtained in the sagittal plane. Other MRI sequences included sagittal
spin-echo T1-weighted imaging (TR/TE, 400/8) and sagittal fast spin-echo
T2-weighted imaging with or without fat suppression (3,000/105). Sagittal
contrast-enhanced fat suppression T1-weighted imaging (550/8) was performed in
four patients with benign compression fractures and in 20 patients with
malignant vertebral tumors with or without collapse.
Data Analysis
Isotropic diffusion images with a high b factor were generated from the
three diffusion directions assessed. Trace ADC maps were generated using the
equation described by Stejskal and Tanner
[8],
 |
Qualitative analysis was performed for benign and malignant compression fractures. Two experienced radiologists evaluated the line scan diffusion-weighted images (b = 1,000 sec/mm2). Image review was performed with special attention paid to the appearance of bone marrow changes in the compression fractures. On the line scan diffusion-weighted images, the signal intensities in the fractured vertebral bodies were classified as hypointense, isointense, hyperintense, or of mixed signal intensity relative to adjacent normal bone marrow in the same patient. Statistical evaluation for the qualitative comparison between benign and malignant compression fractures was performed using the Mann-Whitney U test.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Qualitative Analysis
Clinical cases showing the signal intensities of benign and malignant
compression fractures relative to the adjacent normal vertebrae on the line
scan diffusion-weighted images (b = 1,000 sec/mm2) are shown in
Figs. 2A,
2B,
2C,
2D,
3A,
3B,
3C,
3D,
4A,
4B,
4C,
4D,
4E. The findings in each group
are summarized in Table 1. With
regard to benign compression fractures, most (75%) showed hypointensity
relative to normal vertebrae, followed by mixed (iso- and hypo-) intensity and
isointensity. Hyperintensity was not seen in benign compression fractures. On
the other hand, the malignant compression fractures showed relatively variable
signal intensities. Hyperintensity was seen in one case of malignant
compression fracture. Qualitative analysis showed no statistically significant
difference between benign and malignant compression fractures. The ADC values
(mean ± SD) in each group are also summarized in
Table 1. The mean ADC in each
group was lower in malignant compression fractures than in benign compression
fractures.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our preliminary results indicate that line scan diffusion-weighted imaging of the spine is feasible and provides excellent images without specialized hardware, cardiac gating, or respiratory compensation. Line scan diffusion-weighted imaging uses multiple diffusion-weighted spin-echo column excitations to form a 2D image [18]. This method is insensitive to motion artifacts because the images are constructed column by column and the acquisition time for an individual column is approximately equal to the TE. Patient motion or physiologic pulsatile motion may result in loss of signal from the column of data, although the remainder of the image is unaffected. Most important, in our series, line scan diffusion-weighted images of the thoracic spine were not significantly degraded by cardiac pulsation or respiratory motion. Vertebral diffusion-weighted imaging, unlike single-shot echoplanar imaging, minimizes susceptibility-related image distortions and signal loss, which is an advantage. In addition, this method is ideally suited to scanning of the spine because the rectangular field of view of a sagittally oriented scan matches the aspect ratio of the spinal column, and adjacent structures of little interest do not need to be scanned.
We selected a higher b value (1,000 sec/mm2) than that used in previous reports [1–5]. In the central nervous system, values on the order of 1,000 sec/mm2 have been suggested for cerebral diffusion-weighted imaging. Although the optimal b value for vertebral diffusion-weighted imaging has yet to be determined, maximum b values lower than 500 sec/mm2 may be too low to provide information concerning diffusion. Such low b values may lead to ADC data obtained in normal or abnormal vertebral tissues different than those obtained using higher b values. Zhou et al. [5] reported that benign compression fractures and metastatic lesions had lower ADC values than normal vertebral bodies when a b value of 250 sec/mm2 was used. However, this is unlikely to be true in diffusion studies using higher b values. We found that normal bone marrow showed minimal diffusion at a high b value of 1,000 sec/mm2, and the ADC was much lower than that in abnormal bone marrow. Our results agree with those of previous reports in which higher b values were used [6, 19, 20].
In our study, the ADC in benign compression fractures was statistically higher than that in malignant compression fractures. However, the ADC values overlapped considerably between the two groups (Fig. 1). Thus, our quantitative results did not provide the clear distinction between benign and malignant compression fractures that was found in previous reports. Chan et al. [6] reported no ADC overlap between benign and malignant compression fractures. Zhou et al. [5] also reported that the two distinct disease groups were well separated on the basis of ADC values. In our study, several factors such as the diffusion sequence, the b value, and the time elapsed since occurrence of the fracture differed from those in previous studies [5, 6]. These differences might have been responsible for the difference in results between our study and other studies.
However, it is not clear whether the age of the fracture (< 1 month) has a significant influence on the results. In the inclusion criteria we used, the age of the fracture was less than 1 month, which was longer than those in previous studies (< 14 days) [1, 3, 6]. In benign compression fractures, it might be hypothesized that the ADC would decrease over time because compression fractures can recover normal fatty bone marrow. However, all of our patients with benign compression fractures showed low signal on T1-weighted images at the time of the MRI study, indicating the absence of fatty marrow recovery. In general, the recovery of fatty bone marrow requires several months in patients with benign compression fractures. Yamato et al. [21] reported that during a 3-month period after an injury, the extent of the low-signal areas of benign collapsed vertebrae on T1-weighted images increased in 40% of cases. Complete recovery of fatty marrow was observed at 65–708 days (mean, 290 days). Further investigations are necessary to clarify the influence of the age of the fracture on the ADC.
We found relatively high ADC values (> 1.0 x 10–3 mm2/sec) in four malignant compression fractures, and the ADC values in these cases overlapped with those in cases with benign compression fractures. We see three possible reasons for these results. First, necrotic tissues in malignant bone tumors can increase the ADC. Necrosis in bone tumors causes an increase in the extracellular volume fraction, resulting in a high ADC, as documented in a previous study [22]. In such cases, necrotic portions of malignant vertebrae can be suggested on T2-weighted images or contrast-enhanced images, as seen in Figure 4A, 4B, 4C, 4D, 4E.
Second, a larger fraction of associated bone marrow edema may explain the high ADC in malignant compression fractures. Le Bihan [23] reports that in malignant compression fractures, a mixture of interstitial edema and malignant tumor cellularity occurs and may reduce the specificity of the diffusion analysis. Finally, the hypervascular portion of malignant tumors that increases the proportion of the perfusion effect may be responsible for the high ADC measurement results [24]. The use of perfusion-corrected diffusion-weighted imaging may overcome this problem.
In our qualitative results, most (75%) benign compression fractures showed hypointensity relative to adjacent normal vertebrae on line scan diffusion-weighted images, but only 19% of malignant compression fractures showed hypointensity, whereas some showed variable signal intensities including hypo-, iso-, hyper-, and mixed intensity. However, qualitative analysis showed no significant difference between benign and malignant compression fractures. We found only one case of malignant compression fracture that was hyperintense relative to normal bone marrow. This is not in agreement with the results of Baur et al. [1, 2] or Spuentrup et al. [3], who found that all but a few cases of metastatic fracture showed hyperintensity relative to normal bone marrow. We consider that this difference is probably caused by the higher b value used in our study (b = 1,000 sec/mm2) compared with that used in previous studies (b values ranging from 165 to 598 sec/mm2). Malignant vertebral lesions usually show hyperintensity on T2-weighted images. At low b values, T2 effects particularly modify the signal intensities obtained in diffusion-weighted images. If malignant vertebrae have relatively slow diffusion and relatively high T2 values, diffusion-weighted images at low b values would show hyperintensity of malignant vertebrae relative to normal bone marrow. However, T2 effects are less prominent and diffusion has a greater influence at higher b values such as 1,000 sec/mm2. The ADC of malignant compression fractures is lower than that of benign compression fractures but higher than that of normal vertebrae. Therefore, diffusion-weighted images acquired at b equals 1,000 sec/mm2 can present variable signal intensities in malignant compression fractures depending on the ADC and T2 values of the lesions, as seen in our results.
Qualitative analysis has a number of theoretic disadvantages. One problem is that qualitative analysis cannot completely eliminate the T2 shine-through effect [4–6], as previously mentioned. Another problem is that the fraction of fatty marrow in normal vertebrae can vary from patient to patient. Vertebrae with a larger fraction of fatty marrow would show a lower ADC because the ADC of fat is low [24]. The fraction of fatty marrow is influenced by factors such as the patient's age and hematopoietic status. For example, normal hypercellular marrow in children shows a higher ADC than normal normocellular marrow in adults, but the latter shows a higher ADC than normal hypocellular marrow in adults [25]. Therefore, normal vertebrae can show variable signal intensities on diffusion-weighted images with higher b values and can exhibit variable ADC values according to the degree of fatty marrow in the particular patient. Thus, it is not valid to standardize the signal intensities of fractured vertebrae in comparison with those of adjacent normal vertebrae.
Unlike the promising results obtained in previous studies [1–3, 5, 6], our results showed considerable overlap between benign and malignant compression fractures; therefore, even quantitative assessment could not provide a clear distinction between the two types, although further investigations involving larger numbers of patients are necessary to clarify this point. It has been reported that conventional MRI also has several features that are useful for differentiation, including the marrow space signal intensity with various sequences, the location and extent of the signal abnormalities in the compressed vertebral body, involvement of the pedicles and neural arch structures, the presence of epidural or paravertebral soft-tissue masses, the contour of the dorsal cortex of the involved vertebral body, the presence of other spinal metastases, and the pattern of contrast enhancement [16, 17, 26, 27]. When these criteria are applied, MRI studies have shown accuracy rates in the range of 79–94%, depending on the methods used and the patients selected [16, 17, 26]. However, some of these MRI features are not observed in every case. For example, in patients without a history of cancer but with equivocal clinical or conventional MRI findings, the quantitative assessment of diffusion may provide additional information and a clue to differentiation and may thus prove helpful in reaching a decision as to whether to perform bone biopsy, as we saw in one of our patients (Fig. 3A, 3B, 3C, 3D). Although quantitative assessment alone may have only limited value in terms of the differential diagnosis, we feel that it provides additional useful information as an adjunct to conventional MRI findings.
In conclusion, our preliminary study has shown that ADC values are significantly higher in benign compression fractures than in malignant compression fractures. Such quantitative assessment provides additional information concerning compressed vertebrae and may be a useful adjunct to conventional MRI findings in the differential diagnosis. However, considerable overlap in the ADC values was observed between benign and malignant compression fractures in this study. Although it is necessary to study a larger number of patients before reaching a conclusion, it appears that even quantitative assessment may not provide clear separation between benign and malignant compression fractures.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Satoh, Y. Kitazume, S. Ohdama, Y. Kimula, S. Taura, and Y. Endo Can Malignant and Benign Pulmonary Nodules Be Differentiated with Diffusion-Weighted MRI? Am. J. Roentgenol., August 1, 2008; 191(2): 464 - 470. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Y. Oner, L. Aggunlu, S. Akpek, T. Tali, and A. Celik Diffusion-Weighted Imaging of the Appendicular Skeleton with a Non Carr-Purcell-Meiboom-Gill Single-Shot Fast Spin-Echo Sequence Am. J. Roentgenol., December 1, 2007; 189(6): 1494 - 1501. [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]
|
|
|
|
|
|
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]
|
|
|
|
|
|
A.A.K. A. Razek, A.Y. Kandeel, N. Soliman, H.M. El-shenshawy, Y. Kamel, N. Nada, and A. Denewar Role of Diffusion-Weighted Echo-Planar MR Imaging in Differentiation of Residual or Recurrent Head and Neck Tumors and Posttreatment Changes AJNR Am. J. Neuroradiol., June 1, 2007; 28(6): 1146 - 1152. [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. C. Zajick Jr, W. B. Morrison, M. E. Schweitzer, J. A. Parellada, and J. A. Carrino Benign and Malignant Processes: Normal Values and Differentiation with Chemical Shift MR Imaging in Vertebral Marrow Radiology, November 1, 2005; 237(2): 590 - 596. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Maeda, H. Kato, H. Sakuma, S. E. Maier, and K. Takeda Usefulness of the Apparent Diffusion Coefficient in Line Scan Diffusion-Weighted Imaging for Distinguishing between Squamous Cell Carcinomas and Malignant Lymphomas of the Head and Neck AJNR Am. J. Neuroradiol., May 1, 2005; 26(5): 1186 - 1192. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Nagata, M. Maeda, H. Tsukahara, S. E. Maier, and K. Takeda Brain Stem Hypertensive Encephalopathy Evaluated by Line Scan Diffusion-Weighted Imaging AJNR Am. J. Neuroradiol., May 1, 2004; 25(5): 803 - 806. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
