| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Musculoskeletal Imaging |
1 From the Departments of Radiology (J.L.M., H.K., A.R.), Hematology (M.D.), and Biostatistics (E.L.), Centre Hospitalo-Universitaire Henri Mondor, 51 avenue du Maréchal de Lattre de Tassigny, 94010 Créteil, France; and MR Division, Siemens France, Saint-Denis, France (S.B.). Supported by the Association pour la Recherche contre le Cancer. Received June 18, 2002; revision requested August 20; final revision received March 13, 2003; accepted May 5. Address correspondence to A.R. (e-mail: alain.rahmouni@hmn.ap-hop-paris.fr).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Dynamic contrast material–enhanced MR imaging of the thoracolumbar spine was performed in 71 patients. The maximum percentage of enhancement (Emax), enhancement slope, and enhancement washout were determined from bone marrow enhancement time curves (ETCs). The bone marrow signal intensity on T1-weighted spin-echo MR images was qualitatively classified into three grade categories. Quantitative ETC values were correlated with patient age and bone marrow fat content grade. Statistical analysis included mean t test comparison, analysis of variance, and regression analysis of the correlations between age and quantitative MR parameters.
RESULTS: Emax, slope, and washout varied widely among the patients. Emax values were obtained within 1 minute after contrast material injection and ranged from 0% to 430%. Emax values were significantly higher in patients younger than 40 years than in those aged 40 years or older (P < .001). These values decreased with increasing age in a logarithmic relationship (r = 0.71). Emax values decreased as fat content increased, but some overlap among the fat content grades was noted. Analysis of variance revealed that Emax was significantly related to age (younger than 40 years vs 40 years or older) (P < .001) and fat content grade (P < .001) but not significantly related to sex.
CONCLUSION: Dynamic contrast-enhanced MR imaging patterns of normal spinal bone marrow are dependent mainly on patient age and fat content.
© RSNA, 2003
Index terms: Aging • Bone marrow, 321.121411 • Bone marrow, MR, 321.121411, 321.121413, 321.121416, 321.12143 • Magnetic resonance (MR), contrast enhancement, 321.12143
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The use of gadolinium-enhanced MR imaging, particularly dynamic sequences, for the examination of patients with bone involvement in hematologic diseases has been increasing (7–10). Thus, the purpose of our study was to determine the patterns of dynamic enhancement of normal spinal bone marrow in adults and the changes that occur with aging.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The first group comprised 50 consecutive patients (27 female patients, 23 male patients; age range, 15–95 years; mean age, 51 years ± 20 [SD]) who were examined for benign diseases—specifically, back pain or sciatica recurring after disk surgery, or suspected spinal cord disease. Inclusion criteria included normal bone marrow, as determined on the basis of a normal blood cell count; the absence of diabetes and obesity; no prolonged immobilization; and no history of malignant, hematologic, or systemic diseases. Each patient completed a questionnaire designed to screen for external agents that are known to influence hematologic parameters or cause hematopoietic hyperplasia (11).
The second group comprised 21 patients (11 women, 10 men; age range, 26–76 years; mean age, 53 years ± 11). They were selected from among 163 consecutive patients with lymphoma (n = 24) or plasma cell disorders (n = 139) who had been referred to the MR imaging unit because of back pain before starting treatment for a newly diagnosed hematologic malignancy. All of these patients were recruited from the hematology department of our institution. Twenty-one patients from this group were included in this study because these were all of the patients, from the total of 163 patients, who had a normal blood cell count, normal bone marrow biopsy (iliac crest) results, and no focal bone marrow MR signal intensity abnormalities (ie, areas of high signal intensity on T2-weighted MR images corresponding to areas of low signal intensity on T1-weighted MR images).
Two patients had large cell non–Hodgkin lymphoma, and one had small cell non–Hodgkin lymphoma. These three patients had no extranodal involvement, and their malignancies were designated as stage I (one patient) or stage II (two patients) tumors according to the Ann Arbor classification system. The other 18 patients had the following plasma cell disorders: monoclonal gammopathy of unknown significance (n = 14) or solitary plasmocytoma (n = 4) in the sternum, the rib, the scapula, or a supraclavicular lymph node. All 18 patients had normal hemoglobin and serum calcium values.
MR Imaging
All 71 patients underwent MR imaging of the thoracolumbar spine with a 1.5-T unit (Magnetom SP 63; Siemens, Erlangen, Germany) and a quadrature spine surface coil. The MR imaging protocol was identical for all patients and consisted of T1-weighted (repetition time msec/echo time msec, 500/15) and T2-weighted (2,000/45, 90) spin-echo sequences performed with the following parameters: 256 x 512 matrix, 40-cm field of view, and sagittal, interleaved 4-mm-thick contiguous sections.
Dynamic examinations of gadolinium-induced enhancement were performed with a T1-weighted inversion-recovery, centric k-space reordered, turbo fast low-angle shot (FLASH) sequence and the following parameters: 7/3/300 (repetition time msec/echo time msec/inversion time msec), a 10° flip angle, a 128 x 128 matrix, a 1,200-msec acquisition time, a 40-cm field of view, and a midsagittal 10-mm section thickness.
A bolus of 0.2 mL of gadoterate meglumine (Dotarem; Laboratoire Guerbet, Aulnay-sous-Bois, France) per kilogram of body weight was injected manually through a catheter inserted into an antecubital vein at a rate of approximately 2 mL/sec and followed by a rapid 20-mL saline flush. The turbo FLASH sequences were started when half the contrast medium had been injected. The sequences were repeated 30 times in 3-second intervals and lasted 130 seconds. The postcontrast T1-weighted spin-echo MR sequence was started 4 minutes after the end of the gadoterate meglumine bolus injection.
Qualitative Analysis
Qualitative analysis of the conversion of bone marrow to fat tissue was based on the consensus reading between two radiologists (J.L.M., A.R.), who compared the signal intensity of bone marrow with the signal intensities of extraosseous fat and muscle on T1-weighted spin-echo MR images. The bone marrow signal intensity was classified into one of three grade categories according to the system of Ricci et al (5): Grade 1 indicated homogenous low signal intensity (ie, similar to the signal intensity of muscle) with or without isolated focal fat replacement; grade 2, heterogeneous signal intensity with mixed areas of low and high signal intensity or homogenous intermediate signal intensity lower than that of fat and higher than that of muscle; and grade 3, uniformly high signal intensity (ie, similar to the signal intensity of fat). We assessed bone marrow enhancement visually by comparing pre- and postcontrast T1-weighted spin-echo and turbo FLASH MR images.
Quantitative Analysis
One radiologist (H.K.) quantitatively assessed the bone marrow signal intensity by performing measurements in regions of interest. The regions of interest were placed centrally and anteriorly in the vertebral body to avoid the end plates and possible disk degeneration and to exclude the basivertebral vein plexus and cerebrospinal fluid. The regions of interest were 16 mm in diameter and were at the exact same site on all of the T1-weighted spin-echo MR images and the exact same site on all of the dynamic turbo FLASH MR images. The regions of interest were placed within a single vertebral body at the center of the coil, where the signal-to-noise ratio is highest. All measurements were made from the T8 through L3 vertebral body.
The percentage increase in signal intensity, or enhancement (E), on T1-weighted spin-echo MR images was calculated as follows: E = [(SIpost - SIpre) x 100]/SIpre, where SIpre is the signal intensity measurement in the region of inter-est before contrast medium injection and SIpost is the signal intensity measurement in the region of interest after contrast medium injection. In the dynamic turbo FLASH examinations, SIpre was the region-of-interest signal intensity measurement obtained before the onset of bone marrow enhancement. The percentage increase in signal intensity was graphically plotted against time. A percentage enhancement time curve (ETC) spanning 60 seconds after the onset of aortic enhancement was determined for each patient by directly plotting the E values. No automated curve fitting was applied.
Statistical Analysis
Differences in age distribution according to patient sex were assessed by using the 
2 test. The following values were calculated for each patient to characterize the ETC (7,12): maximum percentage of enhancement (Emax); time to Emax after the onset of aortic enhancement (Tmax); slope—that is, the rate of enhancement increase between 10% and 90% of Emax; and washout—that is, the value of enhancement at 60 seconds after the onset of aortic enhancement.
t Test comparisons of the quantitative covariates were performed to assess differences in mean Emax, Tmax, slope, and washout values between the two groups, between the younger and older patients, between the male and female patients, and between the three grades of bone marrow fat content. Univariate testing was applied by using analysis of variance, with sex and fat content grade as independent variables and age as a covariate. Regression analysis of the relationships between age and the quantitative MR imaging parameters was performed by comparing the logarithmic versus linear models. The nominal significance level for the end points was 5% (P < .05, two-sided analysis).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
|
|
|
Comparison of Groups 1 and 2
There was no significant difference in Emax, Tmax, slope, or washout values between group 1 (50 patients examined for benign diseases) and group 2 (21 patients with hematologic malignancies and normal bone marrow) (Table 2). We therefore considered both groups to have normal bone marrow and thus determined that they could be analyzed together.
|
40 years) (Fig 4). Analysis of variance also revealed that mean Emax values were related to age when the patients younger than 40 years were compared with those aged 40 years or older (P < .001). Emax values varied greatly among the patients younger than 40 years: from 15% in a 37-year-old man to 430% in a 15-year-old girl (Fig 1). In the patients aged 40 years or older, Emax varied between less than 5% in several patients older than 46 years and 137% in a 57-year-old man (Figs 2, 3). The correlation between slope and age was also significant (r = 0.53, P < .001). No correlation was observed between age and either Tmax or washout.
|
|
|
|
Compared with the results of a more recent study (11), in which subjects were divided into those younger than 50 years and those aged 50 years or older, our study results show that women younger than 50 years had higher Emax values than did men younger than 50 years (148% vs 99%), but the difference was not significant. Analysis of variance results showed that mean Emax values were not significantly related to sex. However, the women older than 50 years had significantly lower Emax values than did the men older than 50 years (36% vs 67%, P = .03).
Contrast Enhancement according to Conversion of Bone Marrow to Fat
Analysis of variance results show that mean Emax values were related to fat content grade (P < .001). Mean Emax values were significantly different between patients with grade 1 bone marrow conversion (mean, 172%; range, 40%–430%) and those with grade 2 bone marrow conversion (mean, 83%; range, 0%–320%) (P = .003). Mean Emax values were significantly higher in the patients with grade 2 bone marrow conversion than in those with grade 3 bone marrow conversion (mean, 41%; range, 1%–171%) (P = .005) (Fig 5).
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Turbo FLASH MR sequences have several advantages over T1-weighted spin-echo MR sequences in this setting (17): (a) The range of T1 factors is doubled, from -1 to +1 with inversion-recovery turbo FLASH MR imaging, as compared with 0 to +1 with spin-echo MR imaging. This wider range offers greater T1 contrast and better sensitivity in the depiction of gadolinium enhancement. (b) Turbo FLASH sequences are heavily T1 weighted owing to the short echo time (which can be made as short as possible), which is outside of the time range when fat and water protons are in the opposed phase. (c) Bone marrow enhancement is underestimated with T1-weighted spin-echo MR sequences: We found that the maximum enhancement occurred within the first minute after the bolus contrast medium injection and frequently decreased thereafter owing to washout.
The potential disadvantages of turbo FLASH sequences can be easily overcome: (a) Longitudinal proton relaxation must be allowed before the next radiofrequency is applied, with an interval of 3 seconds between two successive turbo FLASH sequences. (b) Local field inhomogeneities due to bone trabeculae, which prevent the use of gradient-echo sequences in bone marrow examinations, are minimized by using a very short echo time. However, turbo FLASH images are acceptable for quantitative analysis but not for anatomic analysis, because they lack spatial resolution.
Few authors have analyzed bone marrow enhancement with dynamic MR sequences (11,16). In our study, maximum enhancement always occurred within 1 minute (mean, 39 seconds) after contrast medium administration. Our study results confirm those of Baur et al (16), who observed a steep increase in bone marrow enhancement within the first 40–60 seconds after contrast medium administration with use of 20-second gradient-echo FLASH sequences, with an acquisition time of 20 seconds in 10 patients.
Emax values varied widely among our study patients: from less than 5% in the older patients to 430% in a 15-year-old girl. More recently, Chen et al (11), by using a similar MR imaging technique, found that the rate of bone marrow perfusion decreased significantly in patients older than 50 years. However, few young subjects were included in that study, and this possibly explains why the authors reported mean Emax values as low as 38% in men younger than 50 years and as low as 87% in women younger than 50 years. Corresponding values were 99% and 148%, respectively, in our study.
We observed a strong logarithmic relationship between age and Emax in our population of patients aged 15–95 years. Chen et al (11) observed Emax to be more closely associated with sex than with age, with values being significantly higher in the female patients younger than 50 years than in the male patients younger than 50 years. In our study, although Emax values tended to be higher in the female patients younger than 50 years than in the male patients younger than 50 years, the difference was not significant. Our study results confirm, as previously suggested by Baur et al (16), that Emax is significantly higher in younger than in older patients. The influence of aging on Emax values was more pronounced than the influence of sex in our study.
In patients younger than 40 years, the mean Emax value was 162.9%. To our knowledge, such a high value has not been reported previously, probably because less heavily T1-weighted sequences (such as spin echo) were used and image acquisition occurred after the enhancement peak. Contrary to reports that contrast enhancement is persistent, we observed a washout effect in 26 patients and a plateau in the remaining patients (16). Furthermore, the number of patients with washout was probably underestimated in our study because our analysis of washout was restricted to that during the first 60 seconds after the onset of marrow enhancement. Analysis of a longer period is possible with MR imaging systems that are newer than those used in this study.
As shown in Figure 3, washout predominated in the patients younger than 30 years, who had the highest mean Emax value. High Emax values may be explained by abundant hematopoietic marrow characterized by more numerous fenestrated vessels, large vascular pools and channels, and the presence of small, poorly vascularized fat (1,15,18,19).
By using T1-weighted spin-echo MR images, Amano et al (15) found that the reactive hematopoietic marrow in patients with aplastic anemia enhanced markedly after gadolinium-based contrast medium administration, whereas normoplastic marrow showed no enhancement. This washout may have been due to arteriovenous anastomoses, which are favored by the rich vasculature of hematopoietic marrow (15).
We found that Emax decreased as fat content increased, although some overlap among the three grades of bone marrow fat content was seen. Nineteen patients had Emax values that differed markedly from the mean Emax that corresponded to their fat content grade. Thus, the fat content grade of bone marrow was not predictive of the Emax value.
In conclusion, dynamic contrast enhancement of normal bone marrow is strongly influenced by age and fat content. Bone marrow enhancement decreases markedly with increasing age and conversion to fat, despite major intersubject variability. Bone marrow enhancement cannot be adequately assessed with standard T1-weighted spin-echo MR sequences, but it can be readily analyzed by using dynamic ultrafast MR sequences, which yield high temporal resolution and heavily T1-weighted images. The MR imaging criteria for normal bone marrow are based not only on T1-weighted signal intensity analysis data but also on ETC analysis data, both of which warrant further evaluation in patients with bone marrow diseases.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: Emax = maximum percentage of enhancement, ETC = enhancement time curve, FLASH = fast low-angle shot, Tmax = time to Emax after onset of aortic enhancement
Author contributions: Guarantors of integrity of entire study, J.L.M., A.R.; study concepts, J.L.M., A.R.; study design, S.B., J.L.M., M.D.; literature research, J.L.M.; clinical studies, J.L.M., A.R.; data acquisition, S.B., J.L.M., H.K.; data analysis/interpretation, J.L.M., A.R., E.L.; statistical analysis, E.L.; manuscript preparation and definition of intellectual content, A.R., J.L.M.; manuscript editing, A.R.; manuscript revision/review, J.L.M., A.R.; manuscript final version approval, A.R.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y.-J. Liu, G.-S. Huang, C.-J. Juan, M.-S. Yao, W.-P. Ho, and W. P. Chan Intervertebral Disk Degeneration Related to Reduced Vertebral Marrow Perfusion at Dynamic Contrast-Enhanced MRI Am. J. Roentgenol., April 1, 2009; 192(4): 974 - 979. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Lin, A. Luciani, K. Belhadj, P. Maison, A. Vignaud, J.-F. Deux, P. Zerbib, F. Pigneur, E. Itti, H. Kobeiter, et al. Patients with Plasma Cell Disorders Examined at Whole-Body Dynamic Contrast-enhanced MR Imaging: Initial Experience Radiology, March 1, 2009; 250(3): 905 - 915. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Matuszewski, T. Persigehl, A. Wall, N. Meier, R. Bieker, H. Kooijman, B. Tombach, R. Mesters, W. E. Berdel, W. Heindel, et al. Assessment of Bone Marrow Angiogenesis in Patients with Acute Myeloid Leukemia by Using Contrast-enhanced MR Imaging with Clinically Approved Iron Oxides: Initial Experience Radiology, December 1, 2006; 242(1): 217 - 224. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Griffith, D. K. W. Yeung, G. E. Antonio, F. K. H. Lee, A. W. L. Hong, S. Y. S. Wong, E. M. C. Lau, and P. C. Leung Vertebral Bone Mineral Density, Marrow Perfusion, and Fat Content in Healthy Men and Men with Osteoporosis: Dynamic Contrast-enhanced MR Imaging and MR Spectroscopy Radiology, September 1, 2005; 236(3): 945 - 951. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
