HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2352040437v1 235/2/569    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend 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 Kealey, S. M. Articles by Eastwood, J. D. Search for Related Content PubMed PubMed Citation Articles by Kealey, S. M. Articles by Eastwood, J. D.

Assessment of Apparent Diffusion Coefficient in Normal and Degenerated Intervertebral Lumbar Disks: Initial Experience¹

¹ From the Department of Radiology (S.M.K., D.P.B., J.M.P., J.D.E.) and Department of Biostatistics and Bioinformatics (D.D.), Duke University Medical Center, Box 3808, Durham, NC 27710; Department of Veterans Affairs, Durham Veterans Affairs Medical Center, Durham, NC (S.M.K., D.P.B., J.M.P., J.D.E.); and Department of Radiology, Henry Ford Hospital, Detroit, Mich (T.A.). Received March 5, 2004; revision requested May 11; revision received June 8; accepted July 20. Address correspondence to S.M.K. (e-mail keale001@mc.duke.edu).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: To determine prospectively the diffusibility of water in normal lumbar disks in adults by using the mean apparent diffusion coefficient (ADC) and to determine if a relationship exists between disk ADC and magnetic resonance (MR) findings of disk degeneration.

MATERIALS AND METHODS: The study was approved by the Institutional Review Board, and all participants gave written informed consent prior to enrollment. Diffusion-weighted MR imaging of the lumbar spine was performed in 39 patients (all men; mean age, 53 years) and five volunteers (all men; mean age, 31 years). ADC values were recorded for each disk. All disks were visually graded by two independent observers as being normal or as showing at least one of three MR findings of degeneration on sagittal T2-weighted images. Mean ADC values of normal disks were compared with those of degenerated disks and were correlated with age and anatomic location. Data were analyzed by using Kendall correlation statistics, Mantel-Haenszel statistics, and a paired two-tailed Student t test.

RESULTS: The mean ADC value was 2.27 x 10^–3 mm²/sec ± 0.58 (± standard deviation) in normal disks and 2.06 x 10^–3 mm²/sec ± 0.47 in abnormal disks (9% reduction, P = .006). A statistically significant dependence of lumbar disk ADC on anatomic location was reported (analysis of variance, P < .001), with lower ADC values seen in more caudal disks. There was no association between age and mean disk ADC.

CONCLUSION: A statistically significant decrease was seen in the ADC values of degenerated lumbar disks when compared with ADC values of normal disks. More caudal disks, even when normal, showed lower ADC values than more cephalic disks.

© RSNA, 2005

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Diffusion-weighted magnetic resonance (MR) imaging is a well-established modality for the study of cerebral disorders, most notably stroke (1,2). Until recently, however, clinical use of diffusion-weighted imaging as a tool for studying spinal disease processes has been considerably more limited in scope because of the technical challenges of imaging smaller structures, such as the spinal cord, and, likely, because of substantial susceptibility artifacts from bone (3,4). Recent advances in MR imaging technique have helped to make diffusion-weighted imaging of the spine more practical for use in the clinical environment (4,5).

The spinal disk is an avascular structure and, thus, is dependent on diffusion for nutrition. Evidence from prior studies in animals and humans has suggested that degenerative disk disease may be associated with decreased molecular diffusibility (6,7). Diffusion-weighted MR imaging is a noninvasive method of measuring the diffusion of water within tissue in vivo and, as such, is ideally suited for clinical studies designed to determine if pathologic alterations in water diffusion are related to disk disease. Thus, the purpose of our study was to determine prospectively the diffusibility of water in normal lumbar disks in adults by using the mean apparent diffusion coefficient (ADC) and to determine if a relationship exists between disk ADC and MR findings of disk degeneration.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Study Population
The study population consisted of 39 consecutive patients (all men; mean age, 53 years; age range, 29–86 years) and five volunteers (all men; mean age, 31 years; age range, 29–35 years). All 39 patients underwent clinical MR imaging of the lumbar spine at a Veterans’ Administration hospital between January and November 2002. Volunteers underwent MR imaging during the same time frame and had no history of chronic back pain or back surgery. The study was approved by the Institutional Review Board at the Veterans Affairs Medical Center, and all participants gave written informed consent prior to enrollment.

MR Imaging
Each participant underwent MR imaging with a commercial 1.5-T imager (Intera; Philips, Eindhoven, the Netherlands). Imaging included sagittal T1-weighted, T2-weighted, and intermediate-weighted (proton density) imaging and transverse T1-weighted and T2-weighted imaging parallel to the disk spaces. Diffusion-weighted images were obtained in the sagittal plane by using isotropic multishot spin-echo echo-planar MR imaging that provided equal diffusion weighting in three mutually orthogonal directions (right to left, superior to inferior, and anterior to posterior) simultaneously. Diffusion-weighted imaging parameters were as follows: repetition time msec/echo time msec, 1215/80; section thickness, 6 mm; section gap, 1 mm; matrix, 128 x 128; and six signals acquired. Imaging time was 6 minutes 30 seconds. These parameters were chosen to achieve a balance between signal-to-noise ratio and image acquisition time (8). Two acquisitions were performed in each participant—one acquisition without diffusion weighting (b = 0 sec/mm²) and one acquisition with diffusion weighting (b = 400 sec/mm²). A b value of 400 sec/mm² was chosen to minimize diffusion-related loss of signal intensity in disk tissue, which could lead to subsequent inaccuracy in ADC measurement; results from previous studies (9,10) have demonstrated disk ADC values on the order of 1.50 x 10^–3 mm²/sec, a factor that would make the use of higher b values (such as the 1000 sec/mm² commonly used in brain studies) less optimal.

A control standard for disk ADC values was obtained by using a small cylindric water tube that was taped to each participant’s back and included in the field of view for later analysis. By using software that was available on the MR imager, sagittal ADC maps were created and transferred to an imaging workstation (Easy Vision; Philips).

Assessment of degenerative disk disease.—The presence of findings of degenerative disk disease was determined by two neuroradiologists (D.P.B. and J.D.E., with 17 and 11 years experience in MR imaging of the spine, respectively) who worked independently and without knowledge of the findings on diffusion-weighted MR images. Each reader reviewed the intermediate-weighted and T2-weighted sagittal MR images without clinical information or knowledge of the findings on diffusion-weighted images. For every disk, the presence or absence of three findings was noted: (a) loss of disk height, (b) reduction in signal intensity on T2-weighted images compared with that of normal intervertebral disks in the same individual, and (c) loss of distinctness of the intranuclear cleft compared with that of normal disks in the same individual. Normally, disk height increases as one progresses caudally, reaching a maximum height at the L4–5 disk space and then decreasing again at the L5-S1 disk space; this normal pattern was used to determine the presence of a loss of height in any individual disk. Disks with only one of these findings were defined as mildly degenerated, while those with at least two findings were defined as severely degenerated. In cases of disagreement between the two primary readers, a third neuroradiologist (J.M.P. or S.M.K., with 17 and 5 years experience in MR imaging of the spine, respectively) who was blinded to signs of disk degeneration independently examined the image to resolve the disagreement.

Measurement of disk ADC and water tube ADC.—A single operator (T.A., 4 years experience in MR imaging of the spine) placed small, round regions of interest (ROIs) that measured 40–50 mm² within the central portion of each of the five lumbar disks (L1–2 through L5-S1) and within the central portion of the water tube on the ADC map. The ADC values within these ROIs were then recorded. The ratio (ADC_RATIO) of disk ADC to water tube ADC was calculated for each disk.

Comparison of disk ADC values at different anatomic levels.—The mean ADC values at each disk level were analyzed to determine the importance of anatomic location. Mean ADC values at individual disk levels were also compared with each other.

Comparison between ADC values in normal disks and those in abnormal disks.—The mean ADC values of all disks with one or more findings of degeneration were compared with the mean ADC values of all normal-appearing disks. The effect of severity of disk disease on ADC value was estimated by separately comparing the mean ADC values of all normal disks with those of disks with one, two, or all three findings of degeneration described earlier.

Statistical Methods
A mixed linear model capable of determining the significance of anatomic location (ie, disk level) was used to analyze disk ADC values. A paired two-tailed Student t test was used to compare the ADC values obtained at individual disk levels with one another.

Kendall correlation statistics were used to study all disks (normal disks and abnormal disks together) and to compare the mean ADC value and ADC_RATIO of all five disks averaged together with age. Age was also considered as a potential factor in the mixed linear model. Additionally, normal disks were also analyzed alone to control for the effect of degenerative disease on this analysis. Kendall correlation statistics, appropriate to both discrete and continuous data, were chosen by our statistician as being an appropriate method of assessing our data for the presence and degree of correlation. Mantel-Haenszel statistics were used to assess relationships between the discrete findings while adjusting for potential within-participant data correlation. Statistical analysis was performed by using a statistical software program (SAS, version 8.2; SAS Institute, Cary, NC).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Prevalence of Disk Degeneration
A total of 210 intervertebral disks were studied in the 44 participants (39 patients and five volunteers). Ten disks were excluded from assessment. In two instances, the disk was obscured as a result of susceptibility artifacts, and, in eight instances, the disk space was obliterated because of severe narrowing that precluded the placement of ROIs. Of the 210 disks, 111 were normal (ie, showed no findings of degenerative disease on T2-weighted images), and 99 showed one or more of the three findings of degeneration described earlier. The prevalence of degenerative disk disease is listed according to the anatomic level of each disk in Table 1. Figure 1 shows the placement of ROIs on intervertebral disks, with corresponding diffusion-weighted and ADC map images.

View larger version (111K): [in this window] [in a new window] [Download PPT slide]   Figure 1a. MR images of lumbar spine in 57-year-old man. (a) Sagittal T2-weighted MR image with ROI placed over water tube on participant’s back. Intervertebral disk at L4-5 was designated as showing reduction in T2 signal intensity and loss of intervertebral cleft with preservation of disk height. ROI has been placed centrally on intervertebral disk at L4-5. Sagittal plane of acquisition permits optimum placement of ROI, avoiding inclusion of portions of anulus fibrosis and vertebral end plates. (b) Corresponding diffusion-weighted image (b value, 400 sec/mm2; 1215/80; section thickness, 6 mm; section gap, 1 mm; and six signals acquired) and (c) ADC map derived from b, with ROI placed over L4-5 intervertebral disk.

View larger version (74K): [in this window] [in a new window] [Download PPT slide]   Figure 1b. MR images of lumbar spine in 57-year-old man. (a) Sagittal T2-weighted MR image with ROI placed over water tube on participant’s back. Intervertebral disk at L4-5 was designated as showing reduction in T2 signal intensity and loss of intervertebral cleft with preservation of disk height. ROI has been placed centrally on intervertebral disk at L4-5. Sagittal plane of acquisition permits optimum placement of ROI, avoiding inclusion of portions of anulus fibrosis and vertebral end plates. (b) Corresponding diffusion-weighted image (b value, 400 sec/mm2; 1215/80; section thickness, 6 mm; section gap, 1 mm; and six signals acquired) and (c) ADC map derived from b, with ROI placed over L4-5 intervertebral disk.

View larger version (92K): [in this window] [in a new window] [Download PPT slide]   Figure 1c. MR images of lumbar spine in 57-year-old man. (a) Sagittal T2-weighted MR image with ROI placed over water tube on participant’s back. Intervertebral disk at L4-5 was designated as showing reduction in T2 signal intensity and loss of intervertebral cleft with preservation of disk height. ROI has been placed centrally on intervertebral disk at L4-5. Sagittal plane of acquisition permits optimum placement of ROI, avoiding inclusion of portions of anulus fibrosis and vertebral end plates. (b) Corresponding diffusion-weighted image (b value, 400 sec/mm2; 1215/80; section thickness, 6 mm; section gap, 1 mm; and six signals acquired) and (c) ADC map derived from b, with ROI placed over L4-5 intervertebral disk.

No statistically significant relationship was found between age and mean disk ADC value by using either an unadjusted analysis or a mixed model analysis.

Comparison of Disk ADC Values at Different Anatomic Levels
The mean ADC value and ADC_RATIO for each of the five intervertebral disk levels are shown in Tables 2 and 3, respectively. There was a significant relationship between intervertebral disk level and disk ADC value (analysis of variance [ANOVA], P < .001), with lower ADC values seen in more caudal lumbar disks as opposed to more cephalad lumbar disks. ANOVA results from only the 111 normal-appearing disks confirmed a statistically significant reduction in mean ADC value with increasingly caudal disk position (P = .001). Additionally, by using a t test to compare individual disk levels, a statistically significant difference was found between the mean ADC value of normal L1–2 disks and that of normal L5-S1 disks (2.35 x 10^–3 mm²/sec ± 0.59 vs 1.97 x 10^–3 mm²/sec ± 0.51, respectively; 16% reduction; P = .02). A similar relationship was seen when comparing the mean ADC value of normal L2–3 disks with that of normal L5-S1 disks (2.46 x 10^–3 mm²/sec ± 0.55 vs 1.97 x 10^–3 mm²/sec ± 0.51, respectively; 20% reduction; P = .003). This comparison is depicted graphically in Figure 2.

View larger version (16K): [in this window] [in a new window] [Download PPT slide]   Figure 2. Graph of mean ADC values in normal-appearing disks. ADC values are indicated in units of x10–6 mm2/sec. Disk levels 1-5 refer to L1-2 through L5-S1. ADC values of normal disks decrease as one proceeds in caudal direction, a pattern that parallels prevalence of degenerative disk disease in our study population. There was a statistically significant difference in ADC values between L1-2 and L5-S1 disks and between L2-3 and L5-S1 disks, with a trend toward significance between L1-2 and L4-5 disks.

Mantel-Haenszel analysis showed a strong association between anatomic level and the presence of any single disk abnormality (P < .001), with degenerative changes more frequently affecting lower disks than upper disks. A significant relationship was demonstrated between loss of the intranuclear cleft, ADC value (ANOVA, P = .01), and ADC_RATIO (ANOVA, P = .03). There was also a significant relationship between loss of the intranuclear cleft and the anatomic position of the disk (as shown in Table 4) (ANOVA, P = .01). For abnormal disks, ANOVA results demonstrated no significant relationship between mean disk ADC or ADC_RATIO and either (a) loss of disk height alone or (b) reduction in T2 signal intensity alone.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Our finding of a statistically significant decrease in the ADC of disks with one or more findings of degeneration was notable. ADC values in mildly degenerated disks, however, did not differ significantly from those in severely degenerated disks. Early disk degeneration is characterized by a loss of extracellular water and by alterations in the relative proteoglycan content of intervertebral disks (13). Loss of extracellular water in the nucleus pulposus might be expected to result in a reduction in water diffusion and disk ADC values. Further study (eg, animal model or human longitudinal study) could potentially help to determine whether a decreased ADC precedes other MR findings of degeneration and neurologic signs and symptoms.

Although we found that mean ADC values in severely degenerated disks were lower than those obtained in normal disks, other researchers have previously reported normal or elevated ADC values in severely degenerated disks (9,14). In a small series of human cadaveric intervertebral disks, Chiu et al (14) showed an increase in diffusion within severely degenerated disks (Thompson grade V) and suggested that this may be related to clefts within the disks allowing freer microscopic movement of water. It is possible that ADC values in degenerating disks may vary according to the severity of degeneration or the presence of specific types of derangements (eg, the presence of an annular tear).

A notable finding in our study was the significant variation in mean ADC and ADC_RATIO values in disks according to their anatomic location. Specifically, our findings showed a significant correlation between anatomic location (ie, lumbar disk level) and diffusion within normal disks. Our findings demonstrate that more cephalic normal disks had higher ADC values than more caudal normal disks. This difference was seen even when degenerated disks were removed from the analysis. It is interesting to note that this distribution of ADC and ADC_RATIO values in normal-appearing disks mirrors the prevalence of degenerative disk disease observed in our study population, with lower ADC values seen in normal disks at anatomic levels with the highest prevalence of degeneration. We interpret this observation as providing evidence to support the hypothesis that low ADC values in normal disks reflect the relative risk for degeneration. Increased biomechanical stress on lower lumbar disks (compared with biomechanical stress on more cephalic disks) causes a greater loss of water from the disk and may reduce ADC and ADC_RATIO values in lower disks. Further study (eg, in human cadaveric samples), however, will be necessary to prove this finding.

Our control ADC measurement in this study was obtained by taping a small test tube of water on each participant’s back behind the lumbar spine. The mean ADC value of water in our study was 2.33 x 10^–3 mm²/sec ± 0.38. This number lies between the ADC values reported by Mills (15) for water at 20°C (2.0 x 10^–3 mm²/sec) and those reported by Eastwood et al (16) for control water tubes that, because they were placed in contact with subjects’ heads, were more likely to be closer to 37°C (2.8 x 10^–3 mm²/sec). An unexpected finding in our study was that the mean ADC value measured in lumbar disks was similar in magnitude to the mean ADC value measured in water in the control test tubes. We had expected that disk ADC values would be lower than water ADC values. The reason for this finding is not clear, though a temperature effect may be present; the ambient temperature of the MR imaging environment is substantially different from body temperature. It is possible that the ADC of the nucleus pulposus at body temperature (37°C) is similar to the ADC of water at a lower temperature (20°– 37°C). This aspect of disk ADC measurement warrants additional study.

It is notable that the mean ADC values of normal disks in our study were greater than those previously reported by Kurunlahti et al (9) and Kerttula et al (10). In our study, mean ADC values in normal disks ranged from 1.97 x 10^–3 mm²/sec to 2.46 x 10^–3 mm²/sec (mean, 2.27 x 10^–3 mm²/sec); in previous studies, these values ranged from 1.38 x 10^–3 to 1.60 x 10^–3 mm²/sec in the x, y, and z directions. Control ADC values (ie, water ADC values) were not reported in these prior articles, and so it is difficult to determine the precise reasons for the differences between disk ADC values obtained in the present study and those obtained in prior studies. The b values chosen in our study and in prior studies were similar (400 sec/mm² in our study compared with 250–500 sec/mm² in the previous studies). Thus, it is unlikely that the choice of b value was a major factor. Nevertheless, it is possible that differences in technique between our study and prior studies could account at least in part for our higher ADC values. We used sagittal diffusion-weighted images as opposed to the transverse images used in these earlier studies. We chose the sagittal plane for study to minimize the possibility of partial volume effects from vertebral end plate bone (which would artificially decrease measured disk ADC values). Working in the sagittal plane allowed us to place ROIs centrally within the nucleus pulposus while directly observing the position of the end plates so as to avoid including them within the ROI.

Some limitations of our study are evident. Although previous studies (9,10,14) have found minor directional variations in ADC when measured in three orthogonal directions within spinal disks, we were not able to analyze this finding. First, spinal diffusion tensor imaging was not available in our institution at the time this study was performed, and so we were not able to measure anisotropy within the disk in this way. Second, our diffusion pulse sequence had equal diffusion weighting in all three cardinal directions; therefore, a comparison of ADC components in the individual (x, y, and z) directions was not possible. Further study with tensor methods will be needed to discover whether diffusion anisotropy is actually present in the nucleus pulposus. It will be necessary for the technique to distinguish between the nucleus pulposus and the highly anisotropic anulus fibrosus (17) that surrounds it. Similarly, it would be valuable to correlate anisotropy with the severity of degenerative change to determine if there are significant changes in diffusion anisotropy in disks at different stages of the degenerative process.

In conclusion, we report a statistically significant decrease in the ADC values of degenerated lumbar disks compared with the ADC values of normal disks. We also report a statistically significant dependence of lumbar disk ADC on anatomic location.

	FOOTNOTES

 
Abbreviations: ADC = apparent diffusion coefficient, ADC_RATIO = ratio of disk ADC to water tube ADC, ANOVA = analysis of variance, ROI = region of interest

Authors stated no financial relationship to disclose.

Author contributions: Guarantor of integrity of entire study, J.D.E.; study concepts and design, all authors; literature research, J.D.E., S.M.K.; clinical studies, J.D.E., S.M.K., T.A., D.P.B., J.M.P.; data acquisition, J.D.E., S.M.K., T.A., D.P.B., J.M.P.; data analysis/interpretation, J.D.E., S.M.K., D.D.; statistical analysis, J.D.E., S.M.K., D.D.; manuscript preparation, J.D.E., S.M.K., D.D., D.P.B., J.M.P.; manuscript definition of intellectual content, all authors; manuscript editing, J.D.E., S.M.K., D.D., D.P.B., J.M.P.; manuscript revision/review and final version approval, all authors

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Beauchamp NJ, Jr, Barker PB, Wang PY, vanZijl PC. Imaging of acute cerebral ischemia. Radiology 1999; 212:307-324.[Abstract/Free Full Text]
2. Marks MP, de Crespigny A, Lentz D, Enzmann DR, Albers GW, Moseley ME. Acute and chronic stroke: navigated spin-echo diffusion-weighted MR imaging. Radiology 1996; 199:403-408.[Abstract/Free Full Text]
3. Quencer RM. Advances in imaging of spinal cord injury: implications for treatment and patient evaluation. Prog Brain Res 2002; 137:3-8.[Medline]
4. Holder CA. MR diffusion imaging of the cervical spine. Magn Reson Imaging Clin N Am 2000; 8:675-686.[Medline]
5. Baur A, Dietrich O, Reiser M. Diffusion-weighted imaging of the spinal column. Neuroimaging Clin N Am 2002; 12:147-160.[CrossRef][Medline]
6. Nguyen-minh C, Ho KC, Xu R, An H, Haughton VM. Effect of degeneration of the intervertebral disc on the process of diffusion. AJNR Am J Neuroradiol 1997; 18:435-442.[Abstract]
7. Nguyen-minh C, Haughton VM, Papke RA, An H, Censky SC. Measuring diffusion of solutes into intervertebral discs with MR imaging and paramagnetic contrast medium. AJNR Am J Neuroradiol 1998; 19:1781-1784.[Abstract]
8. Holder CA, Muthupillai R, Mukundan S, Eastwood JD, Hudgins PA. Diffusion weighted MR imaging of the normal human spinal cord in vivo. AJNR Am J Neuroradiol 2000; 21:1799-1806.[Abstract/Free Full Text]
9. Kurunlahti M, Jauhiainen J, Karppinen J, Tervonen O. Correlation of diffusion in lumbar intervertebral discs with occlusion of lumbar arteries: a study in adult volunteers. Radiology 2001; 221:779- 786.[Abstract/Free Full Text]
10. Kerttula L, Jauhiainen J, Koivula A, Oikarinen J, Tervonen O. Apparent diffusion coefficients and T2 relaxation time measurements to evaluate disc degeneration: a quantitative MR study of young patients with previous vertebral fracture. Acta Radiol 2001; 42:585-591.[Medline]
11. Buckwalter JA. Ageing and degeneration of the human intervertebral disc. Spine 1995; 20:1307-1314.[Medline]
12. Horner HA, Urban JP. Effect of nutrient supply on the viability of cells from the nucleus pulposus of the intervertebral disc. Spine 2001; 26:2543-2549.[CrossRef][Medline]
13. Resnick D. Degenerative disc disease. In: Resnick D, eds. Diagnosis of bone and joint disorders. 4th ed. Philadelphia, Pa: Saunders, 2002; 1382-1475.
14. Chiu EJ, Newitt DC, Segal MR, Hu SS, Lotz JC, Majumdar S. Magnetic resonance imaging measurement of relaxation and water diffusion in the human lumbar intervertebral disc under compression in vitro. Spine 2001; 26(suppl):E437-E444.[CrossRef][Medline]
15. Mills R. Self-diffusion in normal and heavy water in the range 1–45°. J Phys Chem 1973; 77:685-688.[CrossRef]
16. Eastwood JD, Fiorella DJ, MacFall JF, Delong DM, Provenzale JM, Greenwood RS. Increased brain apparent diffusion coefficient in children with neurofibromatosis type 1. Radiology 2001; 219:354-358.[Abstract/Free Full Text]
17. Hsu EW, Setton LA. Diffusion tensor microscopy of the intervertebral disc anulus fibrosus. Magn Reson Med 1999; 41:992-999.[CrossRef][Medline]

This article has been cited by other articles:

				M. T. Modic and J. S. Ross Lumbar Degenerative Disk Disease Radiology, October 1, 2007; 245(1): 43 - 61. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2352040437v1 235/2/569    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend 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 Kealey, S. M. Articles by Eastwood, J. D. Search for Related Content PubMed PubMed Citation Articles by Kealey, S. M. Articles by Eastwood, J. D.