| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Duke University Medical Center, Box 3808, Durham, NC
27710.
2
Department of Medicine, Division of Neurology, Duke University Medical Center,
Durham, NC 27710.
3
Present address: Department of Neurology, University Hospital, Petersgraben 4,
4031 Basel, Switzerland.
Received March 4, 1999;
accepted after revision January 14, 2000.
The work of S. T. Engelter was supported by a grant from the Swiss Academy
for Medical Science.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. We selected 38 patients with normal MR imaging findings from 332 patients undergoing clinical MR imaging. Diffusion-weighted MR imaging was performed with diffusion gradients applied in three orthogonal directions. For each patient, the average ADC on trace-weighted diffusion images of white matter at prespecified regions of interest and at the thalamus were compared with the patient's age.
RESULTS. For the white matter, ADC sorted by patient age in decades increased with advancing age. Patients at least 60 years old had significantly higher ADC (0.769 ± 0.019 mm2/sec x 10-3) than patients less than 60 years old (0.740 ± 0.013 mm2/sec x 10-3) (p < 0.001). Comparison of individual white matter ADC and age showed a significant increase with advancing age (p < 0.0001). For the thalamus, the average ADC among patients at least 60 years old (0.766 ± 0.015 mm2/sec x 10-3) exceeded the average ADC for patients less than 60 years old (0.745 ± 0.022 mm2/sec x 10-3) (p < 0.05). However, comparison of individual thalamic ADC and patient ages, although showing a trend to higher ADC with increasing age, did not reach statistical significance (p = 0.06).
CONCLUSION. Advancing age is associated with a small but statistically significant increase of water diffusibility in human white matter. A similar trend was present in the thalamus. These increases may reflect mild structural changes associated with normal aging.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
To better understand age-related white matter changes, we undertook a study using diffusion gradients in three orthogonal directions in a series of patients with normal findings on MR imaging. The purpose of this study was to test the hypothesis that water diffusibility of normal-appearing white matter increases with advancing age.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
Diffusion-weighted imaging was performed on a 1.5-T MR imaging system (Signa; General Electric Medical Systems, Milwaukee, WI) using a single-shot multislice spin-echo echoplanar imaging sequence with the following parameters: TR/TE, 12,000/101; flip angle, 90°; field of view, 40 x 20 cm; and matrix size, 128 x 64 mm. Between 20 and 24 axial slices were obtained with a thickness of 5 mm and an interslice gap of 2.5 mm. An inversion recovery pulse of 2200 msec was applied to prevent partial volume averaging of brain tissue with cerebrospinal fluid (CSF), which would result in falsely elevated ADC measurements [18]. This shortcoming can be overcome by application of an inversion recovery pulse that effectively nulls CSF signal and eliminates effects of CSF on ADC calculations in brain tissue [18,19,20].
Diffusion gradients were applied in three orthogonal directions with a maximum b value of 1000 sec/mm2 to obtain diffusion-weighted images in x, y, and z directions (Dx, Dy, Dz). These images were combined to form a composite trace-weighted image with the trace of the diffusion tensor = (Dx x Dy x Dz) 1/3. If contributions of gradient cross-terms are negligible [15], these trace-weighted diffusion-weighted images provide an accurate estimate of the trace of the diffusion tensor [21], which is known to be independent of fiber orientation of tissue. We assumed that gradient cross-terms were negligible in our study on the basis of two factors. First, our MR imaging system was calibrated by the manufacturer to correct for gradient eddy currents and gradient cross-terms that are present to some degree in all MR imaging systems and can cause image distortion in echoplanar diffusion-weighted imaging. This correction reduced geometric distortion to less than one pixel for all diffusion gradients and orientations. Second, we have previously performed diffusion measurements using an isotropic phantom on our MR scanner and found diffusion values to be independent of gradient orientation.
From trace-weighted diffusion-weighted images, average ADC
(ADCav) values were calculated on a pixel-by-pixel basis
using the following equation:
 |
For each patient, three ROIs were obtained in each hemisphere for a total of six ROIs per patient. In each hemisphere, one ROI (87 mm2) was placed in frontal white matter, one in occipital white matter, and one in thalamus (to obtain ADCav of subcortical gray matter) (Fig. 2A,2B). ROIs were drawn on T2-weighted images (b = 0) in a slice containing the basal ganglia and lateral ventricles with an attempt to select the same level in each patient. ROIs were transferred onto the corresponding ADCav map (Fig. 2A,2B) and ADCav values were measured by an observer who was unaware of patient age. Using the Student's t test, we determined that ADCav derived from different locations in the white matter obtained in the same patient did not vary from one another to a statistically significant degree (Table 1). The same process was repeated for values obtained in the thalamus (Table 2). The mean ADCav (±SD between ROIs) for white matter and the thalamus were determined for each patient and plotted against age.
|
|
|
|
Relationship of patient age and mean ADC was examined by analysis of variance and by means of a Kendall's rank correlation coefficient (tau) for white matter and for thalamus. When an association was shown, the relationship between age and water diffusibility was evaluated by two methods. First, patients were sorted by age according to decade. For each decade, ADCav values for white matter and thalamus were calculated and compared with those values for other decades using the unpaired Student's t test. Second, the nature of the relationship was statistically refined by means of post hoc regression models. Linear and quadratic effects in age were considered along with a broken linear model. Break points in these models were determined by graphic inspection and results of grouped t tests.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
The mean ADCav for the thalamus among the entire study population was 0.751 ± 0.02 mm2/sec x 10-3, which was not significantly different from the mean ADCav in white matter, (analysis of variance, p < 0.05), but which suggested that increasing thalamic ADCav are associated with advancing age. However, examination of a correlation coefficient indicated only a weak association between thalamic ADCav and patient age (Kendall's rank correlation coefficient [tau]; r = 0.22, p = 0.06) (Fig. 5). Comparison of the thalamic ADCav in patients 60 or more years old (0.766 ± 0.022 mm2/sec x 10-3) with those less than 60 years old (0.745 ± 0.015 mm2/sec x 10-3) also showed weak evidence of a difference (p < 0.05). When sorted by age according to decade, the thalamic ADCav for the third decade was not significantly different than for all other decades (Fig. 6).
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our data show a small but statistically significant increase of water diffusibility in human white matter with advancing age, confirming previous findings using a single diffusion gradient direction [14]. In a previous study, other investigators found greater differences between younger and older individuals than those in our study and described a relatively wide range of ADC for white matter (0.58-1.23 mm2/sec x 10-3) [14]. With application of three orthogonal diffusion gradients and a CSF-suppression pulse in our study, ADCav values were more tightly clustered within a range of 0.721-0.796 mm2/sec x 10-3. This finding suggests that ADC derived with a single gradient direction might be confounded by white matter anisotropy.
Increase in ADCav from patients in their 20s to those in their 70s was less than 10%, which is a marginal change compared with the 40-50% ADC decrease seen after acute stroke [1,2,3]. Although the changes we report are relatively small, in the future quantitative assessment of such subtle changes in water diffusibility across subject groups may assume increased importance. Quantitative measures in normal-appearing white matter are presently being performed in several disease processes (e.g., multiple sclerosis) [4, 5]. It is important that normative data are available for such studies. Our data indicate that patient age may need to be considered in measuring normative data in control subjects and that proper comparison of quantitative diffusion-weighted imaging data in disease states requires age-matched controls.
Previous studies have shown that, in white matter, directionality of water diffusion is related to orientation of white matter tracts [13]. Initially, this diffusional anisotropy was attributed to myelination alone on the basis of the fact that anisotropy increases with myelination in animals and humans [6,7,8, 25]. However, subsequent animal model studies have shown that anisotropy can also be observed in white matter before histologic evidence of myelination [25] and in unmyelinated nerves [26]. Axonal membranes appear to contribute to the anisotropic effect, whereas axonal neurofibrils and axonal transport mechanisms are noncontributory [26, 27]. In human subcortical white matter, most axons are myelinated [28]. Therefore, it seems likely that membranes around axons and their surrounding myelin contribute to the hindrance of white matter diffusibility. Increased diffusibility with advancing age might therefore reflect loss of structure provided by axons and their myelin sheaths, which has been noted in autopsy studies of aging white matter [10]. An age-related reduction in isolatable myelin was reported by some investigators, who also found an increase in unsaturated acyl chains in myelin lipids [29]. These findings were considered suggestive of decreased stability of the myelin lamellae [23] that might therefore also facilitate water diffusibility. Our data suggest that increase in molecular diffusibility occurs mainly in patients who are more than 60 years old. This finding is corroborated by autopsy data that indicates that myelin contents are relatively stable to the age of 50 years, but thereafter a 10-15% loss of myelinated fibers occurs until the eighth decade [10]. These findings parallel the change in brain volume, which is known to remain relatively constant until the age of 50 years, but begins declining thereafter [11, 30].
We chose the thalamus from several other gray matter structures (e.g., cortical gray matter, caudate nucleus, or lentiform nucleus) that could have been analyzed. The thalamus was chosen because its borders are easier to identify than other gray matter structures on echoplanar images. The margins of gray matter for lentiform nucleus, caudate, and cortical gray matter would have been difficult to determine on our images, and resultant measurements could have been substantially inaccurate because of volume-averaging effects with white matter. We did not record significant differences in ADCav between the white matter and the thalamus, a finding similar to other groups that used a CSF-suppression pulse of the type we used in our study [18,19,20]. This technique is reported to avoid the pitfall of falsely elevated ADC caused by partial volume averaging of CSF, which is thought to cause most differences in ADC between gray and white matter reported in some studies [18].
The age-related increase of water diffusibility seen in our patients is possibly a precursor to hyperintense foci on T2-weighted images commonly seen in white matter of the elderly population [31,32,33]. Such foci have been found to correlate with a variety of histologic changes, including enlargement of perivascular spaces, gliosis, and demyelination resulting from minor perivascular damage [32,33,34]. Many of these changes could potentially increase molecular diffusion. However, a longitudinal study would be necessary to determine whether increased water diffusibility is truly related to white matter hyperintense foci.
One potential explanation for our findings is that loss of myelin integrity with aging causes an increase in the amount of interstitial fluid. This explanation is unlikely because the CSF suppression pulse used with our diffusion-weighted imaging pulse sequences likely nulls not only CSF but also interstitial fluid. However, further studies are needed to more quantitatively define the degree to which interstitial fluid is nulled by a CSF-suppression pulse of the type we used.
In our study, water diffusibility in the thalamus tended to increase with advancing age, although differences with aging were not statistically significant. It is possible that the increase in water diffusibility is also caused by age-related loss of myelin because many white matter pathways exist in the thalamus. If this explanation is true, it is not surprising that the age-related increase in ADCav seen was more evident in the white matter because the thalamus contains proportionately less white matter fiber pathways than the centrum semiovale. However, it is possible that other histologic differences between gray matter and white matter account for a greater increase in water diffusibility in the white matter than in the thalamus. The thalamus contains many histologic components other than neurons and myelinated axons, such as cell bodies and dendrites. These cell bodies and dendrites have a smaller volume than neurons and their myelinated axons (which are primary structural components of white matter) [11]. These differences in volume of each type of microscopic component are thought to cause a slightly greater rate of white matter atrophy compared with gray matter atrophy [11]. Because a lower proportion of the volume of the thalamus is composed of large-volume components (i.e., neurons and myelinated axons), the relative effect of large-volume components should be less in the thalamus than in the white matter. Therefore, if increase in water diffusibility is greater in structures in which a larger proportion of degenerating elements are large-volume components, age-related increase in water diffusibility would be expected to be greater in the white matter than in the thalamus (as was seen in our study).
Because all patients were examined only once, no information about intrasubject variation and its impact on our results is available. To minimize inaccuracy in ADCav measurements, all patients were examined using the same MR imager because a phantom study has recently shown that coefficient of variation of ADCav derived from different MR imagers may approach a value as high as 10%, but may be minimized to 1% using a single system [35].
One potential limitation of our study is that it included patients and not healthy volunteers. Although all patients had clinical symptoms that prompted MR imaging, their MR imaging findings were judged to be normal. Nonetheless, it will be important to determine whether the data from our patients differ from those of healthy subjects. We are planning to study whether our results are reproduced on healthy volunteers.
Some technical limitations of our study are worth noting. First, although calibration of our MR imaging system is presumed to have reduced geometric distortion to less than one pixel, it is understood that the pixel size necessitated by the echo planar technique is large, and structures that need to be coregistered are microscopic. Second, we recognize that, although calibration of our MR imagers by the manufacturer to correct for gradient eddy currents has been performed, slight misregistration at any one voxel location could be present, which might have influenced our results. Furthermore, although our preliminary work using a phantom found that diffusion values were independent of gradient orientation, our phantom was isotropic in design rather than anisotropic. An isotropic phantom could mask effects of subvoxel misregistration. However, because error from such misregistration would be found in both younger and older patients, this factor would not be expected to account for the statistically significant differences we saw between the two populations. Unfortunately, subvoxel registration of the type needed to correct for gradient eddy currents is not readily provided by our manufacturer, and the small error that may have been introduced by this misregistration is unavoidable.
We calculated ADCav using diffusion gradients applied in three orthogonal directions but did not use the entire diffusion tensor because that pulse sequence was not available on our MR imaging system at the time this study was performed. The approach we used is widely used in clinical studies [1, 2] because it is considered an accurate estimate of the trace of the ADC tensor (in the absence of gradient cross-terms) [21] and because acquisition time is shorter compared with that of the entire diffusion tensor, which is based on six directional gradients [17]. Clearly, when available, acquisition of the entire diffusion tensor provides more information for evaluation of diffusion in anisotropic tissues such as white matter structures because it also enables evaluation of diffusional anisotropy itself [17]. Further studies are indicated to determine whether the small, but statistically significant, age-related increase in water diffusibility seen in our study is reproducible using the entire diffusion tensor and whether these age-related changes are reflected in differences in anisotropy indexes as a function of age.
In conclusion, we found evidence of slightly increased water diffusibility in subcortical white matter in older patients compared with younger patients. A similar relation is present for the thalamus. The age-related increase, which seems to be predominantly seen in patients who are more than 60 years old, may reflect mild structural changes associated with normal aging. Our findings indicate that quantitative analysis of diffusion-weighted images can provide information about brain structure that is not available solely by visual inspection of diffusion-weighted images.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Mukherjee, J.I. Berman, S.W. Chung, C.P. Hess, and R.G. Henry Diffusion Tensor MR Imaging and Fiber Tractography: Theoretic Underpinnings AJNR Am. J. Neuroradiol., April 1, 2008; 29(4): 632 - 641. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E V SULLIVAN and A PFEFFERBAUM Neuroradiological characterization of normal adult ageing Br. J. Radiol., December 1, 2007; 80(Special_Issue_2): S99 - S108. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. ANDREONE, M. TANSELLA, R. CERINI, A. VERSACE, G. RAMBALDELLI, C. PERLINI, N. DUSI, L. PELIZZA, M. BALESTRIERI, C. BARBUI, et al. Cortical white-matter microstructure in schizophrenia: Diffusion imaging study The British Journal of Psychiatry, August 1, 2007; 191(2): 113 - 119. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.L. White, W.L. Hadley, Y. Zhang, and M.A. Dogar Analysis of Central Nervous System Vasculitis with Diffusion-Weighted Imaging and Apparent Diffusion Coefficient Mapping of the Normal-Appearing Brain AJNR Am. J. Neuroradiol., May 1, 2007; 28(5): 933 - 937. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. V. Sullivan, E. Adalsteinsson, and A. Pfefferbaum Selective Age-related Degradation of Anterior Callosal Fiber Bundles Quantified In Vivo with Fiber Tracking Cereb Cortex, July 1, 2006; 16(7): 1030 - 1039. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Mascalchi, F. Lolli, R. D. Nave, C. Tessa, R. Petralli, C. Gavazzi, L. S. Politi, M. Macucci, M. Filippi, and S. Piacentini Huntington Disease: Volumetric, Diffusion-weighted, and Magnetization Transfer MR Imaging of Brain Radiology, September 1, 2004; 232(3): 867 - 873. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Provenzale, P. McGraw, P. Mhatre, A. C. Guo, and D. Delong Peritumoral Brain Regions in Gliomas and Meningiomas: Investigation with Isotropic Diffusion-weighted MR Imaging and Diffusion-Tensor MR Imaging Radiology, August 1, 2004; 232(2): 451 - 460. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Head, R. L. Buckner, J. S. Shimony, L. E. Williams, E. Akbudak, T. E. Conturo, M. McAvoy, J. C. Morris, and A. Z. Snyder Differential Vulnerability of Anterior White Matter in Nondemented Aging with Minimal Acceleration in Dementia of the Alzheimer Type: Evidence from Diffusion Tensor Imaging Cereb Cortex, April 1, 2004; 14(4): 410 - 423. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Bilgili and B. Unal Effect of Region of Interest on Interobserver Variance in Apparent Diffusion Coefficient Measures AJNR Am. J. Neuroradiol., January 1, 2004; 25(1): 108 - 111. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Rovaris, G. Iannucci, M. Cercignani, M. P. Sormani, N. De Stefano, S. Gerevini, G. Comi, and M. Filippi Age-related Changes in Conventional, Magnetization Transfer, and Diffusion-Tensor MR Imaging Findings: Study with Whole-Brain Tissue Histogram Analysis Radiology, June 1, 2003; 227(3): 731 - 738. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Lowe, M. D. Phillips, J. T. Lurito, D. Mattson, M. Dzemidzic, and V. P. Mathews Multiple Sclerosis: Low-Frequency Temporal Blood Oxygen Level-Dependent Fluctuations Indicate Reduced Functional Connectivity—Initial Results Radiology, July 1, 2002; 224(1): 184 - 192. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
