- Email: [email protected]
Involvement of pontine transverse and longitudinal fibers in multiple system atrophy: A tractography-based study
Journal of the Neurological Sciences 303 (2011) 61–66
Contents lists available at ScienceDirect
Journal of the Neurological Sciences j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n s
Involvement of pontine transverse and longitudinal fibers in multiple system atrophy: A tractography-based study Takahiro Makino, Shoichi Ito ⁎, Satoshi Kuwabara Department of Neurology, Graduate School of Medicine, Chiba University, Japan
a r t i c l e
i n f o
Article history: Received 25 October 2010 Received in revised form 28 December 2010 Accepted 13 January 2011 Keywords: Multiple system atrophy Pyramidal sign Diffusion tensor tractography Magnetic resonance imaging Cross sign
a b s t r a c t Objective: Pathological studies showed both pontine transverse (cortico-ponto-cerebellar) and longitudinal (corticospinal) fibers degenerate in MSA. The objective was to investigate the association between the development of cross sign, degenerations of pontine fibers, and the frequency of pyramidal signs in MSA. Methods: Patients with MSA (n = 26) and healthy subjects (n = 27) were enrolled in this study. Whole pontine transverse and longitudinal fibers were individually traced by diffusion tensor tractography. FA was calculated along each entire tractography. Cross sign was graded as: 0, no cross sign; 1, anterior–posterior line only; and 2, complete cross sign. T2-hyperintense MCPs was graded as: 0, no change; 1, slight signal change; and 2, severe signal change. FA of pontine fibers in MSA patients and that in healthy subjects was statistically evaluated by ANOVA with an overall statistical significance level of 0.05. The frequency of pyramidal signs in MSA was compared between each cross and MCP grade. Results: FA of pontine transverse fibers in MSA patients decreased with the development of cross sign. FA of Cross 2 was significantly lower than that of healthy subjects (p = 0.003). As regards pontine longitudinal fibers, FA decreased when cross sign was completed. The frequency of pyramidal signs in MCP 2 and 1 was higher than that in MCP 0. Conclusion: Pontine transverse fibers degenerate as cross sign develop, and degenerations of pontine longitudinal fibers begin, or even accelerate when cross sign becomes apparent. Pyramidal signs are frequently present when T2-hyperintense MCPs are clearly observed. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Multiple system atrophy (MSA) is an adult-onset, sporadic, progressive, neurodegenerative disease characterized by cerebellar ataxia, parkinsonism, autonomic failure, and corticospinal disorders [1]. MSA is synonymous with striatonigral degeneration when parkinsonism predominates (MSA-P), and with olivopontocerebellar atrophy when cerebellar signs predominate (MSA-C). The primary pathological changes in MSA patients are glial cell inclusions, gliosis, demyelination and axonal loss of the corticopontocerebellar tract, and neuronal cell loss of the pontine nuclei, putamen, and substantia nigra [2–8]. Loss of axons and Betz cells in the corticospinal tract and loss of Betz cells in the motor cortex have also been reported [5,9,10]. Magnetic resonance imaging (MRI) abnormalities in MSA include cerebellar atrophy, the cross (or “hot-cross-bun”) sign and highintensity signals in the middle cerebellar peduncles (MCPs) on T2weighted images (T2WI) [11–15], which are usually combined with atrophy and signal abnormalities in the dorsolateral putamen [12–16]. ⁎ Corresponding author at: Department of Neurology, Graduate School of Medicine, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba 260-8670, Japan. Tel.: +81 43 226 2126; fax: +81 43 226 2160. E-mail address: [email protected] (S. Ito). 0022-510X/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2011.01.014
In a previous study, the cross sign was considered to be caused by ‘selective’ loss of the transverse pontocerebellar tract, with ‘preservation’ of the corticospinal tracts [11]. However, the corticospinal tract is frequently affected in MSA, and pyramidal signs (hyperreflexia or extensor plantar reflexes) have been identified in 46–54% MSA patients [2,3]. Moreover, high-intensity T2 signals in the corticospinal tract could be observed, especially in advanced cases of MSA [17,18]. Diffusion tensor tractography (DTT) is a recently developed imaging tool that enables us to visualize the neural tracts in the human brain noninvasively [19]. Fractional anisotropy (FA) and apparent diffusion coefficient (ADC) that are used for diffusion tensor analyses may reflect the extent of degenerative axonal and myelin loss of white matter tracts. By using tract-of-interest or tractography-based measurements, these parameters can be calculated along the entire reconstructed neural tracts, and voxel number, which means the number of voxels including within the entire reconstructed tracts, can be measured to evaluate the volume of the reconstructed tracts [20]. In addition, several neural tracts which coexist within a part of the white matter of the human brain can be evaluated independently [20], and this approach may be a useful tool for detecting systemic degeneration such as pontocerebellar degenerations in patients with spinocerebellar degenerations [21]. To the best of our knowledge, the association between the degenerations of the corticopontocerebellar/corticospinal tract and the
62
T. Makino et al. / Journal of the Neurological Sciences 303 (2011) 61–66
frequency of pyramidal signs in MSA patients, with reference to the cross sign, has not been fully discussed in any previous pre-mortem MRI studies. Our aim in this study is to consider tractography-based diffusion tensor abnormalities of pontine transverse and longitudinal fibers to investigate associations between the development of the cross sign, T2-hyperintense MCPs, degenerations of the corticopontocerebellar and corticospinal tracts, and the frequency of pyramidal signs in MSA patients. 2. Materials and methods 2.1. Patients The subjects were 26 consecutive MSA patients (age= 64.3± 8.3 years) consisting of 14 patients with MSA-C (age=66.0±2.1 years) and 12 patients with MSA-P (age=62.6±2.7 years) and the agematched 27 normal subjects (age=65.2±1.2 years). This study was retrospective and included consecutive patients who were initially referred to our hospital. A clinical diagnosis of probable MSA was confirmed according to consensus diagnostic criteria [1]. Informed consent, which was approved by the institutional review board, was obtained from all subjects. Disease duration of the patients with MSA was 3.4±1.5 years, and there were no significant differences in disease duration between MSA-C (3.5±1.8 years) and MSA-P (3.2±1.0 years) patients. Information regarding the presence of pyramidal signs was obtained from medical records. Patients were considered to show pyramidal signs when they exhibited increased tendon reflexes or extensor plantar responses. 2.2. Magnetic resonance imaging acquisition protocol and post-processing Brain MRI was performed with a 1.5 T MRI unit (Signa Horizon, GE, Germany). Diffusion tensor images were acquired using the following
parameters: echo planar imaging (EPI), six motion-probe gradients, b factor = 0 or 1000 s/m2, repetition time= 13,000 ms, echo time = 96.4 ms, field of view = 260 × 208 mm, matrix = 128 × 128, axial slice thickness / gap= 5 / 0 mm, the number of examinations= 1, and acquisition time= 160 s. Axial T2WI was acquired using the following parameters: repetition time= 4000 ms, echo time= 96 ms, the number of examinations = 2; slice thickness / gap= 6.0 / 1.5 mm, field of view = 230 × 230 mm, and matrix = 256 × 256. The axial sections were angled to lie parallel to the antero-posterior commissure (AC-PC) line. DTT of pontine transverse and longitudinal fibers was obtained by using the public domain software dTV II and Volume One 1.72 on a Windows XP platform. The dTV II program was developed by the Image Computing and Analysis Laboratory at the Department of Radiology of the University of Tokyo Hospital, Japan, and is available at http://www.ut-radiology.umin.jp/people/masutani/dTV.htm. First, sagittal and coronal slices were reconstructed from axial slices with b factor = 0 (EPI-T2WI; slice thickness = 1.6 mm). Then, the seed, target, and avoidance areas were set on EPI-T2WI based on the established anatomical facts. Fiber tracking was started from the seed area, and the target area was set to select fibers that penetrated both the seed as well as the target areas, and the avoidance area was set to exclude fibers that penetrated the avoidance area. To trace pontine transverse fibers corresponding to the corticopontocerebellar tract, we set the seed area at entire pontine base on a mid-sagittal pontine slice (Fig. 1A), and the target area at bilateral MCPs was set on a coronal slice (Fig. 1B), in order to select fibers that passed transversely across the pontine base and led to MCPs, that is approximating the corticopontocerebellar tract. To trace pontine longitudinal fibers corresponding to the corticospinal tract, we established the following: seed areas at the bilateral cerebral peduncles at the midbrain on an axial slice (Fig. 1C), a target area at the entire upper pontine base on an axial slice (Fig. 1D), and an
Fig. 1. For the tracing of pontine transverse fibers, a seed area was set at the pontine base on the mid-sagittal image (panel A), and a target area was set at the bilateral middle cerebellar peduncles on the pontine coronal image (panel B) of EPI-T2WIs, without an avoidance area. The corticopontocerebellar tract was successfully visualized (panel E). For the tracing of pontine longitudinal fibers, seed areas were set at the bilateral cerebral peduncles at the midbrain on an axial image (C), a target area was set at the pontine base on an axial image (D), and an avoidance area was set at the bilateral middle cerebellar peduncles on a coronal image (B) of EPI-T2WIs. The corticospinal tracts were successfully visualized (panel F).
T. Makino et al. / Journal of the Neurological Sciences 303 (2011) 61–66
avoidance area at the bilateral MCPs on a coronal slice (Fig. 1B), in order to select fibers that passed both cerebral peduncles and longitudinally through the pontine base, but in a manner in which pontine transverse fibers running alongside pontine longitudinal fibers would be excluded. Finally, the FA values and ADC were calculated along entire tractography of each of the pontine transverse and longitudinal fibers in accordance with the tractography-based measurement [20]. Voxel number was defined as the number of all calculated voxels included in the reconstructed tractography, and this value was calculated in order to evaluate the density of each tract. These procedures were performed again one month later by an experienced neuroradiologist (T. M.) blinded to the diagnosis. The FA and ADC values were not measured in certain ROIs.
63
3. Results 3.1. Intra-rater reliability of tractography The procedure of tracing the pontine transverse and longitudinal fibers enabled successful visualization of the corticopontocerebellar tract and the corticospinal tract, respectively (Fig. 1E and F). As regards intrarater reliability of the pontine transverse fiber analysis, Cronbach's alpha was 0.979 for the FA value, 0.982 for ADC value, and 0.933 for voxel number, and the analysis was assessed as excellent. As regards intra-rater reliability of the pontine longitudinal fibers, Cronbach's alpha was 0.949 for the FA value, 0.957 for ADC value greater than 0.9, and 0.792 for voxel number greater than 0.7. The intra-rater reliability of the FA and ADC was considered as excellent, and that of voxel number was considered as fair.
2.3. Classification of cross signs and T2-hyperintense MCPs 3.2. Cross sign The cross sign in the pontine base on axial T2WI was visually classified into 3 grades as follows: cross grade 0 (Cross 0), no signal change (Fig. 2A); cross grade 1 (Cross 1), presence of an anterior–posterior T2hyperintense line (Fig. 2B); cross grade 2 (Cross 2), presence of cruciform lines and complete cross sign (Fig. 2C). T2-hyperintense MCPs on axial T2WI was visually classified into 3 grades as follows: MCP grade 0 (MCP 0), normal signal intensity signals (Fig. 2A); MCP grade 1 (MCP 1), slight T2 high-intensity signals within a unilateral MCP or the bilateral MCPs (Fig. 2B); and MCP grade 2 (MCP 2), clear T2 high-intensity signals within the bilateral MCPs (Fig. 2C). Classifications were made by two experienced neuroradiologists (T. M. and S. I.), both of whom were blinded to the diagnosis.
2.3.1. Statistical analysis The intra-rater reliability of the FA, ADC, and voxel number of each tractography was evaluated by calculating Cronbach's alpha as a common form of the internal consistency reliability coefficient. For the evaluation of intra-rater reliability, a neuroradiologist (T. M.) re-calculated the FA and ADC values and voxel number one month after the initial study. Analysis of variance (ANOVA) was used to evaluate differences between diffusion parameters in each tract of normal control subjects and MSA patients that were divided by cross grade, MCP grade, or pyramidal signs and then if the results were significant post-hoc pairwise test using Scheffe's F test was performed. Spearman's correlation coefficient was used to analyze the correlation between cross grade, MCP grade, and the diffusion parameters of each tractography. Pearson's correlation coefficient test was used to analyze correlations between disease duration and diffusion parameters of the MSA-C patients. P values less than 0.05 were considered as statistically significant.
Five MSA patients were classified as Cross 0 (all MSA-P), twelve as Cross 1 (6 MSA-C and 6 MSA-P), and nine as Cross 2 (8 MSA-C and 1 MSA-P), according to the classification of cross grades defined above. In the analysis of the pontine transverse fibers (Table 1), the mean FA of the Cross 2 group (mean ± standard deviation: 0.45 ± 0.02) was significantly lower than that of normal subjects (0.47 ± 0.02; p = 0.003; Fig. 3A). The mean ADC of the Cross 2 group (1.42 ± 0.26) was significantly higher than that of normal subjects (1.09 ± 0.10; p b 0.001) and the Cross 1 group (1.17± 0.13, p = 0.004: Fig. 3B). The mean voxel number of the Cross 2 group (9646 ± 3122) and the Cross 1 group (12871 ± 3344) was significantly smaller than that of normal subjects (17,770 ± 3941; p b 0.001 and p = 0.003, respectively; Fig. 3C). There were tendencies for the FA and voxel number to gradually decrease and the ADC to gradually increase in parallel with the development of the cross grade in MSA patients (Fig. 3A–C). In the analysis of the correlation between the diffusion parameters of pontine transverse fibers and the cross grade, the cross grade significantly correlated with FA (r= −0.35, p = 0.042) and voxel number (r= −0.60, p = 0.001), and tended to correlate with the ADC (r= 0.31, p = 0.06). In the analysis of pontine longitudinal fibers (Table 1), the mean ADC of the Cross 2 group (1.26 ± 0.16) was significantly higher than that of normal subjects (1.03± 0.15; p b 0.001) and the Cross 1 group (1.05 ± 0.10; p = 0.007; Fig. 3E). The mean voxel number of the Cross 2 group (9968 ± 2330) was significantly smaller than that of the Cross 1 group (14,453 ± 3013; p = 0.03; Fig. 3F). There was no significant difference between mean FA values of MSA patients in each cross grades and that of normal subjects. FA and voxel number decreased, and ADC increased, only when the cross sign was complete (Fig. 3D–F). In the analysis of the correlation between the diffusion parameters of the pontine
Fig. 2. Classifications of cross grade and MCP grade. The cross sign in the pontine base on axial T2WI was visually classified into 3 grades as follows: grade 0 (Cross 0), no change at the pontine base (A); grade 1 (Cross 1), presence of an anterior-posterior T2-high-intense line (B); and grade 2 (Cross 2), presence of vertical and horizontal T2-high-intense lines and a complete cross sign (C). MCP on axial T2WI was also visually classified into 3 grades as follows: grade 0 (MCP 0), no MCP abnormalities (A); grade 1 (MCP 1), slight T2-hyperintense MCPs (B); and grade 2 (MCP 2), clear T2-hyperintense MCPs (C).
64
T. Makino et al. / Journal of the Neurological Sciences 303 (2011) 61–66
Table 1 Correlation of diffusion abnormalities in pontine fibers with the development cross sign in patients with MSA. Transverse fibers
Cross 2 (n = 9) Cross 1 (n = 12) Cross 0 (n = 5) NC (n = 27)
0.45 ± 0.02 0.46 ± 0.02 0.46 ± 0.03 0.47 ± 0.02
Longitudinal fibers ADC, mm2/s
FA a
a,b
1.42 ± 0.26 1.17 ± 0.13 1.20 ± 0.10a 1.09 ± 0.10
Voxel number
FA a
9646 ± 3122 12,871 ± 3344a 15,327 ± 1761 17,770 ± 3941
0.48 ± 0.02 0.49 ± 0.02 0.49 ± 0.03 0.49 ± 0.02
ADC, mm2/s 1.26 ± 0.16 1.05 ± 0.10 1.06 ± 0.09 1.03 ± 0.15
a,b
Voxel number 9968 ± 2330b 14,453 ± 3013 13,649 ± 4904 13,450 ± 3479
Listed data are mean ± standard deviation. ADC = apparent diffusion coefficient; Cross 0 = cross grade 0; Cross 1 = cross grade 1; Cross 2 = cross grade 2; FA = fractional anisotropy; MSA = multiple system atrophy; NC = normal control subjects. a Significantly different from normal control subjects at p b 0.05. b Significantly different from Cross 1 subjects at p b 0.05.
longitudinal fibers and the cross grade, cross grade significantly correlated with ADC (r = 0.45, p = 0.01) and voxel number (r = −0.38, p = 0.03), and tended to correlate with FA (r= −0.31) although the results were not statistically significant. 3.3. T2-hyperintense MCPs Five patients with MSA were classified as MCP 0 (all MSA-P), eleven as MCP 1 (7 MSA-C and 4 MSA-P), and ten as MCP 2 (7 MSA-C and 3 MSA-P), according to the MCP grades defined above.
In the analysis of the pontine transverse fibers (Table 2), the mean FA of the MCP 2 group (0.45 ± 0.02) and the MCP 1 group (0.45 ± 0.02) were significantly lower than that of normal subjects (0.47 ± 0.02; p b 0.005) and the MCP 0 group (0.48 ± 0.01; p b 0.05). The mean ADC of the MCP 2 group (1.34 ± 0.23) was significantly higher than that of normal subjects (1.08 ± 0.10; p b 0.001). The mean voxel number of the MCP 2 group (9486 ± 2963) and the MCP 1 group (13275 ± 3112) were significantly smaller than that of normal subjects (17770 ± 3941; p b 0.01), and the mean voxel number of the MCP 2 group (9486 ± 2963) was significantly smaller than that of
Fig. 3. Box plots of the diffusion parameters of MSA patients, classified by Cross grade, and those of normal control subjects. Panels A, B, and C respectively show the results of the FA, ADC, and voxel number analyses of the pontine transverse fibers. Panels D, E, and F respectively show the results of the FA, ADC, and voxel number analyses of the pontine longitudinal fibers. In the pontine transverse fiber analyses, FA and voxel number gradually decreased, and the ADC gradually increased in parallel with the progression of the cross grade. In the pontine longitudinal fiber analyses, the diffusion parameters of Cross 2 subjects of MSA patients were significantly different from those of normal control subjects with the exception of the FA value. The horizontal lines of the boxes represent the 25th, 50th (median), and 75th percentiles of the distributions. The vertical lines extending from the boxes stop at the most extreme data point within the 1.5 inter-quartile ranges of boxes; points beyond this range are individually identified. † significantly different from Cross 1, and ‡ significantly different from NC at p b 0.05 by Scheffe's F test.
T. Makino et al. / Journal of the Neurological Sciences 303 (2011) 61–66
65
Table 2 Correlation of diffusion abnormalities in pontine fibers with T2-hyperintense MCPs in patients with MSA. Transverse fibers
MCP 2 (n = 10) MCP 1 (n = 11) MCP 0 (n = 5) NC (n = 27)
Longitudinal fibers ADC, mm2/s
FA a,b
0.45 ± 0.02 0.45 ± 0.02a,b 0.48 ± 0.01 0.47 ± 0.02
1.34 ± 0.23 1.24 ± 0.21 1.14 ± 0.08 1.09 ± 0.10
b
Voxel number
FA a,b
9486 ± 2963 13,275 ± 3112b 15,406 ± 1780 17,770 ± 3941
0.48 ± 0.02 0.48 ± 0.02 0.50 ± 0.01 0.49 ± 0.02
ADC, mm2/s
Voxel number
b
10,666 ± 2745 13,300 ± 3163 15,685 ± 2124 13,450 ± 3479
1.21 ± 0.15 1.01 ± 0.16 1.04 ± 0.08 1.03 ± 0.15
Listed data are mean ± standard deviation. ADC = apparent diffusion coefficient; FA = fractional anisotropy; MCP = middle cerebellar peduncle; MCP 0 = MCP grade 0; MCP 1 = MCP grade 1; MCP 2 = MCP grade 2; MSA = multiple system atrophy; NC = normal control subjects. a Significantly different from MCP 0 subjects at p b 0.05. b Significantly different from normal control subjects at p b 0.05.
the MCP 0 group (15406 ± 1780; p = 0.03). In the analysis of the correlation between the diffusion parameters of the pontine transverse fibers and MCP grade, MCP grade significantly correlated with FA (r = −0.49, p = 0.007), ADC (r = 0.36, p = 0.03), and voxel number (r = −0.67, p b 0.01). In the analysis of pontine longitudinal fibers (Table 2), the mean ADC of the MCP 2 group (1.21 ± 0.15) was significantly higher than that of normal subjects (1.03 ± 0.15; p = 0.017). There was no significant difference between mean FA values and mean voxel number of subjects in each MCP grades and those of normal subjects. In the analysis of the correlation between the diffusion parameters of the pontine longitudinal fibers and MCP grade, MCP grade significantly correlated with ADC (r= 0.42, p = 0.02) and voxel number (r= −0.47, p b 0.01), and tended to correlate with the FA (r= −0.32, p = 0.05). 3.4. Diffusion parameters of pontine longitudinal fibers and pyramidal signs in MSA 8 (30.7%) of the 26 MSA patients showed pyramidal signs (2 MSAC patients and 6 MSA-P patients), whereas 15 (57.7%) of the 26 MSA patients did not exhibit pyramidal signs (9 MSA-C patients and 6 MSA-P patients). No data regarding pyramidal signs could be obtained from 3 MSA patients. With respect to the cross grade, pyramidal signs were present in 2 (40%) patients classified with Cross 0, in 5 (45%) patients with Cross 1, and in 1 (14.3%) patient with Cross 2. According to the MCP grade, pyramidal signs were exhibited in 1 (20%) patient classified with MCP 0, in 4 (40%) patients with MCP 1, and in 3 (37.5%) patients with MCP 2. ADC value in MSA patients lacking pyramidal signs (1.15 ± 0.19) was significantly higher than that in normal subjects (1.03 ± 0.15; p b 0.05). 3.5. Correlation between disease duration and diffusion parameters of pontine fibers In MSA-C patients, the diffusion parameters of the pontine transverse and longitudinal fibers did not significantly correlate with disease duration; however, the following two non-significant trends were observed: 1) the ADC of the pontine transverse fibers tended to positively correlate with disease duration (r= 0.42, p = 0.06), and 2) the voxel number of the pontine longitudinal fibers tended to negatively correlate with disease duration (r= −0.46, p = 0.05). 4. Discussion Our results revealed correlations between the development of cross/MCP grade and decreasing FA, increasing ADC, and decreasing voxel number of the pontine transverse fibers, whereas diffusion abnormalities of the pontine longitudinal fibers appeared when the cross sign and T2-hyperintense MCPs were clearly observable. Namely, it appears that the corticopontocerebellar tract gradually degenerates with the progression of the cross sign and T2-hyperintense MCPs, and
the degeneration of the corticospinal tract begins, or even accelerates, when the cross sign becomes clearly observable. These results confirmed that the cross sign and hyperintense MCPs on T2-weighted images reflected the degeneration of the corticopontocerebellar tract [10], and these findings were in line with previous studies reporting abnormalities in the diffusivity of the corticopontocerebellar and corticospinal tracts in MSA patients [24–32]. To the best of our knowledge, this is the first DTT study to demonstrate an association between the presence of the cross sign, T2-hyperintense MCPs, and abnormalities in diffusivity along the entire paths of the corticopontocerebellar and the corticospinal tracts. In this study, the frequency of pyramidal signs in the Cross 2 and 1 groups, and in the MCP 2 and 1 groups, was almost equal to or higher than that of the Cross 0 and MCP 0 groups. These results suggest that pyramidal signs are frequently present and the degeneration of the corticospinal tract begins, or even accelerates, when the cross sign or T2-hyperintense MCPs become clearly visible on T2WI. Our study identified pyramidal signs in only 35% of the MSA patients, whereas past studies have relatively frequently found pyramidal signs (46– 54%) in patients with MSA [2,3]. This difference is presumably due to the relatively long disease duration (6.1–7.3 years) of subjects examined in previous studies, most of which were actually postmortem analyses. In contrast, the present study was a pre-mortem investigation, in which we examined patients with MSA at an early stage of disease (mean disease duration: 3.4 ± 1.5 years), such that there might have been fewer pyramidal signs than in previous studies. Our study, performed using 1.5-T MRI, failed to show statistically significant differences in diffusion parameters of the pontine longitudinal fibers, with the exception of a difference in the ADC of MSA patients lacking pyramidal signs and that of healthy subjects. FA in the entire corticospinal tract of MSA patients were reported to be lower than that of normal control subjects by 3-T MRI [27]. High magnetic fields may be necessary to reveal subtle diffusion abnormalities in the corticospinal tract in MSA patients. Alternations in diffusivity and diffusion anisotropy might reflect histological changes in normal aging and various disorders [20–24]. Diffusivity in the white matter reflects loss of restricted barriers of the fiber tract. On the other hand, diffusion anisotropy reflects a loss of axonal membrane and myelin sheath. Therefore, pathological changes in the white matter often show decreased diffusion anisotropy accompanied by increased diffusivity. Previous reports have noted that pathological changes in MSA patients included demyelination and axonal loss in the corticopontocerebellar and corticospinal tracts [2–10]. Although no pathological examinations were performed in the current study, our results may indicate such pathological degenerations along the entire paths of the corticopontocerebellar and corticospinal tracts detected as abnormalities of diffusion tensor parameters. We adopted the tractography-based measurements in order to evaluate diffusivity along the entire corticopontocerebellar and corticospinal tracts. In the recent diffusion tensor analyses examining MSA patients [29–32], the diffusion parameters have been measured at
66
T. Makino et al. / Journal of the Neurological Sciences 303 (2011) 61–66
limited small regions of interests (ROIs) selected by observers within the middle cerebellar peduncles, pontine base, and internal capsule. The ROI-based measurements of diffusion parameters have several limitations. First, the results can be influenced by the partial volume effect; for example, an analysis of ROI containing CSF can cause diffusion measurement errors or variables, especially if the cerebellar peduncles are severely atrophied. Second, ROI set within complex regions such as the pontine base may contain several tracts. In contrast with the ROIbased measurements, the tractography-based measurements enable us to calculate diffusion parameters with fewer dispersion or errors, and it was meaningful that significant differences were revealed in spite of the diffusion parameters with little variables in whole tract measurement. In addition, the tractography-based measurements can separate the required tracts from other neural tracts [20], and may be more feasible for clarifying diffusion abnormalities in the cerebellar peduncles in cases of spinocerebellar degeneration [21] and in the corticospinal tract of MSA patients [26], so this method may be able to detect the subtle pathological changes. Moreover, this method enables us to evaluate an entire path of neural tracts, therefore it may be an appropriate tool for analyzing systemic cerebellopetal degeneration such as MSA. In the current study, degeneration of the pontine transverse (corticopontocerebellar) fibers and longitudinal (corticospinal) fibers could be evaluated independently. Additionally, the association between the progression of the cross sign and degeneration of the pontine transverse fibers and longitudinal fibers was confirmed by using diffusion parameters in patients with MSA. If an efficacious therapy for MSA is developed in the future, tractography-based analysis of the pontine fibers could be a feasible tool for evaluating treatment effects objectively. The present study has several limitations. First, some tracking errors could occur when each of the pontine fibers passed closely through the brainstem. Second, only six directions of diffusion gradients with one measurement were made for each direction in order to obtain the diffusion tensor images. Third, slice thickness of diffusion tensor imaging is 5 mm with gapless, and the resolution of the tensor image may be low. Our study could elucidate the abnormalities in diffusivity of the pontine fibers corresponded with the previously reported pathological changes. In a previous study, six motion-probing gradients sufficed to trace optic radiations [33], and the quality of the DTT was improved by using 3.0-T instead of 1.5-T diffusion tensor imaging [34]. Additional 3.0-T diffusion tensor imaging studies, in which several measurements with thinner thickness are performed in each direction, are still needed.
[6]
[7]
[8]
[9]
[10]
[11] [12]
[13] [14]
[15] [16]
[17] [18] [19] [20]
[21]
[22]
[23]
[24]
[25] [26]
5. Conclusions In patients with MSA, degenerations of the corticopontocerebellar tract were showed to progress along with the development of the cross sign and T2-hyperintense MCPs. On the other hand, degenerations of the corticospinal tract were found to begin, or even accelerate with appearance of a complete cross sign. Moreover, the frequency of pyramidal signs in MSA patients was found to increase when T2hyperintense MCPs became apparent. References [1] Gilman S, Wenning GK, Low PA, et al. Second consensus statement on the diagnosis of multiple system atrophy. Neurology 2008;71:670–6. [2] Wenning GK, Tison F, Ben Shlomo Y, et al. Multiple system atrophy: A Review of 203 Pathologically Proven Cases. Mov Disord 1997;12:133–47. [3] Wenning GK, Ben-Shlomo Y, Magalhaes M, et al. Clinicopathological study of 35 cases of multiple system atrophy. J Neurol Neurosurg Psychiatry 1995;58: 160–6. [4] Lowe J, Lennox G, Leigh PN. Disorders of movement and system degenerations. In: Graham DI, Lantos PL, editors. Greenfield's neuropathology. 6th ed. Oxford University Press; 1997. p. 297–300. [5] Inoue M, Yagishita S, Ryo M, et al. The distribution and dynamic density of oligodendroglial cytoplasmic inclusions (GCIs) in multiple system atrophy: a
[27] [28]
[29]
[30]
[31]
[32]
[33]
[34]
correlation between the density of GCIs and the degree of involvement of striatonigral and olivopontocerebellar systems. Acta Neuropathol 1997;93: 585–91. Ozawa T, Paviour D, Quinn NP, et al. The spectrum of pathological involvement of the striatonigral and olivopontocerebellar systems in multiple system atrophy: clinicopathological correlations. Brain 2004;127:2657–71. Wakabayashi K, Mori F, Nishie M, et al. An autopsy case of early (“minimal change”) olivopontocerebellar atrophy (multiple system atrophy-cerebellar). Acta Neuropathol 2005;110:185–90. Yagishita S, Yokoi S, Iwabuchi K, et al. Degeneration of the corticopontine tract in olivopontocerebellar atrophy. Virchows Arch A Pathol Anat Histopathol 1991;418: 99–103. Tsuchiya K, Ozawa E, Haga C, et al. Constant involvement of the Betz cells and pyramidal tract in multiple system atrophy: a clinicopathological study of seven autopsy cases. Acta Neuropathol 2000;99:628–36. Su M, Yoshida Y, Hirata Y, et al. Primary involvement of the motor area in association with the nigrostriatal pathway in multiple system atrophy: neuropathological and morphometric evaluations. Acta Neuropathol 2001;101:57–64. Savoiardo M, Strada L, Girotti F, et al. Olivopontocerebellar atrophy: MR diagnosis and relationship to multisystem atrophy. Radiology 1990;174:693–6. Schrag A, Kingsley D, Phatouros C, et al. Clinical usefulness of magnetic resonance imaging in multiple system atrophy. J Neurol Neurosurg Psychiatry 1998;65: 65–71. Naka H, Oshita T, Murata Y, et al. Characteristic MRI findings in multiple system atrophy: comparison of three subtypes. Neuroradiology 2002;44:204–9. Bhattacharya K, Saadia D, Eisenkraft B, et al. Brain magnetic resonance imaging in multiple-system atrophy and Parkinson disease: a diagnostic algorithm. Arch Neurol 2002;59:835–42. Lee EA, Cho HI, Kim SS, et al. Comparison of magnetic resonance imaging in subtypes of multiple system atrophy. Parkinsonism Relat Disord 2004;10:363–8. Ito S, Shirai W, Hattori T. Putaminal hyperintensity on T1-weighted MR imaging in patients with the Parkinson variant of multiple system atrophy. AJNR Am J Neuroradiol 2009;30:689–92. Limberg N, Jeavons S, Robertson T, et al. Pyramidal tract imaging in multiplesystem atrophy. Mov Disord 2005;20:1527–8. Ramos CG, Silva NF, Leite AB, et al. Multiple system atrophy—unusual MRI findings. Eur J Neurol 2004;11(Suppl 2):255. Catani M, Howard RJ, Pajevic S, et al. Virtual in vivo interactive dissection of white matter fasciculi in the human brain. Neuroimage 2002;17:77–94. Taoka T, Iwasaki S, Sakamoto M, et al. Diffusion anisotropy and diffusivity of white matter tracts within the temporal stem in Alzheimer disease: evaluation of the “tract of interest” by diffusion tensor tractography. AJNR Am J Neuroradiol 2006;27:1040–5. Taoka T, Kin T, Nakagawa H, et al. Diffusivity and diffusion anisotropy of cerebellar peduncles in cases of spinocerebellar degenerative disease. Neuroimage 2007;37: 387–93. Stadlbauer A, Salomonowitz E, Strunk G, et al. Age-related degradation in the central nervous system: assessment with diffusion-tensor imaging and quantitative fiber tracking. Radiology 2008;247:179–88. Vishwas MS, Chitnis T, Pienaar R, et al. Tract-based analysis of callosal, projection, and association pathways in pediatric patients with multiple sclerosis: a preliminary study. AJNR Am J Neuroradiol 2010;31:121–8. Aoki S, Iwata NK, Masutani Y, et al. Quantitative evaluation of the pyramidal tract segmented by diffusion tensor tractography: feasibility study in patients with amyotrophic lateral sclerosis. Radiat Med 2005;23:195–9. Nilsson C, Markenroth Bloch K, Brockstedt S, et al. Tracking the neurodegeneration of parkinsonian disorders—a pilot study. Neuroradiology 2007;49:111–9. Ito M, Watanabe H, Kawai Y, et al. Usefulness of combined fractional anisotropy and apparent diffusion coefficient values for detection of involvement in multiple system atrophy. J Neurol Neurosurg Psychiatry 2007;78:722–8. Ito M, Watanabe H, Atsuta N, et al. Fractional anisotropy values detect pyramidal tract involvement in multiple system atrophy. J Neurol Sci 2008;271:40–6. Kanazawa M, Shimohata T, Terajima K, et al. Quantitative evaluation of brainstem involvement in multiple system atrophy by diffusion-weighted MR imaging. J Neurol 2004;251:1121–4. Shiga K, Yamada K, Yoshikawa K, et al. Local tissue anisotropy decreases in cerebellopetal fibers and pyramidal tract in multiple system atrophy. J Neurol 2005;252:589–96. Nicoletti G, Lodi R, Condino F, et al. Apparent diffusion coefficient measurements of middle cerebellar peduncles differentiate the Parkinson variant of MSA from Parkinson's disease and progressive supranuclear palsy. Brain 2006;129: 2679–87. Kitamura K, Nakayama K, Kosaka S, et al. Diffusion tensor imaging of corticoponto-cerebellar pathway in patients with adult onset ataxic neurodegenerative disease. Neuroradiology 2008;50:285–92. Schocke MF, Seppi K, Esterhammer R, et al. Diffusion-weighted MRI differentiates the Parkinson variant of multiple system atrophy from PD. Neurology 2002;58: 575–80. Yamamoto A, Miki Y, Urayama S, et al. Diffusion tensor fiber tractography of the optic radiation: analysis with 6-, 12-, 40-, and 81- directional motion-probing gradients, a preliminary study. AJNR Am J Neuroradiol 2007;28:92–6. Okada T, Miki Y, Fushimi Y, et al. Diffusion-tensor fiber tractography: intraindividual comparison of 3.0-T and 1.5-T MR imaging. Radiology 2006;238: 668–78.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n s
Involvement of pontine transverse and longitudinal fibers in multiple system atrophy: A tractography-based study Takahiro Makino, Shoichi Ito ⁎, Satoshi Kuwabara Department of Neurology, Graduate School of Medicine, Chiba University, Japan
a r t i c l e
i n f o
Article history: Received 25 October 2010 Received in revised form 28 December 2010 Accepted 13 January 2011 Keywords: Multiple system atrophy Pyramidal sign Diffusion tensor tractography Magnetic resonance imaging Cross sign
a b s t r a c t Objective: Pathological studies showed both pontine transverse (cortico-ponto-cerebellar) and longitudinal (corticospinal) fibers degenerate in MSA. The objective was to investigate the association between the development of cross sign, degenerations of pontine fibers, and the frequency of pyramidal signs in MSA. Methods: Patients with MSA (n = 26) and healthy subjects (n = 27) were enrolled in this study. Whole pontine transverse and longitudinal fibers were individually traced by diffusion tensor tractography. FA was calculated along each entire tractography. Cross sign was graded as: 0, no cross sign; 1, anterior–posterior line only; and 2, complete cross sign. T2-hyperintense MCPs was graded as: 0, no change; 1, slight signal change; and 2, severe signal change. FA of pontine fibers in MSA patients and that in healthy subjects was statistically evaluated by ANOVA with an overall statistical significance level of 0.05. The frequency of pyramidal signs in MSA was compared between each cross and MCP grade. Results: FA of pontine transverse fibers in MSA patients decreased with the development of cross sign. FA of Cross 2 was significantly lower than that of healthy subjects (p = 0.003). As regards pontine longitudinal fibers, FA decreased when cross sign was completed. The frequency of pyramidal signs in MCP 2 and 1 was higher than that in MCP 0. Conclusion: Pontine transverse fibers degenerate as cross sign develop, and degenerations of pontine longitudinal fibers begin, or even accelerate when cross sign becomes apparent. Pyramidal signs are frequently present when T2-hyperintense MCPs are clearly observed. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Multiple system atrophy (MSA) is an adult-onset, sporadic, progressive, neurodegenerative disease characterized by cerebellar ataxia, parkinsonism, autonomic failure, and corticospinal disorders [1]. MSA is synonymous with striatonigral degeneration when parkinsonism predominates (MSA-P), and with olivopontocerebellar atrophy when cerebellar signs predominate (MSA-C). The primary pathological changes in MSA patients are glial cell inclusions, gliosis, demyelination and axonal loss of the corticopontocerebellar tract, and neuronal cell loss of the pontine nuclei, putamen, and substantia nigra [2–8]. Loss of axons and Betz cells in the corticospinal tract and loss of Betz cells in the motor cortex have also been reported [5,9,10]. Magnetic resonance imaging (MRI) abnormalities in MSA include cerebellar atrophy, the cross (or “hot-cross-bun”) sign and highintensity signals in the middle cerebellar peduncles (MCPs) on T2weighted images (T2WI) [11–15], which are usually combined with atrophy and signal abnormalities in the dorsolateral putamen [12–16]. ⁎ Corresponding author at: Department of Neurology, Graduate School of Medicine, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba 260-8670, Japan. Tel.: +81 43 226 2126; fax: +81 43 226 2160. E-mail address: [email protected] (S. Ito). 0022-510X/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2011.01.014
In a previous study, the cross sign was considered to be caused by ‘selective’ loss of the transverse pontocerebellar tract, with ‘preservation’ of the corticospinal tracts [11]. However, the corticospinal tract is frequently affected in MSA, and pyramidal signs (hyperreflexia or extensor plantar reflexes) have been identified in 46–54% MSA patients [2,3]. Moreover, high-intensity T2 signals in the corticospinal tract could be observed, especially in advanced cases of MSA [17,18]. Diffusion tensor tractography (DTT) is a recently developed imaging tool that enables us to visualize the neural tracts in the human brain noninvasively [19]. Fractional anisotropy (FA) and apparent diffusion coefficient (ADC) that are used for diffusion tensor analyses may reflect the extent of degenerative axonal and myelin loss of white matter tracts. By using tract-of-interest or tractography-based measurements, these parameters can be calculated along the entire reconstructed neural tracts, and voxel number, which means the number of voxels including within the entire reconstructed tracts, can be measured to evaluate the volume of the reconstructed tracts [20]. In addition, several neural tracts which coexist within a part of the white matter of the human brain can be evaluated independently [20], and this approach may be a useful tool for detecting systemic degeneration such as pontocerebellar degenerations in patients with spinocerebellar degenerations [21]. To the best of our knowledge, the association between the degenerations of the corticopontocerebellar/corticospinal tract and the
62
T. Makino et al. / Journal of the Neurological Sciences 303 (2011) 61–66
frequency of pyramidal signs in MSA patients, with reference to the cross sign, has not been fully discussed in any previous pre-mortem MRI studies. Our aim in this study is to consider tractography-based diffusion tensor abnormalities of pontine transverse and longitudinal fibers to investigate associations between the development of the cross sign, T2-hyperintense MCPs, degenerations of the corticopontocerebellar and corticospinal tracts, and the frequency of pyramidal signs in MSA patients. 2. Materials and methods 2.1. Patients The subjects were 26 consecutive MSA patients (age= 64.3± 8.3 years) consisting of 14 patients with MSA-C (age=66.0±2.1 years) and 12 patients with MSA-P (age=62.6±2.7 years) and the agematched 27 normal subjects (age=65.2±1.2 years). This study was retrospective and included consecutive patients who were initially referred to our hospital. A clinical diagnosis of probable MSA was confirmed according to consensus diagnostic criteria [1]. Informed consent, which was approved by the institutional review board, was obtained from all subjects. Disease duration of the patients with MSA was 3.4±1.5 years, and there were no significant differences in disease duration between MSA-C (3.5±1.8 years) and MSA-P (3.2±1.0 years) patients. Information regarding the presence of pyramidal signs was obtained from medical records. Patients were considered to show pyramidal signs when they exhibited increased tendon reflexes or extensor plantar responses. 2.2. Magnetic resonance imaging acquisition protocol and post-processing Brain MRI was performed with a 1.5 T MRI unit (Signa Horizon, GE, Germany). Diffusion tensor images were acquired using the following
parameters: echo planar imaging (EPI), six motion-probe gradients, b factor = 0 or 1000 s/m2, repetition time= 13,000 ms, echo time = 96.4 ms, field of view = 260 × 208 mm, matrix = 128 × 128, axial slice thickness / gap= 5 / 0 mm, the number of examinations= 1, and acquisition time= 160 s. Axial T2WI was acquired using the following parameters: repetition time= 4000 ms, echo time= 96 ms, the number of examinations = 2; slice thickness / gap= 6.0 / 1.5 mm, field of view = 230 × 230 mm, and matrix = 256 × 256. The axial sections were angled to lie parallel to the antero-posterior commissure (AC-PC) line. DTT of pontine transverse and longitudinal fibers was obtained by using the public domain software dTV II and Volume One 1.72 on a Windows XP platform. The dTV II program was developed by the Image Computing and Analysis Laboratory at the Department of Radiology of the University of Tokyo Hospital, Japan, and is available at http://www.ut-radiology.umin.jp/people/masutani/dTV.htm. First, sagittal and coronal slices were reconstructed from axial slices with b factor = 0 (EPI-T2WI; slice thickness = 1.6 mm). Then, the seed, target, and avoidance areas were set on EPI-T2WI based on the established anatomical facts. Fiber tracking was started from the seed area, and the target area was set to select fibers that penetrated both the seed as well as the target areas, and the avoidance area was set to exclude fibers that penetrated the avoidance area. To trace pontine transverse fibers corresponding to the corticopontocerebellar tract, we set the seed area at entire pontine base on a mid-sagittal pontine slice (Fig. 1A), and the target area at bilateral MCPs was set on a coronal slice (Fig. 1B), in order to select fibers that passed transversely across the pontine base and led to MCPs, that is approximating the corticopontocerebellar tract. To trace pontine longitudinal fibers corresponding to the corticospinal tract, we established the following: seed areas at the bilateral cerebral peduncles at the midbrain on an axial slice (Fig. 1C), a target area at the entire upper pontine base on an axial slice (Fig. 1D), and an
Fig. 1. For the tracing of pontine transverse fibers, a seed area was set at the pontine base on the mid-sagittal image (panel A), and a target area was set at the bilateral middle cerebellar peduncles on the pontine coronal image (panel B) of EPI-T2WIs, without an avoidance area. The corticopontocerebellar tract was successfully visualized (panel E). For the tracing of pontine longitudinal fibers, seed areas were set at the bilateral cerebral peduncles at the midbrain on an axial image (C), a target area was set at the pontine base on an axial image (D), and an avoidance area was set at the bilateral middle cerebellar peduncles on a coronal image (B) of EPI-T2WIs. The corticospinal tracts were successfully visualized (panel F).
T. Makino et al. / Journal of the Neurological Sciences 303 (2011) 61–66
avoidance area at the bilateral MCPs on a coronal slice (Fig. 1B), in order to select fibers that passed both cerebral peduncles and longitudinally through the pontine base, but in a manner in which pontine transverse fibers running alongside pontine longitudinal fibers would be excluded. Finally, the FA values and ADC were calculated along entire tractography of each of the pontine transverse and longitudinal fibers in accordance with the tractography-based measurement [20]. Voxel number was defined as the number of all calculated voxels included in the reconstructed tractography, and this value was calculated in order to evaluate the density of each tract. These procedures were performed again one month later by an experienced neuroradiologist (T. M.) blinded to the diagnosis. The FA and ADC values were not measured in certain ROIs.
63
3. Results 3.1. Intra-rater reliability of tractography The procedure of tracing the pontine transverse and longitudinal fibers enabled successful visualization of the corticopontocerebellar tract and the corticospinal tract, respectively (Fig. 1E and F). As regards intrarater reliability of the pontine transverse fiber analysis, Cronbach's alpha was 0.979 for the FA value, 0.982 for ADC value, and 0.933 for voxel number, and the analysis was assessed as excellent. As regards intra-rater reliability of the pontine longitudinal fibers, Cronbach's alpha was 0.949 for the FA value, 0.957 for ADC value greater than 0.9, and 0.792 for voxel number greater than 0.7. The intra-rater reliability of the FA and ADC was considered as excellent, and that of voxel number was considered as fair.
2.3. Classification of cross signs and T2-hyperintense MCPs 3.2. Cross sign The cross sign in the pontine base on axial T2WI was visually classified into 3 grades as follows: cross grade 0 (Cross 0), no signal change (Fig. 2A); cross grade 1 (Cross 1), presence of an anterior–posterior T2hyperintense line (Fig. 2B); cross grade 2 (Cross 2), presence of cruciform lines and complete cross sign (Fig. 2C). T2-hyperintense MCPs on axial T2WI was visually classified into 3 grades as follows: MCP grade 0 (MCP 0), normal signal intensity signals (Fig. 2A); MCP grade 1 (MCP 1), slight T2 high-intensity signals within a unilateral MCP or the bilateral MCPs (Fig. 2B); and MCP grade 2 (MCP 2), clear T2 high-intensity signals within the bilateral MCPs (Fig. 2C). Classifications were made by two experienced neuroradiologists (T. M. and S. I.), both of whom were blinded to the diagnosis.
2.3.1. Statistical analysis The intra-rater reliability of the FA, ADC, and voxel number of each tractography was evaluated by calculating Cronbach's alpha as a common form of the internal consistency reliability coefficient. For the evaluation of intra-rater reliability, a neuroradiologist (T. M.) re-calculated the FA and ADC values and voxel number one month after the initial study. Analysis of variance (ANOVA) was used to evaluate differences between diffusion parameters in each tract of normal control subjects and MSA patients that were divided by cross grade, MCP grade, or pyramidal signs and then if the results were significant post-hoc pairwise test using Scheffe's F test was performed. Spearman's correlation coefficient was used to analyze the correlation between cross grade, MCP grade, and the diffusion parameters of each tractography. Pearson's correlation coefficient test was used to analyze correlations between disease duration and diffusion parameters of the MSA-C patients. P values less than 0.05 were considered as statistically significant.
Five MSA patients were classified as Cross 0 (all MSA-P), twelve as Cross 1 (6 MSA-C and 6 MSA-P), and nine as Cross 2 (8 MSA-C and 1 MSA-P), according to the classification of cross grades defined above. In the analysis of the pontine transverse fibers (Table 1), the mean FA of the Cross 2 group (mean ± standard deviation: 0.45 ± 0.02) was significantly lower than that of normal subjects (0.47 ± 0.02; p = 0.003; Fig. 3A). The mean ADC of the Cross 2 group (1.42 ± 0.26) was significantly higher than that of normal subjects (1.09 ± 0.10; p b 0.001) and the Cross 1 group (1.17± 0.13, p = 0.004: Fig. 3B). The mean voxel number of the Cross 2 group (9646 ± 3122) and the Cross 1 group (12871 ± 3344) was significantly smaller than that of normal subjects (17,770 ± 3941; p b 0.001 and p = 0.003, respectively; Fig. 3C). There were tendencies for the FA and voxel number to gradually decrease and the ADC to gradually increase in parallel with the development of the cross grade in MSA patients (Fig. 3A–C). In the analysis of the correlation between the diffusion parameters of pontine transverse fibers and the cross grade, the cross grade significantly correlated with FA (r= −0.35, p = 0.042) and voxel number (r= −0.60, p = 0.001), and tended to correlate with the ADC (r= 0.31, p = 0.06). In the analysis of pontine longitudinal fibers (Table 1), the mean ADC of the Cross 2 group (1.26 ± 0.16) was significantly higher than that of normal subjects (1.03± 0.15; p b 0.001) and the Cross 1 group (1.05 ± 0.10; p = 0.007; Fig. 3E). The mean voxel number of the Cross 2 group (9968 ± 2330) was significantly smaller than that of the Cross 1 group (14,453 ± 3013; p = 0.03; Fig. 3F). There was no significant difference between mean FA values of MSA patients in each cross grades and that of normal subjects. FA and voxel number decreased, and ADC increased, only when the cross sign was complete (Fig. 3D–F). In the analysis of the correlation between the diffusion parameters of the pontine
Fig. 2. Classifications of cross grade and MCP grade. The cross sign in the pontine base on axial T2WI was visually classified into 3 grades as follows: grade 0 (Cross 0), no change at the pontine base (A); grade 1 (Cross 1), presence of an anterior-posterior T2-high-intense line (B); and grade 2 (Cross 2), presence of vertical and horizontal T2-high-intense lines and a complete cross sign (C). MCP on axial T2WI was also visually classified into 3 grades as follows: grade 0 (MCP 0), no MCP abnormalities (A); grade 1 (MCP 1), slight T2-hyperintense MCPs (B); and grade 2 (MCP 2), clear T2-hyperintense MCPs (C).
64
T. Makino et al. / Journal of the Neurological Sciences 303 (2011) 61–66
Table 1 Correlation of diffusion abnormalities in pontine fibers with the development cross sign in patients with MSA. Transverse fibers
Cross 2 (n = 9) Cross 1 (n = 12) Cross 0 (n = 5) NC (n = 27)
0.45 ± 0.02 0.46 ± 0.02 0.46 ± 0.03 0.47 ± 0.02
Longitudinal fibers ADC, mm2/s
FA a
a,b
1.42 ± 0.26 1.17 ± 0.13 1.20 ± 0.10a 1.09 ± 0.10
Voxel number
FA a
9646 ± 3122 12,871 ± 3344a 15,327 ± 1761 17,770 ± 3941
0.48 ± 0.02 0.49 ± 0.02 0.49 ± 0.03 0.49 ± 0.02
ADC, mm2/s 1.26 ± 0.16 1.05 ± 0.10 1.06 ± 0.09 1.03 ± 0.15
a,b
Voxel number 9968 ± 2330b 14,453 ± 3013 13,649 ± 4904 13,450 ± 3479
Listed data are mean ± standard deviation. ADC = apparent diffusion coefficient; Cross 0 = cross grade 0; Cross 1 = cross grade 1; Cross 2 = cross grade 2; FA = fractional anisotropy; MSA = multiple system atrophy; NC = normal control subjects. a Significantly different from normal control subjects at p b 0.05. b Significantly different from Cross 1 subjects at p b 0.05.
longitudinal fibers and the cross grade, cross grade significantly correlated with ADC (r = 0.45, p = 0.01) and voxel number (r = −0.38, p = 0.03), and tended to correlate with FA (r= −0.31) although the results were not statistically significant. 3.3. T2-hyperintense MCPs Five patients with MSA were classified as MCP 0 (all MSA-P), eleven as MCP 1 (7 MSA-C and 4 MSA-P), and ten as MCP 2 (7 MSA-C and 3 MSA-P), according to the MCP grades defined above.
In the analysis of the pontine transverse fibers (Table 2), the mean FA of the MCP 2 group (0.45 ± 0.02) and the MCP 1 group (0.45 ± 0.02) were significantly lower than that of normal subjects (0.47 ± 0.02; p b 0.005) and the MCP 0 group (0.48 ± 0.01; p b 0.05). The mean ADC of the MCP 2 group (1.34 ± 0.23) was significantly higher than that of normal subjects (1.08 ± 0.10; p b 0.001). The mean voxel number of the MCP 2 group (9486 ± 2963) and the MCP 1 group (13275 ± 3112) were significantly smaller than that of normal subjects (17770 ± 3941; p b 0.01), and the mean voxel number of the MCP 2 group (9486 ± 2963) was significantly smaller than that of
Fig. 3. Box plots of the diffusion parameters of MSA patients, classified by Cross grade, and those of normal control subjects. Panels A, B, and C respectively show the results of the FA, ADC, and voxel number analyses of the pontine transverse fibers. Panels D, E, and F respectively show the results of the FA, ADC, and voxel number analyses of the pontine longitudinal fibers. In the pontine transverse fiber analyses, FA and voxel number gradually decreased, and the ADC gradually increased in parallel with the progression of the cross grade. In the pontine longitudinal fiber analyses, the diffusion parameters of Cross 2 subjects of MSA patients were significantly different from those of normal control subjects with the exception of the FA value. The horizontal lines of the boxes represent the 25th, 50th (median), and 75th percentiles of the distributions. The vertical lines extending from the boxes stop at the most extreme data point within the 1.5 inter-quartile ranges of boxes; points beyond this range are individually identified. † significantly different from Cross 1, and ‡ significantly different from NC at p b 0.05 by Scheffe's F test.
T. Makino et al. / Journal of the Neurological Sciences 303 (2011) 61–66
65
Table 2 Correlation of diffusion abnormalities in pontine fibers with T2-hyperintense MCPs in patients with MSA. Transverse fibers
MCP 2 (n = 10) MCP 1 (n = 11) MCP 0 (n = 5) NC (n = 27)
Longitudinal fibers ADC, mm2/s
FA a,b
0.45 ± 0.02 0.45 ± 0.02a,b 0.48 ± 0.01 0.47 ± 0.02
1.34 ± 0.23 1.24 ± 0.21 1.14 ± 0.08 1.09 ± 0.10
b
Voxel number
FA a,b
9486 ± 2963 13,275 ± 3112b 15,406 ± 1780 17,770 ± 3941
0.48 ± 0.02 0.48 ± 0.02 0.50 ± 0.01 0.49 ± 0.02
ADC, mm2/s
Voxel number
b
10,666 ± 2745 13,300 ± 3163 15,685 ± 2124 13,450 ± 3479
1.21 ± 0.15 1.01 ± 0.16 1.04 ± 0.08 1.03 ± 0.15
Listed data are mean ± standard deviation. ADC = apparent diffusion coefficient; FA = fractional anisotropy; MCP = middle cerebellar peduncle; MCP 0 = MCP grade 0; MCP 1 = MCP grade 1; MCP 2 = MCP grade 2; MSA = multiple system atrophy; NC = normal control subjects. a Significantly different from MCP 0 subjects at p b 0.05. b Significantly different from normal control subjects at p b 0.05.
the MCP 0 group (15406 ± 1780; p = 0.03). In the analysis of the correlation between the diffusion parameters of the pontine transverse fibers and MCP grade, MCP grade significantly correlated with FA (r = −0.49, p = 0.007), ADC (r = 0.36, p = 0.03), and voxel number (r = −0.67, p b 0.01). In the analysis of pontine longitudinal fibers (Table 2), the mean ADC of the MCP 2 group (1.21 ± 0.15) was significantly higher than that of normal subjects (1.03 ± 0.15; p = 0.017). There was no significant difference between mean FA values and mean voxel number of subjects in each MCP grades and those of normal subjects. In the analysis of the correlation between the diffusion parameters of the pontine longitudinal fibers and MCP grade, MCP grade significantly correlated with ADC (r= 0.42, p = 0.02) and voxel number (r= −0.47, p b 0.01), and tended to correlate with the FA (r= −0.32, p = 0.05). 3.4. Diffusion parameters of pontine longitudinal fibers and pyramidal signs in MSA 8 (30.7%) of the 26 MSA patients showed pyramidal signs (2 MSAC patients and 6 MSA-P patients), whereas 15 (57.7%) of the 26 MSA patients did not exhibit pyramidal signs (9 MSA-C patients and 6 MSA-P patients). No data regarding pyramidal signs could be obtained from 3 MSA patients. With respect to the cross grade, pyramidal signs were present in 2 (40%) patients classified with Cross 0, in 5 (45%) patients with Cross 1, and in 1 (14.3%) patient with Cross 2. According to the MCP grade, pyramidal signs were exhibited in 1 (20%) patient classified with MCP 0, in 4 (40%) patients with MCP 1, and in 3 (37.5%) patients with MCP 2. ADC value in MSA patients lacking pyramidal signs (1.15 ± 0.19) was significantly higher than that in normal subjects (1.03 ± 0.15; p b 0.05). 3.5. Correlation between disease duration and diffusion parameters of pontine fibers In MSA-C patients, the diffusion parameters of the pontine transverse and longitudinal fibers did not significantly correlate with disease duration; however, the following two non-significant trends were observed: 1) the ADC of the pontine transverse fibers tended to positively correlate with disease duration (r= 0.42, p = 0.06), and 2) the voxel number of the pontine longitudinal fibers tended to negatively correlate with disease duration (r= −0.46, p = 0.05). 4. Discussion Our results revealed correlations between the development of cross/MCP grade and decreasing FA, increasing ADC, and decreasing voxel number of the pontine transverse fibers, whereas diffusion abnormalities of the pontine longitudinal fibers appeared when the cross sign and T2-hyperintense MCPs were clearly observable. Namely, it appears that the corticopontocerebellar tract gradually degenerates with the progression of the cross sign and T2-hyperintense MCPs, and
the degeneration of the corticospinal tract begins, or even accelerates, when the cross sign becomes clearly observable. These results confirmed that the cross sign and hyperintense MCPs on T2-weighted images reflected the degeneration of the corticopontocerebellar tract [10], and these findings were in line with previous studies reporting abnormalities in the diffusivity of the corticopontocerebellar and corticospinal tracts in MSA patients [24–32]. To the best of our knowledge, this is the first DTT study to demonstrate an association between the presence of the cross sign, T2-hyperintense MCPs, and abnormalities in diffusivity along the entire paths of the corticopontocerebellar and the corticospinal tracts. In this study, the frequency of pyramidal signs in the Cross 2 and 1 groups, and in the MCP 2 and 1 groups, was almost equal to or higher than that of the Cross 0 and MCP 0 groups. These results suggest that pyramidal signs are frequently present and the degeneration of the corticospinal tract begins, or even accelerates, when the cross sign or T2-hyperintense MCPs become clearly visible on T2WI. Our study identified pyramidal signs in only 35% of the MSA patients, whereas past studies have relatively frequently found pyramidal signs (46– 54%) in patients with MSA [2,3]. This difference is presumably due to the relatively long disease duration (6.1–7.3 years) of subjects examined in previous studies, most of which were actually postmortem analyses. In contrast, the present study was a pre-mortem investigation, in which we examined patients with MSA at an early stage of disease (mean disease duration: 3.4 ± 1.5 years), such that there might have been fewer pyramidal signs than in previous studies. Our study, performed using 1.5-T MRI, failed to show statistically significant differences in diffusion parameters of the pontine longitudinal fibers, with the exception of a difference in the ADC of MSA patients lacking pyramidal signs and that of healthy subjects. FA in the entire corticospinal tract of MSA patients were reported to be lower than that of normal control subjects by 3-T MRI [27]. High magnetic fields may be necessary to reveal subtle diffusion abnormalities in the corticospinal tract in MSA patients. Alternations in diffusivity and diffusion anisotropy might reflect histological changes in normal aging and various disorders [20–24]. Diffusivity in the white matter reflects loss of restricted barriers of the fiber tract. On the other hand, diffusion anisotropy reflects a loss of axonal membrane and myelin sheath. Therefore, pathological changes in the white matter often show decreased diffusion anisotropy accompanied by increased diffusivity. Previous reports have noted that pathological changes in MSA patients included demyelination and axonal loss in the corticopontocerebellar and corticospinal tracts [2–10]. Although no pathological examinations were performed in the current study, our results may indicate such pathological degenerations along the entire paths of the corticopontocerebellar and corticospinal tracts detected as abnormalities of diffusion tensor parameters. We adopted the tractography-based measurements in order to evaluate diffusivity along the entire corticopontocerebellar and corticospinal tracts. In the recent diffusion tensor analyses examining MSA patients [29–32], the diffusion parameters have been measured at
66
T. Makino et al. / Journal of the Neurological Sciences 303 (2011) 61–66
limited small regions of interests (ROIs) selected by observers within the middle cerebellar peduncles, pontine base, and internal capsule. The ROI-based measurements of diffusion parameters have several limitations. First, the results can be influenced by the partial volume effect; for example, an analysis of ROI containing CSF can cause diffusion measurement errors or variables, especially if the cerebellar peduncles are severely atrophied. Second, ROI set within complex regions such as the pontine base may contain several tracts. In contrast with the ROIbased measurements, the tractography-based measurements enable us to calculate diffusion parameters with fewer dispersion or errors, and it was meaningful that significant differences were revealed in spite of the diffusion parameters with little variables in whole tract measurement. In addition, the tractography-based measurements can separate the required tracts from other neural tracts [20], and may be more feasible for clarifying diffusion abnormalities in the cerebellar peduncles in cases of spinocerebellar degeneration [21] and in the corticospinal tract of MSA patients [26], so this method may be able to detect the subtle pathological changes. Moreover, this method enables us to evaluate an entire path of neural tracts, therefore it may be an appropriate tool for analyzing systemic cerebellopetal degeneration such as MSA. In the current study, degeneration of the pontine transverse (corticopontocerebellar) fibers and longitudinal (corticospinal) fibers could be evaluated independently. Additionally, the association between the progression of the cross sign and degeneration of the pontine transverse fibers and longitudinal fibers was confirmed by using diffusion parameters in patients with MSA. If an efficacious therapy for MSA is developed in the future, tractography-based analysis of the pontine fibers could be a feasible tool for evaluating treatment effects objectively. The present study has several limitations. First, some tracking errors could occur when each of the pontine fibers passed closely through the brainstem. Second, only six directions of diffusion gradients with one measurement were made for each direction in order to obtain the diffusion tensor images. Third, slice thickness of diffusion tensor imaging is 5 mm with gapless, and the resolution of the tensor image may be low. Our study could elucidate the abnormalities in diffusivity of the pontine fibers corresponded with the previously reported pathological changes. In a previous study, six motion-probing gradients sufficed to trace optic radiations [33], and the quality of the DTT was improved by using 3.0-T instead of 1.5-T diffusion tensor imaging [34]. Additional 3.0-T diffusion tensor imaging studies, in which several measurements with thinner thickness are performed in each direction, are still needed.
[6]
[7]
[8]
[9]
[10]
[11] [12]
[13] [14]
[15] [16]
[17] [18] [19] [20]
[21]
[22]
[23]
[24]
[25] [26]
5. Conclusions In patients with MSA, degenerations of the corticopontocerebellar tract were showed to progress along with the development of the cross sign and T2-hyperintense MCPs. On the other hand, degenerations of the corticospinal tract were found to begin, or even accelerate with appearance of a complete cross sign. Moreover, the frequency of pyramidal signs in MSA patients was found to increase when T2hyperintense MCPs became apparent. References [1] Gilman S, Wenning GK, Low PA, et al. Second consensus statement on the diagnosis of multiple system atrophy. Neurology 2008;71:670–6. [2] Wenning GK, Tison F, Ben Shlomo Y, et al. Multiple system atrophy: A Review of 203 Pathologically Proven Cases. Mov Disord 1997;12:133–47. [3] Wenning GK, Ben-Shlomo Y, Magalhaes M, et al. Clinicopathological study of 35 cases of multiple system atrophy. J Neurol Neurosurg Psychiatry 1995;58: 160–6. [4] Lowe J, Lennox G, Leigh PN. Disorders of movement and system degenerations. In: Graham DI, Lantos PL, editors. Greenfield's neuropathology. 6th ed. Oxford University Press; 1997. p. 297–300. [5] Inoue M, Yagishita S, Ryo M, et al. The distribution and dynamic density of oligodendroglial cytoplasmic inclusions (GCIs) in multiple system atrophy: a
[27] [28]
[29]
[30]
[31]
[32]
[33]
[34]
correlation between the density of GCIs and the degree of involvement of striatonigral and olivopontocerebellar systems. Acta Neuropathol 1997;93: 585–91. Ozawa T, Paviour D, Quinn NP, et al. The spectrum of pathological involvement of the striatonigral and olivopontocerebellar systems in multiple system atrophy: clinicopathological correlations. Brain 2004;127:2657–71. Wakabayashi K, Mori F, Nishie M, et al. An autopsy case of early (“minimal change”) olivopontocerebellar atrophy (multiple system atrophy-cerebellar). Acta Neuropathol 2005;110:185–90. Yagishita S, Yokoi S, Iwabuchi K, et al. Degeneration of the corticopontine tract in olivopontocerebellar atrophy. Virchows Arch A Pathol Anat Histopathol 1991;418: 99–103. Tsuchiya K, Ozawa E, Haga C, et al. Constant involvement of the Betz cells and pyramidal tract in multiple system atrophy: a clinicopathological study of seven autopsy cases. Acta Neuropathol 2000;99:628–36. Su M, Yoshida Y, Hirata Y, et al. Primary involvement of the motor area in association with the nigrostriatal pathway in multiple system atrophy: neuropathological and morphometric evaluations. Acta Neuropathol 2001;101:57–64. Savoiardo M, Strada L, Girotti F, et al. Olivopontocerebellar atrophy: MR diagnosis and relationship to multisystem atrophy. Radiology 1990;174:693–6. Schrag A, Kingsley D, Phatouros C, et al. Clinical usefulness of magnetic resonance imaging in multiple system atrophy. J Neurol Neurosurg Psychiatry 1998;65: 65–71. Naka H, Oshita T, Murata Y, et al. Characteristic MRI findings in multiple system atrophy: comparison of three subtypes. Neuroradiology 2002;44:204–9. Bhattacharya K, Saadia D, Eisenkraft B, et al. Brain magnetic resonance imaging in multiple-system atrophy and Parkinson disease: a diagnostic algorithm. Arch Neurol 2002;59:835–42. Lee EA, Cho HI, Kim SS, et al. Comparison of magnetic resonance imaging in subtypes of multiple system atrophy. Parkinsonism Relat Disord 2004;10:363–8. Ito S, Shirai W, Hattori T. Putaminal hyperintensity on T1-weighted MR imaging in patients with the Parkinson variant of multiple system atrophy. AJNR Am J Neuroradiol 2009;30:689–92. Limberg N, Jeavons S, Robertson T, et al. Pyramidal tract imaging in multiplesystem atrophy. Mov Disord 2005;20:1527–8. Ramos CG, Silva NF, Leite AB, et al. Multiple system atrophy—unusual MRI findings. Eur J Neurol 2004;11(Suppl 2):255. Catani M, Howard RJ, Pajevic S, et al. Virtual in vivo interactive dissection of white matter fasciculi in the human brain. Neuroimage 2002;17:77–94. Taoka T, Iwasaki S, Sakamoto M, et al. Diffusion anisotropy and diffusivity of white matter tracts within the temporal stem in Alzheimer disease: evaluation of the “tract of interest” by diffusion tensor tractography. AJNR Am J Neuroradiol 2006;27:1040–5. Taoka T, Kin T, Nakagawa H, et al. Diffusivity and diffusion anisotropy of cerebellar peduncles in cases of spinocerebellar degenerative disease. Neuroimage 2007;37: 387–93. Stadlbauer A, Salomonowitz E, Strunk G, et al. Age-related degradation in the central nervous system: assessment with diffusion-tensor imaging and quantitative fiber tracking. Radiology 2008;247:179–88. Vishwas MS, Chitnis T, Pienaar R, et al. Tract-based analysis of callosal, projection, and association pathways in pediatric patients with multiple sclerosis: a preliminary study. AJNR Am J Neuroradiol 2010;31:121–8. Aoki S, Iwata NK, Masutani Y, et al. Quantitative evaluation of the pyramidal tract segmented by diffusion tensor tractography: feasibility study in patients with amyotrophic lateral sclerosis. Radiat Med 2005;23:195–9. Nilsson C, Markenroth Bloch K, Brockstedt S, et al. Tracking the neurodegeneration of parkinsonian disorders—a pilot study. Neuroradiology 2007;49:111–9. Ito M, Watanabe H, Kawai Y, et al. Usefulness of combined fractional anisotropy and apparent diffusion coefficient values for detection of involvement in multiple system atrophy. J Neurol Neurosurg Psychiatry 2007;78:722–8. Ito M, Watanabe H, Atsuta N, et al. Fractional anisotropy values detect pyramidal tract involvement in multiple system atrophy. J Neurol Sci 2008;271:40–6. Kanazawa M, Shimohata T, Terajima K, et al. Quantitative evaluation of brainstem involvement in multiple system atrophy by diffusion-weighted MR imaging. J Neurol 2004;251:1121–4. Shiga K, Yamada K, Yoshikawa K, et al. Local tissue anisotropy decreases in cerebellopetal fibers and pyramidal tract in multiple system atrophy. J Neurol 2005;252:589–96. Nicoletti G, Lodi R, Condino F, et al. Apparent diffusion coefficient measurements of middle cerebellar peduncles differentiate the Parkinson variant of MSA from Parkinson's disease and progressive supranuclear palsy. Brain 2006;129: 2679–87. Kitamura K, Nakayama K, Kosaka S, et al. Diffusion tensor imaging of corticoponto-cerebellar pathway in patients with adult onset ataxic neurodegenerative disease. Neuroradiology 2008;50:285–92. Schocke MF, Seppi K, Esterhammer R, et al. Diffusion-weighted MRI differentiates the Parkinson variant of multiple system atrophy from PD. Neurology 2002;58: 575–80. Yamamoto A, Miki Y, Urayama S, et al. Diffusion tensor fiber tractography of the optic radiation: analysis with 6-, 12-, 40-, and 81- directional motion-probing gradients, a preliminary study. AJNR Am J Neuroradiol 2007;28:92–6. Okada T, Miki Y, Fushimi Y, et al. Diffusion-tensor fiber tractography: intraindividual comparison of 3.0-T and 1.5-T MR imaging. Radiology 2006;238: 668–78.