J Neurol DOI 10.1007/s00415-015-7728-1
ORIGINAL COMMUNICATION
Significance of the hot-cross bun sign on T2*-weighted MRI for the diagnosis of multiple system atrophy Kazushi Deguchi1,2 • Kazuyo Ikeda3 • Kodai Kume2 • Tadayuki Takata2 Yohei Kokudo2 • Masaki Kamada3 • Tetsuo Touge4 • Naomi Honjo5 • Tsutomu Masaki1
•
Received: 10 February 2015 / Revised: 26 March 2015 / Accepted: 27 March 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Although the sensitive detection of putaminal iron deposition by T2*-weighted imaging (T2*-WI) is of diagnostic value for multiple system atrophy (MSA), the diagnostic significance of the pontine hot-cross bun (HCB) sign with increased ferritin-bound iron in the background remains unknown. We retrospectively evaluated the cases of 33 patients with cerebellar-form MSA (MSA-C) and 21 with MSA of the parkinsonian form (MSA-P) who underwent an MRI study with a 1.5-T system. Visualization of the HCB sign, posterior putaminal hypointensity and putaminal hyperintense rim on T2*-WI was assessed by two neurologists independently using an established visual grade, and were compared with those on T2-weighted imaging (T2-WI). The visual grade of pontine and putaminal signal changes was separately assessed for probable MSA (advanced stage) and possible MSA (early stage). T2*-WI demonstrated significantly higher grades of HCB sign than T2-WI (probable MSA-C, n = 27, & Kazushi Deguchi [email protected] 1
Department of Gastroenterology and Neurology, Kagawa University Faculty of Medicine, 1750-1 Ikenobe, Miki-cho, Kita-gun, Kagawa 761-0793, Japan
2
Department of Neurology, Kagawa University Hospital, 1750-1 Ikenobe, Miki-cho, Kita-gun, Kagawa 761-0793, Japan
3
Department of Intractable Neurological Research, Kagawa University Faculty of Medicine, 1750-1 Ikenobe, Miki-cho, Kita-gun, Kagawa 761-0793, Japan
4
Department of Health Sciences, Kagawa University Faculty of Medicine, 1750-1 Ikenobe, Miki-cho, Kita-gun, Kagawa 761-0793, Japan
5
Department of Radiology, Osaka Neurosurgical Hospital, 378-1 Sanmyo-cho, Takamatsu, Kagawa 761-8083, Japan
p \ 0.001; possible MSA-C, n = 6, p \ 0.05; probable MSA-P, n = 13, p \ 0.01). The visual grade of the HCB sign on T2*-WI in the possible MSA-C patients was comparable to that in the probable MSA-C patients. Although the HCB sign in MSA-P was of lower visual grade than in MSA-C even on T2*-WI, some patients showed evolution of the HCB sign preceding the appearance of the putaminal changes. These findings suggest that T2*-WI is of extreme value for detecting the HCB sign, which is often cited as a hallmark of MSA. The appearance of the HCB sign on T2*-WI might not only support but also improve the diagnosis of MSA. Keywords Multiple system atrophy MRI T2*-weighted imaging Hot-cross bun sign Iron
Introduction Multiple system atrophy (MSA) is a sporadic, progressive, adult-onset neurodegenerative disorder characterized by varying combinations of poorly levodopa-responsive parkinsonism, cerebellar ataxia, autonomic failure and pyramidal signs. The diagnosis criteria for MSA comprise three categories reflecting differing levels of certainty: definite, probable and possible. Definite MSA is based on the neuropathologic findings of glial cytoplasmic inclusions, and probable MSA requires rigorously defined autonomic failure in addition to poor levodopa-responsive parkinsonism or cerebellar ataxia. When an individual does not fulfill the diagnostic criteria for probable MSA, structural and functional neuroimaging might support the diagnosis of possible MSA [1]. Conventional magnetic resonance imaging (MRI) has significant advantages over other imaging modalities,
123
J Neurol
including its widespread availability, relative ease in the interpretation of MRI images, and the lack of a need for complicated post-acquisition processing [2]. On MRI, atrophy of the putamen, middle cerebellar peduncle (MCP), pons, or cerebellum on conventional MRI is used as an additional feature of possible MSA, and signal changes on T2-weighted images (T2-WIs) using 1.5-T MRI such as posterior putaminal hypointensity, a hyperintense lateral putaminal rim, the hot-cross bun (HCB) sign and MCP hyperintensities can be helpful [1–3]. Although the sensitivity for detecting MSA characteristic signal changes by conventional MRI remains suboptimal, it may be possible to improve the sensitivity by modifying technical aspects of MRI [3]. Indeed, it has been demonstrated that by increasing the sensitivity to the iron deposition, T2*-weighted gradient echo (GE) sequences easily reveal putaminal iron deposition as a neuropathologic hallmark of MSA compared to T2-weighted fast spin echo (FSE) sequences [4–6]. The characteristics of T2*weighted images (T2*-WIs) might help to differentiate the parkinsonian form of MSA (MSA-P) from Parkinson’s disease (PD) [4]. Patients with MSA-P have varying degrees of olivopontocerebellar (OPC) pathology as well as predominant striatonigral (StrN) pathology [7]. In addition to the putaminal iron deposition, an elevated concentration of ferritin-bound iron in the basis pontis of MSA-P patients was recently demonstrated [8]. Although this finding suggests that the characteristic pontine signal changes (as typified by the HCB sign) might be sensitively detected on T2*-WIs along with the putaminal change, the pontine MRI findings in MSA-P patients have not drawn much attention. This could be because the HCB sign in MSA-P is less prevalent in the early stage and is completed later [9, 10]. In contrast to MSA-P, pontine signal changes including the HCB sign might appear in a relatively early stage of illness in the cerebellar form of MSA (MSA-C) [9, 10]. However, MRI findings in patients with MSA-C have scarcely been investigated [3], although as much as one-third of all MSA patients in North America and Europe have MSA-C [11, 12]. Although differentiating MSA in its disease course from PD or idiopathic late-onset cerebellar ataxia (ILOCA) can be extremely challenging [13], detailed evaluations of the HCB sign by T2*-WI might facilitate the differentiation of MSA-P and PD, and that of MSA-C and ILOCA. The present study was conducted to clarify whether T2*-weighted GE sequences suitable for detecting iron deposition give a clear picture of the HCB sign, which has an elevated concentration of ferritin-bound iron as its backdrop. In addition, we assessed whether the detection of the HCB sign by T2*-weighted GE sequences can contribute to the diagnosis of MSA-P along with the putaminal signal changes. We also evaluated whether the HCB sign
123
on T2*-WIs provides better information for the diagnosis of MSA-C.
Methods Participants We recruited 54 consecutive patients (29 men and 25 women) who were diagnosed as having MSA between November 2007 and November 2014. All of the patients fulfilled the consensus criteria of probable or possible MSA [1]. The mean age of the patients was 67 ± 8 years (range 48–83 years), and the mean duration of disease was 3.1 ± 2.2 years (range 1–12 years). The Unified MSA Rating Scale (UMSARS) scores for the patients were 19 ± 12 in Part 1 (historical review), 20 ± 11 in Part 2 (motor examination scale), and 3 ± 1 in Part 4 (global disability scale) [14]. Of the 54 patients, 33 had MSA-C and the other 21 patients had MSA-P. The main clinical features in MSA-C and MSA-P are shown in Table 1. This retrospective study was approved by our institutional ethics committee, and the research content was put on the Kagawa University Hospital website. MRI acquisition Brain MRI studies were performed with a 1.5 T system (VISART/EX, Toshiba, Tokyo; Achieva 1.5 T A-series, Philips Medical Systems, Best, The Netherlands) using a standard head coil. Images were recorded in the transverse plane using T2-weighted FSE sequences [4000–4500 ms repetition time (TR), 90–106 ms echo time (TE)] and T2*weighted GE sequences [500–732 ms TR, 15–23 ms TE, 18°–20° flip angle (FA)]. The slice thickness was 7 mm with a 1.5-mm gap. Matrices were 512 9 512. Assessment of signal intensity changes Hot-cross bun sign For the pontine intensity changes, we assessed the HCB sign which was graded as follows: no change appeared = 0; a vertical T2 high-intensity line began to appear = 1; only a clear vertical line appeared = 2; a horizontal line began to appear following a vertical line appearance = 3; both a clear horizontal and a vertical line appeared (cross line completed) = 4; and the ventral pons in front of the horizontal line showing T2 high intensity or the ventral pons decreased in size with pontine base atrophy = 5 [9]. The assessment was performed on two sequential slices where pontine signals were detected. The mean value of the grades in two slices was used as the grade of the pontine intensity changes.
J Neurol Table 1 Clinical features in MSA-C and MSA-P patients MSA-C
MSA-P
Probable cases (n = 27)
Possible cases (n = 6)
Probable cases (n = 13)
Possible cases (n = 8)
Urinary incontinence
23 (85)
NA
11 (85)
NA
Urinary disturbance
26 (96)
4 (67)
12 (92)
7 (88)
CSystolic 30 or diastolic 20 mmHg
16 (59)
NA
8 (62)
CSystolic 20 or diastolic 10 mmHg
3 (11)
3 (50)
1 (8)
2 (25)
15 (88)
3 (75)
3 (75)
4 (80)
Bradykinesia
4 (15)
1 (17)
13 (100)
8 (100)
Rigidity
4 (15)
1 (17)
12 (92)
7 (88)
Autonomic failure
Orthostatic hypotension
Erectile dysfunction Parkinsonism
NA
Tremor
2 (7)
0 (0)
7 (54)
4 (50)
Postural instability
4 (15)
1 (17)
11 (85)
7 (88)
Gait ataxia
27 (100)
6 (100)
4 (31)
0 (0)
Dysarthria
27 (100)
6 (100)
4 (31)
0 (0)
Limb ataxia
27 (100)
5 (83)
4 (31)
0 (0)
Oculomotor dysfunction
21 (78)
5 (83)
3 (8)
0 (0)
5 (19)
0 (0)
2 (15)
1 (13)
Cerebellar ataxia
Pyramidal tract sign Babinski sign with hyperreflexia
Values are number of patients (% of all subjects). Urinary disturbance means otherwise unexplained urinary urgency or frequency, or incomplete bladder emptying MSA-C multiple system atrophy of the cerebellar form, MSA-P multiple system atrophy of the parkinsonian form, NA not available
Putaminal findings Since the consensus criteria of MSA include atrophy of the putamen on MRI as an additional feature of possible MSA [1], we assessed the putaminal intensity changes in only the patients with MSA associated with putaminal atrophy. The appearance of posterolateral putaminal linearization was judged as putaminal atrophy [15]. The putaminal hypointense signal change was graded as follows: signal of putamen higher = 0; equal = 1; hypointense = 2; and markedly hypointense = 3 compared with the signal intensity of the pallidum [4]. The putaminal hyperintense rim with a posterolateral location and discontinuity of the rim, which is identifiable as MSA [16] was graded as follows: no change appeared = 0; slit-like high intensity which appeared on only one side of the putamen = 1; high intensity which appeared at both sides but one side was much weaker than the other = 2; and high intensity equally at both sides = 3 [9]. Decisions regarding the grading of signal intensity changes The signal intensity changes were assessed independently by two experienced neurologists (K. I. and K. D.) who were
blind to the patients’ clinical phenotype of MSA. The averaged grading by the two investigators was used. Statistical analysis Data are expressed as mean ± standard deviation (SD). Differences in the grade of the putaminal and pontine intensity changes between T2-WIs and T2*-WIs were assessed using the Wilcoxon signed-rank test. The Mann– Whitney test was performed to analyze intergroup differences. p values \0.05 were considered significant. All statistical analyses were performed using Stat Flex Ver.6 software.
Results Clinical profile of participants At the initial MRI assessment, 40 patients were diagnosed as having probable MSA (MSA-C, n = 27; MSA-P, n = 13). The other 14 patients were classified as having possible MSA (MSA-C, n = 6; MSA-P, n = 8). The initial MRI in these probable and possible MSA groups was conducted at 3.5 ± 2.3 and 1.9 ± 1.4 years from the
123
J Neurol
onset of clinical symptoms, respectively. Since possible MSA is thought to represent a transitional stage that evolves into probable MSA [17], the grading of MRI findings in the probable MSA group and the possible MSA group was assessed separately. In fact, five patients with MSA-C and six patients with MSA-P evolved into probable MSA over the subsequent few years. For the assessment of the development of signal abnormalities, repeated MRI was performed in 16 patients with probable MSA and seven patients with possible MSA. The time interval between the first and second MRIs was 18 ± 9 months. Assessment of the hot-cross bun sign The pontine signal intensity changes are shown in Table 2. Among the MSA-C patients, varying degrees of the HCB sign were observed in 26 of the 27 probable MSA cases (96 %) and in all of the possible MSA cases. T2*-WI showed a significantly higher grade of the HCB sign compared to T2-WI (probable MSA-C, p \ 0.001; possible MSA-C, p \ 0.05). The grading of the HCB sign on T2*and T2-WIs did not show a significant difference between the probable and possible MSA-C patients. Among the MSA-P patients, 10 of the 13 probable cases (77 %) and three of the eight possible cases (38 %) had the HCB sign. T2*-WI showed a significantly higher grade of the HCB sign compared to T2-WI in the probable cases (p \ 0.01) but not in the possible cases. The HCB sign showed an almost complete cross shape in the MSA-C patients, but not in the MSA-P patients. The difference in the grades of the HCB sign between the probable and possible MSA-P patients showed a trend toward significance on T2*-WIs (p = 0.0542), but it did not reach significance on T2-WIs.
Assessment of putaminal findings The putaminal signal intensity changes are shown in Tables 3 and 4. Putaminal atrophy was observed in eight of the 27 patients with probable MSA-C (30 %), 11 of the 13 patients with probable MSA-P (85 %) and seven of the eight patients with possible MSA-P (88 %). None of the possible MSA-C patients showed putaminal atrophy. In MSA-P, all of the probable cases with putaminal atrophy showed apparent putaminal hypointensity on T2*WI, and they demonstrated a significant higher grade of the putaminal hypointense signal change on T2*-WI compared to T2-WI (p \ 0.01) (Table 3). Three probable MSA-C patients (38 %) and three possible MSA-P (43 %) patients with putaminal atrophy showed apparent putaminal hypointensity only on T2*-WI, but the differences of visual grade between T2*- and T2-WI did not reach significance (Table 3). In MSA-P, the probable cases showed significantly higher grades of putaminal hypointensity on T2*- and T2-WIs compared to the possible cases (p \ 0.01). On the other hand, neither MSA-C nor MSA-P showed significant differences in the grade of hyperintense rim between T2*- and T2-WI (Table 4). In the MSA-P group, there was a significant difference in the grade of the hyperintense rim on T2-WIs between the probable and possible cases (p \ 0.05), but not on T2*-WIs. Repeated MRI findings Figure 1 shows repeated MRI findings in a patient with MSA-C whose case evolved from possible MSA-C into probable MSA-C. At the first MRI, the HCB sign was extremely clear on T2*-WI, but not on T2-WI (Fig. 1a, c). On a second MRI performed 15 months later, it was slightly more easy to determine the HCB on T2-WI (Fig. 1b), but the HCB sign did not catch up with the visual grade on the first T2*-WI (Fig. 1c). Similar findings were observed in another MSA-C patient.
Table 2 Assessment of the hot-cross bun (HCB) sign in MSA-C and MSA-P patients T2*-WI
T2-WI
p value
Probable cases (n = 27)
2.8 (1.3)
1.9 (1.5)
\0.001
MSA-C
Table 3 Assessment of putaminal hypointensity in MSA-C and MSA-P patients T2*-WI
T2-WI
p value
1.9 (1.1)
1.0 (0.6)
NS
Probable cases (n = 11)
2.6 (0.6)
1.6 (0.5)
\0.01
Possible cases (n = 7)
1.4 (0.6)
0.8 (0.4)
NS
Possible cases (n = 6)
2.9 (0.7)
1.4 (1.0)
\0.05
MSA-C
MSA-P Probable cases (n = 13)
1.7 (1.6)
0.8 (1.1)
\0.01
Probable cases (n = 8) MSA-P
Possible cases (n = 8)
0.7 (1.4)
0.4 (0.9)
NS
Values are means (SD). The p values were obtained by Wilcoxon signed-rank test MSA-C multiple system atrophy of the cerebellar form, MSA-P multiple system atrophy of the parkinsonian form, T2*-WI T2*weighted image, T2-WI T2-weighted image, NS not significant
123
Putaminal hypointensity was assessed in patients with putaminal atrophy. Values are means (SD). The p values were obtained by Wilcoxon signed-rank test. Abbreviations are explained in the Table 2 footnote
J Neurol Table 4 Assessment of putaminal hyperintense rim in MSA-C and MSA-P patients T2*-WI
T2-WI
p value
0.4 (1.1)
0.4 (0.7)
NS
MSA-C Probable cases (n = 8) MSA-P Probable cases (n = 11)
1.3 (1.3)
1.5 (1.2)
NS
Possible cases (n = 7)
0.4 (0.5)
0.1 (0.4)
NS
Putaminal hyperintense rim was assessed in patients with putaminal atrophy. Values are means (SD). The p values were obtained by Wilcoxon signed-rank test. Abbreviations are explained in the Table 2 footnote
Three of the 21 patients with MSA-P (14 %) did not show putaminal atrophy on the first MRI performed within 2 years from the onset of clinical symptoms. Nevertheless, one of these three patients demonstrated an extremely clear HCB sign (T2*-WI, grade 4; T2-WI, grade 2.5) on the first MRI. The remaining two patients showed no apparent HCB sign on the first MRI, but a completed HCB sign with mild putaminal atrophy was observed on the follow-up MRI performed almost a full year later. Although a patient with cerebellar ataxia less than 1 year from the onset was categorized into the probable MSA-C group based on rigorously defined autonomic failure, he did not show pontine or putaminal signal changes on T2*- or T2-WI. This patient demonstrated a
distinct HCB sign without putaminal atrophy on T2*-WI re-examined at 19 months after the first MRI. Although the pons and cerebellum on T2-WI showed equivalent isointense signal changes at the first (Fig. 1a) and second MRI (Fig. 1b), the pons on T2*-WI showed a distinct hypointense signal change relative to the cerebellar hemisphere (Fig. 1c). The hypointense signal change extended into the cerebellar dentate nuclei on the second MRI (Fig. 1d).
Discussion The three principal findings of the present study were: (i) T2*-WI demonstrated an apparent HCB sign compared to T2-WI in patients with MSA-C, and the degree of HCB sign on T2*-WI in the possible MSA-C (early stage) patients was equivalent to that of the probable MSA-C (advanced stage) patients; (ii) visualization of the HCB sign in the patients with MSA-P was inferior compared to the visualization in the patients with MSA-C even with T2*-WI, but some patients showed evolution of the HCB sign preceding the appearance of the putaminal atrophy and signal change; and (iii) posterior putaminal hypointensity was not common except in the probable MSA-P patients, even on T2*-WI. Assessments of the HCB sign on T2*-WI have not been adequately conducted. In our study, the HCB sign on T2*WI was significantly more apparent than that on T2-WI in
Fig. 1 Initial MRI findings (a, c) and re-examined MRI findings (b, d) in a patient with MSA-C (72-year-old woman). a, b Axial T2-weighted images (T2-WI). c, d Axial T2*weighted images (T2*-WI). She was diagnosed as possible MSA-C at the initial MRI study, and then evolved into probable MSA-C at the re-examined MRI approx. 1 year later. The pontine basis behind the hotcross bun (HCB) sign showed hypointense signal change compared to the cerebellar hemisphere on T2*-WI. T2*-WI demonstrated a clear HCB sign that is superior to that on T2-WI at any stage
123
J Neurol
the MSA-C patients. This finding was confirmed not only in the probable cases (corresponding to the advanced stage), but also in the possible cases (corresponding to the early stage). In addition, the advantage of T2*-WI over T2WI was observed throughout the course of the same patients with MSA-C. Although the pathologic basis of the HCB is composed of gliosis as well as loss of neurons and myelinated fibers, the high signal intensity producing the HCB on T2- and T2*-WIs could primarily reflect the increased local water content associated with astrocytosis [18]. On the other hand, the background of the HCB on T2*-WI showed clearly low intensity relative to the cerebellum. Since elevated concentrations of ferritin-bound iron in the basis pontis have been demonstrated in MSA-P [8], the significant hypointensity in the pons but not in the cerebellum on T2*-WI might reflect the increased iron deposition in the pons at a density comparable to that of the observed changes in the striatum of MSA. Thus, enhancing the contrast between water and iron on T2*-WI could greatly contribute to the improved detection of the HCB sign. None of the 100 age-matched cases (69 ± 12 years, range 41–86 years) who underwent MRI examinations (1.5 T) for neurological symptoms such as headache in our institution showed the HCB sign against a background of clear low intensity on T2*-WIs (data not shown). This finding demonstrates that the HCB on T2*-WIs in our study is a real pathological sign, unlike a pseudo-hyperintense putaminal rim sign in normal subjects on a 3-T MRI device [19]. In our study, the HCB sign on T2*-WI in the patients with MSA-P was relatively mild compared to that in the MSA-C patients. However, it is interesting to see that three of the 21 MSA-P patients (14 %) demonstrated the HCB sign preceding the appearance of putaminal changes. This unexpected finding is consistent with a previous study using T2-WI that found that 2 of 33 patients with MSA-P (6 %) within the first 2 years of disease had only the HCB sign without putaminal hypointensity and/or putaminal hyperintense rim [20]. It was also shown that the HCB sign preceded the appearance of a putaminal slit in four of 19 patients with MSA-P (21 %) [9]. These findings might be explained by a previous report that parkinsonism in MSA is caused at a high rate even by mild pathological changes in the StrN region [7]. The sensitivity of MRI could be insufficient for detecting cases of mild StrN lesion. Taken together, these findings suggest that assessments of pontine lesions by T2*-WI may facilitate the clinical diagnosis in patients with MSA-P as well as those with MSA-C. The grading of the HCB sign on T2*-WIs in the possible MSA-C patients was equivalent to that in the probable MSA-C patients, whereas the difference in the grading of the HCB sign on T2*-WI between the possible and
123
probable MSA-P patients showed a trend toward significance. Since cerebellar signs in MSA do not appear without moderate to severe pathological changes in the OPC region [7], the level of the OPC lesion even in the stage of possible MSA-C might be sufficient to form the HCB sign on T2*-WI. On the other hand, cerebellar signs in patients initially classified as having MSA-P do not become dominant over parkinsonism during follow-up [10]. This finding suggests that the OPC lesion in MSA-P is slowly progressive and remains at a less severe level. Thus, the HCB sign in MSA-P could become clear in the course of progression from possible to probable MSA-P. In the present patients with probable MSA-P, T2*-WI demonstrated significantly lower putaminal signal intensities compared to those shown by T2-WI. This finding is consistent with previous reports that T2*-WI has a great advantage for the detection of putaminal iron deposition in MSA-P [4–6]. In patients with possible MSA-P, however, the change in the putaminal signal intensity remained relatively mild even with sensitive T2*-WI for iron deposition. In addition, putaminal hypointensity was reported to be uncommon in MSA-C patients in whom putaminal atrophy appears at the later stage of the illness [9, 10]. These findings suggest that the assessment of putaminal signal intensity on T2*-WI might be of little help in the early diagnosis of MSA-P and MSA-C. The putaminal signal loss combined with the presence of a putaminal hyperintense rim indicating gliosis suggests the diagnosis of MSA with high specificity [5]. However, the frequency of the coexistence of both findings in our study remained low (MSA-C, 13 %; MSA-P, 56 %). The relatively low detection rate might be influenced by the magnetic field strength of the MRI device (1.5 T) used in this study [21]. There are several limitations to this study. First, all of the patients in this study were clinically diagnosed based on the consensus criteria [1]. Except for the cases of two patients confirmed by autopsy, we could not completely exclude MSA-mimicking disorders such as progressive supranuclear palsy with cerebellar ataxia [22]. Second, the HCB sign on T2-WI can be seen in other disorders, such as spinocerebellar ataxias [23], parkinsonism due to vasculitis [24], variant Creutzfeldt–Jakob disease [25] and cerebellar ataxia in human immunodeficiency virus (HIV) infection [26]. Since these disorders are relatively rare, we have not assessed the difference in the detection of the HCB sign on T2*-WIs or the pathological basis between MSA and the other disorders. Third, 60 % of the MSA patients in this study had MSA-C, which is predominant in Japanese MSA patients [10]. Therefore, to establish the usefulness of the HCB sign in the diagnosis of MSA-P, further assessments to identify the frequency and the time of appearance of the HCB sign in MSA-P should be carried out with larger groups of patients with MSA-P.
J Neurol
In summary, our findings demonstrated that T2*-WI gave a clear picture of the HCB sign compared to conventional T2-WI. The HCB sign on T2*-WI was observed with clarity not only in probable MSA-C but also in possible MSA-C patients. In addition, some patients with MSA-P showed that the HCB sign preceded the appearance of putaminal signal changes on T2*-WI. The detection of putaminal hypointensity remained insufficient in MSA-C and possible MSA-P, even with sensitive T2*-WI for iron deposition. These findings suggest that the diagnosis of MSA might be supported and improved by the assessment of the HCB sign on T2*-WI. Although the correlations between T2*-WI and pathological changes remain unknown, the assessment of the HCB sign on T2*-WI might be a powerful tool for the diagnosis of MSA. Acknowledgments This work was partly supported by JSPS KAKENHI Grant No. 24591297.
10.
11.
12.
13.
14.
15.
Conflicts of interest On behalf of all authors, the corresponding author states that there is no conflict of interest.
16.
Ethical standard The study was approved by the appropriate ethics committee and has therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki.
17.
References 1. Gilman S, Wenning GK, Low PA, Brooks DJ, Mathias CJ, Trojanowski JQ et al (2008) Second consensus statement on the diagnosis of multiple system atrophy. Neurology 71:670–676 2. Chandran V, Stoessl AJ (2014) Imaging in multiple system atrophy. Neurol Clin Neurosci 2:178–187 3. Brooks DJ, Seppi K (2009) Proposed neuroimaging criteria for the diagnosis of multiple system atrophy. Mov Disord 24:949–964 4. Kraft E, Trenkwalder C, Auer DP (2002) T2*-weighted MRI differentiates multiple system atrophy from Parkinson’s disease. Neurology 59:1265–1267 5. von Lewinski F, Werner C, Jorn T, Mohr A, Sixel-Doring F, Trenkwalder C (2007) T2*-weighted MRI in diagnosis of multiple system atrophy. A practical approach for clinicians. J Neurol 254:1184–1188 6. Arabia G, Morelli M, Paglionico S, Novellino F, Salsone M, Giofre L et al (2010) An magnetic resonance imaging T2*weighted sequence at short echo time to detect putaminal hypointensity in Parkinsonisms. Mov Disord 25:2728–2734 7. Ozawa T, Paviour D, Quinn NP, Josephs KA, Sangha H, Kilford L et al (2004) The spectrum of pathological involvement of the striatonigral and olivopontocerebellar systems in multiple system atrophy: clinicopathological correlations. Brain 127:2657–2671 8. Visanji NP, Collingwood JF, Finnegan ME, Tandon A, House E, Hazrati LN (2013) Iron deficiency in parkinsonism: regionspecific iron dysregulation in Parkinson’s disease and multiple system atrophy. J Parkinsons Dis 3:523–537 9. Horimoto Y, Aiba I, Yasuda T, Ohkawa Y, Katayama T, Yokokawa Y et al (2002) Longitudinal MRI study of multiple system
18.
19.
20.
21.
22.
23.
24.
25. 26.
atrophy - when do the findings appear, and what is the course? J Neurol 249:847–854 Yabe I, Soma H, Takei A, Fujiki N, Yanagihara T, Sasaki H (2006) MSA-C is the predominant clinical phenotype of MSA in Japan: analysis of 142 patients with probable MSA. J Neurol Sci 249:115–121 Geser F, Seppi K, Stampfer-Kountchev M, Kollensperger M, Diem A, Ndayisaba JP et al (2005) The European Multiple System Atrophy-Study Group (EMSA-SG). J Neural Transm 112:1677–1686 Gilman S, May SJ, Shults CW, Tanner CM, Kukull W, Lee VM et al (2005) The North American Multiple System Atrophy Study Group. J Neural Transm 112:1687–1694 Lin DJ, Hermann KL, Schmahmann JD (2014) Multiple system atrophy of the cerebellar type: clinical state of the art. Mov Disord 29:294–304 Wenning GK, Tison F, Seppi K, Sampaio C, Diem A, Yekhlef F et al (2004) Development and validation of the Unified Multiple System Atrophy Rating Scale (UMSARS). Mov Disord 19:1391–1402 Ito S, Shirai W, Hattori T (2007) Evaluating posterolateral linearization of the putaminal margin with magnetic resonance imaging to diagnose the Parkinson variant of multiple system atrophy. Mov Disord 22:578–581 Tha KK, Terae S, Tsukahara A, Soma H, Morita R, Yabe I et al (2012) Hyperintense putaminal rim at 1.5 T: prevalence in normal subjects and distinguishing features from multiple system atrophy. BMC Neurol 12:39. doi:10.1186/1471-2377-12-39 Deguchi K, Sasaki I, Touge T, Tsukaguchi M, Takeuchi H, Kuriyama S (2004) Abnormal baroreceptor-mediated vasopressin release as possible marker in early diagnosis of multiple system atrophy. J Neurol Neurosurg Psychiatry 75:110–115 Takao M, Kadowaki T, Tomita Y, Yoshida Y, Mihara B (2007) ‘Hot-cross bun sign’ of multiple system atrophy. Intern Med 46:1883 Fujii S, Matsusue E, Kinoshita T, Sugihara S, Ohama E, Ogawa T (2007) Hyperintense putaminal rim at 3T reflects fewer ferritin deposits in the lateral marginal area of the putamen. AJNR 28:777–781 Ishigame K, Yagishita A (2010) Diagnostic imaging of multiple system atrophy. Neurol Med (Tokyo, Japan) 73:340–345 (in Japanese) Watanabe H, Ito M, Fukatsu H, Senda J, Atsuta N, Kaga T et al (2010) Putaminal magnetic resonance imaging features at various magnetic field strengths in multiple system atrophy. Mov Disord 25:1916–1923 Kanazawa M, Shimohata T, Toyoshima Y, Tada M, Kakita A, Morita T et al (2009) Cerebellar involvement in progressive supranuclear palsy: a clinicopathological study. Mov Disord 24:1312–1318 Lee YC, Liu CS, Wu HM, Wang PS, Chang MH, Soong BW (2009) The ‘hot cross bun’ sign in the patients with spinocerebellar ataxia. Eur J Neurol 16:513–516 Muqit MM, Mort D, Miskiel KA, Shakir RA (2001) ‘‘Hot cross bun’’ sign in a patient with parkinsonism secondary to presumed vasculitis. J Neurol Neurosurg Psychiatry 71:565–566 Soares-Fernandes JP, Ribeiro M, Machado A (2009) ‘‘Hot cross bun’’ sign in variant Creutzfeldt-Jakob disease. AJNR 30:E37 Oji Y, Hatano T, S-i Kubo, Mori A, Nakazato T, Hattori N (2013) Cerebellar ataxia and the ‘‘hot cross bun’’ sign in association with human immunodeficiency virus infection. Neurol Clin Neurosci 1:127
123
ORIGINAL COMMUNICATION
Significance of the hot-cross bun sign on T2*-weighted MRI for the diagnosis of multiple system atrophy Kazushi Deguchi1,2 • Kazuyo Ikeda3 • Kodai Kume2 • Tadayuki Takata2 Yohei Kokudo2 • Masaki Kamada3 • Tetsuo Touge4 • Naomi Honjo5 • Tsutomu Masaki1
•
Received: 10 February 2015 / Revised: 26 March 2015 / Accepted: 27 March 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Although the sensitive detection of putaminal iron deposition by T2*-weighted imaging (T2*-WI) is of diagnostic value for multiple system atrophy (MSA), the diagnostic significance of the pontine hot-cross bun (HCB) sign with increased ferritin-bound iron in the background remains unknown. We retrospectively evaluated the cases of 33 patients with cerebellar-form MSA (MSA-C) and 21 with MSA of the parkinsonian form (MSA-P) who underwent an MRI study with a 1.5-T system. Visualization of the HCB sign, posterior putaminal hypointensity and putaminal hyperintense rim on T2*-WI was assessed by two neurologists independently using an established visual grade, and were compared with those on T2-weighted imaging (T2-WI). The visual grade of pontine and putaminal signal changes was separately assessed for probable MSA (advanced stage) and possible MSA (early stage). T2*-WI demonstrated significantly higher grades of HCB sign than T2-WI (probable MSA-C, n = 27, & Kazushi Deguchi [email protected] 1
Department of Gastroenterology and Neurology, Kagawa University Faculty of Medicine, 1750-1 Ikenobe, Miki-cho, Kita-gun, Kagawa 761-0793, Japan
2
Department of Neurology, Kagawa University Hospital, 1750-1 Ikenobe, Miki-cho, Kita-gun, Kagawa 761-0793, Japan
3
Department of Intractable Neurological Research, Kagawa University Faculty of Medicine, 1750-1 Ikenobe, Miki-cho, Kita-gun, Kagawa 761-0793, Japan
4
Department of Health Sciences, Kagawa University Faculty of Medicine, 1750-1 Ikenobe, Miki-cho, Kita-gun, Kagawa 761-0793, Japan
5
Department of Radiology, Osaka Neurosurgical Hospital, 378-1 Sanmyo-cho, Takamatsu, Kagawa 761-8083, Japan
p \ 0.001; possible MSA-C, n = 6, p \ 0.05; probable MSA-P, n = 13, p \ 0.01). The visual grade of the HCB sign on T2*-WI in the possible MSA-C patients was comparable to that in the probable MSA-C patients. Although the HCB sign in MSA-P was of lower visual grade than in MSA-C even on T2*-WI, some patients showed evolution of the HCB sign preceding the appearance of the putaminal changes. These findings suggest that T2*-WI is of extreme value for detecting the HCB sign, which is often cited as a hallmark of MSA. The appearance of the HCB sign on T2*-WI might not only support but also improve the diagnosis of MSA. Keywords Multiple system atrophy MRI T2*-weighted imaging Hot-cross bun sign Iron
Introduction Multiple system atrophy (MSA) is a sporadic, progressive, adult-onset neurodegenerative disorder characterized by varying combinations of poorly levodopa-responsive parkinsonism, cerebellar ataxia, autonomic failure and pyramidal signs. The diagnosis criteria for MSA comprise three categories reflecting differing levels of certainty: definite, probable and possible. Definite MSA is based on the neuropathologic findings of glial cytoplasmic inclusions, and probable MSA requires rigorously defined autonomic failure in addition to poor levodopa-responsive parkinsonism or cerebellar ataxia. When an individual does not fulfill the diagnostic criteria for probable MSA, structural and functional neuroimaging might support the diagnosis of possible MSA [1]. Conventional magnetic resonance imaging (MRI) has significant advantages over other imaging modalities,
123
J Neurol
including its widespread availability, relative ease in the interpretation of MRI images, and the lack of a need for complicated post-acquisition processing [2]. On MRI, atrophy of the putamen, middle cerebellar peduncle (MCP), pons, or cerebellum on conventional MRI is used as an additional feature of possible MSA, and signal changes on T2-weighted images (T2-WIs) using 1.5-T MRI such as posterior putaminal hypointensity, a hyperintense lateral putaminal rim, the hot-cross bun (HCB) sign and MCP hyperintensities can be helpful [1–3]. Although the sensitivity for detecting MSA characteristic signal changes by conventional MRI remains suboptimal, it may be possible to improve the sensitivity by modifying technical aspects of MRI [3]. Indeed, it has been demonstrated that by increasing the sensitivity to the iron deposition, T2*-weighted gradient echo (GE) sequences easily reveal putaminal iron deposition as a neuropathologic hallmark of MSA compared to T2-weighted fast spin echo (FSE) sequences [4–6]. The characteristics of T2*weighted images (T2*-WIs) might help to differentiate the parkinsonian form of MSA (MSA-P) from Parkinson’s disease (PD) [4]. Patients with MSA-P have varying degrees of olivopontocerebellar (OPC) pathology as well as predominant striatonigral (StrN) pathology [7]. In addition to the putaminal iron deposition, an elevated concentration of ferritin-bound iron in the basis pontis of MSA-P patients was recently demonstrated [8]. Although this finding suggests that the characteristic pontine signal changes (as typified by the HCB sign) might be sensitively detected on T2*-WIs along with the putaminal change, the pontine MRI findings in MSA-P patients have not drawn much attention. This could be because the HCB sign in MSA-P is less prevalent in the early stage and is completed later [9, 10]. In contrast to MSA-P, pontine signal changes including the HCB sign might appear in a relatively early stage of illness in the cerebellar form of MSA (MSA-C) [9, 10]. However, MRI findings in patients with MSA-C have scarcely been investigated [3], although as much as one-third of all MSA patients in North America and Europe have MSA-C [11, 12]. Although differentiating MSA in its disease course from PD or idiopathic late-onset cerebellar ataxia (ILOCA) can be extremely challenging [13], detailed evaluations of the HCB sign by T2*-WI might facilitate the differentiation of MSA-P and PD, and that of MSA-C and ILOCA. The present study was conducted to clarify whether T2*-weighted GE sequences suitable for detecting iron deposition give a clear picture of the HCB sign, which has an elevated concentration of ferritin-bound iron as its backdrop. In addition, we assessed whether the detection of the HCB sign by T2*-weighted GE sequences can contribute to the diagnosis of MSA-P along with the putaminal signal changes. We also evaluated whether the HCB sign
123
on T2*-WIs provides better information for the diagnosis of MSA-C.
Methods Participants We recruited 54 consecutive patients (29 men and 25 women) who were diagnosed as having MSA between November 2007 and November 2014. All of the patients fulfilled the consensus criteria of probable or possible MSA [1]. The mean age of the patients was 67 ± 8 years (range 48–83 years), and the mean duration of disease was 3.1 ± 2.2 years (range 1–12 years). The Unified MSA Rating Scale (UMSARS) scores for the patients were 19 ± 12 in Part 1 (historical review), 20 ± 11 in Part 2 (motor examination scale), and 3 ± 1 in Part 4 (global disability scale) [14]. Of the 54 patients, 33 had MSA-C and the other 21 patients had MSA-P. The main clinical features in MSA-C and MSA-P are shown in Table 1. This retrospective study was approved by our institutional ethics committee, and the research content was put on the Kagawa University Hospital website. MRI acquisition Brain MRI studies were performed with a 1.5 T system (VISART/EX, Toshiba, Tokyo; Achieva 1.5 T A-series, Philips Medical Systems, Best, The Netherlands) using a standard head coil. Images were recorded in the transverse plane using T2-weighted FSE sequences [4000–4500 ms repetition time (TR), 90–106 ms echo time (TE)] and T2*weighted GE sequences [500–732 ms TR, 15–23 ms TE, 18°–20° flip angle (FA)]. The slice thickness was 7 mm with a 1.5-mm gap. Matrices were 512 9 512. Assessment of signal intensity changes Hot-cross bun sign For the pontine intensity changes, we assessed the HCB sign which was graded as follows: no change appeared = 0; a vertical T2 high-intensity line began to appear = 1; only a clear vertical line appeared = 2; a horizontal line began to appear following a vertical line appearance = 3; both a clear horizontal and a vertical line appeared (cross line completed) = 4; and the ventral pons in front of the horizontal line showing T2 high intensity or the ventral pons decreased in size with pontine base atrophy = 5 [9]. The assessment was performed on two sequential slices where pontine signals were detected. The mean value of the grades in two slices was used as the grade of the pontine intensity changes.
J Neurol Table 1 Clinical features in MSA-C and MSA-P patients MSA-C
MSA-P
Probable cases (n = 27)
Possible cases (n = 6)
Probable cases (n = 13)
Possible cases (n = 8)
Urinary incontinence
23 (85)
NA
11 (85)
NA
Urinary disturbance
26 (96)
4 (67)
12 (92)
7 (88)
CSystolic 30 or diastolic 20 mmHg
16 (59)
NA
8 (62)
CSystolic 20 or diastolic 10 mmHg
3 (11)
3 (50)
1 (8)
2 (25)
15 (88)
3 (75)
3 (75)
4 (80)
Bradykinesia
4 (15)
1 (17)
13 (100)
8 (100)
Rigidity
4 (15)
1 (17)
12 (92)
7 (88)
Autonomic failure
Orthostatic hypotension
Erectile dysfunction Parkinsonism
NA
Tremor
2 (7)
0 (0)
7 (54)
4 (50)
Postural instability
4 (15)
1 (17)
11 (85)
7 (88)
Gait ataxia
27 (100)
6 (100)
4 (31)
0 (0)
Dysarthria
27 (100)
6 (100)
4 (31)
0 (0)
Limb ataxia
27 (100)
5 (83)
4 (31)
0 (0)
Oculomotor dysfunction
21 (78)
5 (83)
3 (8)
0 (0)
5 (19)
0 (0)
2 (15)
1 (13)
Cerebellar ataxia
Pyramidal tract sign Babinski sign with hyperreflexia
Values are number of patients (% of all subjects). Urinary disturbance means otherwise unexplained urinary urgency or frequency, or incomplete bladder emptying MSA-C multiple system atrophy of the cerebellar form, MSA-P multiple system atrophy of the parkinsonian form, NA not available
Putaminal findings Since the consensus criteria of MSA include atrophy of the putamen on MRI as an additional feature of possible MSA [1], we assessed the putaminal intensity changes in only the patients with MSA associated with putaminal atrophy. The appearance of posterolateral putaminal linearization was judged as putaminal atrophy [15]. The putaminal hypointense signal change was graded as follows: signal of putamen higher = 0; equal = 1; hypointense = 2; and markedly hypointense = 3 compared with the signal intensity of the pallidum [4]. The putaminal hyperintense rim with a posterolateral location and discontinuity of the rim, which is identifiable as MSA [16] was graded as follows: no change appeared = 0; slit-like high intensity which appeared on only one side of the putamen = 1; high intensity which appeared at both sides but one side was much weaker than the other = 2; and high intensity equally at both sides = 3 [9]. Decisions regarding the grading of signal intensity changes The signal intensity changes were assessed independently by two experienced neurologists (K. I. and K. D.) who were
blind to the patients’ clinical phenotype of MSA. The averaged grading by the two investigators was used. Statistical analysis Data are expressed as mean ± standard deviation (SD). Differences in the grade of the putaminal and pontine intensity changes between T2-WIs and T2*-WIs were assessed using the Wilcoxon signed-rank test. The Mann– Whitney test was performed to analyze intergroup differences. p values \0.05 were considered significant. All statistical analyses were performed using Stat Flex Ver.6 software.
Results Clinical profile of participants At the initial MRI assessment, 40 patients were diagnosed as having probable MSA (MSA-C, n = 27; MSA-P, n = 13). The other 14 patients were classified as having possible MSA (MSA-C, n = 6; MSA-P, n = 8). The initial MRI in these probable and possible MSA groups was conducted at 3.5 ± 2.3 and 1.9 ± 1.4 years from the
123
J Neurol
onset of clinical symptoms, respectively. Since possible MSA is thought to represent a transitional stage that evolves into probable MSA [17], the grading of MRI findings in the probable MSA group and the possible MSA group was assessed separately. In fact, five patients with MSA-C and six patients with MSA-P evolved into probable MSA over the subsequent few years. For the assessment of the development of signal abnormalities, repeated MRI was performed in 16 patients with probable MSA and seven patients with possible MSA. The time interval between the first and second MRIs was 18 ± 9 months. Assessment of the hot-cross bun sign The pontine signal intensity changes are shown in Table 2. Among the MSA-C patients, varying degrees of the HCB sign were observed in 26 of the 27 probable MSA cases (96 %) and in all of the possible MSA cases. T2*-WI showed a significantly higher grade of the HCB sign compared to T2-WI (probable MSA-C, p \ 0.001; possible MSA-C, p \ 0.05). The grading of the HCB sign on T2*and T2-WIs did not show a significant difference between the probable and possible MSA-C patients. Among the MSA-P patients, 10 of the 13 probable cases (77 %) and three of the eight possible cases (38 %) had the HCB sign. T2*-WI showed a significantly higher grade of the HCB sign compared to T2-WI in the probable cases (p \ 0.01) but not in the possible cases. The HCB sign showed an almost complete cross shape in the MSA-C patients, but not in the MSA-P patients. The difference in the grades of the HCB sign between the probable and possible MSA-P patients showed a trend toward significance on T2*-WIs (p = 0.0542), but it did not reach significance on T2-WIs.
Assessment of putaminal findings The putaminal signal intensity changes are shown in Tables 3 and 4. Putaminal atrophy was observed in eight of the 27 patients with probable MSA-C (30 %), 11 of the 13 patients with probable MSA-P (85 %) and seven of the eight patients with possible MSA-P (88 %). None of the possible MSA-C patients showed putaminal atrophy. In MSA-P, all of the probable cases with putaminal atrophy showed apparent putaminal hypointensity on T2*WI, and they demonstrated a significant higher grade of the putaminal hypointense signal change on T2*-WI compared to T2-WI (p \ 0.01) (Table 3). Three probable MSA-C patients (38 %) and three possible MSA-P (43 %) patients with putaminal atrophy showed apparent putaminal hypointensity only on T2*-WI, but the differences of visual grade between T2*- and T2-WI did not reach significance (Table 3). In MSA-P, the probable cases showed significantly higher grades of putaminal hypointensity on T2*- and T2-WIs compared to the possible cases (p \ 0.01). On the other hand, neither MSA-C nor MSA-P showed significant differences in the grade of hyperintense rim between T2*- and T2-WI (Table 4). In the MSA-P group, there was a significant difference in the grade of the hyperintense rim on T2-WIs between the probable and possible cases (p \ 0.05), but not on T2*-WIs. Repeated MRI findings Figure 1 shows repeated MRI findings in a patient with MSA-C whose case evolved from possible MSA-C into probable MSA-C. At the first MRI, the HCB sign was extremely clear on T2*-WI, but not on T2-WI (Fig. 1a, c). On a second MRI performed 15 months later, it was slightly more easy to determine the HCB on T2-WI (Fig. 1b), but the HCB sign did not catch up with the visual grade on the first T2*-WI (Fig. 1c). Similar findings were observed in another MSA-C patient.
Table 2 Assessment of the hot-cross bun (HCB) sign in MSA-C and MSA-P patients T2*-WI
T2-WI
p value
Probable cases (n = 27)
2.8 (1.3)
1.9 (1.5)
\0.001
MSA-C
Table 3 Assessment of putaminal hypointensity in MSA-C and MSA-P patients T2*-WI
T2-WI
p value
1.9 (1.1)
1.0 (0.6)
NS
Probable cases (n = 11)
2.6 (0.6)
1.6 (0.5)
\0.01
Possible cases (n = 7)
1.4 (0.6)
0.8 (0.4)
NS
Possible cases (n = 6)
2.9 (0.7)
1.4 (1.0)
\0.05
MSA-C
MSA-P Probable cases (n = 13)
1.7 (1.6)
0.8 (1.1)
\0.01
Probable cases (n = 8) MSA-P
Possible cases (n = 8)
0.7 (1.4)
0.4 (0.9)
NS
Values are means (SD). The p values were obtained by Wilcoxon signed-rank test MSA-C multiple system atrophy of the cerebellar form, MSA-P multiple system atrophy of the parkinsonian form, T2*-WI T2*weighted image, T2-WI T2-weighted image, NS not significant
123
Putaminal hypointensity was assessed in patients with putaminal atrophy. Values are means (SD). The p values were obtained by Wilcoxon signed-rank test. Abbreviations are explained in the Table 2 footnote
J Neurol Table 4 Assessment of putaminal hyperintense rim in MSA-C and MSA-P patients T2*-WI
T2-WI
p value
0.4 (1.1)
0.4 (0.7)
NS
MSA-C Probable cases (n = 8) MSA-P Probable cases (n = 11)
1.3 (1.3)
1.5 (1.2)
NS
Possible cases (n = 7)
0.4 (0.5)
0.1 (0.4)
NS
Putaminal hyperintense rim was assessed in patients with putaminal atrophy. Values are means (SD). The p values were obtained by Wilcoxon signed-rank test. Abbreviations are explained in the Table 2 footnote
Three of the 21 patients with MSA-P (14 %) did not show putaminal atrophy on the first MRI performed within 2 years from the onset of clinical symptoms. Nevertheless, one of these three patients demonstrated an extremely clear HCB sign (T2*-WI, grade 4; T2-WI, grade 2.5) on the first MRI. The remaining two patients showed no apparent HCB sign on the first MRI, but a completed HCB sign with mild putaminal atrophy was observed on the follow-up MRI performed almost a full year later. Although a patient with cerebellar ataxia less than 1 year from the onset was categorized into the probable MSA-C group based on rigorously defined autonomic failure, he did not show pontine or putaminal signal changes on T2*- or T2-WI. This patient demonstrated a
distinct HCB sign without putaminal atrophy on T2*-WI re-examined at 19 months after the first MRI. Although the pons and cerebellum on T2-WI showed equivalent isointense signal changes at the first (Fig. 1a) and second MRI (Fig. 1b), the pons on T2*-WI showed a distinct hypointense signal change relative to the cerebellar hemisphere (Fig. 1c). The hypointense signal change extended into the cerebellar dentate nuclei on the second MRI (Fig. 1d).
Discussion The three principal findings of the present study were: (i) T2*-WI demonstrated an apparent HCB sign compared to T2-WI in patients with MSA-C, and the degree of HCB sign on T2*-WI in the possible MSA-C (early stage) patients was equivalent to that of the probable MSA-C (advanced stage) patients; (ii) visualization of the HCB sign in the patients with MSA-P was inferior compared to the visualization in the patients with MSA-C even with T2*-WI, but some patients showed evolution of the HCB sign preceding the appearance of the putaminal atrophy and signal change; and (iii) posterior putaminal hypointensity was not common except in the probable MSA-P patients, even on T2*-WI. Assessments of the HCB sign on T2*-WI have not been adequately conducted. In our study, the HCB sign on T2*WI was significantly more apparent than that on T2-WI in
Fig. 1 Initial MRI findings (a, c) and re-examined MRI findings (b, d) in a patient with MSA-C (72-year-old woman). a, b Axial T2-weighted images (T2-WI). c, d Axial T2*weighted images (T2*-WI). She was diagnosed as possible MSA-C at the initial MRI study, and then evolved into probable MSA-C at the re-examined MRI approx. 1 year later. The pontine basis behind the hotcross bun (HCB) sign showed hypointense signal change compared to the cerebellar hemisphere on T2*-WI. T2*-WI demonstrated a clear HCB sign that is superior to that on T2-WI at any stage
123
J Neurol
the MSA-C patients. This finding was confirmed not only in the probable cases (corresponding to the advanced stage), but also in the possible cases (corresponding to the early stage). In addition, the advantage of T2*-WI over T2WI was observed throughout the course of the same patients with MSA-C. Although the pathologic basis of the HCB is composed of gliosis as well as loss of neurons and myelinated fibers, the high signal intensity producing the HCB on T2- and T2*-WIs could primarily reflect the increased local water content associated with astrocytosis [18]. On the other hand, the background of the HCB on T2*-WI showed clearly low intensity relative to the cerebellum. Since elevated concentrations of ferritin-bound iron in the basis pontis have been demonstrated in MSA-P [8], the significant hypointensity in the pons but not in the cerebellum on T2*-WI might reflect the increased iron deposition in the pons at a density comparable to that of the observed changes in the striatum of MSA. Thus, enhancing the contrast between water and iron on T2*-WI could greatly contribute to the improved detection of the HCB sign. None of the 100 age-matched cases (69 ± 12 years, range 41–86 years) who underwent MRI examinations (1.5 T) for neurological symptoms such as headache in our institution showed the HCB sign against a background of clear low intensity on T2*-WIs (data not shown). This finding demonstrates that the HCB on T2*-WIs in our study is a real pathological sign, unlike a pseudo-hyperintense putaminal rim sign in normal subjects on a 3-T MRI device [19]. In our study, the HCB sign on T2*-WI in the patients with MSA-P was relatively mild compared to that in the MSA-C patients. However, it is interesting to see that three of the 21 MSA-P patients (14 %) demonstrated the HCB sign preceding the appearance of putaminal changes. This unexpected finding is consistent with a previous study using T2-WI that found that 2 of 33 patients with MSA-P (6 %) within the first 2 years of disease had only the HCB sign without putaminal hypointensity and/or putaminal hyperintense rim [20]. It was also shown that the HCB sign preceded the appearance of a putaminal slit in four of 19 patients with MSA-P (21 %) [9]. These findings might be explained by a previous report that parkinsonism in MSA is caused at a high rate even by mild pathological changes in the StrN region [7]. The sensitivity of MRI could be insufficient for detecting cases of mild StrN lesion. Taken together, these findings suggest that assessments of pontine lesions by T2*-WI may facilitate the clinical diagnosis in patients with MSA-P as well as those with MSA-C. The grading of the HCB sign on T2*-WIs in the possible MSA-C patients was equivalent to that in the probable MSA-C patients, whereas the difference in the grading of the HCB sign on T2*-WI between the possible and
123
probable MSA-P patients showed a trend toward significance. Since cerebellar signs in MSA do not appear without moderate to severe pathological changes in the OPC region [7], the level of the OPC lesion even in the stage of possible MSA-C might be sufficient to form the HCB sign on T2*-WI. On the other hand, cerebellar signs in patients initially classified as having MSA-P do not become dominant over parkinsonism during follow-up [10]. This finding suggests that the OPC lesion in MSA-P is slowly progressive and remains at a less severe level. Thus, the HCB sign in MSA-P could become clear in the course of progression from possible to probable MSA-P. In the present patients with probable MSA-P, T2*-WI demonstrated significantly lower putaminal signal intensities compared to those shown by T2-WI. This finding is consistent with previous reports that T2*-WI has a great advantage for the detection of putaminal iron deposition in MSA-P [4–6]. In patients with possible MSA-P, however, the change in the putaminal signal intensity remained relatively mild even with sensitive T2*-WI for iron deposition. In addition, putaminal hypointensity was reported to be uncommon in MSA-C patients in whom putaminal atrophy appears at the later stage of the illness [9, 10]. These findings suggest that the assessment of putaminal signal intensity on T2*-WI might be of little help in the early diagnosis of MSA-P and MSA-C. The putaminal signal loss combined with the presence of a putaminal hyperintense rim indicating gliosis suggests the diagnosis of MSA with high specificity [5]. However, the frequency of the coexistence of both findings in our study remained low (MSA-C, 13 %; MSA-P, 56 %). The relatively low detection rate might be influenced by the magnetic field strength of the MRI device (1.5 T) used in this study [21]. There are several limitations to this study. First, all of the patients in this study were clinically diagnosed based on the consensus criteria [1]. Except for the cases of two patients confirmed by autopsy, we could not completely exclude MSA-mimicking disorders such as progressive supranuclear palsy with cerebellar ataxia [22]. Second, the HCB sign on T2-WI can be seen in other disorders, such as spinocerebellar ataxias [23], parkinsonism due to vasculitis [24], variant Creutzfeldt–Jakob disease [25] and cerebellar ataxia in human immunodeficiency virus (HIV) infection [26]. Since these disorders are relatively rare, we have not assessed the difference in the detection of the HCB sign on T2*-WIs or the pathological basis between MSA and the other disorders. Third, 60 % of the MSA patients in this study had MSA-C, which is predominant in Japanese MSA patients [10]. Therefore, to establish the usefulness of the HCB sign in the diagnosis of MSA-P, further assessments to identify the frequency and the time of appearance of the HCB sign in MSA-P should be carried out with larger groups of patients with MSA-P.
J Neurol
In summary, our findings demonstrated that T2*-WI gave a clear picture of the HCB sign compared to conventional T2-WI. The HCB sign on T2*-WI was observed with clarity not only in probable MSA-C but also in possible MSA-C patients. In addition, some patients with MSA-P showed that the HCB sign preceded the appearance of putaminal signal changes on T2*-WI. The detection of putaminal hypointensity remained insufficient in MSA-C and possible MSA-P, even with sensitive T2*-WI for iron deposition. These findings suggest that the diagnosis of MSA might be supported and improved by the assessment of the HCB sign on T2*-WI. Although the correlations between T2*-WI and pathological changes remain unknown, the assessment of the HCB sign on T2*-WI might be a powerful tool for the diagnosis of MSA. Acknowledgments This work was partly supported by JSPS KAKENHI Grant No. 24591297.
10.
11.
12.
13.
14.
15.
Conflicts of interest On behalf of all authors, the corresponding author states that there is no conflict of interest.
16.
Ethical standard The study was approved by the appropriate ethics committee and has therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki.
17.
References 1. Gilman S, Wenning GK, Low PA, Brooks DJ, Mathias CJ, Trojanowski JQ et al (2008) Second consensus statement on the diagnosis of multiple system atrophy. Neurology 71:670–676 2. Chandran V, Stoessl AJ (2014) Imaging in multiple system atrophy. Neurol Clin Neurosci 2:178–187 3. Brooks DJ, Seppi K (2009) Proposed neuroimaging criteria for the diagnosis of multiple system atrophy. Mov Disord 24:949–964 4. Kraft E, Trenkwalder C, Auer DP (2002) T2*-weighted MRI differentiates multiple system atrophy from Parkinson’s disease. Neurology 59:1265–1267 5. von Lewinski F, Werner C, Jorn T, Mohr A, Sixel-Doring F, Trenkwalder C (2007) T2*-weighted MRI in diagnosis of multiple system atrophy. A practical approach for clinicians. J Neurol 254:1184–1188 6. Arabia G, Morelli M, Paglionico S, Novellino F, Salsone M, Giofre L et al (2010) An magnetic resonance imaging T2*weighted sequence at short echo time to detect putaminal hypointensity in Parkinsonisms. Mov Disord 25:2728–2734 7. Ozawa T, Paviour D, Quinn NP, Josephs KA, Sangha H, Kilford L et al (2004) The spectrum of pathological involvement of the striatonigral and olivopontocerebellar systems in multiple system atrophy: clinicopathological correlations. Brain 127:2657–2671 8. Visanji NP, Collingwood JF, Finnegan ME, Tandon A, House E, Hazrati LN (2013) Iron deficiency in parkinsonism: regionspecific iron dysregulation in Parkinson’s disease and multiple system atrophy. J Parkinsons Dis 3:523–537 9. Horimoto Y, Aiba I, Yasuda T, Ohkawa Y, Katayama T, Yokokawa Y et al (2002) Longitudinal MRI study of multiple system
18.
19.
20.
21.
22.
23.
24.
25. 26.
atrophy - when do the findings appear, and what is the course? J Neurol 249:847–854 Yabe I, Soma H, Takei A, Fujiki N, Yanagihara T, Sasaki H (2006) MSA-C is the predominant clinical phenotype of MSA in Japan: analysis of 142 patients with probable MSA. J Neurol Sci 249:115–121 Geser F, Seppi K, Stampfer-Kountchev M, Kollensperger M, Diem A, Ndayisaba JP et al (2005) The European Multiple System Atrophy-Study Group (EMSA-SG). J Neural Transm 112:1677–1686 Gilman S, May SJ, Shults CW, Tanner CM, Kukull W, Lee VM et al (2005) The North American Multiple System Atrophy Study Group. J Neural Transm 112:1687–1694 Lin DJ, Hermann KL, Schmahmann JD (2014) Multiple system atrophy of the cerebellar type: clinical state of the art. Mov Disord 29:294–304 Wenning GK, Tison F, Seppi K, Sampaio C, Diem A, Yekhlef F et al (2004) Development and validation of the Unified Multiple System Atrophy Rating Scale (UMSARS). Mov Disord 19:1391–1402 Ito S, Shirai W, Hattori T (2007) Evaluating posterolateral linearization of the putaminal margin with magnetic resonance imaging to diagnose the Parkinson variant of multiple system atrophy. Mov Disord 22:578–581 Tha KK, Terae S, Tsukahara A, Soma H, Morita R, Yabe I et al (2012) Hyperintense putaminal rim at 1.5 T: prevalence in normal subjects and distinguishing features from multiple system atrophy. BMC Neurol 12:39. doi:10.1186/1471-2377-12-39 Deguchi K, Sasaki I, Touge T, Tsukaguchi M, Takeuchi H, Kuriyama S (2004) Abnormal baroreceptor-mediated vasopressin release as possible marker in early diagnosis of multiple system atrophy. J Neurol Neurosurg Psychiatry 75:110–115 Takao M, Kadowaki T, Tomita Y, Yoshida Y, Mihara B (2007) ‘Hot-cross bun sign’ of multiple system atrophy. Intern Med 46:1883 Fujii S, Matsusue E, Kinoshita T, Sugihara S, Ohama E, Ogawa T (2007) Hyperintense putaminal rim at 3T reflects fewer ferritin deposits in the lateral marginal area of the putamen. AJNR 28:777–781 Ishigame K, Yagishita A (2010) Diagnostic imaging of multiple system atrophy. Neurol Med (Tokyo, Japan) 73:340–345 (in Japanese) Watanabe H, Ito M, Fukatsu H, Senda J, Atsuta N, Kaga T et al (2010) Putaminal magnetic resonance imaging features at various magnetic field strengths in multiple system atrophy. Mov Disord 25:1916–1923 Kanazawa M, Shimohata T, Toyoshima Y, Tada M, Kakita A, Morita T et al (2009) Cerebellar involvement in progressive supranuclear palsy: a clinicopathological study. Mov Disord 24:1312–1318 Lee YC, Liu CS, Wu HM, Wang PS, Chang MH, Soong BW (2009) The ‘hot cross bun’ sign in the patients with spinocerebellar ataxia. Eur J Neurol 16:513–516 Muqit MM, Mort D, Miskiel KA, Shakir RA (2001) ‘‘Hot cross bun’’ sign in a patient with parkinsonism secondary to presumed vasculitis. J Neurol Neurosurg Psychiatry 71:565–566 Soares-Fernandes JP, Ribeiro M, Machado A (2009) ‘‘Hot cross bun’’ sign in variant Creutzfeldt-Jakob disease. AJNR 30:E37 Oji Y, Hatano T, S-i Kubo, Mori A, Nakazato T, Hattori N (2013) Cerebellar ataxia and the ‘‘hot cross bun’’ sign in association with human immunodeficiency virus infection. Neurol Clin Neurosci 1:127
123
