Clinical Neurology and Neurosurgery 118 (2014) 32–36
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Cerebellar dentate nucleus in progressive supranuclear palsy Nobuhiro Sawa a , Hiroshi Kataoka a,∗ , Takao Kiriyama a , Tesseki Izumi a , Toshiaki Taoka b , Kimihiko Kichikawa b , Satoshi Ueno a a b
Department of Neurology, Nara Medical University, Kashihara, Nara, Japan Department of Radiology, Nara Medical University, Kashihara, Nara, Japan
a r t i c l e
i n f o
Article history: Received 14 December 2012 Received in revised form 19 September 2013 Accepted 20 December 2013 Available online 4 January 2014 Keywords: Progressive supranuclear palsy Cerebellar dentate nucleus MRI Ataxia Parkinsonism
a b s t r a c t Objectives: Some patients with progressive supranuclear palsy (PSP) present with cerebellar dysfunction. Severe degeneration of the cerebellar dentate nucleus (CDN) was evident in these patients. We evaluated signal intensity on MRI in the CDN of PSP patients with or without cerebellar ataxia. Patients and methods: We reviewed the clinical histories and brain MRI studies of 28 patients with clinically probable PSP. Three disease control groups were studied: a group of 28 sex- and age-matched patients with Parkinson’s disease (PD), a group of 15 patients with multiple system atrophy with predominant parkinsonian features (MSA-P), and 15 control subjects. Turbo spin-echo sequences for T2-weighted images were used using a 1.5 T magnetic resonance imager. Results: Eight patients with PSP (28%) and one patient with MSA-P (6%) had heterogeneous regions in the CDN. This finding was not evident in the patients with PD or controls. Three out of four PSP patients with cerebellar ataxia had heterogeneous regions in the CDN and other one patient with cerebellar ataxia as the initial and principal symptoms had no heterogeneous regions in the CDN. Conclusion: Heterogeneous regions in the CDN on MRI do not always reflect cerebellar ataxia in PSP patients, and this finding might be an additional marker to support a probable diagnosis of PSP. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Progressive supranuclear palsy (PSP) is characterized by postural instability and supranuclear gaze palsy [1]. Postural instability has been suggested to be responsible for grumose degeneration in the cerebellar dentate nucleus (CDN) [2,3]. This pathological change is characterized by eosinophilic, granular, and amorphous material around dentate neurons in the CDN in PSP patients and is associated with myelinated fiber loss, tau pathology, and microgliosis in the dentatorubrothalamic tract [2,3]. A recent study reported evidence of such degeneration in the CDN of PSP patients, some of whom had cerebellar dysfunction as an initial principal symptom [4]. We recently reported that patients with clinically probable PSP showed increased signals in the superior cerebellar peduncle on conventional magnetic resonance imaging (MRI) [5]. Such signals were absent in patients with Parkinson’s disease (PD) or multiple system atrophy with predominant parkinsonian features (MSA-P) [5]. We hypothesized that pathological degeneration in the CDN might contribute to increased signal intensity on MRI in PSP patients. When
∗ Corresponding author at: Department of Neurology, Nara Medical University, 840 Shijo-cho, Kashihara, Nara 634-8522, Japan. Tel.: +81 744 29 8860; fax: +81 744 24 6065. E-mail address: [email protected] (H. Kataoka). 0303-8467/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.clineuro.2013.12.011
increased MRI signal intensity is evident in the CDN, it remains unclear whether this finding is useful for differential diagnosis from other parkinsonian syndromes or whether it is specific to PSP with cerebellar ataxia. We thus evaluated signal intensity on MRI in the CDN of PSP patients with or without cerebellar ataxia. 2. Materials and methods We reviewed the clinical histories and brain MRI studies of 28 patients with clinically probable PSP (15 men, 13 women; age range 56–79 years, mean 70.5 ± 4.9 years) according to the National Institute of Neurologic Disorders and Stroke-SPSP diagnostic criteria [2]. Three disease control groups were studied: a group of 28 sex- and age-matched patients with PD, a group of 15 patients who had MSAP (age range 57–83 years, mean 68.6 ± 7.0 years), and 15 control subjects (age range 58–78 years, mean 69.4 ± 5.0 years). Clinically definite PD was diagnosed according to both the Calne and Gelb criteria [6,7]. Clinically possible MSA-P was diagnosed according to recent diagnostic criteria [8]. All patients with MSA-P had orthostatic hypotension, defined as a drop in systolic blood pressure of more than 20 mmHg or a drop in diastolic blood pressure of more than 10 mmHg on standing up. The heart-mediastinum 123 Imetaiodobenzylguanidine uptake ratio was significantly decreased in all patients with PD, but not decreased in all patients with MSA-P [9]. The control subjects had no history of central nervous system
N. Sawa et al. / Clinical Neurology and Neurosurgery 118 (2014) 32–36
33
Fig. 1. All evaluated axial T2-weighted magnetic resonance images (left panel) were obtained at the level of the cerebellar dentate nucleus on coronal (middle panel) or sagittal magnetic resonance images (right panel) (white arrows). This patient showed heterogeneous regions in the cerebellar dentate nucleus.
disorders, abnormal neurological signs, or abnormal cranial MRI findings. We excluded patients who had any of the following conditions: another atypical parkinsonian syndrome such as normal pressure hydrocephalus and large vessel disease, infarction, multiple lacunar infarctions, or tumor on cranial MRI. Patients with an infarct in the cerebellar lobes or with white matter lesions of grade 3 according to Fazekas scale [10] on cranial MRI were also excluded. There was no different in age and sex between patients with PSP and patients with MSA-P (P = 0.307 and 0.343, respectively) or controls (P = 0.492 and 1.0, respectively). Cerebellar ataxia was defined as the presence of gait ataxia plus at least either of cerebellar dysarthria, limb ataxia or sustained gaze-evoked nystamgus [8]. Neurological findings were assessed by at least two experienced neurologists.
3.1. MRI evaluation Patients were examined with a 1.5-T magnetic resonance imager. Because two patients with PSP and one patient with MSA-P were studied retrospectively, various techniques were used for imaging, but the scan slice thickness and interslice gap were similar in all patients. These patients did not show signal changes in the CDN as described as below, and the two patients with PSP had no cerebellar ataxia. In the other 26 patients with PSP and 14 patients with MSA-P and in all patients with PD and all controls, conventional imaging techniques were used as follows: turbo spinecho sequences for T2-weighted images (T2WI) (TR 4400 ms; TE 110 ms; 5-mm slice thickness, with a 1-mm interslice gap), FLAIR images (TR 9000 ms; TE 120 ms; TI 2200 ms; 5-mm slice thickness, with a 1-mm interslice gap), and T1-weighted images (T1WI) (TR 450 ms; TE 9 ms; 5-mm slice thickness, with a 1-mm interslice gap) using a 1.5 T magnetic resonance imager. Signal intensity changes on these axial MRI scans were assessed by visual inspection by two experienced reviewers who were blinded to the clinical data and diagnosis, and agreement was reached by consensus. These axial images were obtained at the level of CDN as shown in Fig. 1. Heterogeneous regions in the CDN were defined as some apparent high-signal intensities mixed on a background of low-signal intensities in the CDN on T2WI, as shown in Fig. 2. These MRI images were analyzed in a blinded, randomized fashion to control for bias. When the reviewers unanimously agreed that there were heterogeneous regions in the CDN, the reviewers were also requested to indicate whether the lesions were apparent by marking a position
Fig. 2. Schema of the heterogeneous regions in the cerebellar dentate nucleus on a T2-weighted magnetic resonance image. Some high-intensity regions were imposed against a background of low intensity in the cerebellar dentate nucleus.
along a continuous 50-mm horizontal visual analog scale line (left [0 mm], absolutely not detectable; right [50 mm], absolutely apparent) as described previously [11,12]. A detailed description of the bar chart is provided in the supplemental material. Increased signals of the superior cerebellar peduncle on FLAIR were defined as previously [5]. In patients with PSP, midbrain atrophy on mid-sagittal and axial T1-weighted images was assessed by an experienced neuro-radiologist. The interval between the onset of neurological symptoms and MRI examinations in PSP was 41.4 ± 32.7 months. For statistical analysis, differences were examined with the use of unpaired t-tests or Fisher’s exact probability tests for categorical data (SPSS software, version 18). P values of less than 0.05 were considered to indicate statistical significance. 3. Results Eight patients with PSP (28%) and one patient with MSA-P (6%) had heterogeneous regions in the CDN. This finding was not evident in the patients with PD or controls (Fig. 3). The sensitivity and specificity of heterogeneous regions in the CDN for PSP were 28% and 98%, respectively. There was no difference in the duration of disease between PSP patients with and those without heterogeneous regions in the CDN (48.0 ± 29.8 vs. 40.1 ± 34.5 months, P = 0.579). The heterogeneous regions in the CDN showed the same findings on FLAIR and on T2WI, and TIWI showed iso-signal intensity.
34
N. Sawa et al. / Clinical Neurology and Neurosurgery 118 (2014) 32–36
Fig. 3. The heterogeneous regions in the cerebellar dentate nucleus on T2-weighted magnetic resonance images were evident in patients with progressive supranuclear palsy (white arrows) (panel A). These regions were absent in multiple system atrophy with predominant parkinsonian features (panel B), Parkinson’s disease (panel C), and control subjects (panel D).
Four patients with PSP had cerebellar ataxia. All of these patients presented with gait ataxia, cerebellar dysarthria, and limb ataxia, and one patient had cerebellar oculomotor dysfunction (Table 1). One patient (Patient 1) had cerebellar ataxia as the initial and principal symptoms and received an initial diagnosis of spinocerebellar
degeneration, but cerebellar ataxia was not prominent. In three patients (Patient 1, 2 and 3), the results of genetic analysis for DRPLA and SCA-1, 2, and 3 were negative. Cerebellar ataxia in the four patients had persisted for 13–24 months (Table 1). When the cerebellar ataxia disappeared, all patients had typical PSP features. All
Table 1 Clinical features in patients who had progressive supranuclear palsy with cerebellar ataxia. Hasegawa dementia scale (HDS-R) [21], based on nine questions including tasks such as stating one’s age, orientation with respect to date and place, memorizing three words, calculation (subtraction), reciting figures in reverse order, repeating the last three words of a sequence, and memorizing five articles and the names of 10 vegetables. A score of ≥24 was considered normal.
Age/sex Initial symptom Duration from onset of disease to initial examinations PSP diagnosis other than cerebellar ataxiaa Cerebellar ataxia
Magnitude of cerebellar ataxia MMSE Genetic analysis for DRPLA, SCA1, 3, and 6 Duration from onset of disease to MRI evaluation Hummingbird sign Heterogeneous regions in the cerebellar dentate nucleus Increased signals in the superior cerebellar peduncle Cerebellar atrophy Duration of follow-up Cerebellar ataxia (duration) Cerebellar atrophy on MRI PSP diagnosis
Patient 1
Patient 2
Patient 3
Patient 4
64/F Dizziness 29 mo
71/M Gait difficulty 77 mo
71/M Gait difficulty 89 mo
76/F Speech disorder 61 mo
Not diagnosed
Probable
Probable
Probable
Dys. and decom. in limbs, dysarthria, saccadic eye movements, nystamgus, cerebellar gait Moderate 22b Negative
Dys. and decom. in limbs, dysarthria, explosive speech, cerebellar gait
Dys. and decom. in limbs, dysarthria, explosive speech, cerebellar gait
Dys. and decom. in limbs, dysarthria, cerebellar gait
Mild 24 Negative
Mild 29 Negative
Mild 25 N.D.
34 mo
78 mo
89 mo
65 mo
+ −
+ +
+ +
+ +
+
+
+
−
− 9 yr Disappeared (24 mo) Not appeared Probable
− 5 yr Disappeared (13 mo) Not appeared Probable
− 3 yr Disappeared (20 mo) Not appeared Probable
− 2 yr Disappeared (14 mo) Not appeared Probable
PSP, progressive supranuclear palsy; MMSE, Mini-Mental State Examination; SCA, spinocerebellar ataxia; DRPLA, dentatorubral-pallidoluysian atrophy; MRI, magnetic resonance image; dys.: dysmetria; decom., decomposition; N.D.: not done; mo, months; yr, years. a According to the National Institute of Neurologic Disorders and Stroke-SPSP diagnostic criteria [2]. b Hasegawa dementia scale (HDS-R) [21].
N. Sawa et al. / Clinical Neurology and Neurosurgery 118 (2014) 32–36
35
Table 2 Characteristic findings of magnetic resonance imaging in patients with progressive supranuclear palsy with/without cerebellar ataxia. PSP
With cerebellar ataxia (n = 4)
Without cerebellar ataxia (n = 24)
Age Sex (male/female) Duration from onset of disease to MRI evaluation
70.5 (65.7) 2/2 66.5 (23.7)
71.2 (5.8) 10/14 37.4 (32.5)
MRI findings Hummingbird sign Heterogeneous regions in the cerebellar dentate nucleus Increased signals in the superior cerebellar peduncle*
4 3 3
19 5 1**
PSP, progressive supranuclear palsy; MRI, magnetic resonance image; n, number. * Significant difference between two groups. ** Available for 19 patients.
patients had supranuclear vertical gaze, postural instability, frequent falls, and asymmetric rigidity, with no tremor or asymmetric onset. During follow-up, all patients showed cognitive decline and dysphagia and did not respond to levodopa. Brain MRI in all patients showed atrophy in the midbrain, with no evidence of atrophy in the cerebellum on the first or follow-up MRI. These four patients were thus given diagnoses of PSP according to the National Institute of Neurologic Disorders and Stroke-SPSP diagnostic criteria [2]. Heterogeneous regions in the CDN were evident on MRI in three patients (Patient 2, 3 and 4). Midbrain atrophy was evident in all patients, and three patients (Patient 1, 2 and 3) showed increased signals of the superior cerebellar peduncle. None of the patients with cerebellar ataxia had vascular lesions, and the grade of white matter lesions was 1 in all patients according to Fuzakau scale [10]. There was no significant difference in age (70.5 ± 4.9 vs. 70.5 ± 5.0 years, P = 0.976) and sex (P = 1.0) between PSP patients with and those without cerebellar ataxia, as shown in Table 2. There was also no difference in the duration of disease between these two groups (66.5 ± 23.7 vs. 38.3 ± 32.9 months, P = 0.116). The heterogeneous regions in the CDN (P = 0.058) and increased signals of the superior cerebellar peduncle (P = 0.009) were more frequent in PSP patients with cerebellar ataxia than in those without cerebellar ataxia. Five patients without cerebellar ataxia had heterogeneous regions in the CDN, but the mean distance between the checked point and the left margin of the analog scale was significantly shorter in these patients than in those with cerebellar ataxia (16.7 ± 6.7 mm vs. 33.1 ± 6.3 mm, P < 0.001). Midbrain atrophy was evident in 23 of 28 patients with PSP, and four of 25 patients with PSP had increased signals of the superior cerebellar peduncle. At the initial neurological evaluations, cognitive decline with a score of
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Cerebellar dentate nucleus in progressive supranuclear palsy Nobuhiro Sawa a , Hiroshi Kataoka a,∗ , Takao Kiriyama a , Tesseki Izumi a , Toshiaki Taoka b , Kimihiko Kichikawa b , Satoshi Ueno a a b
Department of Neurology, Nara Medical University, Kashihara, Nara, Japan Department of Radiology, Nara Medical University, Kashihara, Nara, Japan
a r t i c l e
i n f o
Article history: Received 14 December 2012 Received in revised form 19 September 2013 Accepted 20 December 2013 Available online 4 January 2014 Keywords: Progressive supranuclear palsy Cerebellar dentate nucleus MRI Ataxia Parkinsonism
a b s t r a c t Objectives: Some patients with progressive supranuclear palsy (PSP) present with cerebellar dysfunction. Severe degeneration of the cerebellar dentate nucleus (CDN) was evident in these patients. We evaluated signal intensity on MRI in the CDN of PSP patients with or without cerebellar ataxia. Patients and methods: We reviewed the clinical histories and brain MRI studies of 28 patients with clinically probable PSP. Three disease control groups were studied: a group of 28 sex- and age-matched patients with Parkinson’s disease (PD), a group of 15 patients with multiple system atrophy with predominant parkinsonian features (MSA-P), and 15 control subjects. Turbo spin-echo sequences for T2-weighted images were used using a 1.5 T magnetic resonance imager. Results: Eight patients with PSP (28%) and one patient with MSA-P (6%) had heterogeneous regions in the CDN. This finding was not evident in the patients with PD or controls. Three out of four PSP patients with cerebellar ataxia had heterogeneous regions in the CDN and other one patient with cerebellar ataxia as the initial and principal symptoms had no heterogeneous regions in the CDN. Conclusion: Heterogeneous regions in the CDN on MRI do not always reflect cerebellar ataxia in PSP patients, and this finding might be an additional marker to support a probable diagnosis of PSP. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Progressive supranuclear palsy (PSP) is characterized by postural instability and supranuclear gaze palsy [1]. Postural instability has been suggested to be responsible for grumose degeneration in the cerebellar dentate nucleus (CDN) [2,3]. This pathological change is characterized by eosinophilic, granular, and amorphous material around dentate neurons in the CDN in PSP patients and is associated with myelinated fiber loss, tau pathology, and microgliosis in the dentatorubrothalamic tract [2,3]. A recent study reported evidence of such degeneration in the CDN of PSP patients, some of whom had cerebellar dysfunction as an initial principal symptom [4]. We recently reported that patients with clinically probable PSP showed increased signals in the superior cerebellar peduncle on conventional magnetic resonance imaging (MRI) [5]. Such signals were absent in patients with Parkinson’s disease (PD) or multiple system atrophy with predominant parkinsonian features (MSA-P) [5]. We hypothesized that pathological degeneration in the CDN might contribute to increased signal intensity on MRI in PSP patients. When
∗ Corresponding author at: Department of Neurology, Nara Medical University, 840 Shijo-cho, Kashihara, Nara 634-8522, Japan. Tel.: +81 744 29 8860; fax: +81 744 24 6065. E-mail address: [email protected] (H. Kataoka). 0303-8467/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.clineuro.2013.12.011
increased MRI signal intensity is evident in the CDN, it remains unclear whether this finding is useful for differential diagnosis from other parkinsonian syndromes or whether it is specific to PSP with cerebellar ataxia. We thus evaluated signal intensity on MRI in the CDN of PSP patients with or without cerebellar ataxia. 2. Materials and methods We reviewed the clinical histories and brain MRI studies of 28 patients with clinically probable PSP (15 men, 13 women; age range 56–79 years, mean 70.5 ± 4.9 years) according to the National Institute of Neurologic Disorders and Stroke-SPSP diagnostic criteria [2]. Three disease control groups were studied: a group of 28 sex- and age-matched patients with PD, a group of 15 patients who had MSAP (age range 57–83 years, mean 68.6 ± 7.0 years), and 15 control subjects (age range 58–78 years, mean 69.4 ± 5.0 years). Clinically definite PD was diagnosed according to both the Calne and Gelb criteria [6,7]. Clinically possible MSA-P was diagnosed according to recent diagnostic criteria [8]. All patients with MSA-P had orthostatic hypotension, defined as a drop in systolic blood pressure of more than 20 mmHg or a drop in diastolic blood pressure of more than 10 mmHg on standing up. The heart-mediastinum 123 Imetaiodobenzylguanidine uptake ratio was significantly decreased in all patients with PD, but not decreased in all patients with MSA-P [9]. The control subjects had no history of central nervous system
N. Sawa et al. / Clinical Neurology and Neurosurgery 118 (2014) 32–36
33
Fig. 1. All evaluated axial T2-weighted magnetic resonance images (left panel) were obtained at the level of the cerebellar dentate nucleus on coronal (middle panel) or sagittal magnetic resonance images (right panel) (white arrows). This patient showed heterogeneous regions in the cerebellar dentate nucleus.
disorders, abnormal neurological signs, or abnormal cranial MRI findings. We excluded patients who had any of the following conditions: another atypical parkinsonian syndrome such as normal pressure hydrocephalus and large vessel disease, infarction, multiple lacunar infarctions, or tumor on cranial MRI. Patients with an infarct in the cerebellar lobes or with white matter lesions of grade 3 according to Fazekas scale [10] on cranial MRI were also excluded. There was no different in age and sex between patients with PSP and patients with MSA-P (P = 0.307 and 0.343, respectively) or controls (P = 0.492 and 1.0, respectively). Cerebellar ataxia was defined as the presence of gait ataxia plus at least either of cerebellar dysarthria, limb ataxia or sustained gaze-evoked nystamgus [8]. Neurological findings were assessed by at least two experienced neurologists.
3.1. MRI evaluation Patients were examined with a 1.5-T magnetic resonance imager. Because two patients with PSP and one patient with MSA-P were studied retrospectively, various techniques were used for imaging, but the scan slice thickness and interslice gap were similar in all patients. These patients did not show signal changes in the CDN as described as below, and the two patients with PSP had no cerebellar ataxia. In the other 26 patients with PSP and 14 patients with MSA-P and in all patients with PD and all controls, conventional imaging techniques were used as follows: turbo spinecho sequences for T2-weighted images (T2WI) (TR 4400 ms; TE 110 ms; 5-mm slice thickness, with a 1-mm interslice gap), FLAIR images (TR 9000 ms; TE 120 ms; TI 2200 ms; 5-mm slice thickness, with a 1-mm interslice gap), and T1-weighted images (T1WI) (TR 450 ms; TE 9 ms; 5-mm slice thickness, with a 1-mm interslice gap) using a 1.5 T magnetic resonance imager. Signal intensity changes on these axial MRI scans were assessed by visual inspection by two experienced reviewers who were blinded to the clinical data and diagnosis, and agreement was reached by consensus. These axial images were obtained at the level of CDN as shown in Fig. 1. Heterogeneous regions in the CDN were defined as some apparent high-signal intensities mixed on a background of low-signal intensities in the CDN on T2WI, as shown in Fig. 2. These MRI images were analyzed in a blinded, randomized fashion to control for bias. When the reviewers unanimously agreed that there were heterogeneous regions in the CDN, the reviewers were also requested to indicate whether the lesions were apparent by marking a position
Fig. 2. Schema of the heterogeneous regions in the cerebellar dentate nucleus on a T2-weighted magnetic resonance image. Some high-intensity regions were imposed against a background of low intensity in the cerebellar dentate nucleus.
along a continuous 50-mm horizontal visual analog scale line (left [0 mm], absolutely not detectable; right [50 mm], absolutely apparent) as described previously [11,12]. A detailed description of the bar chart is provided in the supplemental material. Increased signals of the superior cerebellar peduncle on FLAIR were defined as previously [5]. In patients with PSP, midbrain atrophy on mid-sagittal and axial T1-weighted images was assessed by an experienced neuro-radiologist. The interval between the onset of neurological symptoms and MRI examinations in PSP was 41.4 ± 32.7 months. For statistical analysis, differences were examined with the use of unpaired t-tests or Fisher’s exact probability tests for categorical data (SPSS software, version 18). P values of less than 0.05 were considered to indicate statistical significance. 3. Results Eight patients with PSP (28%) and one patient with MSA-P (6%) had heterogeneous regions in the CDN. This finding was not evident in the patients with PD or controls (Fig. 3). The sensitivity and specificity of heterogeneous regions in the CDN for PSP were 28% and 98%, respectively. There was no difference in the duration of disease between PSP patients with and those without heterogeneous regions in the CDN (48.0 ± 29.8 vs. 40.1 ± 34.5 months, P = 0.579). The heterogeneous regions in the CDN showed the same findings on FLAIR and on T2WI, and TIWI showed iso-signal intensity.
34
N. Sawa et al. / Clinical Neurology and Neurosurgery 118 (2014) 32–36
Fig. 3. The heterogeneous regions in the cerebellar dentate nucleus on T2-weighted magnetic resonance images were evident in patients with progressive supranuclear palsy (white arrows) (panel A). These regions were absent in multiple system atrophy with predominant parkinsonian features (panel B), Parkinson’s disease (panel C), and control subjects (panel D).
Four patients with PSP had cerebellar ataxia. All of these patients presented with gait ataxia, cerebellar dysarthria, and limb ataxia, and one patient had cerebellar oculomotor dysfunction (Table 1). One patient (Patient 1) had cerebellar ataxia as the initial and principal symptoms and received an initial diagnosis of spinocerebellar
degeneration, but cerebellar ataxia was not prominent. In three patients (Patient 1, 2 and 3), the results of genetic analysis for DRPLA and SCA-1, 2, and 3 were negative. Cerebellar ataxia in the four patients had persisted for 13–24 months (Table 1). When the cerebellar ataxia disappeared, all patients had typical PSP features. All
Table 1 Clinical features in patients who had progressive supranuclear palsy with cerebellar ataxia. Hasegawa dementia scale (HDS-R) [21], based on nine questions including tasks such as stating one’s age, orientation with respect to date and place, memorizing three words, calculation (subtraction), reciting figures in reverse order, repeating the last three words of a sequence, and memorizing five articles and the names of 10 vegetables. A score of ≥24 was considered normal.
Age/sex Initial symptom Duration from onset of disease to initial examinations PSP diagnosis other than cerebellar ataxiaa Cerebellar ataxia
Magnitude of cerebellar ataxia MMSE Genetic analysis for DRPLA, SCA1, 3, and 6 Duration from onset of disease to MRI evaluation Hummingbird sign Heterogeneous regions in the cerebellar dentate nucleus Increased signals in the superior cerebellar peduncle Cerebellar atrophy Duration of follow-up Cerebellar ataxia (duration) Cerebellar atrophy on MRI PSP diagnosis
Patient 1
Patient 2
Patient 3
Patient 4
64/F Dizziness 29 mo
71/M Gait difficulty 77 mo
71/M Gait difficulty 89 mo
76/F Speech disorder 61 mo
Not diagnosed
Probable
Probable
Probable
Dys. and decom. in limbs, dysarthria, saccadic eye movements, nystamgus, cerebellar gait Moderate 22b Negative
Dys. and decom. in limbs, dysarthria, explosive speech, cerebellar gait
Dys. and decom. in limbs, dysarthria, explosive speech, cerebellar gait
Dys. and decom. in limbs, dysarthria, cerebellar gait
Mild 24 Negative
Mild 29 Negative
Mild 25 N.D.
34 mo
78 mo
89 mo
65 mo
+ −
+ +
+ +
+ +
+
+
+
−
− 9 yr Disappeared (24 mo) Not appeared Probable
− 5 yr Disappeared (13 mo) Not appeared Probable
− 3 yr Disappeared (20 mo) Not appeared Probable
− 2 yr Disappeared (14 mo) Not appeared Probable
PSP, progressive supranuclear palsy; MMSE, Mini-Mental State Examination; SCA, spinocerebellar ataxia; DRPLA, dentatorubral-pallidoluysian atrophy; MRI, magnetic resonance image; dys.: dysmetria; decom., decomposition; N.D.: not done; mo, months; yr, years. a According to the National Institute of Neurologic Disorders and Stroke-SPSP diagnostic criteria [2]. b Hasegawa dementia scale (HDS-R) [21].
N. Sawa et al. / Clinical Neurology and Neurosurgery 118 (2014) 32–36
35
Table 2 Characteristic findings of magnetic resonance imaging in patients with progressive supranuclear palsy with/without cerebellar ataxia. PSP
With cerebellar ataxia (n = 4)
Without cerebellar ataxia (n = 24)
Age Sex (male/female) Duration from onset of disease to MRI evaluation
70.5 (65.7) 2/2 66.5 (23.7)
71.2 (5.8) 10/14 37.4 (32.5)
MRI findings Hummingbird sign Heterogeneous regions in the cerebellar dentate nucleus Increased signals in the superior cerebellar peduncle*
4 3 3
19 5 1**
PSP, progressive supranuclear palsy; MRI, magnetic resonance image; n, number. * Significant difference between two groups. ** Available for 19 patients.
patients had supranuclear vertical gaze, postural instability, frequent falls, and asymmetric rigidity, with no tremor or asymmetric onset. During follow-up, all patients showed cognitive decline and dysphagia and did not respond to levodopa. Brain MRI in all patients showed atrophy in the midbrain, with no evidence of atrophy in the cerebellum on the first or follow-up MRI. These four patients were thus given diagnoses of PSP according to the National Institute of Neurologic Disorders and Stroke-SPSP diagnostic criteria [2]. Heterogeneous regions in the CDN were evident on MRI in three patients (Patient 2, 3 and 4). Midbrain atrophy was evident in all patients, and three patients (Patient 1, 2 and 3) showed increased signals of the superior cerebellar peduncle. None of the patients with cerebellar ataxia had vascular lesions, and the grade of white matter lesions was 1 in all patients according to Fuzakau scale [10]. There was no significant difference in age (70.5 ± 4.9 vs. 70.5 ± 5.0 years, P = 0.976) and sex (P = 1.0) between PSP patients with and those without cerebellar ataxia, as shown in Table 2. There was also no difference in the duration of disease between these two groups (66.5 ± 23.7 vs. 38.3 ± 32.9 months, P = 0.116). The heterogeneous regions in the CDN (P = 0.058) and increased signals of the superior cerebellar peduncle (P = 0.009) were more frequent in PSP patients with cerebellar ataxia than in those without cerebellar ataxia. Five patients without cerebellar ataxia had heterogeneous regions in the CDN, but the mean distance between the checked point and the left margin of the analog scale was significantly shorter in these patients than in those with cerebellar ataxia (16.7 ± 6.7 mm vs. 33.1 ± 6.3 mm, P < 0.001). Midbrain atrophy was evident in 23 of 28 patients with PSP, and four of 25 patients with PSP had increased signals of the superior cerebellar peduncle. At the initial neurological evaluations, cognitive decline with a score of
