Abstract
INTRODUCTION
Human prion diseases are rare transmissible diseases caused by the accumulation of pathologic prion protein (PrPSc) in the central nervous system. About 85% of Creutzfeldt-Jakob disease cases are sporadic (sCJD) [1, 2]. The polymorphism at codon 129 of the prion protein gene (PRNP) and the PrPSc types 1 and 2 belong to a molecular classification of sCJD that correlates well with the clinical and neuropathological phenotype of sCJD and thus replaces previous attempts at classification of CJD [3]. M129 V polymorphism of the PRNP and PrPSc type influence different disease features such as disease duration, clinical course, age at onset, and neuropathological findings. However, only few studies have investigated neuropsychological symptoms in CJD. Data on PrPSc type or M129 V polymorphism are mostly missing, and the number of patients investigated is often rather limited to case series [4, 5] or single case reports [6–9]. Therefore, the aim of our study was to perform a detailed analysis of neuropsychological symptoms in a large group of patients with definite sCJD stratified by M129 V genotype and PrPSc type.
Evaluation of neuropsychological symptoms in CJD is very difficult. In particular, the classical sCJD subtype is rapidly progressive, and most patients are physically impaired to undergo complex, sometimes stressful, and protracted testing. Communication with sCJD patients is often difficult or even impossible due to early aphasia and prominent cognitive deficits. The quality and amount of data is therefore limited. In the present study, we were able to analyze data collected using the Mini-Mental State Examination (MMSE) [10], which CJD Surveillance Unit physicians performed routinely in CJD patients if it was still possible at the time of examination. Further neuropsychological tests are not routinely recommended and were performed on the basis of the clinical presentation. Therefore, data on tests other than MMSE in our CJD patients are very inconsistent and limited. We think that data on neuropsychological symptoms in sCJD may provide important information for early CJD diagnosis and patient management.
PATIENTS AND METHODS
Study design
The present study is a retrospective study. Patients with suspected CJD were directed to the CJD Surveillance Unit (Göttingen, Germany) and examined by study physicians on site. All patients included in the study were evaluated in person by study physicians. Additionally, medical records were also evaluated in all patients.
Cerebrospinal fluid (CSF), blood samples, and copies of the important diagnostic tests such as electroencephalography (EEG), magnetic resonance imaging (MRI), and laboratory tests were gathered. After informed consent was obtained, detailed questionnaires were filled out together with the relatives of the patient. The patients were classified according to established diagnostic criteria [11–13]. Neuropsychological symptoms were defined according to Hartje and Poeck [14] and were collected by direct examination by study physicians or extracted from medical documentation. A neuropsychological symptom was considered to be present if it was found by the study physicians during examination or clearly described or mentioned in the medical documentation. MMSE was routinely performed, when possible, by study physicians to obtain data. 23 other tests were performed in 61 patients; none of these single tests were done in more than three patients.
The study was approved by the Ethics Committee of the Medical Faculty of the University of Göttingen.
MRI and EEG findings
The MRI images were analyzed with respect to the hyperintensities of the basal ganglia. The basal ganglia were considered hyperintense when a signal increase was found in the putamen, the caudate, or the globus pallidus in relation to normointense brain tissue. Since most of the patients received T2- or FLAIR-weighted MRI images only, the frequency of cortical changes was not assessed due to the low sensitivity of T2 and FLAIR to these abnormalities. The EEGs were analyzed according to establishedcriteria [15].
Neuropathological and molecular studies
Western blot analysis, immunohistochemistry, and analysis of the PRNP were performed by standard methods [16–18]. Only CJD patients without PRNP mutations, with known PrPSc type, and no evidence of iatrogenic CJD recruited from January 1993 to December 2004 were included in this study. Patients with mixed PrPSc types were excluded.
CSF analysis
The 14-3-3 protein analysis was performed at least twice in each CSF sample as reported previously [19]. Tau-protein was measured by INNOTEST hTau ELISA, and Aβ-peptide 1-42 by INNOTEST β-Amyloid 1-42 ELISA (INNOGENETICS N.V., Ghent, Belgium). NSE and S-100B were quantified by a commercially available immunoluminometric assay (LIAISON NSE and LIAISON Sangtec 100, DiaSorin S.p.A. Saluggia, Italy). Phospho-Tau was measured according to the manufacturer’s instructions (Innotest Phospho-Tau (181P), Innogenetics, Heiden-Westfalen, Germany).
Statistical analysis
Significances (p) were tested by the SIGMASTAT 3.1 software (Systat Software Inc., Point Richmond, USA) using Student t-test/Mann-Whitney Rank Sum test/Kruskal-Wallis One Way Analysis of Variance on Ranks or Chi-square test/Fisher Exact test. Correlations (r) were tested by the SIGMASTAT 3.1 software using Pearson test. A p-value < 0.05 was considered as statistically significant.
RESULTS
Study collective
All consecutive 248 patients with definite sCJD, known M129 V polymorphism, and PrPSc type registered in Göttingen from January 1993 to December 2003 were included in this study. There were 140 female and 108 male patients (ratio = 1.3:1). Sex distribution was significantly different between sCJD subtypes. There were significantly more male patients in VV1 subtype compared to other subtypes (p = 0.004 in MM1, 0.013 in MV2, 0.0026 in VV2 patients). Data on sex, age, and disease duration stratified by molecular subtype are shown in Table 1. VV1 patients were significantly younger at disease onset than other subtypes (p < 0.001 for all subtypes). MV1 patients had the shortest (4.5 months), and VV1 (17.5 months) the longest disease duration (p = 0.001). MM1 (61%) was the most frequent molecular subtype while VV1 was therarest (4%).
Distribution of molecular subtypes
Distribution of molecular subtypes
m, months; y, years; SD, standard deviation.
Frequency of neuropsychological symptoms and time of onset
Neuropsychological symptoms were very frequent in our patients (96%). There was no statistically significant difference in regard to frequency between the subtypes. MV2 patients showed the latest onset of neuropsychological abnormalities among the subtypes, and this finding was statistically significant in comparison with all sCJD subtypes except VV2 (for MM1 p < 0.001, for MM2 p = 0.035, for MV1 p = 0.02, for VV1 p = 0.028). In VV2, neuropsychological symptoms occurred significantly later than in all other subtypes except MV2 (p for MM1 = 0.004, for MM2 p = 0.048, for MV1 p = 0.005, for VV1 p = 0.039).
Apraxia
Apraxia was found in 140 patients (56.5%). In VV1 patients, apraxia was very common (90%), in VV2 significantly less so (44%, p = 0.013). In MM1 patients, apraxia was found already after one month and thus occurred significantly earlier than in all other subtypes except MV1 (for MM2 p = 0.003, for MV2 p < 0.001, for VV1 p = 0.007, for VV2 p = 0.019). In MV2 patients, apraxia was found as late as 6 months after disease onset and significantly later than in MV1 (p = 0.041) and VV2 (p = 0.019) patients. Further differentiation was only possible in part due to insufficient data. Ideatoric apraxia was documented in 23/141 patients (16%), ideomotoric apraxia in 13 patients (5%), and facial apraxia in 8patients (3%).
Aphasia
Aphasia was documented in 157 patients (63%). Broca aphasia was reported in 71 patients (29% of total patients, 45% of all patients with aphasia). Sensory aphasia was observed in 50 patients (20% of total patients, 32% of all patients with aphasia), and global aphasia in 33 patients (13% of total patients, 21% all patients with aphasia). There was no significant difference in frequency between different subtypes. In MM1 patients, aphasia was found significantly earlier than in MM2 (p = 0.019), MV2 (p < 0.01), and VV2 (p = 0.004) patients. Moreover, in MV2 patients, aphasia occurred significantly later than in MV1 (p = 0.003) and VV2 (p = 0.004) patients.
Acalculia
Acalculia was reported in 44 patients (18%). There was no significant difference in frequency between different subtypes. In MV2 patients, acalculia occurred after disease duration of 6.5 months and therefore significantly later than in MM1 (p = 0.02), VV1 (p = 0.004), and VV2 (p = 0.009) patients.
Alexia
Alexia was found in 38 patients (15%). There was no significant difference in frequency between different subtypes. In MM1 patients, alexia occurred significantly earlier than in MV2 patients (p = 0.011).
Agraphia
Agraphia was documented in 44 patients (18%). There was no significant difference in frequency between different subtypes. Agraphia occurred in MV2 patients as late as after 8 months and significantly later than in MM1 (p = 0.009), MM2 (p = 0.026), and MV1 patients (p = 0.015).
Spatial disorientation
108 patients (43.5%) showed spatial disorientation. Hemispatial neglect was reported in 6 patients (2% of total patients, 6% of all patients with spatial disorientation), and visuo-constructive processing disturbance in 34 patients (13.7% of total patients, 31.5% of patients with spatial disorientation). In MM1 patients, spatial disorientation was observed significantly less frequently than in MM2 (p = 0.023), MV2 (p = 0.023), and VV1 patients (p < 0.001). In VV1 patients, spatial disorientation was found significantly more often than in MV1 (p = 0.002), MV2 (p = 0.019), and VV2 patients (p = 0.012). In MV2 patients, spatial disorientation occurred significantly later than in VV1 patients (p = 0.029).
Agnosia
Agnosia was found in 53 patients (21%). There was no significant difference in frequency or time of occurrence between different subtypes. 25 patients (10% of total patients, 47% of all patients with agnosia) showed anosognosia.
Frontal lobe syndrome
Frontal lobe syndrome was common in sCJD and was found in 186 patients (75%). There was no significant difference in frequency or time of occurrence between different subtypes. Data on drive, attention, and emotion are shown separately in Table 2.
Neuropsychological symptoms, their frequency, and time point of occurrence
SD, standard deviation; p, statistical significance, p values in the text (time point of occurrence and/or frequency); n.s., not significant; s., significant.
Amnesia
Amnesia was very common in sCJD and was found in 224 patients (90%). There was no significantdifference in frequency of occurrence between different subtypes. In MM1 patients, amnesia occurred already in the first month and was therefore significantly earlier than in MV2, VV1, and VV2 patients (p < 0.001 each). In MV1 patients, amnesia also occurred significantly earlier than in MV2 (p = 0.014) and VV1 (p = 0.022) patients. In VV2 patients, amnesia was found significantly earlier than in VV1 patients (p = 0.042).
Lack of drive was found in 131 (53%) of the sCJD patients investigated. VV2 patients showed this symptom significantly more often MM1 patients (p = 0.030) and also significantly later (p = 0.021).
Impaired emotion
The patients showed impaired emotion in the form of affective lability. It was found in 84 patients (33%). Impaired emotion was significantly less frequent in VV2 than in MM1 (p = 0.016), MM2 (p < 0.001), MV1 (p = 0.025), MV2 (p = 0.001), and VV1 patients (p = 0.043). In MM1, impaired emotion was found significantly earlier than in MV2 (p = 0.001) and VV1 patients (p = 0.028).
Impaired attention
Impaired attention was as common as amnesia in sCJD and was found in 224 patients (90%). There was no significant difference in frequency of occurrence between different subtypes. In MM1 patients, impaired attention occurred significantly earlier than in MV2, VV1, and VV2 patients (p < 0.001 for each). In MV1 patients, attention deficits also occurred significantly earlier than in MV2 (p = 0.014) and VV1 (p = 0.022) patients. In VV2 patients impaired attention was found significantly earlier than in VV1 patients (p = 0.042).
Neuropsychological tests
Neuropsychological tests were performed in 97 patients (39%, Table 3). Data on age at the time of testing as well as disease duration are shown in Table 3. Distribution of MMSE (Mini-Mental State Examination) results stratified by subtype is shown in Table 4. MMSE was performed in 65 patients (26%), and other tests in 61 patients (25%) but the data from the other tests could not be statistically evaluated as there were 23 different tests (legend Table 4), and no one test was performed in more than 3 patients.Neuropsychological examination was less frequently statistically significant in MM1 than in MV2 (p < 0.001), VV1 (p = 0.010), and VV2 (p < 0.001) patients. It was also less frequent in MV1 than in MV2 patients (p = 0.006).
Neuropsychological tests
n, number; SD, standard deviation.
Frequency of neuropsychological testing and distribution of Mini-Mental State Examination results stratified by molecular subtype
MMSE, Mini-Mental State Examination; n, number; SD, standard deviation. *First MMSE in patients with more than one MMSE. **Clock Drawing Test, SKT, SIDAM, SISCO, DigitSpan, TMT, CERAD, DemTect, Mattis Dementia Rating Scale, 3 Words Test, Serial Substraction, HAWIE, Raven, MWT, Benton, d2, KAI, WIP, AAT, AAKT, Token, MMPI, FPI.
MMSE
MMSE was performed in 77 patients (79% of neuropsychologically examined patients). There were no statistically significant differences in the percentage of MMSE investigated patients. However, multiple MMSE examinations were significantly more frequent in MV2 compared with MM1 (p = 0.004) subtype.
Characterization of molecular subtypes
MM1 subtype
Impaired attention and amnesia (86% each), frontal lobe syndrome (72%), aphasia (62%), and apraxia (54%) were the most common deficits and occurred as early as after median disease duration of one month (Table 2). MM1 patients showed the most neuropsychological deficits already in the first third of disease course, and only few neuropsychological deficits appeared in the second third(Table 5).
Comparison of occurrence of psychiatric symptoms stratified by molecular subtypes and disease thirds
MM2 subtype
Impaired attention and amnesia (100% each), frontal lobe syndrome (92%), apraxia (77%), aphasia, and spatial disorientation (69% each) were the most frequent neuropsychological symptoms. Frontal lobe syndrome occurred already after 0.5 months, amnesia and impaired attention after 2 months, aphasia after 3 months, and apraxia after 5 months. MM1 subtype was the only subtype in which all neuropsychological symptoms appeared at a very early phase of the disease course (Table 5).
MV1 subtype
Amnesia and impaired attention (92% each), frontal lobe syndrome (75%), aphasia (75%), and apraxia (67%) were the most common neuropsychological deficits. Amnesia and impaired attention, frontal lobe syndrome and aphasia occurred at disease onset and apraxia after 2.5 months (Table 2). All neuropsychological symptoms except spatial disorientation, which was observed in the second third of the disease course, were found in the first third (Table 5).
MV2 subtype
Amnesia and impaired attention (92% each), frontal lobe syndrome (87%), aphasia (73%), and apraxia (60%) were the most frequent neuropsychological symptoms. Frontal lobe syndrome was found after 2 months, amnesia and impaired attention after4 months, aphasia after 5.5 months and apraxia after 6 months (Table 2). Frontal lobe syndrome and spatial disorientation were found already in the first third of disease duration. Sensory and global aphasia occurred as late as in the last third (Table 5).
VV1 subtype
Amnesia and impaired attention (100% each), spatial disorientation (100%), apraxia (90%), and aphasia (70%) were the most common neuropsychological disturbances. Spatial disorientation was found in all patients and occurred in the first third of the disease course (Table 5), in the median after 2.5 months. Apraxia was reported after 4 months, amnesia and impaired attention relatively late after 5 months,and aphasia after 5.5 months (Table 2). Only agraphia and impaired emotion were reported in the last third.
VV2 subtype
Amnesia and impaired attention (97% each), frontal lobe syndrome (78%), aphasia (50%), and apraxia (44%) were the most frequent neuropsychological symptoms. Frontal lobe syndrome occurred after 2 months, aphasia and apraxia after 2.5 months (Table 2). While spatial disorientation, amnesia and impaired attention were observed after 3 months and thus in the second third of disease duration, amnestic aphasia, anosognosia, and frontal lobe syndrome were found during the first third and agraphia occurred as late as in the last disease third (Table 5).
DISCUSSION
In this study, we present the first detailed evaluation of neuropsychological deficits in a large group of definite sCJD patients with known molecular subtype. Despite the fact that the most CJD patients develop neuropsychological deficits during disease course, data on this aspect of CJD are very limited. Because of the very rapidly progressive course, frequently patients cannot be assessed by means of neuropsychological test batteries, and communication is barely possible because of early aphasia and prominent cognitive deficits. The quality and amount of data are therefore limited. We are aware of the methodological limitations of our study, as the data on tests other than MMSE in our CJD patients were very inconsistent and few. To supplement the data, we have also included the results of clinical investigation by our study physicians.
The number of neuropsychological investigations per molecular subtype provided us with at least some clues as to the type of dementia as well as its temporal evolution. The frequency and complexity of tests performed also depended on the cognitive capacity of the patients, meaning that there was a certain limit to the tests that could be used. Interestingly, MV2 patients showed statistically significant later onset of neuropsychological symptoms compared to all subtypes except VV2. Generally, we observed a higher proportion of neuropsychologically tested patients in atypical molecular subtypes such as MV2, MM2, and VV1. These patients also showed the longest median disease duration. Slower disease and especially dementia progression seemed to be associated with longer survival time. As for dementia and neuropsychological deficits in general, this association is clearly understandable. Less handicapped patients experience as rule fewer restrictions in locomotion and thus have a lower risk of pneumonia, which is one of the most common causes of death in CJD and in dementia patients in general [20]. They also can remain in their familiar environment for a longer period of time.
As shown in Table 5, all neuropsychological deficits evaluated occurred in the first third of the disease course. However, in MV2 and VV2 patients, pronounced aphasia first occurred as late as in the second (VV2) or even last third of the disease course (MV2), and amnesia was first observed in VV2 patients in the second third. Thus, neuropsychological tests can also be applied in the later disease course. Clinically, less rapid dementia progression was particularly observed in MV2 patients. Neuropsychological tests are often used in patients presenting with neuropsychological deficits, such as MM2 patients [21].
Interestingly, many different tests are performed in Germany for the same disease (23 tests in 61 patients, and not one of these tests was used in more than three patients). This variety of tests may be partly explained by different neuropsychological deficit patterns. MMSE evaluation should be routinely performed in suspected CJD patients, but other tests should only be conducted in atypical cases or at very early disease stages. Unfortunately, it is still not possible to differentiate between sCJD and Alzheimer’s disease by means of neuropsychological testing, so that additional diagnostic methods (MRI, CSF markers) are essential.
Due to the general methodological problems described above, there are only two previous studies in sCJD patients that applied consistent data analysis and standardized tests [4, 5]. The data in both of these studies were evaluated without consideration of PrPSc type or M129 V polymorphism. In contrast to our present study, Cordery et al. aimed todifferentiate between the neuropsychological features of sCJD, variant CJD, and inherited prion diseases but not between different sCJD subtypes. Only 10 sCJD patients were investigated. Similar to our study, all but one sCJD patient showed severe cognitive decline, with deficits in each of five cognitive domains tested (memory for words and faces, nominal, perceptual, and executive functions, including frontal executive functions; no single neuropsychological symptoms were named as in our study).
Snowden et al. investigated neuropsychological characteristics in six CJD patients, and only four of them were sCJD patients. As in our study, the authors reported a wide range of neuropsychological deficits. Two of four patients presented with neuropsychological deficits (one patient with anomia, one with agraphia). The common features described were more generally psychiatric than neuropsychological: fluctuations in the efficiency of their performance, “intrusion errors” and perseverative errors correspond to the psychiatric term “formal disorders of thought” more than to the term “neuropsychological deficits”, and capacity for self reflection and preservation of awareness of illness is known in psychiatry as preserved insight.
There are also some case reports describing neuropsychological features of sCJD or even disease manifestation with neuropsychological deficits or prominent neuropsychological deficits dominating the clinical course [6–9]. However, none of this case reports deals with PrPSc type or M129 V polymorphism. The authors report clinical manifestation of CJD as primary progressive aphasia in one case [9] or isolated aphasia [6], agraphia as first presenting symptom [7], and prominent deficits in all cognitive domains in a patient during disease course [8].
In conclusion, our data contribute to a better differentiation of neuropsychological findings between molecular subtypes of sCJD and thus facilitate more accurate diagnosis of the disease. We hope the present data will help characterize the clinical picture of sCJD more clearly, especially in the atypical subtypes. Particularly from an epidemiological standpoint, accurate diagnosis of sCJD is veryimportant.
Footnotes
ACKNOWLEDGMENTS
The authors would like to express their special thanks to the patients and their families included in the study as well as their physicians from all over Germany who provided clinical and neuropathological data on patients and also sent CSF and blood samples. The assistance of Mrs. Ehrlich and Schneider-Dominco is gratefully acknowledged. This study was funded by the Robert Koch-Institute through funds of the Federal Ministry of Health (grant number 1369-341).
