Official reprint from UpToDate

www.uptodate.com 2014 UpToDate

Print

Back

Creutzfeldt-Jakob disease
Authors
Henry G Brown, MD, PhD
John M Lee, MD, PhD
Section Editor
Steven T DeKosky, MD, FAAN, FACP
Deputy Editor
Janet L Wilterdink, MD
All topics are updated as new evidence becomes available and our peer
review process is complete.
Literature review current through: Aug 2014. | This topic last updated:
Aug 09, 2013.
INTRODUCTION Prion diseases are neurodegenerative diseases that
have long incubation periods and progress inexorably once clinical
symptoms appear. Five human prion diseases are currently recognized:
kuru, Creutzfeldt-Jakob disease (CJD), variant Creutzfeldt-Jakob disease
(vCJD), Gerstmann-Strussler-Scheinker syndrome (GSS), and fatal familial
insomnia (FFI). CJD accounts for more than 90 percent of sporadic prion
disease [1].
These human prion diseases share certain common neuropathologic
features including neuronal loss, proliferation of glial cells, absence of an
inflammatory response, and the presence of small vacuoles within the
neuropil, which produces a spongiform appearance.
The clinical manifestations and diagnosis of Creutzfeldt-Jakob disease will
be reviewed here. Variant Creutzfeldt-Jakob disease, Kuru, GSS, and FFI,
and the biology of prions and the genetics of prion diseases are discussed
separately. (See "Variant Creutzfeldt-Jakob disease" and "Diseases of the
central nervous system caused by prions" and "Biology and genetics of
prions".)
EPIDEMIOLOGY AND CLASSIFICATION Creutzfeldt-Jakob disease

(CJD) is the most frequent of the human prion diseases, although it is still
rare. Sporadic (sCJD), familial (fCJD), iatrogenic (iCJD), and variant forms of
CJD (vCJD) are all recognized. The vast majority of CJD cases are sporadic
(85 to 95 percent), while 5 to 15 percent are due to fCJD; iCJD generally
accounts for less than 1 percent [2,3].
Approximately one case of sporadic CJD occurs per 1,000,000 population
per year with a worldwide distribution. The mean age for the onset of
disease is between 57 and 62 years, although rare cases in young adults
and those over 80 years of age have been described [3-7]. Patients with
vCJD and iCJD tend to be much younger, which led to an early appreciation
that the mode of transmission might be different. fCJD patients have only a
slightly younger age of onset compared with sCJD [8]. There is no gender
predilection for CJD. In the United States, the incidence of CJD appears to
be less in African Americans, American Indians, and Alaska natives
compared with the white population, however this observation may have
resulted from ascertainment bias [9,10].
The incidence of CJD is increased 30 to 100-fold in certain geographic
regions including areas of North Africa, Israel, Italy, and Slovakia, due
primarily to clusters of fCJD [8].
Sporadic CJD A number of epidemiologic studies have attempted to
identify potential risk factors for sCJD. In one reappraisal of three casecontrol studies, family history of CJD (adjusted odds ratio [OR] 19.1) or
medical history of psychosis (OR 9.9) were the only significant factors
associated with disease [11]. A second study in Australia found that a history
of multiple surgical procedures and residence for more than 10 years on a
farm were significant risk factors for sCJD [12]. This latter report used
community rather than hospital-based controls. Neither study documented
an increased risk of sCJD with receipt of blood products.
Small clusters of sCJD cases have been reported [13-15]. A retrospective
case-control study of all sCJD cases from 1990 to 1998 in the United
Kingdom concluded that sCJD cases lived closer together than might be
expected in the absence of a disease-clustering mechanism; the results
suggested that some sCJD cases may result from exposure to a common
external factor [16]. However, the significance of these possible clusters
remains unclear. The results of one statistical analysis have suggested that
these apparent clusters may result from higher intensity of surveillance in
localized geographic regions [17].
Variant CJD Variant CJD (vCJD) is discussed separately. (See "Variant
Creutzfeldt-Jakob disease".)

Familial CJD Overall, familial CJD contributes to less than 15 percent of

CJD cases. Familial CJD is discussed in more detail separately. (See
"Biology and genetics of prions", section on 'Familial CJD'.)
Iatrogenic CJD Iatrogenic CJD is believed to be controlled by current
practices; new cases are likely to be those with very long incubation periods
[18]. Iatrogenic CJD (iCJD) has followed administration of cadaveric human
pituitary hormones (growth hormone and gonadotrophin) [19-21], dural graft
transplants [22-24], use of dural mater in radiographic embolization
procedures [25], corneal transplants [26,27], liver transplants [28], the use of
contaminated neurosurgical instruments or stereotactic depth electrodes
[29], and secondary infection with variant CJD through transfusion of
infected blood products. Dural graft transplants and use of cadaveric
pituitary hormones account for the vast majority of cases of iCJD [18].
It is possible that other surgical procedures also play a role in the
transmission of CJD. In a case control study, the risk of sporadic CJD was
increased among individuals who had ever undergone surgery (OR 2.0; 95%
CI:1.3-2.1) with highest rates in the category called "other surgery" which
included skin stitches, nose/throat operations and removal of
growths/cysts/moles [30]. This is consistent with two other case control
studies that also found an association between CJD and prior surgery
[12,31,32]. In one of these studies, latency from surgery to clinical disease
onset was shorter with surgeries involving the central nervous system or eye
and longer for systemic surgeries [32]. In another study, three or more
procedures was associated with a higher risk than fewer procedures [12]. It
is possible that recall bias plays a role in this apparent association.
Because prions are resistant to conventional physical decontamination
methods, disinfection and sterilization of prion-contaminated medical
instruments requires specific procedures. It is important to keep the
instruments moist after use (either by immersion in water or a prionicidal
detergent), because sterilization does not appear to be as effective on dried
tissue. Prior to sterilization the instruments should be decontaminated in an
automated dishwasher. Options for sterilization that appear to be effective
include [33]:

Autoclave at 134C for 18 minutes in a prevacuum sterilizer

Autoclave at 132C for 1 hour in a gravity displacement sterilizer
Immerse in 1 N NaOH for 1 hour; remove and rinse in water then transfer to
an open pan and autoclave for 1 hour at 121C in a gravity displacement
sterilizer or at 134*C in a porous or prevacuum sterilizer
Immerse in 1 N NaOH for 1 hour; then heat in a gravity displacement

sterilizer at 121*C for 30 minutes, then clean and subject to routine

sterilization
If the instrument is not fully accessible to the actions of the steam sterilizant,
it should be discarded. It should also be noted that the combination of NaOH
and steam sterilization can be harmful to surgical instruments. Flash
sterilization or low temperature methods for sterilization should not be used.
Proven surgical routes for transmission have now been largely eliminated
following changes in procedures used to prepare dural grafts [34] and use of
recombinant-derived hormones rather than hormones derived from
cadaveric pituitary pools. An international CJD surveillance team reported
that, in 2000, all incident cases of iatrogenic CJD were due to long
incubation periods from infections acquired before 1985 [35]. An exception
may be the use of Lyodura dural grafts; those manufactured before 1987
(before new procedures were instituted) were not recalled internationally,
and continued to be used until 1993 [36]. Cases of CJD related to these
grafts continue to accrue, in some cases 25 years after surgery. In addition,
there are four reported cases of transfusion-associated variant CJD [35].
(See "Variant Creutzfeldt-Jakob disease", section on 'Transfusion-related
vCJD'.)
The incubation period for iCJD is unknown and probably depends upon the
mode of transmission. One study estimated a mean time of 9 to 10 years
based upon mathematical models of the incubation time for iCJD acquired
after administration of human growth hormone in a population of 1361
recipients, 55 of whom had developed iCJD [37]. A similar mean incubation
period has been reported for dural graft transplants [34]. However,
incubation periods as short as five years and as long as 30 years have been
reported in other patients [37-39].
Transfusion-related CJD Available epidemiologic studies suggest that a
history of preceding transfusion does not increase the risk of developing
sCJD [40,41]. However, human-to-human transmission of prion diseases
can occur under special circumstances, and low levels of infectivity have
been noted when whole blood, serum, or buffy coat derived from patients
with sCJD is inoculated into animals.
No definite cases of transfusion-associated iCJD are known to have
occurred. One study reviewed 436 recipients of 36 blood donors who
subsequently developed CJD [42]. After 2096 person years of follow-up, no
recipient had developed CJD; 144 recipients survived more than five years
after transfusion; 68 of these had received blood donated less than 60
months before the donor developed CJD. There is one report of

development of sCJD in a patient 8 to 10 months after albumin infusions

during coronary artery bypass surgery [43]; however, this is likely to
represent a de novo rather than transfusion-acquired case of sCJD.
Transfusion-associated vCJD has been reported. (See "Variant CreutzfeldtJakob disease", section on 'Transfusion-related vCJD'.)
Beginning in November, 1999, leukocyte depletion of all blood components
was mandated in the United Kingdom in order to reduce the concentration of
transmissible agents [44]. Other risk-reduction measures are in development
[45-49].
Infections in health care workers CJD infection in health care workers
is extremely rare [50]. A 1993 publication reports a CJD infection in a sixth
physician (neurosurgeons, pathologists) and 24th health care worker
(including laboratory technicians, nurses) to date [51]. Several cases have
since been reported, but such reports have not been systematically collected
[52-54]. Not all cases were biopsy-proven. In most cases, a link to a prior
exposure is difficult to establish given the long incubation period and the
dependence on the recall of relatives and/or coworkers. Case-control studies
have not found a significant association between CJD and health care
profession [54-56].
Physical contact with CJD patients entails no risk of transmission and
special precautions are not required in their care. However, special
precautions should be employed in the handling of cerebrospinal fluid as
well as biopsy tissue; all materials and instruments used must be
decontaminated according to established protocols [57-59]. Routine
sterilization measures are not adequate for decontamination of CJD. (See
"General principles of infection control".)
Possible animal origins of sCJD All available evidence suggests that a
single bovine prion strain responsible for BSE has infected humans to cause
vCJD. (See "Variant Creutzfeldt-Jakob disease", section on 'Relationship
with BSE'.) However, additional bovine prion strains have now been reported
to cause spongiform encephalopathy in cattle [60-62]. One of these strains is
associated with atypical amyloid deposition, a neuropathological feature not
found in typical BSE [62]. The PrPSc molecular signature of this amyloidotic
form of BSE (called BASE) is similar to that of sCJD subtype MV2. (See
'Molecular subtypes of sCJD' below.)
The significance of this PrPSc molecular similarity is unclear. One
explanation is that some cases of sCJD with the PrPSc MV2 pattern might
be the result of exposure to BASE. However, it is important to recognize that
glycosylation and migration patterns of PrPSc may not be a reliable indicator
of prion protein origin. Until further evidence accumulates, the role, if any, of

BASE exposure as a source of sCJD should be considered speculative [63].

NEUROPATHOLOGY On gross examination, most cases show some
atrophy. However, some, even those associated with long clinical histories,
appear normal in this respect. Atrophic brains usually show ventricular
enlargement. The atrophy may include the deep gray structures: the
caudate, putamen, and thalamus. In contrast to Alzheimer disease, the
hippocampus is spared. The cerebellum may also show atrophy of the folia,
usually due to loss of gray matter.
The main histologic features of prion disease are (picture 1):

Spongiform change
Neuronal loss (particularly of cortical layers III-V) without inflammation
Accumulation of the abnormal prion protein
Spongiform change The vacuoles that create the spongiform change in
CJD are round and between 20 to 50 microns in diameter (picture 2).
Electron microscopy has revealed that the vacuolization is an intraneuronal
process, and that synaptic loss is another common ultrastructural feature.
Spongiform vacuolation can occur in other diseases. Neurodegenerative
diseases associated with severe neuronal loss produce status spongiosus
as a common feature in end stages. This is a coarse microvacuolization with
astrocytosis. The vacuoles are usually larger and more irregular than those
seen in CJD, and they are usually concentrated in the upper layers (I-III) of
the cerebral cortex. Microvacuolization in frontotemporal dementias is
usually restricted to the upper cortical layers and to the frontal and temporal
lobes. Transcortical vacuolization in dementia with Lewy bodies is usually
restricted to the medial temporal lobe. Acute hypoxic ischemic and other
metabolic encephalopathies can also give this microscopic appearance.
Finally, inadequate tissue fixation or poor processing can cause artifactual
changes that resemble prion vacuoles. The clinical history along with prion
protein immunohistochemistry generally clearly distinguishes among these
entities.
Prion protein immunohistochemistry Accumulation of the prion protein
is detected by immunohistochemical techniques (picture 3). The prion
protein can be distributed in several patterns, such as a perivacuolar
distribution, a diffuse synaptic distribution, or in plaques.
The plaque type can be associated with specific prion protein diseases, such
as kuru, Gerstmann-Strussler-Scheinker syndrome, or vCJD. (See "Variant

Creutzfeldt-Jakob disease" and "Diseases of the central nervous system

caused by prions", section on 'Neuropathology'.)
The pattern and distribution of prion protein deposition may be specific to the
subtype of sCJD. (See 'Molecular subtypes of sCJD' below.) As examples,
MM1 and MV1 subtypes have been associated with the synaptic pattern of
PrP immunostaining, MV2 and VV2 with plaque formation, especially
prominent in the deep gray structures and cerebellum, and MM2 with
perivacuolar and loose plaque-like formations [64]. Iatrogenic CJD related to
dura mater grafts has also been associated with plaque formation [65].
CLINICAL FEATURES
Symptoms and signs Rapidly progressive mental deterioration and
myoclonus are the two cardinal clinical manifestations of sCJD. Death
usually occurs within one year of symptom onset [66].

Mental deterioration may be manifest as dementia, behavioral abnormalities,

and deficits involving higher cortical function. Concentration, memory, and
judgment difficulties are frequent early signs [67]. Mood changes such as
apathy and depression are common; euphoria, emotional lability, and anxiety
occur less frequently. Sleep disturbances, particularly hypersomnia, but also
insomnia, are also common, and may be a presenting sign [68,69]. With
disease progression, dementia becomes dominant in most patients and can
advance rapidly.
Myoclonus, especially provoked by startle, is present in more than 90
percent of patients at some point during the illness but may be absent at
presentation, even when dementia is profound. sCJD should always be
considered in a patient with the combination of a rapidly progressive
dementia and myoclonus.
Extrapyramidal signs such as hypokinesia and cerebellar manifestations,
including nystagmus and ataxia, occur in approximately two-thirds of
patients and are the presenting symptoms in 20 to 40 percent [67]. In
particular, iatrogenic CJD related to human gonadotrophin and growth
hormone treatment as well as to dura mater grafts has a propensity to
manifest as a largely isolated cerebellar syndrome early in the disease
course [21,65,70].
Signs of corticospinal tract involvement develop in 40 to 80 percent of
patients, including such findings as hyperreflexia, extensor plantar
responses (Babinski sign), and spasticity.

Younger patients with sCJD have clinical features that are somewhat distinct
from the more typical older patient. In one case series of 52 patients younger
than 50 years, psychiatric symptoms were more prominent and the clinical
course more prolonged than in older patients, features that may suggest
variant CJD [71,72]. (See "Variant Creutzfeldt-Jakob disease".) However,
test results, CSF protein markers, and neuroimaging, are consistent with
sCJD.
Some clinical findings, although compatible with CJD, should raise the
suspicion of an alternative diagnosis, especially if they are among the more
prominent features of the illness. These include cranial nerve abnormalities,
sensory abnormalities, and involvement of the peripheral nervous system.
Disturbances of pupillary responses, extraocular movements, trigeminal
neuropathy, and vestibulocochlear dysfunction have all been reported in
isolated cases but are not characteristic. Sensory signs and symptoms are
common in vCJD but are otherwise extremely atypical in human prion
diseases.
However, a number of variants or subtypes of disease have been defined
based upon focal neurologic findings reflecting predominant involvement of
individual brain regions. Examples of these include forms with mainly visual
(Heidenhain variant), cerebellar (Oppenheimer-Brownell variant), thalamic,
and striatal features [72,73]. Variants of sCJD have also been classified
based according to the genotype of the prion protein gene (PRNP) and the
molecular properties of the pathological prion protein (PrPSc) as discussed
in the following section.
Molecular subtypes of sCJD Clinical phenotypes of sporadic CJD have
been associated with molecular subtypes determined by the PRNP gene
codon 129 genotype and the pathologic prion protein (PrPSc) type. The
PRNP genotype is homozygous or heterozygous for methionine (M) or
valine (V) at codon 129. The PrPSc type is determined by Western blot
analysis and classified in the Parchi/Gambetti nomenclature as type 1 or
type 2 depending on the size and electrophoretic mobility of the protease
resistant core fragment (PrPres) [74,75].
Using this molecular classification, six clinical phenotypes of sCJD have
been described [75,76]. The frequency and duration of illness are based on
a series of 300 cases in North America and Europe [75] and an additional
2451 patients were analyzed for diagnostic test characteristics [77]:

MM1 and MV1 (myoclonic, Heidenhain variant) account for about 70 percent
of cases and correlate with the "classic CJD" phenotype of advanced age at

onset, a rapidly progressive dementia with early and prominent myoclonus,

and a short duration of illness (mean 3.9 months). The MM1 phenotype is
the one most commonly associated with periodic sharp wave complexes
(PSWC) on electroencephalogram (EEG).
VV2 (ataxic variant) accounts for 15 percent or less of sCJD and presents
with ataxia at onset, often as an isolated feature, late dementia, and a longer
duration of illness (mean 7 to 9 months) [78].
MV2 (Kuru plaque variant) accounts for 9 percent and presents with ataxia,
progressive dementia with prominent psychiatric features, and longer
duration (mean 17.1 months) [79]. The 14-3-3 protein in the CSF is a
relatively insensitive marker for the MV2 variant (about 70 percent), and
PSWC are only infrequently seen on EEG [77,79-81].
MM2 can present as either a thalamic variant or a cortical variant. Some, but
not all, patients have a young age at onset, and the disease course is
typically long, with a median disease duration of 14 months in one study
[82]. The 14-3-3 protein has been reported to be present in 61 to 91 percent
of patients with MM2, and periodic sharp wave complexes (PSWCs) on EEG
are more often absent than in other MM and MV subtypes [75,77,80-82].
The clinical features of MM2 type sCJD may resemble those of variant CJD.
(See "Variant Creutzfeldt-Jakob disease", section on 'Clinical features'.)The
thalamic MM2 variant accounts for 2 percent of cases, and mean disease
duration is 15.6 months. Insomnia, psychomotor hyperactivity, ataxia, and
cognitive impairment are the predominate manifestations, and this
phenotype resembles that of fatal familial insomnia (FFI) [83]. (See
"Diseases of the central nervous system caused by prions", section on 'Fatal
familial insomnia'.)The cortical MM2 variant accounts for 2 percent of cases,
with a mean disease duration of 15.7 months. Dementia is the predominate
manifestation, while cerebellar and visual signs are rarely described at
presentation [84].
VV1 accounts for 1 percent of cases and is notable for progressive dementia
and longer duration (mean 15.3 months). A case series of nine patients with
this subtype confirmed the slower, more prolonged course (median 21
months) [85]. All patients had elevated CSF levels of the 14-3-3 protein, but
none had PSWCs on EEG, and cortical rather than basal ganglia
abnormalities were more common on MRI.
The classification for PrPSc is complicated by the existence of at least two
nomenclatures, although the Parchi and Gambetti nomenclature for PrPSc
[74,75] described above may be the most widely used. The alternate
Collinge nomenclature distinguishes four rather than two PrPSc subtypes
[86-88]. The Collinge PrPSc types 1 and 2 are thought to correspond with
Parchi and Gambetti PrPSc type 1, and Collinge types 3 and 4 with Parchi

and Gambetti type 2 [89].

These molecular subtypes have also been associated with distinctive
patterns of cerebral gene expression in CJD involving upregulation of genes
involved in immune and stress responses and in cell death and
downregulation of genes involved in synaptic function [90]. Another study
found that more than one molecular subtype may be present in 20 percent of
individuals with sCJD and that PrPSc type might be influenced by both
genetic and brain region-specific factors [91].
The ongoing identification of rare molecular subtypes of CJD that are not
included in these classification schemes may require their subsequent
modification [92]. As an example, it has been increasingly recognized that
some patients with CJD have both type 1 and type 2 PrPSc, and that
patients who have both types have a distinct clinical and pathologic
phenotype [93]. In a series of 34 patients with the MM genotype, 20
individuals had both types (MM1+2) [94]. These patients had histopathologic
features that were intermediate between patients with the MM1 and MM2
subtypes.
DIFFERENTIAL DIAGNOSIS Creutzfeldt-Jakob disease (CJD) must be
distinguished from other dementias. Occasionally Alzheimer disease (AD)
and frontotemporal dementia is associated with myoclonus and a more
rapidly progressive course than is typical and is therefore mistaken for CJD
[95,96]. Prominent ataxia or parkinsonism when present may suggest
dementia with Lewy bodies, progressive supranuclear palsy, or multiple
systems atrophy [95,97,98]. However, even when more rapidly progressive
than is usual, the neurodegenerative dementias virtually never have a
disease course that progresses to death within 12 months, a time-frame that
is typical for CJD [99].
Other entities that have been mistaken for CJD include [95,98,100-107]:

Autoimmune disorders, including paraneoplastic syndromes, demyelinating

disease, central nervous system vasculitis, sarcoidosis, and Hashimoto's
encephalopathy
Infections, including viral and postviral encephalitis, fungal and tubercular
meningitis, and HIV
Malignancy, including lymphoma, gliomatosis cerebri, and paraneoplastic
syndromes
Toxic and metabolic encephalopathies, including alcoholic cerebral
degeneration, and methylmalonic acidemia,
Cerebrovascular disease, including cerebral amyloid angiopathy, MELAS,

and CADISIL
Psychiatric disease, including depression
DIAGNOSIS Brain biopsy remains the gold standard diagnostic test for
Creutzfeldt-Jakob disease (CJD). However, a number of tests can be helpful
in providing clinical support for the diagnosis.
Diagnostic criteria A number of criteria for the diagnosis of sCJD have
been proposed [2,108-110]. Appropriate clinical and laboratory features
generally are sufficient for a "probable" diagnosis of sCJD.
The Centers for Disease Control and Prevention (CDC) outline the following
criteria for probable sporadic CJD [111]:

Progressive dementia and

At least two out of the following four clinical features: myoclonus; visual or
cerebellar disturbance; pyramidal/extrapyramidal dysfunction; akinetic
mutism and
Atypical electroencephalogram (EEG) during an illness of any duration,
and/or a positive 14-3-3 cerebrospinal fluid (CSF) assay with a clinical
duration to death less than two years, and/or magnetic resonance imaging
(MRI) high signal abnormalities in caudate nucleus and/or putamen on
diffusion-weighted imaging (DWI) or fluid attenuated inversion recovery
(FLAIR) and
Routine investigations should not suggest an alternative diagnosis
However, a definitive diagnosis requires these features in combination with
one or more of the following neuropathologic findings:

Loss of neurons, gliosis, spongiform degeneration, or plaques positive for

PrPSc on histopathology of brain tissue
Positive PrPSc staining following pretreatment of brain tissue with proteinase
K to destroy PrPC reactivity
Positive histoblotting of brain tissue extracts for PrPSc after treatment with
proteinase to destroy PrPC reactivity
Transmission of characteristic neurodegenerative disease to experimental
animals
Demonstration of PRNP gene mutations

Immunohistochemistry techniques such as ELISA and conformationdependent immunoassay (CDI) can detect the presence and levels of
disease-causing PrPSc in human brain tissue at autopsy or in biopsy
samples. (See "Biology and genetics of prions", section on 'Demonstration of
PrPSc'.) The CDI method appears to have a much higher sensitivity for the
diagnosis of sCJD disease compared with routine neuropathologic
examination and immunohistochemistry [112], but its specificity remains to
be established.
MR imaging Of the available neuroimaging modalities, magnetic
resonance imaging (MRI) is the most helpful in the diagnosis of CJD (image
1 and image 2) [113].
The recommended MR imaging protocol for assessment of patients with
suspected CJD include T2 and proton density axial images with 3 mm slice
thickness, FLAIR axial and sagittal images at 3 mm slice thickness, T1
images, and diffusion-weighted images (DWI) [114].
Although the data are limited to small studies and case reports, the following
observations have been made regarding MRI in the evaluation of CJD:

Abnormally increased T2 and FLAIR signal intensity in the putamen and

head of the caudate are the most common finding on conventional MRI
sequences in patients with CJD [114-118]. Less commonly, areas of T2 and
FLAIR signal hyperintensity are seen in the globus pallidus, thalamus,
cerebral and cerebellar cortex, and white matter. Laminar lesions in the
cerebral cortex and cerebellum may be observed.
Proton density and DWI are more sensitive than T2-weighted MRI and
FLAIR for the detection of CJD-related lesions. DWI is the most sensitive
technique, especially for cortical changes. This is also the most sensitive test
in the early stages of disease, showing abnormalities before the onset of
characteristic clinical findings such as myoclonus and periodic sharp wave
complexes seen on electroencephalogram (see 'Electroencephalogram'
below) [119-123]. In autopsy studies, decreased apparent diffusion
coefficient (ADC) correlates with areas of spongiform change with gliosis
and neuronal loss [124]. These findings are not entirely specific for CJD and
may be confused with stroke, vasculitis, or reversible posterior
leukoencephalopathy [125,126]. However, one study found that a pattern of
FLAIR/DWI hyperintensity and restricted diffusion could reliably distinguish
CJD from other causes of rapidly progressive dementia, with a high degree
of sensitivity [127].
MRI abnormalities may vary with the clinical syndrome and molecular
subtype. Patients with suspected sCJD and increased T2 signal in the

caudate and putamen are more likely to have early dementia, shorter
survival, and VV2, MV2, or MM1 codon 129 genotypes than those without
high basal ganglia signal [68,128]. Thalamic hyperintensities occur most
often in VV2 and MV2 subtypes, and wide cortical signal increase is most
common in the VV1 and MV1 subtypes [128]. Patients with the MM2
thalamic form of sCJD (see 'Molecular subtypes of sCJD' above) may have
nearly normal brain by MRI and DWI, or show only late atrophy or white
matter change [129].
The MRI features of vCJD differ from those of sCJD. (See "Variant
Creutzfeldt-Jakob disease".)
Diagnostic utility Independent interpretation of MRI films in 193 patients
with CJD by three radiologists were associated with 82 to 86 percent
agreement as to whether findings were typical or atypical for CJD [130]. High
degrees of interrater reliability have been found in other studies as well
[121].
An analysis of 1036 patients accrued between 1992 and 2002 with
pathologically confirmed CJD found that MRI studies had an overall
sensitivity of 46 percent when obtained at least six months after disease
onset [77]. This finding is consistent with an earlier, smaller study of
conventional MRI [115]. However, this study found a high specificity of MRI
findings of 93 percent.
DWI increases the sensitivity of MRI, particularly for cortical changes [120122,131]. DWI abnormalities can appear early in the course of CJD, and in
some cases (with familial CJD) even before the onset of clinical signs
[120,132]. In studies that used patients with other causes of rapidly
progressive dementia as controls, MRI with DWI had sensitivities that
ranged between 83 to 92 percent, higher than for EEG abnormalities and
similar to 14-3-3 protein in the CSF [120,121,131]. Specificities in these
studies ranged from 87 to 95 percent, requiring some caution in MRI
interpretation [133].
Evolution of MRI abnormalities There appears to be a characteristic
progression of MRI signal change from early to late disease in sCJD and
iCJD. Although the data are limited to small observational studies, the
following MRI patterns have been reported [114,123,125,134]:

Early CJD is characterized by increased DWI signal in cortex or deep gray

matter (particularly the caudate nucleus and anterior putamen) or both. The
abnormal signal may be unilateral or bilateral, focal, multifocal or diffuse, and

asymmetrical or symmetrical. The corresponding apparent diffusion

coefficient (ADC) values are decreased, suggesting restricted diffusion and
tissue injury.
Intermediate CJD is characterized on DWI by progression of
unilateral/asymmetrical lesions to greater contralateral/symmetrical
involvement and progression of caudate lesions to involve the putamen
(image 1). FLAIR images are very likely to show high signal abnormalities,
but with less prominence than DWI. Generalized atrophy and ventricular
dilatation may be apparent on all sequences.
Late or terminal CJD is characterized by prominent generalized atrophy and
ventricular dilatation. Limited data suggest that loss of abnormal cortical and
basal ganglionic DWI high signal may occur in some but not all cases.
In the early and intermediate stages of CJD, FLAIR appears to be more
sensitive than T2 imaging [135], but sensitivity and conspicuity are best with
DWI (image 1 and image 2). The relative sensitivities of these methods in
late CJD are unclear. While a positive scan can bolster the diagnosis, a
negative MRI should not be interpreted to exclude CJD in a compatible
clinical case, particularly if DWI is not performed.
Other neuroimaging studies MRI is better than computed tomography
(CT) in detecting abnormalities in patients with sCJD [113,114]. A head CT
scan is generally normal and serves mainly to exclude other diagnoses.
However, serial CT scans performed over several months may show rapid
ventricular enlargement and progressive cortical atrophy in some patients
[136].
Abnormal positron emission tomography (PET), single photon emission
computed tomography (SPECT), and MRI spectroscopy have also been
described in patients with sCJD [137-139]. As an example, thalamic
hypometabolism on PET scan or hypoperfusion on cerebral blood flow
SPECT have been reported with the MM2 thalamic variant (see 'Molecular
subtypes of sCJD' above) [129]. However, these tests have not been
evaluated sufficiently to assess their clinical utility [113].
Electroencephalogram An electroencephalogram (EEG) can provide
supportive but not definitive evidence for CJD. A characteristic EEG pattern
of periodic synchronous bi- or triphasic sharp wave complexes (PSWC) is
observed in 67 to 95 percent of patients with sCJD at some time during the
course of the illness.
PSWCs have a very high specificity for the diagnosis of sCJD. Objective
diagnostic EEG criteria proposed in 1996 were found to have a sensitivity
and specificity of 67 and 86 percent, respectively for the diagnosis of CJD

[140]. In addition, interobserver variability was very low. The PSWCs typical
of sCJD were characterized by the following features [140]:

Strictly periodic cerebral potentials, the majority with a duration of 100 to 600
milliseconds and an intercomplex interval of 500 to 2000 milliseconds
Generalized and lateralized complexes permitted
At least five repetitive intervals with a duration difference of <500
milliseconds required to exclude semiperiodic activity
In a subsequent report from the same investigators, a much larger series of
autopsy confirmed (n=150) or autopsy excluded (n=56) cases of CJD were
studied [141]. Objective EEG criteria for sCJD showed a sensitivity and
specificity of 64 and 91 percent, respectively, and positive and negative
predictive values were 95 and 49 percent. In this study, there were five falsepositive EEG results; four had Alzheimer disease, and one had vascular
dementia. However, applying combined EEG criteria with clinical criteria
yielded an overall specificity of 98 percent and resulted in only one falsepositive case among the 56 excluded from CJD by autopsy.
The mechanism of PSWCs is speculative, but attention has been called to
the similarity of PSWCs in sCJD to the EEG pattern seen in preterm
newborns [141-143]; cortical degeneration due to sCJD may erode the
normal physiologic sleep architecture, which is replaced by activity driven
from an underlying midline pacemaker, possibly thalamic, and involved with
the ascending reticulothalamocortical activating system [144].
PSWCs are helpful in the differentiation of sCJD from other prion disease
[141].

PSWCs are occasionally found in patients with fCJD, although PSWCs are
found more commonly in patients with fCJD who have the codon 200
mutation [145].
PSWCs are not found in patients with new variant CJD (vCJD). (See "Variant
Creutzfeldt-Jakob disease".)
PSWCs are also not found in patients with kuru, Gerstmann-StrusslerScheinker syndrome, or fatal familial insomnia. (See "Diseases of the central
nervous system caused by prions", section on 'Gerstmann-StrusslerScheinker Syndrome' and "Diseases of the central nervous system caused
by prions", section on 'Kuru'.)

Other factors may contribute to the sensitivity of the PSWC finding.

Molecular subtype. In particular, PSWCs may be more commonly absent in

the thalamic variant of MM2 sCJD, as well as the MV2 and VV2 subtypes
[77,79,146].
Disease duration. PSWCs may not be recorded in the initial stages of the
illness. The probability of recording PSWCs corresponds to the amount of
neuronal loss, and serial EEG recording may be useful in patients suspected
of having sCJD when initial EEG recordings are negative [142]. PSWCs
typically disappear in later stages of sCJD, which is characterized by low
voltage activity followed by electrocerebral inactivity [77,147].
Age. One analysis of 2083 patients with pathologically confirmed CJD found
that the presence of PSWCs steadily increased with age, from 22 percent in
individuals <50 years old to 67 percent in those >70 years [77].
Drugs such as barbiturates and benzodiazepines can mask PSWCs.
Iatrogenic CJD may also be less likely to be associated with PSWCs [65].
Protein markers Several reports have suggested that abnormal CSF
proteins may serve as a marker of this illness.
14-3-3 protein Detection of 14-3-3 protein in CSF should be considered
an adjunctive rather than absolute test for the diagnosis of prion diseases as
studies have reported mixed results regarding its sensitivity and specificity
[77,148-152]. A 2012 systematic review sponsored by the American
Academy of Neurology reported an overall sensitivity of 92 percent and a
specificity of 80 percent in diagnosing sCJD [153]. A negative test does not
exclude the diagnosis, especially in cases of possible fCJD or nonclassical
sCJD, and a positive result can occur in nonprion diseases. However, a
positive test increases the probability of CJD when other clinical features are
suggestive but not diagnostic [153]. Pathological studies of brain material to
detect protease resistant PrPSc remain the gold standard for the diagnosis
of prion diseases (see "Biology and genetics of prions", section on
'Demonstration of PrPSc').
The National Prion Disease Pathology Surveillance Center
(www.cjdsurveillance.com) based at Case Western Reserve University is the
only laboratory in the United States that tests for the 14-3-3 protein. This
center helps to monitor and diagnose cases of CJD, as well as act as a
reporting center to the CDC.
False positive elevations in CSF 14-3-3 have been noted in patients with a
variety of neurologic diseases including herpes simplex encephalitis, hypoxic

encephalopathy, cerebral metastases, paraneoplastic disease, and

metabolic encephalopathies [81,149,154,155]. False-positive results are less
likely in neurodegenerative disease [81]. Another report examined
expression of this protein in cell cultures of neural and nonneural tissues and
found the protein in all cell types, suggesting that the protein may be a
marker of brain cell death rather than CJD [155]. Seven distinct isoforms of
the 14-3-3 protein, with two phosphorylated subtypes, have been found in
neurons. It has been suggested that CSF assays for one or more of the
specific isoforms may have improved specificity for distinguishing sCJD from
other dementias compared with the standard commercial 14-3-3 assay
[156,157]. However, elevated CSF levels of most 14-3-3 isoforms are
described in a variety of other conditions including Alzheimer disease,
vascular dementia, metabolic and viral encephalopathies, and
paraneoplastic syndromes, indicating that none of these is clearly diagnostic
of prion disease.
The test characteristics of 14-3-3 protein test may be influenced by several
additional factors:

In one preliminary report, based on the NIH experience, the sensitivity was
significantly lower (62 percent) in patients with fCJD. Based upon this report,
caution should be used in interpreting negative 14-3-3 test results in patients
with possible fCJD.
The clinical and molecular heterogeneity of sCJD may also affect the results
of 14-3-3 protein testing. Studies in patients with sCJD have found that the
sensitivity of CSF 14-3-3 protein was high in the classical subtypes MMI and
MV1 (91 to 100 percent) and relatively low in most of the nonclassical
subtypes, especially MM2 and MV2 (57 to 68 percent) [77,79-81]. These
results suggest that the CSF 14-3-3 protein test may be helpful for diagnosis
of the classical subtypes of sCJD but may be falsely negative for the
nonclassical subtypes. Small numbers of patients for most of the
nonclassical subtypes in this study limit the strength of these findings. In one
study, a second lumbar puncture (LP) at later stages of the disease
improved the sensitivity of this test [81].
In one analysis of 1032 assays for the 14-3-3 protein, a positive result was
significantly less likely when obtained more than 12 months after disease
onset, compared with earlier in the disease (72 versus 92 percent) [77].
Others A variety of other CSF diagnostic tests have been reported in
small series, including the S100 protein [82], neuron specific enolase [158],
thymosin 4 [159], and tau protein [157,160-162]. In one large case series,

tau had superior accuracy and specificity when compared to 14-3-3 protein
as a diagnostic test for CJD, although both tests produced a significant
number of false negative and false positive results [161]. The National Prion
Disease Pathology Surveillance Center now includes tau in addition to 14-33 levels in their analysis of CSF.
Prion protein diagnostic assays performed on blood samples, are in
development [163,164], and in one case series, elevated plasma levels of
several acute phase reactants were noted in patients with CJD [165]. All of
these tests are of unproven diagnostic utility at present.
Studies are also examining the diagnostic utility of CSF test combinations in
detecting CJD [81,152,162]. The presence of any of the following markers:
14-3-3, tau, S100, and neuron specific enolase was associated with high
sensitivity of CSF analysis for detecting CJD [81]. However, the sensitivity
still remained low (50 to 85 percent) for the nonclassical MV2 subtype of
CJD, and overall specificity was also modest (57 to 88 percent).
Routine tests Routine laboratory studies are normal in CJD with the
occasional exception of liver function tests [166]. The CSF contains no cells
and usually has normal glucose. An elevated CSF protein may occur in
about 40 percent of patients [133].
Extraneural PrPSc Establishing the diagnosis of prion diseases by
detecting abnormal prions outside of the CNS remains a major research
goal. Although a number of methods have been reported, none have been
validated as diagnostic tests in human prion diseases.

One study reported that protease-resistant prion protein isoform was

detected in the urine of hamsters, cattle, and humans affected with prion
disease [167]. However, a subsequent study failed to confirm the presence
of PrPSc by Western blot analysis in urine from patients with sporadic,
familial, or variant CJD [168].
In nine patients with sCJD, PrPSc was detected at autopsy in the olfactory
mucosa and cilia, but not in the respiratory mucosa and not in olfactory
tissue from control patients or those with other neurodegenerative conditions
[169]. This small study raises the possibility that a nasal biopsy might
provide tissue for diagnosis.
Another report, using differential precipitation to concentrate PrPSc prior to
performing a Western blot, found evidence of this protein in spleen and
skeletal muscle specimens in approximately one-third of 36 patients with
known sCJD; all patients had positive tests for PrPSc in brain tissue [170].
Patients with PrPSc detectable in spleen, muscle, or both specimens had a

longer history of the disease and were more likely to have uncommon
molecular variants.
PrPSc deposits were detected at autopsy in dorsal root ganglia and
superficial peroneal nerve from one of three patients with sCJD [171].
The development of a streptomycin precipitation protocol holds promise in its
ability to improve the rapid detection of PrPSc in extraneural tissue; as yet,
this has only been demonstrated in cerebral tissues [172].
It is important to recognize that none of the assays described above have
yet been subjected to anything but the most rudimentary evaluation on
human material, and no information exists about their sensitivity or
specificity, even in non-human prion diseases.
TREATMENT AND PROGNOSIS There is no effective treatment for CJD
which is uniformly fatal. Death usually occurs within one year of symptom
onset with a median disease duration of six months [66,98,173].
An overview of past and ongoing efforts to find a treatment for CJD and
other prion diseases is presented separately. (See "Diseases of the central
nervous system caused by prions", section on 'Treatment of prion diseases'.)
INFORMATION FOR PATIENTS UpToDate offers two types of patient
education materials, The Basics and Beyond the Basics. The Basics
patient education pieces are written in plain language, at the 5th to 6th grade
reading level, and they answer the four or five key questions a patient might
have about a given condition. These articles are best for patients who want a
general overview and who prefer short, easy-to-read materials. Beyond the
Basics patient education pieces are longer, more sophisticated, and more
detailed. These articles are written at the 10th to 12th grade reading level
and are best for patients who want in-depth information and are comfortable
with some medical jargon.
Here are the patient education articles that are relevant to this topic. We
encourage you to print or e-mail these topics to your patients. (You can also
locate patient education articles on a variety of subjects by searching on
patient info and the keyword(s) of interest.)

Basics topic (see "Patient information: Creutzfeldt-Jakob disease (The

Basics)")
SUMMARY AND RECOMMENDATIONS

Creutzfeldt-Jakob disease (CJD) is the most frequent of the human prion

diseases, although it is still rare. Approximately one case of sporadic CJD
occurs per 1,000,000 population per year. The mean age of onset is
between 57 and 62 years. (See 'Epidemiology and classification' above.)
Iatrogenic CJD has followed administration of cadaveric human pituitary
hormones, dural graft transplants, use of dural mater in radiographic
embolization procedures, corneal transplants, liver transplants, and the use
of contaminated neurosurgical instruments or stereotactic depth electrodes.
No definite cases of transfusion-associated CJD are known to have
occurred; although transfusion-related variant CJD has been described.
(See 'Iatrogenic CJD' above and "Variant Creutzfeldt-Jakob disease", section
on 'Transfusion-related vCJD'.)
The main histologic features of prion disease are spongiform change,
neuronal loss (particularly of cortical layers III-V) without inflammation, and
accumulation of the abnormal prion protein. Accumulation of the prion
protein is detected by immunohistochemical techniques. (See
'Neuropathology' above.)
Rapidly progressive mental deterioration, often with behavioral
abnormalities, and myoclonus are the two cardinal clinical manifestations of
sCJD. Extrapyramidal signs such as hypokinesia and cerebellar
manifestations are also common. (See 'Clinical features' above.)
Clinical phenotypes of sporadic CJD have been associated with molecular
subtypes determined by the PRNP gene codon 129 genotype and the
pathologic prion protein (PrPSc) type. (See 'Molecular subtypes of sCJD'
above.)
CJD is distinguished from more common causes of dementia by its rapidly
progressive course with prominent myoclonus and gait disturbance. Other
autoimmune, infectious, malignant, and toxic-metabolic etiologies should be
considered in the differential diagnosis. (See 'Differential diagnosis' above.)
While brain biopsy is the gold standard test for diagnosis, it is often
unnecessary. A typical clinical presentation with corroborating findings on
magnetic resonance imaging, electroencephalography, and cerebrospinal
fluid are in most cases sufficient to exclude other causes and establish CJD
as the probable diagnosis. (See 'Diagnosis' above.)

MRI should be performed according to a suggested protocol and in typically

shows abnormal signal in the putamen and head of the caudate. Sensitivity
and specificity for typical MRI findings range between 83 to 92 percent and
87 to 95 percent respectively. (See 'MR imaging' above.)

A finding of periodic sharp wave complexes on EEG has a high specificity for
the diagnosis of CJD, but a low sensitivity. (See 'Electroencephalogram'
above.)
The 14-3-3 protein test in cerebrospinal fluid is a specific test finding for
CJD, but its sensitivity can be low, particularly in some molecular subtypes.
(See '14-3-3 protein' above.)

There is no effective treatment for CJD which is uniformly fatal. Death

usually occurs within one year of symptom onset
Use of UpToDate is subject to the Subscription and License Agreement.
Topic 5084 Version 11.0

GRAPHICS
Low power photomicrograph of classic CJD
showing fine vacuolation distributed throughout
the cortical gray matter

Courtesy of Henry G Brown, MD, PhD.

Graphic 80949 Version 2.0

High power photomicrograph of classic CJD

demonstrating the typical size variation of CJD
vacuoles

Courtesy of Henry G Brown, MD, PhD.

Graphic 70414 Version 2.0

Medium power photomicrograph of prion protein

immunoperoxidase preparation revealing a
diffuse and granular synaptic distribution
pattern

(Immunoperoxidase preparation courtesy of the National Prion Disease

Surveillance Center). Courtesy of Henry G Brown, MD, PhD.
Graphic 50744 Version 1.0

Serial MRI of sporadic CJD

Head MRI of a 34-year-old woman with pathologically proven

sporadic CJD. The initial T2 MRI (Figure A) shows barely
perceptible asymmetric hyperintensity in the bilateral caudate and
putamen, more prominent on the right. Initial diffusion-weighted
MRI (DWI) shows more obvious hyperintensity in the bilateral
caudate and putamen, again more prominent on the right (Figure
B). DWI also shows slight hyperintensity in the medial frontal lobe
cortices (Figure C). MRI two weeks later shows more apparent
increased signal intensity within the bilateral basal ganglia on T2
(Figure D) and especially FLAIR images (Figure G) with
characteristic posterior progression and involvement of the
putamen. DWI images show increased signal intensity within the
left basal ganglia (Figure E) and medial frontal lobe cortices

(Figure F), with new involvement of the posterior right frontal lobe
(Figure F). The FLAIR image (Figure H) shows slightly increased
signal intensity in the medial frontal lobes and possibly in the
posterior right frontal lobe, but the conspicuity of these abnormal
regions is better appreciated on the corresponding DWI image
(Figure F).
Courtesy of Eric D Schwartz, MD.
Graphic 56359 Version 2.0

MRI of heidenhain variant CJD

Head MRI of a 65-year-old man with biopsy proven sporadic CJD

(Heidenhain variant). Figure A is a T2 MRI that shows essentially
normal signal intensity within the basal ganglia and throughout the
brain. Figure B is a proton density MRI that shows increased
cortical signal intensity within the right frontal lobe and bilateral
medial occipital lobes. Figure C is a diffusion weighted MRI (DWI)
that reveals more conspicuous involvement of the right frontal lobe
and more extensive hyperintensities of the bilateral occipital lobes

with no involvement of the basal ganglia. Figure D is a DWI rostral

to the basal ganglia that shows extensive cortical involvement
throughout the brain with more prominence posteriorly, as may be
expected with the Heidenhain variant.
Courtesy of Eric D Schwartz, MD.
Graphic 58441 Version 2.0