Conditions that can look similar

How often is suspected CJD something else?

In a series reported by Chitravas and colleagues and cited in a 2013 review of rapidly progressive dementia, 1,106 specimens were sent for confirmation of suspected CJD. Three hundred and fifty two of them, 32 percent, showed evidence of an alternative diagnosis. Of those, 71 specimens, 7 percent of the whole group, showed treatable disorders that had gone undiagnosed: 26 immune-mediated, 25 neoplastic, 14 infectious and 6 metabolic or toxic. A Mayo Clinic series of patients without CJD reported autoimmune encephalitis, frontotemporal dementia, dural arteriovenous fistulas, cerebral amyloid angiopathy and systemic lupus.

Misdiagnosis runs in both directions. The same review notes patients initially thought to have other dementias who were found to have sporadic CJD at autopsy.

What is on the differential?

The 2013 review organises causes with the mnemonic VITAMIN C: vascular, infectious, traumatic, autoimmune, metabolic, idiopathic or iatrogenic, neoplasm and congenital. It gives most attention to the autoimmune dementias, which are treatable, and to CJD. Named autoimmune entities include neuropsychiatric lupus, paraneoplastic encephalopathies and steroid-responsive encephalopathy with autoimmune thyroiditis, also called Hashimoto encephalopathy. Treatments described are steroids, intravenous immunoglobulin, plasma exchange and immunomodulators, with treatment of the underlying cancer taking priority in paraneoplastic disease. Anti-voltage-gated potassium channel syndromes usually respond well to immunomodulation, while anti-Hu syndromes are almost always associated with an underlying malignancy and do not respond unless the cancer is treated.

The 2021 international biomarker review lists rapidly progressive neurodegenerative diseases, immune-mediated encephalitis, status epilepticus and cerebral ischaemia among differential diagnoses, together with inflammatory and neoplastic central nervous system disease. Alzheimer disease, dementia with Lewy bodies and frontotemporal lobar degeneration are discussed because they affect the specificity of CSF markers. The review also reminds clinicians to consider iatrogenic and variant CJD in suspected cases. The NHS adds Alzheimer disease, Parkinson disease and brain tumours as the conditions a neurologist will rule out first.

Which features should push the search wider?

The 2013 review names cognitive impairment in a patient under 50 as warranting extensive evaluation for a possible autoimmune disorder. Other pointers it gives are acute or subacute onset, new-onset seizures, sudden psychosis, respiratory impairment, involvement of other parts of the nervous system such as neuropathy or myelopathy, and a recently diagnosed cancer. It also warns that an apparently rapid course sometimes reflects subtle cognitive impairment that had been present for years.

The investigations it recommends are blood tests, systemic imaging with body CT or whole-body PET to screen for malignancy, and CSF analysis covering glucose, protein, cell count, flow cytometry, cytology, microbiology, viral PCR, oligoclonal bands, IgG index, 14-3-3 and total tau, alongside MRI. Where malignancy is suspected but unconfirmed, it suggests repeating the evaluation about every six months.

What can mislead the tests themselves?

CSF 14-3-3 and total tau are proxy markers of neuronal injury and are not prion specific. The US surveillance centre states that many conditions can cause these tests to be positive or elevated outside prion disease, naming Alzheimer disease, frontotemporal dementia, central nervous system infections, stroke, seizures and head trauma. Reported specificity for 14-3-3 in the 2021 review ranges from 40 to 92 percent depending on the control group used.

RT-QuIC is prion specific, with reported specificity of 98.5 percent at the US centre and 99 to 100 percent in the studies reviewed in 2021, and a positive result carrying a predictive value above 98 percent. False positives are described as possible in subarachnoid haemorrhage, encephalitis and stroke. The centre recommends that a positive result in an unexpected clinical setting still be read alongside the rest of the work-up.

What if RT-QuIC is negative and prion disease still fits?

The US centre sets out a sequence. Re-examine the diagnosis, checking whether the presentation suggests an atypical prion disease or another condition, and confirm the standard work-up for mimics is complete. Check the other tests: periodic sharp wave complexes on EEG are fairly specific but often absent, and MRI showing diffusion-weighted hyperintensity in the basal ganglia or in at least two cortical areas supports prion disease. Then consider the atypical forms, which it characterises as fatal familial insomnia with sleep disturbance and dysautonomia, Gerstmann-Straussler-Scheinker disease with early cerebellar or parkinsonian features, and variant CJD with early psychiatric or sensory symptoms, an exposure history and the pulvinar sign on FLAIR imaging. Genetic PRNP testing may be warranted.

Causes of a false negative given by the centre are atypical prion disease types, discoloured CSF, patient age and disease duration. Reported sensitivity is 90.3 percent across all forms of prion disease, so a negative result requires clinical reassessment instead of closing the question.

The 2021 review warns separately about imaging. Its MRI criteria require restricted diffusion in the caudate, caudate and putamen, caudate, putamen and thalamus, or at least two cortical regions, with no subcortical white matter involvement and no isolated restricted diffusion confined to the thalamus.

Where this leads

Brain biopsy is reserved in the 2021 review for cases where non-invasive tests are inconclusive and a treatable alternative is suspected, which is also how the NHS describes its use. The criteria sets that frame all of this are compared in diagnostic criteria, the tissue confirmation route is in autopsy and confirming a diagnosis, and the section index is for clinicians. The plain-language page on rapidly progressive dementia is written for families who have been given that phrase without an explanation.