Comment on validation of diagnostic criteria for variant Creutzfeldt‐Jakob disease

There is no doubt that the development of diagnostic criteria has contributed greatly to epidemiological research in prion diseases, and Heath and colleagues emphasize this in surveillance studies of variant Creutzfeldt-Jakob disease (vCJD). We caution, however, against a more broad application in clinical practice, particularly in governing decisions about clinical diagnosis, communication with patients/caregivers, and access to experimental therapies. The physician looking after a young patient with an unexplained rapidly progressive neuropsychiatric syndrome, dementia, or ataxia needs to make prompt clinical decisions. There are treatable alternative diagnoses, and an early firm diagnosis is essential. The pulvinar sign on magnetic resonance imaging is often not identified when patients are first imaged, and a requirement for a clinical duration of 6 months or greater makes a probable diagnosis impossible in the early stages of disease. Physicians who have cared for families affected by vCJD are aware of the complicated psychological issues generated by the perceived mismanagement of the bovine spongiform encephalopathy epidemic, which are often exacerbated by a delay or equivocation about diagnosis. Several families also choose experimental intracerebroventricular pentosan polysulfate therapy, which requires neurosurgery. In the context of these issues, the role of tonsillar biopsy is underemphasized by Heath et al and the criteria. In our experience of 60 biopsies, by far the largest series worldwide, tonsillar biopsy has 100% sensitivity and specificity, at any stage of the disease. Prion protein deposition in the tonsil can be patchy, and at least 20 germinal centers need to be examined. The number examined in 1 French case reported by Heath et al may not have been adequate to avoid a false-negative result. It is notable that of the 6 most recent patients suspected clinically of having vCJD in the United Kingdom, 3 did not meet epidemiological criteria for probable vCJD while alive. Two of these patients would have been misdiagnosed as sporadic CJD according to the updated clinical diagnostic criteria for sporadic Creutzfeldt-Jakob Disease criteria; typical vCJD was diagnosed at autopsy in both. In a third patient, with a heterozygous codon 129 genotype reported by Kaski et al, the pulvinar sign was not thought to be present by all neuroradiologists, and no tissue was examined. It is reasonable to expect that tonsillar biopsy may have made the correct diagnosis in each of these cases. Given experience with transfusion-associated secondary vCJD, vCJD prions are likely to be present in significant titer in human blood, a diagnostic blood test based on detection of the infectious agent is clearly possible in principle, and if technologically achieved, will necessitate a complete revision of how we approach diagnosis in this disease.