Tuesday, October 28, 2008

14-3-3 protein: What is it good for?

The neurology residents and I got into a discussion today about the utility of the 14-3-3 protein immunoassay on CSF fluid. I published a post on this assay many months ago in which I called it essentially useless. Now that I've done some more research on the topic, I've revised my opinion and feel that it is a decent confirmatory test -- but only in those cases in which there is a fairly high pre-test likelihood of prion disease based on observation of an evolving mental deterioration along with some type of involuntary movement or periodic EEG activity. In a prospective study of 110 patients done in The Netherlands (see reference below), it was found that the 14-3-3 assay was 97% sensitive and 87% specific. It must be emphasized that these results are based on a population of patients who have other clinical features that point to a prion disease. Of course, routine CSF tests should also be run on these patients because, as the authors state, "abnormalities of the routine examination of CSF almost exclude CJD as a diagnosis". They go on to say that "conditions associated with increased cell counts or total protein (e.g., meningitis and encephalitis) are known sources of false-positive 14-3-3 test-results. Therefore, it is reasonable to consider other diagnoses first if total protein or cell count is increased in patients with initially suspected CJD."
(Reference: Lemstra AW, et al. 14-3-3 testing in diagnosing Creutzfeldt-Jakob disease: A prospective study in 112 patients. Neurology 2000;55:514-516.)

1 comment:

Anonymous said...

It's not totally useless, but it can't make a diagnosis by itself. When considered in the context of the history, MRI, and EEG, it can help confirm a diagnosis of prion disease or make it much less likely. As always, pathology is the gold standard for making the diagnosis: Donahue JE, Hanna PA, Hariharan S, “Autopsy-proven Creutzfeldt-Jakob disease in a patient with a negative 14-3-3 assay and nonspecific EEG and MRI,” Neurol Sci 2004; 24: 411-413.

Dr. Diamandis develops network to help pathologists interface with AI computational scientists

A neuropathology colleague in Toronto (Dr. Phedias Diamandis) is developing some amazing AI-based tools for pathology and academia. He hel...