This site uses cookies to measure how you use the website so it can be updated and improved based on your needs and also uses cookies to help remember the notifications you’ve seen, like this one, so that we don’t show them to you again. If you could also tell us a little bit about yourself, this information will help us understand how we can support you better and make this site even easier for you to use and navigate.

A combination of CSF tau ratio and midsaggital midbrain-to-pons atrophy for the early diagnosis of progressive supranuclear palsy


Borroni, Barbara, Malinverno, Matteo, Gardoni, Fabrizio, Grassi, Mario, Parnetti, Lucilla, Agosti, Chiara, Alberici, Antonella, Premi, Enrico, Bonuccelli, Ubaldo, Gasparotti, Roberto, Calabresi, Paolo, Di Luca, Monica, Padovani, Alessandro


Journal Of Alzheimer's Disease: JAD, Volume: 22, No.: 1, Pages.: 195-203

Year of Publication



Cerebrospinal fluid (CSF) tau ratio decrease (33kDa/55kDa forms) and mid-saggital midbrain-to-pons (MP) atrophy have been suggested as diagnostic markers for progressive supranuclear palsy (PSP). The usefulness of their combined evaluation has never been tested. We evaluated the CSF tau ratio and the MP atrophy as a combined marker for early identification of PSP. A total of 87 subjects, namely 18 PSP, 25 controls (CON), 16 corticobasal syndrome (CBS), and 28 frontotemporal dementia (FTD), were included. Each subject underwent a lumbar puncture and a conventional MRI scan to assess CSF tau 33 kDa/55 kDa ratio and mid-saggital MP measure, respectively. CSF tau ratio and MP ratio were significantly reduced in PSP patients when compared to CON, CBS, and FTD (p< 0.001). Data-based "optimal" combination of CSF tau ratio and MP measure was defined, and the combined marker TrMp= sqrt[3]{CSF Tau ratio}} × {MP ratio} was considered. Considering the combined marker, the difference between the area under the curve (dAUC) of the receiver operating characteristic analysis in PSP versus the various subgroups was higher by about 10% than that obtained by each marker individually. In PSP versus others, a proposed "best" cut-off of TrMP = 0.182 resulted in 94.2% sensitivity and 84.0% specificity. When patients with onset of symptoms ≤ 2 years were included, TrMP resulted significantly decreased in PSP compared to CBS (p< 0.001) and FTD (p< 0.001). The combined marker increases the discriminative power in identifying PSP and suggests that the interplay of different markers should be considered in future trials to enhance diagnostic accuracy from the early stages.;


aged, atrophy, basal ganglia diseases, biological markers, cerebrospinal fluid, diagnosis, diagnosis, differential, early diagnosis, female, frontotemporal dementia, humans, male, mesencephalon, metabolism, middle aged, neuropsychological tests, pathology, pons, supranuclear palsy progressive, tau proteins

Countries of Study


Types of Dementia

Fronto Temporal (also known as Pick’s Disease)

Types of Study

Non randomised controlled trial

Type of Interventions

Diagnostic Target Identification

Diagnostic Targets

Biological Testing