Maybe nothing is more boring to patients and their families than squabbles among doctors about how to classify diseases. But here goes.
You may have read that PSP is one of the “frontotemporal dementias.” The FTDs are an umbrella category of diseases with deficits involving degeneration of the frontal and temporal lobes. The results are trouble planning, forming new ideas, multi-tasking, obeying rules and adapting to circumstances. Some types of FTD also (or mostly) have problems with speech and language. Yes, PSP includes some of those things to some degree, but unfortunately, that’s only one of many parts of PSP.
The protein aggregating in the brain cells in the various FTDs can be tau, as in PSP, but only in a minority. Even in those few with tau, the distribution of the aggregates is different from that of PSP. The majority of FTDs don’t even have tau – instead, they have the proteins TDP-43, FUS or ubiquitin. So, it has always irked me to hear PSP classified as an FTD.
But now, I’ve got backup: In August 2022, a group of leading neuropathologists published a revised set of criteria for making a diagnosis of PSP at autopsy.
This replaces a set of criteria from 1996, antedating modern methods of tissue staining (necessary for viewing through the microscope) certain observations about the pathology of PSP. Now, here’s the critical part: the new criteria don’t require, or even accept, abnormalities in the temporal lobes in support of the diagnosis of PSP.
The new criteria, called the “Rainwater Charitable Foundation Criteria” for the philanthropy funding the project, are very simple. They require both of these:
- Neurofibrillary tangles or pre-tangles, at least mild in frequency, in two or more of the following regions: globus pallidus, subthalamic nucleus and substantia nigra
- Tufted astrocytes, at least mild in frequency, in either peri-Rolandic cortices or putamen
In English:
- Neurofibrillary tangles: mature aggregates of tau protein
- Pretangles: aggregates of tau protein that aren’t (yet) sufficiently well-formed to be called tangles
- Globus pallidus: part of the basal ganglia, an important area for control of movement
- Subthalamic nucleus: another movement-control area, a cluster of brain cells so-called because it’s just under the thalamus
- Substantia nigra: yet another movement-control area, the one where dopamine is made; It’s also a critical one for Parkinson’s.
- Tufted: containing a type of tau aggregate with a sort of fluffy appearance
- Astrocytes: the main type of glia, which are non-electrical brain cells
- Peri-Rolandic cortices: the folds of the cerebrum running down each side of the brain in front of and behind a long in-folding called the Rolandic fissure. The pre-Rolandic cortex is part of the frontal lobe and serves motor control. The post-Rolandic cortex is part of the parietal lobe and serves the sense of touch.
- Putamen: another movement control area of the basal ganglia
The fact that involvement of the temporal lobe is so mild and inconsistent in PSP as not to merit a place in the new diagnostic criteria should finally put an end to the notion that PSP should be classified as one of the fronto-temporal dementias.
I was gratified to discover recently that the Memory and Ageing Center at UCSF, possibly the leading such institution in the world, now specifically states on its website’s home page that PSP, while sharing some symptoms with the FTDs, is not one of them.
So, why does this matter? Because PSP is sufficiently different from the FTDs that it deserves to be researched and treated on its own. Its sufferers and their families need a type of support not generally relevant to the FTDs. Similarly, those serving the urgent and important needs of the FTD community should not be distracted by efforts aimed at PSP.
PSP Blog 
Dr Golbe: this is interesting. I understand and appreciate that PSP is not FTD. (To my mind, it would also be helpful if it was not referenced as “atypical parkinsonism” because when layfolk (including many therapists) hear that they inevitably think “oh its just a variation of PD” and treat patients accordingly without understanding important distinctions like the much shorter prognosis). For us non-medical family members and caregivers, however, could you explain how this new criteria relates to what is referenced as the “hummingbird sign” and deterioration of the upper midbrain that has previously been referenced as being main area of PSP degeneration. Thanks.
Hi, embreedo:
You’re referring to my post about PSP’s pathological criteria, which apply to autopsied brains. The “gold standard” definition of a disease depends heavily on molecular and cellular abnormalities that can only be ascertained at autopsy — the pathological criteria. On the other hand, a disease’s clinical criteria apply to living patients and use things like the history, the physical exam, blood tests and MRI. The hummingbird sign, of course, is an MRI finding, but even in PSP’s latest clinical criteria, the hummingbird sign is not used. That’s because there are too many false negatives in the early stages of the disease and even a few false positives in cases where corticobasal degeneration mimics the symptoms of PSP. Don’t forget — the areas of damage determine both the clinical and MRI features of a disease, so a disease that mimics PSP’s clinical features is also likely to mimic its MRI features, including the hummingbird sign. Dr. G