Atypical FDG-PET

[ExecutiveDysfunction]

Hello,

I’m a neuropsychology fellow looking for some recommendations for literature. We recently saw a patient with slightly reduced inferior basal temporal and cerebellar hypometabolism on FDG-PET concerned about cognitive decline. I’m familiar with patterns suggestive of other dementias but this is out of my wheel house.

Can anyone recommend literature on such a pattern, either degenerative or developmental, that might be consistent with such findings?

---

Members don't see this ad.

 

[deathmerchant]

Hello,

I’m a neuropsychology fellow looking for some recommendations for literature. We recently saw a patient with slightly reduced inferior basal temporal and cerebellar hypometabolism on FDG-PET concerned about cognitive decline. I’m familiar with patterns suggestive of other dementias but this is out of my wheel house.

Can anyone recommend literature on such a pattern, either degenerative or developmental, that might be consistent with such findings?

As far as i know, there is no known condition with just temporal and cerebellar hypo metabolism. Temporal hypometabolism can be seen in AD, FTD, PCA etc. Cerebellar hypo metabolism can be seen with chronic alcohol abuse, antiepileptic use etc. Sometimes secondary cerebellar hypo metabolism can be seen with primary Basal ganglia or frontal disorders.

PSP (Cerebellar variant) could theoretically cause cerebellar hypo metabolism esp unilateral.

Also, hypometabolism can be seen if there is a structural damage like stroke/encephal (?Vascular dementia), calcification etc. So compare with MRI brain and clinical history.

Also fyi, PET scan findings for dementia have low sensitivity and specificity. The images are also limited by technical aspects and reader variability, esp a report that says "mild hypometabolism"

 

[Thama]

Yeah, the utility of FDG PET for diagnosis of neurodegenerative disease is extremely limited. It can be a reasonable supportive piece of info if you already have a high suspicion and the pattern seen is strong, but otherwise I'd say this result has very low diagnostic utility.

Why was PET even pursued here?

 

[ExecutiveDysfunction]

Thanks for the responses.

I agree, these results are mild and definitely not something I want to hang my hat on, just wanted to make sure I wasn’t missing anything.

Referral was for symptoms that could fit with a primary progressive aphasia or posterior cortical atrophy presentation. Our neurologists often like to refer for PET with these cases to shed a little more light on dementia differentials when it’s not as straightforward of a diagnosis.

 

[neglect]

I've grown distrustful of hypometabolism on FDG-PET. Sure, sometimes it fits neatly and CAN add value. But more often it does not. It also tends to fail spectacularly in exactly the cases where you need it most: those with unclear phenotypes (are they normal, are they conversion, are they mild FTD?). But @ExecutiveDysfunction, if the question is one of PPA vs. PCA, you guys were TOTALLY lost to begin with, which also makes me think that pretty much every case is not straightforward for you. Sorry for being truthful, but you get it right? PPA and PCA are very different. PPA will present with a slowly progressive aphasia. PCA with well articulated visual agnosia.

Back to glucose PET: my personal experience is that subtle changes in metabolism can be seen with just about anything. I think I've seen the same abnormalities in cases of worried well and in SNAP. When it is classic, then you'll see classic features. But then you didn't need the test, did you?

My personal bias is to determine proteinopathy always. Is this amyloid and tau? Or is this DLB? Or is this one of the tauopathies? FDG-PET does NOT determine this, but an LP might. If amnestic and fairly early, then enroll into a clinical trial and see what the amyloid PET scan shows. If not, then there's very little to be done anyway aside from trying to treat cognitive problems and n-psych aspects.

 

[ExecutiveDysfunction]

Appreciate the thoughts. For what it’s worth it’s a very odd presentation of symptoms that are fairly subtle, although I don’t plan to discuss here in a great deal of depth, but the symptoms are related to numerical processing which manifests as something that you can either conceptualize as an early manifestation of phonologic processing deficits (logopenic PPA) or an early sign of dyscalculia (among other signs possibly consistent with PCA). Or worried well of course, but the best I can do is consider the possibilities in light of reported decline. I can see how it would appear that I have no idea what the in the world I’m doing based on the brief description though.

To your other points though, I agree the PET is probably superfluous in this situation and doesn’t add much to the picture but as PET isn’t my expertise, I hoped to make sure I was correct in holding this info lightly.

 

[neglect]

Appreciate the thoughts. For what it’s worth it’s a very odd presentation of symptoms that are fairly subtle, although I don’t plan to discuss here in a great deal of depth, but the symptoms are related to numerical processing which manifests as something that you can either conceptualize as an early manifestation of phonologic processing deficits (logopenic PPA) or an early sign of dyscalculia (among other signs possibly consistent with PCA). Or worried well of course, but the best I can do is consider the possibilities in light of reported decline. I can see how it would appear that I have no idea what the in the world I’m doing based on the brief description though.

To your other points though, I agree the PET is probably superfluous in this situation and doesn’t add much to the picture but as PET isn’t my expertise, I hoped to make sure I was correct in holding this info lightly.

That's a concise rendition of why neuropsychology usually serves to achieve paralysis by analysis. You have to just call it. And if you cannot dissociate what you're actually testing with a math question, then you have to look past it to the company it keeps: and call it. And then looking for a glucose PET scan for answers - oh god. When you're lost at sea, you don't ask a fish for north.

But nevermind me. I've grown VERY jaded about neuropsych with the exceptions of a few. Mostly they seem to quibble over nonsense and become overly concerned about minority outreach. Not my gig.

 

[ExecutiveDysfunction]

That's a concise rendition of why neuropsychology usually serves to achieve paralysis by analysis. You have to just call it. And if you cannot dissociate what you're actually testing with a math question, then you have to look past it to the company it keeps: and call it. And then looking for a glucose PET scan for answers - oh god. When you're lost at sea, you don't ask a fish for north.

But nevermind me. I've grown VERY jaded about neuropsych with the exceptions of a few. Mostly they seem to quibble over nonsense and become overly concerned about minority outreach. Not my gig.

Who hurt you?

But seriously, you know very little information about this case and have broadly jumped into generalizations about my and my fields work, based on extremely limited information. If we’re discussing anecdotal data, both of our fields are subject to ineffective practice and having less than definitive answers, but that’s no reason to denigrate the utility of either. Besides, it wasn’t my call to have a PET done, I’m just trying to integrate the data at hand.

As for the topic, the pattern is likely meaningless. As for whether the symptoms are indicative of something, we shall see.

 

[neglect]

Who hurt you?

“He is best who is trained in the severest school.”

Thankfully MANY people hurt me. Some hurt me kindly, some hurt me behind my back, but I love and respect those who told me the unabashed truth to my face. They made me sharper and harder and better. Now I return the favor. You HAVE TO CALL IT.

But seriously, you know very little information about this case and have broadly jumped into generalizations about my and my fields work, based on extremely limited information. If we’re discussing anecdotal data, both of our fields are subject to ineffective practice and having less than definitive answers, but that’s no reason to denigrate the utility of either. Besides, it wasn’t my call to have a PET done, I’m just trying to integrate the data at hand.

As for the topic, the pattern is likely meaningless. As for whether the symptoms are indicative of something, we shall see.

The lesson from dealing with the brain is that we actually know a ton. We know proteins build up in brains and mess them up.

You have a mortal obligation to the patient to give them your best thoughts. Think about every word in that sentence. Your role entails a moral obligation. You are in a position of authority, they are a patient, ie, clueless and at your mercy. You give. You dispense. You are the fulcrum where it tip into different baskets. And you owe them YOUR best thoughts. Not the idiots. Your own thoughts. Your accumulated wisdom. To do so, you have to call it.

 

[Telamir]

Where I trained there was a lot of research into dementia and a lot of emphasis on diagnoses. This may be an unpopular opinion but I always felt that it was kind of useless. The molecular tests are quite expensive, the FDG PET scans are very expensive, relatively nonspecific, and ultimately have no solid impact on management. The difference between AD vs PPA (and its variants) vs PCA in the context of treatment is negligible and families can be out THOUSANDS of dollars in testing between molecular analysis and new imaging modalities. I always felt it was a bit masturbatory.

 

[neglect]

Where I trained there was a lot of research into dementia and a lot of emphasis on diagnoses. This may be an unpopular opinion but I always felt that it was kind of useless. The molecular tests are quite expensive, the FDG PET scans are very expensive, relatively nonspecific, and ultimately have no solid impact on management. The difference between AD vs PPA (and its variants) vs PCA in the context of treatment is negligible and families can be out THOUSANDS of dollars in testing between molecular analysis and new imaging modalities. I always felt it was a bit masturbatory.

Actually think this is the popular opinion.

But to differ: the course of disease with PPA is very different than AD and PCA (a variant of AD). So there's that. While it might not be aweome to know that PPA will pretty much remain in language areas and render one essentially mute, I think this information should be shared in a candid manner if the patient and family want to know.

The tests are only important in clinical trials which enrich for fibrilullar amyloid. Then it really matters. Clinically it doesn't seem to matter at all. The best case the IDEAS trial showed was amyloid knowledge "significantly influenced clinical management of patients," but obviously this was limited to the current meds, which are not useful. Also, those with MCI were given Aricept/Exelon! So it increased off label usage!

The IDEAS trial also looked at economic value, hoping to show that those with + PET scans use LESS healthcare resources. I don't think this will show adn could show the reverse. But what do I know, right?

 

[Telamir]

I actually appreciate hearing some dissent with what I said above. To be honest whenever I expressed something similar I got a high and mighty response instead.

I understand that sometimes we do extensive workups on diseases/complaints we know to be relentlessly progressive or untreatable if only to give peace of mind/an answer to patients, and I get that. I also understand there are differences in progression, much like in LBD, MSA A/C/P, and PD which would be helpful for patients to know. In which patients would you order this workup vs neuropsychologic testing that will say "yup it's dementia", along with the usual workup (B12, TSH, Folate, RPR, etc)?

I also agree with you I think dementia medications leave much to be desired. I remember this trial for the mab to bind amyloid...we had a couple of CAARI induced by the treatment in the hospital and the drug ended up not working anyway.

 

[neglect]

I actually appreciate hearing some dissent with what I said above. To be honest whenever I expressed something similar I got a high and mighty response instead.

I understand that sometimes we do extensive workups on diseases/complaints we know to be relentlessly progressive or untreatable if only to give peace of mind/an answer to patients, and I get that. I also understand there are differences in progression, much like in LBD, MSA A/C/P, and PD which would be helpful for patients to know. In which patients would you order this workup vs neuropsychologic testing that will say "yup it's dementia", along with the usual workup (B12, TSH, Folate, RPR, etc)?

I also agree with you I think dementia medications leave much to be desired. I remember this trial for the mab to bind amyloid...we had a couple of CAARI induced by the treatment in the hospital and the drug ended up not working anyway.

Really highlights the importance of working these things out for oneself, then staying true to the style you’ve developed.

Clinically I agree with every word. I VERY often step BACK and avoid doing a costly WU (sending off paraneoplastic panels and genetic tests and PET scans) when just taking a history and doing a cognitive test reveals the answer. And I’m much more comfortable with uncertain cases once reversible causes have been ruled out, then just focusing on symptomatic management. So there are VERY few cases in which I order massive work ups other than an anatomic head image, B12, TSH, CBC and CMP. And n-psych is best if they actually say “this is a dementia,” but usually they use big words and say nothing and none of it is helpful.

Your honest questions bring out emotionally fraught, ‘high and mighty’ responses from those who are fragile. Why are they fragile? Because at some level they know that they’re up to no good, pulling off a bit of a scam, making mountains out of molehills, making things overly complex and they are panicked that someone wil find out, IMO. They are basically on NIH welfare studying nonsense like disparities in dementia care or APOE/amyloid/dx disclosure and distress or another n=20-50 single site study that goes nowhere. I can’t believe these things get funded, and frankly neither can they. Ask them what happens if their pilot study is +, you get the same answer as you would if it were neg.

In terms of trials, it does matter. They must have amyloid to get into an AD trial. While many antibodies have failed, one looks like it made a difference, called BAN2401. A confirmatory trial is underway and the rate of ARIA (Amyloid Related Imaging Abnormalities) looked manageable.

Also, if you quote me or @ me, then I’ll know it and will respond.

 

[oopsy]

@ExecutiveDysfunction

Not sure about the particulars of the patterns you're seeing, but for the cerebellar hypometabolism, particularly if unilateral, you could consider diaschesis. These are classically opposite the cortical hypometabolism and thought to reflect a pathologic disconnection of the two. I don't do too many PETs in my practice, but have seen it once in the workup of an unusual stroke syndrome. I think the radiologists see it a lot, so I doubt it wouldn't have been commented upon/noticed already, but FYI. Two references below.

Crossed Cerebellar Diaschisis: Three Case Reports Imaging Using a Tri-Modality PET/CT–MR System

Crossed cerebellar diaschisis (CCD) describes a depression of oxidative metabolism glucose and blood flow in the cerebellum secondary to a supratentorial lesion in the contralateral cerebral hemisphere. PET/MR has the potential to become a powerful tool ...

www.ncbi.nlm.nih.gov

Crossed cerebellar diaschisis | Radiology Reference Article | Radiopaedia.org

Crossed cerebellar diaschisis refers to a depression in function, metabolism, and perfusion affecting a cerebellar hemisphere occurring as a result of a contralateral focal supratentorial lesion, classically an infarct. Clinical presentation ...

radiopaedia.org