Current Series: Amyloidosis

John L. Berk, MD: That sort of leads into the next topic here, which is the diagnostic work-up for these patients and the importance and considerations of alternative diagnoses. Jim, are there particular ways we should be going about the diagnosis in an effort to minimize the number of cases that are delayed in diagnosis or misdiagnosed?

P. James B. Dyck, MD: I’m sitting at it from a perspective of a person in a peripheral nerve clinic. Everybody I’m seeing is pretty much coming in with peripheral neuropathy. So my work-up will probably be more intensive and extensive than many people’s. But I would do nerve conductions, an EMG [electromyogram]. Most amyloidosis is going to be an axonal neuropathy. There will be exceptions. Will there be more demyelinating? That will be helpful.

Most of them are length-dependent. I like to do an autonomic reflex screen test showing autonomic involvement. Most people won’t have that at their disposal, but it’s a very good way of showing certain manifestations that would be typical of amyloid neuropathy. I like to do quantitative sensory testing. It shows small fiber involvement, unmyelinated fiber involvement, large fiber involvement. These are all very useful things. I like to do an abdominal fat pad biopsy. And I like to do the genetic testing. I’m also looking for other types of neuropathy. I’m not just keying in on amyloid, I’m trying to find out the cause of their neuropathy in general.

There are neuropathies that can mimic amyloidosis. Diabetes can be a mimic of amyloidosis. As we’ve mentioned, CIDP [chronic inflammatory demyelinating polyradiculoneuropathy] can mimic amyloidosis, and many other neuropathies, early on, can look like amyloidosis. I sort of made this point earlier, but it really has been striking to me that early on in amyloidosis it looks a lot like the idiopathic sensory motor axonal neuropathies that we can see from a myriad of different causes.

John L. Berk, MD: Michael, EMGs. Exactly what nerve fibers are we testing, and is it possible that you can have a normal battery of EMG results when there is early, familial TTR [transthyretin] amyloidosis?

Michael J. Polydefkis, MD: EMG nerve conduction tests, large myelinated sensory, and motor fibers. So the classic distal length-dependent neuropathy. It’s very common for all neuropathies, and also very common for hATTR [hereditary transthyretin amyloidosis]. In addition, there are other types of fibers that can’t be captured with nerve conduction testing. And those are the small unmyelinated nerve fibers. Patients with involvement of those fibers often present with pain, burning, dysesthesias in their feet, and that’s a hallmark of small fiber neuropathy. That can be diagnosed by skin biopsy, quantitative sensory testing, or other techniques. But those are the 2 main ones. At Johns Hopkins [University], we tend to use skin biopsy with intraepidermal nerve fiber density. That can have the advantage in that you could also detect amyloid in the skin. So you have the potential to test 2 things.

John L. Berk, MD: So if a neurologist based in the community is interested in doing skin punch biopsies, how would they get those properly assessed? And then parenthetically, what is the sensitivity of these biopsies for amyloid in proven cases?

Michael J. Polydefkis, MD: There are a number of laboratories that now offer this as a diagnostic test. So I think it’s easy to perform in the community. In terms of the sensitivity of the hit rate that would detect the amyloid, it’s very much dependent on the patient population. So if you have symptomatic patients with hATTR, upwards of 80% of skin biopsies will be positive for amyloid. So in my experience, it’s better than a fat pad biopsy.

Of course, if you look at asymptomatic patients and how you define asymptomatic, the hit rate’s lower. In a group of patients in their early 50s with some symptoms, maybe some tingling but not overt neuropathy, we saw a 20% positivity rate.

John L. Berk, MD: Well, 80% sensitivity is pretty earth shattering, from my standpoint. At Boston University, with extensive experience with the abdominal fat pad aspirates, we get 85% sensitivity in AL [amyloid light chain] cases. But in terms of the TTR cases, the familial ones, it’s about 50% sensitivity and that drops to about 25% in the wild-type TTR cases. So if you have punch biopsy sensitivities of 80%, that really is phenomenal and quite aggressive.

P. James B. Dyck, MD: Is that 2 sites, Michael, or 3 sites?

Michael J. Polydefkis, MD: We typically do 3 biopsies, but the distal length is at the highest yield. And I’m speaking to hereditary disease, so wild-type is much lower.

John L. Berk, MD: Yes. So Jim, what’s your experience with sural nerve biopsies or other nerves that are clearly involved by the neuropathy? Because you would anticipate, naively perhaps, that the sensitivity would be quite high for amyloid deposit detection?

P. James B. Dyck, MD: I think that’s exactly right, and it’s a similar experience to Michael’s. When you have a symptomatic person with neuropathy that you can show on electrophysiological testing and on examination, the success rate is 80%-plus on nerve biopsy. When you have somebody who has a positive genetic test and has anxiety and tingling in their feet, and they’re very worried that they might have it, it’s often very low. It’s often more in the 20% sort of thing. So I’m not very keen on doing a nerve biopsy on somebody who I can’t show has a definite neuropathy.