Episode 100: Autoimmune encephalitis testing with Tammy Smith

We're gonna get real nerdy today about autoimmune encephalitis and the associated testing. But if this is your cup of tea, you can learn a whole lot more over at the Intensive Care Academy at icu101.com. That's our other project where we do educational videos. Continuing education credits are available. There's a couple more free podcast feeds there. I think you'll dig in. And we just launched recently a neuro lesson on autoimmune encephalitis. Go check it out. Visit icu101.com and for the month of April, 2026, there is a sale. You can get 20% off new subscriptions. If you check out with the promo code APR26, APR26 for 20% off. I see you 101.com. Check it out. On to the show. Hey, everyone. I'm Brandon Odo. And I'm Brian Bowling. And this is Critical Care Scenarios. The podcast is about autoimmune encephalitis. And it's about autoimmune encephalitis. And it's about autoimmune encephalitis. And it's about autoimmune encephalitis. And it's about autoimmune encephalitis. And it's about autoimmune encephalitis. And it's about autoimmune encephalitis. So, here are the two scenarios. The podcast where we use clinical cases, narrative storytelling, and expert guests to unpack how Critical Care is practiced in the real world. Hello, everyone. Welcome back to the podcast. This is Brian Bowling with you. And with me, as always, Brandon Odo. Yeap. Without we talk about a neurology topic today that's really, really neurology. I say that because a lot of neurocritical care is really just for a good critical care. Really nerdy, I think, is the word you meant. Yeah, yeah, a little bit nerdy maybe. And Brandon and I are pretty huge nerds, but we are not smart enough for a topic like this. So we had to get some help. So we reached out and got Dr. Tammy Smith, who is an assistant professor of neurology, appropriately enough, at the University of Utah. Even better, she's in the division of neuroimmunology and autoimmune neurology. She's a clinical consultant at Arup Laboratories as well. So this is right up her alley and very much not up ours. So we're going to talk about a little bit of neuroimmunology and autoimmune stuff. Welcome to the podcast. Thanks for inviting me to join you today. Yeah, what we'd love to explore is autoimmune encephalitis and specifically the process of testing for it. And we've talked about this disease a little bit before with Casey Albin, but the testing for this spectrum of diseases tends to be a little bit baffling to I think to most clinicians. And you have some particular insight because you work with, what do you guys say, ARUP or Arup or? We do say each letter individually. That's what I say. Okay. But yeah, okay, that's so great. This has already been worthwhile. Okay, so let's say you're in the hospital, you're consulting on neurology. There's a 49-year-old female who turns up in the ED there. Pretty healthy history of diabetes, hypothyroidism. But she'd been having about three days of, the family described this sort of some odd behaviors. They have a hard time kind of putting a finger on until they found her unconscious in her kitchen with some twitching movements. They bring her to the ED, she's unresponsive, so she's intubated. She has a bit of a low-grade fever and a leukocytosis. They do a CT of her head, which is unremarkable. So they send her up to the neuro ICU with suspicion for seizures. They apply a continuous EEG. It does show ongoing generalized seizure activity. She's already on some propofol. They escalate it to general anesthetic doses. They load her with levitarastim. And they do a lumbar puncture, looking for things like encephalitis and meningitis. And they send off the usual routine labs. But they also send an antibody panel assessing for autoimmune encephalitis. And in every hospital I've been in, this process looks similar. You go through your electronic orders and you find a little box that sends a panel. And sometimes it's always a send out test. Sometimes you have to jump through some additional hoops to even send the thing. I've never been anywhere that does these in-house. And generally, and you tell me if there's any I'm missing, but these either go to ARUP, which is in Utah, I think, or the Mayo Clinic, which has their own panel. Is there any others that you know about? Yeah. So LabCorp, Quest, Athena, these guys also have their own version of these panels. And those are the major national clinical reference laboratories in the United States. There are a few oddball labs that offer testing for this that I've had patients come to me with from the community that I would not recommend. So either the Mayo Clinic. Joe's antibody to assay. It really is. You know, when people come in with a big list of tests that they're chiropractor ordered. And I just don't really understand where they're getting them tested or the methodology. So I would love to just have a little insight into the tail of these panels, because this is not how we test for most diseases. Usually, you suspect a disease, you send a confirmatory test for that disease. It's not so much this like shotgun approach where you're like, I think it's one of these sorts of things. So we'll send like 30 different assays that test for all of those. I mean, certainly people are a little broad in their testing sometimes, but it's usually not our goal. And yet in this case, we've sort of just given up. And so how did it come to be that this is how we do testing? And then, you know, how did it kind of become centralized to these particular reference labs? Yeah. Yeah. So how did it come to be? I mean, we could go back to like the 1950s when a lot of these antibodies were first being discovered. But what I will say is that in the old days, the first antibodies that were discovered were antibodies that were closely associated with cancer. And so because of that, the first panels that were available specifically tested for these cancer associated antibodies, which are also known as perineoplastic antibodies. But in the case that you described, right, I think you were describing a pretty classic case for what might be an MDA receptor encephalitis, right? A woman, maybe some behavioral changes found down now in status, refractory status, need to be intubated. Certainly there are other antibodies that can cause a similar phenotype. But an MDA receptor encephalitis is probably the most common of those rare antibody syndromes. And so while an MDA receptor encephalitis can be associated with an ovarian teratoma, so some people might put it in that perineoplastic category, most cases are idiopathic or not associated with cancer. So the newer antibodies that have been added to testing panels are not strictly perineoplastic. And so I would say the first step, if you're taking care of a patient in a hospital system and you think you want to send this testing, is to find out what lab you're hospital contracts with, which of those major reference laboratories. Just like healthcare, right, your patient comes to you and says, you know, how much will this procedure cost with my insurance? And we can't tell them, right? And so if you ask me how much an antibody test in any given reference lab, it depends on where your hospital contracts with. So it sounds like one thing you're implying is that it can be expensive. Oh, these tests are some of the most expensive laboratory tests you'll send on your patients. And they take a long time to result, right? I mean, I know, I know often when these get sent out, it takes up to two weeks to result. And so you want to make sure you're ordering the right test. And depending on what lab it's going to, the name of that test might be different. So step one, know where your laboratory testing is going to. And then I would say step two is to look at that laboratory's website and look at their testing options. That's so funny because I mean, people often, I mean, billing is always baffling, but even getting to the right hospital or the right specialist can be a challenge. But now we're talking about sending a test to a whole nother system, which I'm sure can present. I mean, maybe your insurance says, well, your hospital is in our network, but the Mayo Clinic laboratory is not or something like that. I think your labs will all be covered. Yeah, your labs will be covered the same way. So really the question is what the hospital system is contracting with for their lab. I was going to say, so as a clinician, you probably don't really have a choice, do you? Because your hospital lab probably contracts with... You can't shop it around. Yeah, so you can't just be like, well, I've really heard good things about AARUP. So I mean, I wouldn't work with AARUP if I didn't think they did high quality testing, right? They also happen to be right down the street from my hospital. So I can get results on my patients much faster than if I send something to Minnesota. Right. So that's really helpful for me to get high quality test results quickly and affordably. So yeah. What was I? So you can send wherever you want, but you'll have a fight with your hospital administrators because of the expense. Okay. Yeah, there would have to be some reason probably why you shouldn't use the normal place. And I'm sure this is not a good time for shared decision making with families. Like, well, we're thinking of testing for this, but maybe call your insurance and see... I mean, these are really ill patients. So this is probably more just an opportunity for people to think twice if this is something they really want, not just do the click everything kind of situation. Yeah. I mean, I think one of the pitfalls that I often see is that when doing a broad workup for a patient that's found down is, I think people order all the tests upfront, right? In critical care medicine, you've got really sick people and you want an answer quickly. But there are some signs that testing for auto antibodies might be appropriate and there are some signs that they might not be appropriate to send. And I think the first thing I ask all of my trainees is to think about why they're sending the testing before they send it. I don't want it to be a knee-jerk reaction that every patient is getting this testing done because no test is 100% specific. And probably you've seen in your own practice, I know I certainly have this testing might be ordered for a patient who maybe was found down but has known substance use disorder and was having withdrawal seizures, something that's quite clearly one clinical picture. And then the testing comes back two weeks later when they're fully recovered without any other signs or symptoms of an autoimmune encephalitis and they have a very low positive in their serum of one of these antibodies. And now everyone's scratching their heads saying, should we steroid this guy? And that's not appropriate, right? When a test is not perfectly specific, you can get positives that are not related to the clinical phenotype you're seeing. And so sending it in the right clinical context is actually really important. So I mean, the standard like diagnostic reasoning lesson, it would be, well, think harder to get a good history, think about the picture and then test for something specific or maybe a few things. And the lesson I've kind of drawn from these panels is that this is a situation where in real life, most of the time that's just too hard. There's just too many possibilities for any reasonable diagnostician to pick the right test without missing it fairly often. And so this is just one of those, this doesn't apply there. You have to be that guy who's kind of broad. I mean, do you think that's fair to say? Or are there times when you can say, ah, this is classic for a certain antibody phenomenon, let's just send that test? Because oftentimes I haven't even seen those available as individual tests only as part of the panel. So ARIP is one of the few clinical reference laboratories that will allow you to send a single antibody test for any of the antibodies you want to look for. Generally I don't do that unless I'm following up an antibody titer in a patient, which I also don't generally do. But it's nice to know that you can order a single antibody test. It's much cheaper and your turnaround time will be faster if you're looking for one specific thing. But in most cases when your patient is really sick in the ICU, you want to order a panel of antibodies that tests for any antibody that could cause the clinical phenotype you're looking at. So both ARIP and the Mayo Clinic have this matrix, this antibody matrix you can Google search for. I consult it regularly. It's on the shelf right behind me. I keep a print out, but I also pull it up to make sure I'm looking at the most current version online when I send antibody testing. And so for example, if your patient comes in with encephalitis, you want to order an encephalitis panel. If they're in refractory status epilepticus, you want to order an epilepsy panel. If they're having new movement disorders you're trying to understand, then it would be a movement disorder panel. These panels have a lot of overlap. So if your patient has a movement disorder, epilepsy, and encephalopathy, you do not need to send three panels. You want to pick one panel that reflects the clinical phenotype. And so for example, NMDA receptor antibodies can cause any of those phenotypes and they're included in every one of those panels. So you pick the single panel that best fits the clinical phenotype of their patient that you're looking to test for. And we generally recommend that you send that in both serum and CSF. And well, I'll circle back to that. So you should be broad, but not infinitely broad. There are, for all these reference labs, there are some narrowing of the general phenotype of what you're working up that will get you, you know, add a certain test and remove others so that you're getting closer. And like you said, there's a lot of overlap. Now I guess the challenge is that if there's overlap, you don't want to miss something, but you don't want to over test. And that may require you to really look at what tests are being done and to understand which you want to say, oh, there's no way it's that we don't need that one. I definitely want to make sure I cover that. So we'll know. And then for a lot of us who are not true specialists, that may be kind of challenging. Right. I mean, part of the reason we're being broad is because we don't have a deep understanding of all this, not that we know exactly what we want and we're just picking it out. True. Yeah. So I get that. I will say most of these reference labs also offer a very broad panel. I think of it as the one ring to rule them all. So clinicians who do feel in over their head with these autoimmune autoimmune diseases will often pick that panel because they say, oh, it has the most antibodies on it. But the problem is that really broad panel has antibodies unrelated to the clinical phenotype of the patient in front of you. And that's where you get these low positives in antibodies that then everyone can see their chart when they recover and they're looking and they're saying, do I have this autoimmune disease and you're trying to figure out what to do with that low positive. All right. So maybe rather than spending a hundred grand and getting some weird positives, ask somebody for some guidance. Yeah. I was going to say, ARIP, the Mayo Clinic, they have people you can call and talk to if you're trying to figure out the right panel too. So they should have client services available. Is there like a hotline or something that? Yeah, 20.7. Is it 20.7? Oh, OK. Yeah. Wow, that's very interesting. So you can get some help figuring out what exactly. Even if you're in that small community hospital without a sub specialist in this area. Yeah. So you said typically you'll want to send a panel from both the CSF and the serum. And these are different tests. It's not the same test from two sources. They generally have the same antibodies in the matching panel. But the reason you would want to send it from both, one, obviously it might be hard to get a tap right away, right? So you could always get the blood. Yeah. Yeah. If you want to get some something cooking quickly, then the serum is always available. But the CSF is usually more specific for the disease. However, it's not always as sensitive. So the antibody titers will often be higher in serum. So if you have a classic clinical phenotype of a patient in front of you, for example, with LGI1 encephalitis. So that often has a lot of cognitive symptoms and then these stereotyped facial brachial dystonic seizures happening up to hundreds of times a day. And those tend to not respond to ASDs. They respond to steroids. So if you're looking at this patient, you're like, wow, this guy is classic textbook LGI1. And you send that only from CSF and it's negative. I wouldn't be that surprised, right? It can happen. But if you send it from serum, then you might get the antibody that you're looking for and sort of put the whole clinical picture together. That's very interesting. So I would have thought intuitively that the CSF would be more sensitive because these are sort of a phenomenon of the brain, but it's the opposite. It sounds like you sort of generate the antibodies in your blood. And if they're in the CSF, then it's more likely to be relevant. It varies, right? But we do think that there's a peripheral immune process being kicked off, but you're right. The pathology is from the antibody getting into the central nervous system. So I would say there are occasions when we know that there is intrathecal synthesis of these antibodies. So, you know, a cell gets in there and it's getting super activated and pumping them out in the CNS. But part of the LGI1 issue is just that test is not one of the more sensitive tests, the way it's done commercially. And so just kind of knowing that then if you go to the serum, it tends to be a higher titer and then since it's a lower sensitivity test, if it's a higher titer in the serum, you might get the answer you're looking for. Are these tests that all sensitive to timing when they're done in the disease course or treatment? If there's already been some therapies applied? I mean, they're PCRs generally, right? No, they're actually, they're either tissue based assays or they're cell based assays. So they're actually like, um, how would I describe it? So PCR is looking for genetic material from like a cootie generally, right? So the, a lot of them meningitis and cephalitis panels that you send for infectious causes are a PCR. These are actually taking a patient sample that has put all their antibodies and binding them to either a tissue or a cell expressing that antigen and then detecting that antibody using another antibody. A little tricky. So, uh, let's say you had somebody who you didn't get your testing done right away and you started to treat them empirically. Maybe you're doing polsto steroids, Plex things like that. And then you sent off your panel. Is it less sensitive now or you should usually get a result? Of course, you won't always get a result anyway, but you're not much less likely to know. Yeah. Yeah. So, um, if you've Plex them every day that you Plex them, you're removing a bunch of their immunoglobulin, right? That's the role of the Plex. So you're going to decrease the sensitivity for even pathogenic antibodies like these. So, um, it's not that I wouldn't bother to test it. If I found out it hadn't been tested before, but I would make a note in the chart that the patient had already undergone three rounds of plasma exchange. For example, before we discovered that we, we should have sent this and it wasn't sent. Um, and also if you send CSF, despite being Plexed, we think that it's probably going to be retained there better. So if they've been Plexed, maybe choose to send CSF, you know, over serum, you know, re-tap them to get that sent off. Okay. And then the, it should at least be easy to get serum done early, even if for some reason you weren't doing the tap steroids or IVIG shouldn't be such a problem. Well, so steroids, um, I would say steroids, probably not a huge problem. Um, at least initially, um, over time, obviously we think that they would probably suppress the production of those new antibodies. IVIG kind of causes another problem. So IVIG is immune globulin from a bunch of healthy donors that's pooled. And so if you take a whole lot of healthy people, some of them are going to have some of these antibodies floating around just a little bit for whatever reason. Um, and so often if people have gotten a lot of IVIG and then we send this antibody panel, we might get a low positive result for something like GAD 65 that's really common in the general population. And so now you have to interpret that in the context of the patient. Do I think they have a GAD 65 autoimmune disease? Or do I think that this came from the IVIG I gave them? So more chance for the false positive. It's like you're testing the whole population now. No, this is one person. Yeah. And, and you know, the level is going to kind of help impact that too. Right. If it's a lower level, you'll say, okay, this is probably contamination. If it's sky high, then you say, okay, that's probably still part of the patient's immune repertoire. Well, that's everything I want to ask. Generally the results you'll get will not just be qualitative, but quantitative. They'll give you a degree of positivity for most of these. Yeah. Occasionally you'll just get a qualitative result for some technical reason, but you do want a quantitative result from the lab you send to. And I, I'm going to guess that there's no like numeric cutoff that everyone can apply that's like, oh, this is relevant. But is it fair to say that the higher the tighter, the more likely it is to be clinically relevant? We do say that the higher tighter is, is certainly, let's see, if I had a really high tighter result for a patient, but the clinical phenotype was a little iffy, I'm going to really think hard about that tighter. Um, and, and why I got that. And if I can explain it away, if I have a very low tighter, but it's a classic phenotype for that disease, then I'm still going to believe it. So the, the tighter does not necessarily correlate with the symptoms, like the severity of the disease. It often does, but not always. So I've seen patients with NMDA receptor encephalitis with very low titers, um, that have very severe disease. That's a good pearl. Yeah. Ultimately your goal is still just to say they have it or they don't. It's not like, oh, they got real bad disease. They must have really impressive labs. It's the other, um, okay. Yeah. Treat the patient, not the lab, right? So look at how the patients behaving and responding to, to therapy rather than targeting a certain antibody tighter result. It seems like this, this science of identifying these antibodies and the related syndromes is very much still evolving. I mean, I don't know if we could say it's in the early stage, but it's certainly not at the end stage. Like they continued to identify new antibodies and, um, I would imagine they get introduced to these panels at times. Is that fair to say if someone does some science and identify something relevant and they say, we should add it to our test. They do. Um, there's often a lag though. So, um, it's helpful to have colleagues who are trained in auto immunology who are maybe staying current on the most recent antibodies. Um, and then they might know where to send the testing to on a research basis before it gets fully integrated into the panels. So, I mean, this is a little bit of infectious diseases, right? Oh, you know, someone in France has identified a new strain of staff or something like that. I mean, there's just sort of many things that are maybe not evolving in the case of antibodies, but are yet to be discovered. I mean, it certainly seems like there are many times and you'll send these panels and everything will be negative. And I don't know to what extent that is because the testing is imperfect or because the patient may have an antibody we haven't found yet. Is that, I mean, there's probably still many that are undiscovered. Yeah. There's probably many that are undiscovered. Um, and we might find one that's quite common. However, I think we probably discovered most of the more common of these rare antibodies. Um, but I will say that, you know, there was a nice paper that the Mayo Clinic put out of all of their testing that was done by panels and how often the results are positive. So they don't, this is just people sending reference testing to their lab. They don't know anything about the clinical picture of the patients, but 95% of the panels sent to their lab are negative for every antibody tested. And we've done our own internal look at AARP and we've seen very similar numbers. And so I would say that I don't think that 95% of patients have an unclassified or undiscovered antibody. I would say these panels are probably being sent too often in a lot of cases. Sure. So there's probably a little bit of both. People could be a little more selective with who they're testing. And in some cases the test is just not perfect. I mean, is the testing good enough that if a team got this result back and maybe a couple of weeks later and in a sick or highly suspicious patient, they may have been empirically treating this whole time with some aggressive therapies. Can you say they probably don't have anything if the testing was negative or does it depend on the patient? You know, if they still have a lot of symptomatology, you might have to say, well, I think we're missing whatever they have. Whereas if they're good, then all right, they're good. Yeah. I mean, there are criteria for antibody negative autoimmune encephalitis. So if the clinical picture fits with that, with those criteria, that's the gross criteria, I think that paper came out in 2016, I think. Maybe it was more recently. So the gross criteria are a good way to kind of look and say, does this patient have all of these criteria met such that I think that they have an antibody negative autoimmune encephalitis and want to move forward with further treatment? I think one of the most important clinical data points, independent of the antibody testing results is how a patient is responding to initial immunotherapy. So when a patient has no response to first line high dose IV methylpred, plasma exchange and or IVIG, you keep looking at their imaging and it's worse instead of better. Or they're clinically progressing or they're having ongoing seizures despite all your treatments. That's a time to go back to the drawing board and say, did I miss something else? Is it a genetic syndrome? Is it a strange infection? We're overlooking. But if everything, if they're responding to first line immunotherapy and you have collaborating data, MRI findings consistent with autoimmune encephalitis, mild pleocytosis in their CSF, mild elevated protein, maybe a clinical history of the subacute onset of symptoms that are sort of pathodemonic for a lot of these diseases, then you might say, I have enough data, regardless of the antibody results to continue moving down this path of treating empirically for autoimmune encephalitis. How quickly would you expect to see a clinical response to therapy? I mean, is it like the next day they should be getting a little better? Or I mean, there may be a lag, which makes it a little harder to tell. I mean, there's certainly a lag. Steroids, I mean, you've probably seen sort of the Lazarus effect with someone with an autoimmune disease who gets their high dose IV methylpred. Within a couple of days, they often look a little bit better. Certainly people with NMDA receptor encephalitis might not markedly improve on first line treatment, right? And but the first line treatments, I think, is important to recognize, continue to work for a period of time. And often we want to jump the gun and move on to the next level of treatment without giving those first line treatments time to work. So IV methylpred, wrap it on, relatively wrap it off, right? Lingering effect for a few days, something like plasma exchange. When you do five rounds of plasma exchange, those immunoglobulins stay depleted in the patient's blood for up to a month. So at about a month out, they're recovered 80% of their amount of immunoglobulin. So there's probably a long tail of benefit from the plasma exchange. And the plasma exchange also is removing other inflammatory markers or other inflammatory molecules that are contributing to the process. And then something like IV-IG, you probably know, has a half life of about 28 days, depending on the formulation somewhere between 23 and 30. And so that's also going to have a long tail of benefit. And so when empirically treating with patients, even if we don't see an improvement within a couple of days, I usually recommend we give it a week or two to see how they're going before we think that we need to move right to second line immunotherapy. Though we're currently doing clinical trials to try to better understand if early second line immunotherapy, specifically an NMDA receptor encephalitis, improves outcomes. So that's the extinguished clinical trial that's currently enrolling in the United States and then Spain and the Netherlands. And hopefully in a few years, we'll be reporting out on whether early immunotherapy is necessary or whether patients can sort of be watched and allow that first line immunotherapy to work and then make a treatment decision a little further along. We talked about how many of these antibodies are a pair of neoplastic syndromes. They occur in the setting of cancer and neoplasm. I'm curious when you think workup for that is appropriate. In other words, going to look for a source of malignancy in a patient who is not known to have one. Because in some cases, this is the first evidence that they have cancer. Would you wait until you have an antibody that is associated with malignancy and then go look? Or would you look sooner? And then I'm curious what sort of evaluation you think makes sense. Yeah, so it definitely kind of depends on a couple of different things. In pediatric patients, there's certainly less malignancy than in older adults. But let's see, I want to touch on a couple of things there. While some of these are pair neoplastic, those are some of the most rare. So things like NMDA, LGI1, Casper2, GAD65, these are more common. And while they can be associated with cancer, they're not considered like a classic pair neoplastic antibody. So a true near pair neoplastic, you would say it's like there's cancer somewhere if we have to find it. And then some is like there's an association, but it's not like there must be. Exactly. So the classic high risk antibodies for pair neoplastic disease, greater than 70 percent of the time and really probably greater than 90 percent of the time you will find a cancer if they have those antibodies. But those are quite rare. And I would say that I do a basic screen for a lot of patients. Again, it depends on their clinical presentation, but a CT chest abdomen pelvis is certainly reasonable when someone has a condition that might be associated with one of these antibodies. And if that is unrevealing, if I get the antibody result, let's say it's an anti-yo, which is highly associated with gynecologic malignancy, I'm not going to stop looking until I find the cancer. So I'll go as far as a PET scan if we don't see it on the initial screen. For something like KELCH 11, that antibody is really highly associated with testicular cancer, right? So you're going to go looking specifically for that. So you might target your cancer screening based on the antibody. And so I think a basic chest abdomen and pelvis is a good screen if you think there is malignancy. But then if you have an antibody result that comes back, just go to the literature and see what that antibody is closely associated with. And then talk to your body imaging people to figure out what the best study is to really look for that cancer. Yeah. I mean, pets are so hard to get in our sicker in patients. You know, for NMDA receptor encephalitis, especially people talk about, you know, assessing the ovaries or in some cases the testicles and we can do ultrasounds, but, you know, I've heard some talk that you could be microscopic tumors at times and people will talk about blindly removing things or surgically biopsying. I mean, how hard do you look for things like this? Yeah. So for classic NMDA receptor encephalitis, right, about 30 to 40 percent of those patients will have an ovarian teratoma. In the older patients, sometimes we see a lung cancer. So again, I kind of target it based on the age of the patient. In the much younger patients, we often don't find cancer. I will do a screen. And then, you know, one thing to know is that, for example, for an ovarian teratoma, the read might say that there's a compli- we think there's like a cystic component on the left ovary, right? But we don't think it's cancer. If they pull that out and look at it carefully in NMDA receptor encephalitis, it's very often cancer. So I would, in a patient, go after that just like I consider it source control, right? You're always worried about in your infections getting source control. If you have an immune response that's going, hey, why are making someone sick? And you have a lesion, a suspicious lesion. Again, I talk back and forth with radiology. We really discuss this carefully, but I would usually go after that. Now, as far as like a prophylactic orchiectomy, we'll write these tiny micro calcifications that are found in the testicles sometimes. You know, that's a conversation with patients and their families if they're really sick and the antibody is highly associated with testicular cancer. And we can't really find anything on our scans. We would have a discussion about would this help get source control if it's so closely associated with testicular cancer. Yeah. Risk benefit, right? Like if you're really convinced that this is killing them and you don't have another source, even if you're not certain that this will fix them. Yeah. Certainly something to discuss with them. Yeah. It's tricky. It's complicated for sure. Yeah. So the last thing I want to touch on is specific syndromes. And the reason that I find this challenging is because there are just so many of these and it's really not clear to me which, if any, we should have some grasp of as, as generalists. I mean, most people are not you. This is not their realm. It's a pretty specialized area, even for, for people like neurologists who, who kind of dabble in this realm. And there's just certain clinical topics that I'm sort of, I'm a little curious about everything and you don't know much. I'll say, oh, I should learn more. And then you try and you just like, you just like slam into it like a cognitive brick wall. There's just so much there. And the easy answer is to say, well, what's the low hanging fruit? Are there a, a few of these that I should know more about because they're more important to us. They make more likely to be, make patients sicker, more common. And I'm not, that I'm not really sure if that's true. It just seems like there's just a zillion. Maybe an MDA, I mean, we talk about that more, maybe it's a little more common than some and more likely to cause critical illness. But otherwise, I don't know, are there some of these syndromes that you think are important to know? And we can limit this to the realm of critical care because that's our topic today. There might be a different answer in the clinic or somewhere else, but. Yeah, that's, that's a good question. I mean, if we're just talking about frequency, an MDA, LGI one, a GAD 65 and Casper two are the ones that are most commonly seen. If you just look at that, that same paper I mentioned from the Mayo Clinic, that 95% of the tests were negative for everything. If you look at the ones that were positive for something, those are the ones that are most commonly positive. And again, based on age, that the one, and whether it's serum or CSF, that the pieces of the pie that represented by each of those antibodies will kind of change in size. So in young patients, the vast majority are going to be an MDA receptor encephalitis. In older patients, you kind of get a mix of those. But I think it's, it's, it's too much to ask yourself to know every syndrome, certainly. But I think broadly having a framework to think about, could it be one of these diseases? So we think clinically about subacute onset of symptoms. Um, and those symptoms usually involve like behavior changes, um, confusion episodes, maybe changes in speech, changes in wakefulness, either sleeping too much or more often not sleeping at all. So if you get a history that like in the past one week to two months, these behavior changes have happened. Um, that's, that's something that should tickle in your brain. Like maybe this is one of these autoimmune conditions. Um, if the MRI comes back and they show, they say something about those temporal lobes or the radiologists, as it could be one of these encephalitis, maybe like, okay, that's another piece of data to think about. If they're seizing and you can't get control of it, you know, that's one of those things that we often think like these seizures in these diseases tend to respond a lot better to steroids than to classic anti-seizure medications. Um, so if you're like, wow, we can't get control of these seizures, despite all the things, all the tools you have in your critical care, medicine tool, toolkit, keep that in the back of your head. If they have a known history of cancer, that's helpful to know, right? To, to say this could be one of these more perineoplastic rare things. Um, if they have a pleosytosis, some of those, uh, CSF markers that are sent all the time, um, those would be reasons to think about it maybe being autoimmune as well. So I think it's almost, it's too much certainly to learn all these conditions. And it's almost, uh, may, might give you false confidence if you're deeply knowledgeable about those four, because you're still going to miss some of these rare ones. And so I like to think more about a framework of could it be autoimmune? And then what tools can I use to kind of narrow that down? Oh, it's reassuring to know that we could probably think about it in a fairly, you know, broad syndromic way. And then of course sending the tests and hopefully they'll help. And I mean, by and large, is it fair to say that the treatments are going to be similar, at least in the acute course for most of these, if they're, if they're causing the sort of sequela we're talking about, um, maybe there's prognostic implications, you know, how long it's going to last, will it occur? And we talked about some downstream things like a perineoplastic workup, but I mean, are there certain diseases where like, Oh, that's what it is. We have to give them such and such a map or some specific therapy. Right now, no. We certainly, that's the dream, right? In the future is that we have more targeted therapies. But again, I think it's worth thinking back, right? NMDA receptor encephalitis was described less than 20 years ago. Um, so we've, we still don't have a full, fully completed clinical trial for the treatment of NMDA receptor encephalitis. And that's one of the more common ones that you'll see. Um, so we have a lot of work to do going forward, but in general, you know, once you're, once you've done a basic screen for infectious causes, um, slam in them with steroids, uh, and then choosing IVIG or plasma exchange based on, you know, whatever factors in your patient might contribute to that or availability at your facility, right? Cause some, some facilities can't get the plex, um, going reasonably quickly. And so people might go to IVIG first. I, I can get plasma exchange very readily. And I do like to reach for that first, um, as long as a patient doesn't have such bad autonomic instability that I think it's going to be problematic. Um, but yeah, for any of these, so you should think first line immunotherapy before you even know for sure what it is. So that's going to be steroids, plasma exchange and or IVIG. And I think of that as like firefighting one-on-one, right? You have to redo your firefighting trading every year, um, for your hospital. Right. And so I think this is the first step and then you'll figure out what else is going on. Right. Stop the burn. Exactly. I mean, that's why our clinical trial for NMDA receptor and stuff. The latest is called extinguish, right? We're going to put out the fire. Okay. Brian, what else? So, uh, kind of just kind of along the lines of Bremen's last point, um, you know, I, I didn't do your critical care and we're going to big academic medical center and I never see this stuff. Um, now that may be because our NRIC tends to be neurosurgery and vascular neurology and a little bit of status epileptics. It's up thrown in. Maybe these things are going to the MICU. I don't know. Um, but I imagine there's a lot of people out there who are like, yeah, that sounds great. I'll read up on those four things just cause I'm curious, but I'm probably never going to see this. And then one day they will. So besides the, okay, I read up on the low hanging fruit. When do people in a, let's say a non-academic center or community hospital, when do they need to call for help? Is this something that if I'm in, you know, Smith town community hospital and I have a neurologist, is that neurologist going to be able to help me or am I going to have to call the big university center down the road? Are they going to be able to help me? At what point do patients need to get transferred to other centers? Um, things like that. I mean, I imagine that you could probably, if nothing else, listen to this podcast and go, okay, well, I'll do that thing. She said with the workup and when that stuff comes back, I'm going to go, I don't know what to do with this, but then I have something that can call a neurologist for. But at what point do I need to reach out for help? And ultimately, like what point do I need to say, I cannot manage this patient. They need to go up the road to the big neuro place. Yeah. I think everyone has different levels of comfort with these diseases, right? Um, but I would say if a patient, uh, so some of these patients can have severe autonomic dysfunction, right? If that autonomic dysfunction is really challenging to manage in your setting, then it makes sense to get someone with a subspecialty expertise onto that. Um, if you are not seeing a response to your first line treatments and you're considering second line treatments, I think at least having a conversation with an academic colleague, um, to see if you're missing anything, do you have source control? Did you do the right studies to look for infection? Was the testing that was sent the right testing? Um, this is actually one of the biggest mistakes I see in clinical settings is people don't consult that they don't find out what testing they sent. And so they send a panel of antibodies and they think that they have checked for all of the things. And so it comes back negative and it becomes lore in the patients chart that we checked and it was negative. So the first question I have, if a colleague reaches out to me from the community is exactly what testing was sent. Exactly what testing was sent, what lab was it sent to and what antibodies resulted because these, these panels have really goofy names and the names are different at different clinical reference laboratories. And so it's really important. I just want to really stress to make sure that the right testing was sent. And you ask yourself, you know, when the results come back, did that even include NMDA? If I think this is NMDA receptor is the latest because the panel called perineoplastic at the Mayo Clinic and ARUP does not include NMDA because that's not really one of the classic perineoplastics. And so you, you want to strike that word perineoplastic from your brain and try to look at what panel you, you need for the clinical phenotype. Again, whether that's encephalopathy, epilepsy, movement disorders. It's hard. There are not a lot of specialists in this field. I would say I'm never mad when someone reaches out to me with a question. That's part of the reason I am a consultant at ARUP is so that clients can reach out when they're not sure what to do. So while ARUP is a clinical reference lab and most of the people who work, there are pathologists, they recognize the importance of having a neurologist with subspecialty trading available to, for their clients to ask questions to. So I would say just don't be shy about asking questions. I don't know any colleague who would turn you down. So yeah, and the Mayo Clinic, of course, has consultants as well. And I'm not as familiar with the process of lab core questitheta, but really don't be shy about reaching out to colleagues to ask. If we say you're doing everything great, then fine. But I think it's one of those cases to not be shy and not say, oh, we can handle this. We know what's going on. Just just have a conversation with a colleague. Well, like you said, apparently we can call ARUP and they can at least give some guidance on on testing. Yeah. The other time you might want to maybe transfer a patient, obviously, would be therapy too. Like if you want to do plasma exchange and you don't have that available in your center. Exactly. Yeah. I mean, the tricky thing is I work at these academic centers. So I forget what the resources are and or are not in some of these community settings. So for sure. Yeah. Transfer surgery, right? Like if you don't have the right surgery imaging modalities to look for malignancy, if you're concerned for that. Continuous EEG monitoring and a patient who's not waking up, right? That's another reason that I see patients come to our center from the community. I mean, even a lumbar puncture in some of the smallest hospitals. Yeah. Yeah. And then comfort with the use of strong immunosuppressive medications, right? So not everyone's comfortable with using even rotoxamab or cyclophosphamide as their second line therapy. So I think it's it's helpful to get someone who can manage that and make sure the patient has the appropriate screening done before you hit them with that. All right, Tammy, well, let's call it a wrap there. Thank you so much for help with it. I think this has shed some light on a sort of obscure topic. And it's just going to be still a bit of a scare in a lot of ways, but at least we understand a little bit of the man behind the curtain. I guess the one point I just want to reemphasize for folks is it's so easy in the ICU to get used to just like, oh, this patient's here because they're having seizures. Like your whole history is what they saw in the ED. If you at least take like two minutes, talk to a family and just ask a little bit about what's going on. You may get some real clues to what has been happening that can lead you in some of these directions. I mean, you don't have to be Sherlock Holmes here, but literally ask like two questions. I mean, I know you've already forgotten in school how you do HMPs because all our patients are unconscious, but like the barest effort is sometimes versus just like staring at the chart and be like, oh, the white counts elevated. It could be anything. I mean, that's all we've got. But thank you for your help. Remind everyone, this is just our own opinions. We don't represent any of our institutions here. And no specific clinical advice, just some general educational guidance. I'll talk to you next time.