All right. Thank you, Dr. Lee. Thank you to my co-presenters for the opportunity to be up here with you today. So I just want to start with a show of hands in the audience. How many of you have taken care of a Stevens-Johnson's patient? Excellent. Okay. How many of you have taken care of a patient who was called Stevens-Johnson's but was not actually Stevens-Johnson's? Fantastic. Okay. So my people. So again, I'm Laura Johnson. I am an Associate Professor of Surgery at Emory and really a burn surgeon at heart. So that's going to inform some of the discussion that we have this morning. I have yet to develop a financial disclosure relevant to this topic, but if you have an idea for one, please let me know. Just a brief word on how long this has been a problem in medicine. So the first described cases of Stevens-Johnson's were actually described by a surgeon and a pediatrician in 1922, and you can really see that it was not just the skin component of that problem. It was skin and stomach mucosa and the ocular involvement, and it took another 30 years until a dermatologist actually described the other end of this spectrum, which is toxic epidermal necrolysis. But it's taken time for us as a group of physicians to really understand where this body of disease comes from, and I still don't think we have a great answer for it, which I'll mention in a minute. From a diagnosis and definition standpoint, Stevens-Johnson's TEN really depends on the extent of the body surface area that's involved in the disease process. So anything less than 10% involvement is going to be Stevens-Johnson's. Anything greater than 30% is going to be toxic epidermal necrolysis, and anything in the middle, that 10% to 30%, is going to be thrown into this category of Stevens-Johnson's TEN overlap. You can see that pictorially represented on the picture to your right, but it's really not often that obvious when they first show up in the hospital. It is a very rare presentation, right? Under 10 patients per 1 million that we're going to see are going to present with this diagnosis, so there are some chronic infectious disease subpopulations who have a higher incidence of this problem, and it is unfortunately a particularly morbid problem when patients show up. If you look at the SCORE 10 screening criteria for what actually contributes to mortality, though, you can understand why. If a patient is older, if they have a hematologic malignancy or other malignancy, and they have a large surface area involved, they're already set up for a significant mortality based on those pre-existing problems. We think that this is a type 4 hypersensitivity reaction, as Dr. May has already presented, and there's clearly some evidence that activation of the FAS ligand is causing this keratinocyte apoptosis. There's a lot of literature suggesting that, and by a lot, I mean the people who have looked at this specifically, have found a significant number of cytotoxic CD8 T cells in the blister fluid that is presenting in these patients, and so this is driving a lot of the current thoughts on systemic interventional therapy. But there are still subgroups of patients presenting with Stevens-Johnson's who don't necessarily meet these criteria. The inciting agents for this, again, the history is so critical when we evaluate these patient populations, and unfortunately, it's internationally variable how these patients present. So whereas the European literature sees a preponderance of allopurinol as the inciting agent, there's genetic predisposition to Stevens-Johnson's in the Asian population relative to exposure to carbamazepine, and in the United States, you can see a lot of cases reported mainly with antibiotic exposure. The duration of the exposure to agent is also going to inform somewhat of the chance of developing Stevens-Johnson's, so there are some drugs that only require a short exposure to actually prompt this response. Those tend to be your antibiotics, whereas there are other agents that you can have longer term exposure to, some of the mood-stabilizing agents, the anti-epileptics, where it's not necessarily the immediate exposure, but often that first two-month interval after initiation of the drug, or a rapid change in dosing. So you've been on it for two months, the patient's been on it for two to three months, and then their primary provider changes the dose. That's when you're going to see some of this reactive change. So what does it present as? And here is, again, part of the diagnostic challenge in dealing with Stevens-Johnson's in TEN is it presents like an upper respiratory tract infection. People have a low-grade fever, they have a sore throat, sometimes their eyes are dry, and so they go to their primary care provider and are put on antibiotics. Seven to ten days later, they present with a rash, and so it's often unclear whether that antibiotic is going to be the inciting agent, or whether, in fact, this was a presentation of a problem that preceded the antibiotic. Typically the rash is going to show up trunk and face first, though it can sometimes show up on the palms and the soles of the feet. And we've all heard about this Nikolsky sign, which you see being demonstrated in the video here, where tangential mechanical pressure on these areas of erythema start to evidence the epidermal-dermal split that is the real problem in the management of these patients. So what does it look like when it shows up? I think many of you have said that you've seen these patients before. These are examples of how these patients are going to present, certainly some of the more extreme examples that we've seen. But you can really understand this description of a violacious central core with ultimately blistering and this visible, you know, often healthy dermis underneath. We do need to be careful as we are managing patients with different levels of pigment in their skin that we're very sensitive to the fact that sometimes these present differently than the classic textbook teaching. And so here you see patients with a higher Fitzpatrick score from a pigment perspective looking at these rashes, but ultimately everyone loses their epidermis and the underlying dermis is seen as you see in that bottom picture. So what do we do about this? And really the first step is to stop the inciting agent. Often for our patients this means we have to stop every drug that they're on while we're working through the history to identify which temporally is most likely to be the problem. If you don't have an inpatient dermatologist, I would strongly recommend you advocate for one. Many of us have dermatologists who have been with us a long time and are expert in their own right, but there is a subspecialty now of inpatient dermatology and they are invaluable in the management of hospitalized patients. Both the European group and the American group have actually put out consensus guidelines on whether or not, on how to manage these patients. Much of what I'll be talking about subsequently comes from those recommendations. It starts with a multidisciplinary team. Not a surprise to anybody in critical care, that multidisciplinary team though needs to include a dermatologist to make sure that we're really addressing the skin components well. And I write up their experts in complication of complex epidermal loss. My bias as a burn surgeon is that that's going to be a burn surgeon or one of my colleagues around the country, but there are many people in critical care who have spent a lot of time thinking about that, who aren't necessarily in burn centers and can provide valuable insight. It certainly includes nursing, nursing therapists, pharmacists who are used to dealing with patients with these complex changes in, again, the largest organ system in the body. And so really that multidisciplinary approach is critical. We've known for at least 20 years that early referral decreases mortality and certainly early admission to the ICU decreases mortality. Again, I think this speaks to the expert nursing care that you get in these places. And both of the papers that I cited specifically speak to one-to-one nursing care, at least in the acute phase of these patients' presentation. So why do burn centers get these patients a lot? And while this isn't a resuscitation curve for a Stevens-Johnson's patient, I put it up there to remind us that at a meeting where we're also talking about papers on low-volume resuscitation for sepsis, that this is a patient population that actually gets a lot of crystalloid volume during their acute presentation phase. Running 800 to 900 cc's of fluid an hour is Tuesday in a burn unit, but not necessarily something that we typically do in other parts of critical care. And so it's important to recognize that this is a patient population that's really going to need aggressive fluid resuscitation, focused ultimately on effective end-organ perfusion. One of the other places where we really can help is in managing nutritional goals for these patients. We know from the burn literature that changes in resting metabolic rate are substantially increased off of our baseline, anybody in this room, and even our normal ICU admissions. This is old data, but it's held up over the years. And so really understanding that and making sure that we're aggressively repleting patients' nutritional requirements is critical. Sometimes it is literally the thing keeping the patient in the hospital because they cannot take enough oral intake without supplementation through nasogastric feeding to be able to leave and heal their wounds. So this is a critical part of what we do. Finally, you know, pain control is something that burn surgeons deal with a lot. We really are looking to aggressively maintain the functional capacity of our patients. And so we get them up moving, walking through the ICU as often as possible, as frequently as possible, and as soon as possible. And we do this by looking at multimodal pain therapy. So again, a topic that is increasingly discussed in medicine. In this particular subset of patients, though, we have to be very careful to avoid NSAIDs because there is a subset of Stevens-Johnson TEN patients who are triggered by exposure to an NSAID. So what do we actually do about the skin? And here you'll find, you know, often dermatologists and burn surgeons will present pistols at dawn because there's differing approaches to what we actually want to do to these patients. You see this reflected in the survey data in the supplements from one of these studies from the dermatology groups. There's a variety of approaches to how people think the skin should be managed. But ultimately, the goal is to keep the patient clean, minimize infection, to limit their fluid losses, and to really optimize patient comfort so they can move through the healing process as effectively as possible. Almost every center who manages these patients will have their own combination of debreeding the blister, non-debreeding the blister, topical versus non-topical therapies for this. And it really doesn't matter. To my mind, the important thing is that it is consistent and repeatable and makes sure that it addresses those previous three components of skin care that I mentioned. I will note there's really no role for prophylactic systemic antimicrobials in this patient population. And again, I go back to the burn literature where the organ at risk is the organ we're staring at. So we can do antimicrobial therapy topically in a way that is much more effective than anything that these patients are going to get systemically. If these patients do present with an infection, though, it's infections that we see pretty commonly in other patient populations who've lost their skin, namely Staph aureus and Pseudomonas. And in patients who have persistent non-healing lesions, again, in the perioral region or the periuregenital region, we need to be thinking about other reasons for them to have persistent open wounds. What else is involved, though? It's not just skin as we think about it. It's every surface that has keratinocytes. So one of the most commonly involved additional surface areas is going to be the eyes. And I'll tell you that vision loss is the number one cause of lawsuits after patients have Stevens-Johnson's TEN. So vision loss is something that we can impact by early and aggressive involvement of our ophthalmologic colleagues. And really, how it presents doesn't necessarily determine how they're going to end up long term. But the more aggressively and early we can get that consultation, the more likely we are to have a positive impact for these patients' long-term vision. It's important to remember that eye care alone is a reason for these patients to be in a monitored setting. If you are giving someone eyedrops every hour, which, depending on the grade of the injury, these patients could actually need, they are going to need to be in a monitored setting for the nursing care alone. And then there's other interventions that our ophthalmology colleagues can perform, including amniotic membrane and other devices that will help, again, temporarily seal the wounds in a way like we do with the rest of the skin and really help these patients optimize their vision outcome. So what else has keratinocytes on it? The mouth and the urogenital space, which are not spaces we spend a lot of time thinking about in the critical care world. It's really important to make sure that, again, for pain control reasons, these patients have good anesthetic agents mixed in with their oral care and that we really work on consistent, persistent oral care so that they're, even though it may hurt, they're staying clean. As one of my gerontology professors used to say a long time ago, the mouth is the gateway to the GI tract. So if the mouth is unhealthy, the rest of the GI tract is unhealthy. Additionally, urogenital care is something that needs to be addressed early and aggressively. All patients who have problems with urination need to have a Foley placed, and if the question is how long does the Foley stay in place, a great metric is, is the mouth healed? Because if the oral mucosa of the mouth is healed, then it's probably time for the Foley to come out. In women, we need to be also thoughtful of the vaginal mucosa, and we need to be thinking about how can we best optimize patient outcomes for minimization of infertility and scarring. And that's, again, early consultation with our gynecologists. So that's systemic care, right, and supportive care is a very active process in the management of patients who have lost their skin. But what else can we do? What can we do to minimize, perhaps, the duration or the severity of the damage that's done? And unfortunately, again, given the relatively rare incidence of this disease, there's not been a lot of randomized prospective studies looking at interventions. But ultimately, at the moment, the best we know is that the following four agents are actually helpful. So we know that steroids, short-course, high-dose steroids, can actually decrease the severity and the duration of open wound. There's data on the use of IVIG. There's data on the use of TNF-alpha inhibitors, or TANRCEPT is actually being studied currently in a large multi-center prospective trial run out of Vanderbilt, looking to see whether or not this actually makes a difference. And then the use of cyclosporine has been well-reported in the literature. Again, I want to be clear, by evidence and well-reported, I'm saying that there are more than case reports supporting the use of these agents. And really, we're still looking for what is the best combination to promote, again, optimal safety for our patients. I will also note that the dose ranges up here are essentially an average of what's reported in the literature. You'll find smaller doses and higher doses of some of these agents. So when you go looking for the literature, again, it's important to pick a dose that your team feels comfortable with and that can be followed consistently. Inevitably, of course, there were going to be reports of Stevens-Johnson's after COVID. So I just want to assure everyone that, yes, in fact, it happened. And most of them were, in fact, related to COVID itself rather than the treatment. So it's worth noting that there were no reported cases of Stevens-Johnson's after exposure to the Johnson & Johnson vaccine. And then finally, diagnosis, right? Identifying this disease early means that we can potentially engage in our patient's care and stop some of these more systemic problems. And so it's interesting to look at where computer machine learning is moving. And there's actually a nice work looking at, again, deep neural networks for identification of Stevens-Johnson's. And you can see substantially better assessment than even our board-certified dermatologists using this model. So look for this coming to an iPhone near you soon. Finally, we need to think about the long-term implications of this disease, right? It's not just they're in the ICU and then they're done. There are many patients who have long-term ocular sequela of this. Infertility and sexual dysfunction can be a problem plaguing these patients forever. And there's a substantial percentage of patients who end up with depression, anxiety, and PTSD. Because let me tell you, once a patient has a diagnosis of Stevens-Johnson's or TEN, many providers in the community are afraid to put them on any medication ever again. And so they often spend years searching for a physician who will take care of them and manage their chronic health conditions. So as a community, this is really an opportunity for us to own this upfront. Again, with our dermatology colleagues to identify the inciting agent and to provide longitudinal care that keeps this into account. These are my references. And again, I want to thank all of you for being here and listening.

Stevens-Johnson syndrome

SJS

skin reaction

mucous membranes

type 4 hypersensitivity reaction