This podcast is brought to you by the GOSH Learning Academy.
Emma F: Hello and welcome to Master the MRCPCH. In this podcast, we tap into the expertise here at Great Ormond Street Hospital, giving you an overview of a topic on the RCPCH curriculum. You may be revising for an exam, or just fancy brushing up on a need to know topic. I'm Emma, an anaesthetic registrar and the digital learning fellow at GOSH. Today, I'm delighted to be joined by the brilliant Dr. Alexis Bouvier, who is talking to me about Kawasaki disease, an important clinical topic that also comes up under several areas of the theory curriculum, including cardiology, dermatology, and musculoskeletal disease. We hope you enjoy the episode.
Thank you so much, Alexis, for coming on the show today.
Alexis B: Not to worry. Thank you for having me.
Emma F: Firstly, could I ask, what would you like people to get out of this podcast?
Alexis B: So with regards to Kawasakis, it's about getting an increased awareness of it, so that it's in your mind for part of a differential diagnosis of fever and rash in unwell looking children, and then the features to make a clinical diagnosis of Kawasaki's Disease or Kawasaki syndrome.
Emma F: So I guess to start with I'm just gonna ask the most basic question. What is Kawasaki disease?
Alexis B: So Kawasaki disease or Kawasaki syndrome, you may find it written or talked about in both terms, is also known as Mucocutaneous Lymph-node Syndrome. And it kind of does what it says on the tin. It affects your skin, it affects your mucosa, it affects your lymph nodes. And it was initially described in Japan in the sixties. And essentially it's an acute multisystem vasculitis. We're seeing it a bit more in the UK. Now, whether that's because we're recognizing it, whether that's because it's actually of increasing prevalence and incidence in the UK, is slightly uncertain. But at the moment we're probably in about four to eight per hundred thousand. And when you compare that to the much higher likelihood of getting it in the East Asian population, you're going from four to eight, up to 120 to 250 plus per hundred thousand in a Japanese population. So whilst we are seeing it more, it's still nothing in the remote ballpark of what they're seeing over in the East Asian populations, including in, say, your Hawaiian/East Asian population on that western side of America, for example.
But essentially it's a multisystem vasculitis. There's no obvious cause so far that we know of, but there is a feeling that it's likely a result of some kind of respiratory viral RNA trigger. And that's because we tend to see it more in the winter and early spring, which is the same period of time over the year, seasonal-wise, that we see our respiratory illnesses spike in children. Now having said that, Kawasakis itself is not contagious. Whilst it may come from some kind of viral illness trigger, and those kind of coughs and colds and whatnots are as contagious as they will ever be, Kawasaki's itself is not contagious. However impressive the rash, however impressive the fever, however impressive the unwellness of the child is, Kawasaki's itself is not contagious. And the reason that it plays up so much, and the core pathophysiology of it, is it's a combined and excessive, innate and adaptive immune response. Really, everything, every part of your immune system goes into overdrive, for lack of a better term, and causes you to become unwell.
It tends to be a preschool, definitely under five, likely under two years-old illness. And we see it slightly more in boys, not so much that it would automatically change the way you are assessing children for it, but we see it just a little bit more in boys than we do in girls. And finally, the main reason why we want to think about Kawasakis is because it's the commonest cause of acquired heart disease in paediatrics where 25% of Kawasakis can get some form of coronary or cardiac complications. So we don't want to miss it because we don't want to risk any of the increased mortality or morbidity anymore than absolutely necessary from what they've already got of the illness.
Emma F: So just going back to the epidemiology, you've mentioned that it's more common in younger children, so under fives, more common in boys, and then there's this big discrepancy between our population and the East Asian countries where they see a lot more of it. Are children of East Asian origin within the UK more likely to get it, or are there any other high-risk groups within the UK that are more likely to get it, other than what you've already mentioned?
Alexis B: From the research and evidence I was able to look at previously on this, it referenced specifically your sort of Hawaiian, Japanese Pacific, Japanese and East Asian from that side. But taking that on, I think there would be cause to think about it more in populations of Eastern Asian ethnicity within the UK, considering we've got such a multicultural population.
Emma F: How does a child with Kawasaki disease typically present? Is there a typical presentation?
Alexis B: So this is where we come into it being a clinical diagnosis. There's no one blood test that gives you the diagnosis of Kawasaki's disease. And in fact, there is no one imaging diagnosis that gives you the diagnosis of Kawasaki's Disease, as we'll get onto a little bit later.
Generally speaking, we would teach to think about it if you've got five days in a row of high fevers, so 38 pluses often up to the 39/40s, and that's usually the first symptom is a good going fever. And it's why it's so important to ask about a fever history, of the difference between a parent saying their child feels hot versus actually having checked their temperature and how high it is. And then how many days in a row are they having fevers, as opposed to some days yes, some days no.
So you start off with this high fever, then you often get a rash as the next thing. And it can be a quite non-specific, macular-papular, red, blanching rash, or standard child rash to an extent, especially over the trunk and then, to a lesser extent, over the groin and perineum. This comes on early, and can last for days to weeks, often longer than the fever. You often find that these children have cracked lips, and this is more than just your dry cracked lips that we might see in mild dehydration. These are really uncomfortable-looking, properly cracked lips, plus or minus a red tongue. You can have peeling extremities, what we might call desquamation, and they can also have oedema and or erythema of the digits of your fingers, of your toes that can also be quite uncomfortable. They can get some good-going, impressive cervical lymphadenopathy, often unilateral. And then they can also be quite irritable and miserable and inconsolable, almost that you might see in like a meningitic-type picture.
You can get some other symptoms - diarrhoea, vomiting, abdominal pain, lethargy, some uncomfortable joints, irritated joints, arthralgia, arthritis. And with the abdominal GI perspective, you can sometimes get a little bit of jaundice if there's liver involvement within that vasculitis. But generally speaking, it's about the fever, the rash, the lips and tongue, the extremities, your fingers and toes, the cervical lymph nodes, and just a generally unwell, really unhappy looking child. Those are the main presenting features of it.
Emma F: And other than what you've already mentioned, are there any particular features to inquire about in the history or look for on examination of the child?
Alexis B: So those features are what we use to give ourselves a diagnosis, so those are the main ones to look at. Obviously you would be doing a general examination to see if there was any obvious source of the fever, or source of a rash, or source of any other pain or any other illness that the child might be presenting with. But those are the main ones that you're going to do. With regards to fever and to rashes, you got to be thinking about travel, unwell contacts, what kind of other infectious triggers this child may have been subject to.
But really it's those key features that lead us into the diagnosis, which is a clinical one, whereby you need five days of fever, and four or more of the following features: Discharge - you may see it written or spoken about as non-supportive, or non purulent conjunctivitis, so not the kind of gunky eyes that you might give some chloramphenicol for. Oral mucosal changes is number two, so your red and cracked lips, your red tongue, your red throat, and this is often a really floridly red throat, not your slightly angry throat of a cough, viral upper respiratory-tract infection, tonsillitis picture. Then number three is peripheral extremity changes, so the desquamation, the oedema, the erythema, the tenderness, the pain in your digits, in your hands and toes. Then you've got the rash, which is number four. This diffuse, polymorphic, often quite non-specific, macular-papular type rash. And then number five is the cervical lymphadenopathy. It's often unilateral, often even solitary, usually decent-sized, more than 1.5 to two centimeters, and quite often at, or towards, the anterior cervical triangle.
If you've got five days or more of consecutive good-going fevers, and four of those features out of those five, then you have enough for a diagnosis clinically of Kawasaki's disease.
The other way of diagnosing Kawasaki's disease, you only need three of the features, but you need to get an Echo to show some form of cardiac or coronary abnormalities. Now that's going to be more difficult to do in a DGH where, outside of Great Ormond Street, most of paediatrics is. But that is something that we would always try and get as soon as possible to confirm. And finally, a lot of these children end up being labelled as incomplete Kawasakis. They've got some, but not enough, of the features on either history or examination to fully give you a diagnosis, but they look unwell and there's nothing else to clearly explain their presentation. And we see more and more probably 'incomplete' Kawasakis being diagnosed. And again, whether that's a case of having it on our list of potential differential diagnosis, having it to consider, and therefore being able to think about, "Is this something we need to be looking at? Is this something we need to be treating?"
Emma F: That's a really helpful run through of the diagnosis. In terms of diagnostic tests or investigations, I know you mentioned earlier that there's no one specific test or investigation. What investigations or tests would you perform in a child with suspected Kawasakis?
Alexis B: So if you think that in front of you, you've got an unwell, unhappy looking, hot, red, rashy child, you are gonna be thinking about doing your baseline bloods, your FBC, your CRP, as your two core ones. The FBCs going to likely show you raised platelets and white cells, non-specific for inflammation. It may also show you an anaemia. Your CRP and/or ESR, if you do that, are going to be raised. But again, that's non-specific. And then if you do the rest of your "baseline blood tests", in inverted commas, your U&Es, your LFTs, your bone profile, which people tend to do, the LFTs are probably the one that's going to show you something. You can see raised abnormal AST and ALT, as well as gamma GT, and you can see a low albumin, which often you also see in signs of inflammation or unwell children. So again, non-specific blood tests, which just give you an impression of inflammation, plus or minus infection.
Sometimes children get urine samples, which might show you a sterile pyuria, so white cells without any obvious organism being grown. Similarly, a lot of these more inconsolable, irritable, younger children might end up having lumbar punctures, and you might show signs of aseptic meningitis, so again, lots of white cells, signs of inflammation, but no bugs. And you might see signs of pancreatitis if you do extended abdominal-type blood tests, such as your lipases or amylases, depending on where you work.
One of the best investigations, that sometimes you can't quite get as easily at a DGH as we might be able to here at Great Ormond Street, is an Echo. And the reason for that is that some of the most concerning complications of Kawasakis, are the cardiac ones and the vascular ones. And what we look for are coronary artery aneurysms mainly. You can get aneurysms of the other arteries, but really it's those coronary arteries, that first part of where the blood supply comes back off the aorta into supplying the heart muscle itself. Those are the ones you're looking at. And if you've got that, plus three features within the fever, rash etc. that we spoke about just before, then you can give yourself a diagnosis of Kawasakis.
Other things you can find on the Echo would include pericardial effusions or valvular insufficiencies, so problems with some of your valves, and you'll get different valves impacted in the acute phase versus the late-onset phase. So depending on when you do your Echo, and at what point of the repeat, more surveillance-type echoes you'd be getting into.
You can also get arrhythmias on ECG and signs of myo- or peri-carditis on ECG as well. I won't go into all of that at this point cause that's probably a talk for a cardiology specialist, rather than myself. But realistically you're gonna get an inflammatory, nonspecific infectious picture in your bloods, you might get a myo/peri-carditis, slightly unhappy heart, on an ECG. But it's your Echo that's going to be your core investigation for a suspected Kawasakis, whether incomplete or whether it's got enough for a clinical diagnosis, you'd be getting an Echo anyway.
Emma F: And what proportion of children with Kawasakis have cardiac manifestations that you can see on an Echo?
Alexis B: 25% of them get some form of coronary or cardiac complication. So a normal Echo, ie a negative, absolutely clear and fine, Echo does not exclude Kawasakis. You would just need to have more of your features to have a diagnosis, or sometimes you therefore end up in that incomplete Kawasaki's bracket. But the mortality if untreated is about 2 to 3%, so that's why we have to think about it. Most five day fevers, most 6, 7, 8, 9 day fevers are not going to be Kawasakis. They're gonna be some other viral infection or bacterial infection, or they're still more likely to be a non-Kawasaki rheumatological problem, for example. But if we miss these, there's a two to 3% mortality, which is too much for us to not at least consider it. And consider checking, and consider discussing it with a cardiology specialist, whether they could do an Echo if we felt it warranted.
Emma F: And are there any other conditions that might mimic Kawasakis or present in a similar way, that it would be important to exclude?
Alexis B: Absolutely fever and rash is, if not half, a large chunk of what you'd be seeing in acute, unwell paediatrics, in A+E or in GP-land. Rash and fever - Measles, Scarlet Fever, Toxic Shocks, Scalded Skin, especially the latter two when you start to get into the more unwell looking children. All that infectious differential needs to be considered. You might also consider EBV, which can be tested for with serology, and in the post-Covid era, PIMS-TS, which can present in a Kawasaki's-like way, as one of the three branches of PIMS-TS, but that opens up its whole own podcast in its own right.
And then, outside of the immediate infectious perspective, one other condition that can cause fever and rash and unwell looking children, but is sometimes forgotten about is JIA, or Juvenile Idiopathic Arthritis. If that is presenting with arthralgia or arthritis, so joint pains and joint problems, that might push you more towards a diagnosis of JIA, but it's about doing your thorough, systemic assessment of a child, to look for the features of Kawasakis, but also to avoid missing anything else. For the more thorough your examination, the better chance you are finding the positives and of finding the relevant negatives in an examination as well.
Emma F: Moving on now to management. How do you manage Kawasaki? Are there any treatments available?
Alexis B: The two main treatments for Kawasakis are IV Ig, so intravenous immunoglobulin, and aspirin. The IV Igis given as an anti-inflammatory agent, and to reduce cardiac complications, so you try to give that as soon as possible. You play around with the fact that it's often unavailable out of hours and on weekends, et cetera, but there are ways of working around that. And some children need more than one dose or more than one course of it. The aspirin is both an anti-inflammatory and antiplatelet aggregation, and analgesic and an antipyretic. So quite a bit of effect from one simple, over-the-counter drug.
A lot of these children will end up getting antibiotics, especially in the more incomplete Kawasaki cohort, where they've got a fever, they've got a rash, they look unwell, they've got CRP, they've got white cells, they've got all this, but they don't have enough for a full-blown diagnosis. They look unwell, they get a couple of days of antibiotics, hopefully, pending blood cultures, and being able to stop and rationalise as quickly, and as smoothly as possible.
And then there's talk and research being done as towards adding in steroids as a routine part of Kawasaki treatment, with the hypothesis that it will help to prevent coronary artery aneurysms occurring, or how bad they can get. And in particular, there's a trial being run over the last year or so called the KD-CAP trial, including within centers in North Central London, if nothing else. And so it'll be interesting to see what results they get and whether things like IV Methylpred, I think it is, might become part of our core Kawasaki's treatment, alongside the IV Ig and the Aspirin.
And then, as for duration, you're gonna be guided by your cardiologists with your Echoes, you're gonna be guided by improvement and resolution clinically, in terms of fevers and just general behaviour of a child, general appearance of a child, and your inflammatory markers and whatever other forms of biochemistry had been impacted, like LFTs, for example. You'd want to see those start to improve, but you'd be in daily liaisons with your cardiologists, plus or minus your infectious disease immunologist. Depending on where you work, there may be different people to call as part of a protocol.
Emma F: Would IV Ig be warranted in children with incomplete Kawasaki if there was no manifestations on an Echo that confirmed the diagnosis? What would your threshold be for starting it in those children?
Alexis B: So incomplete Kawasakis is a tricky one. Not all Kawasaki's presentations will have cardiac findings at diagnosis. So just because you have a normal echo doesn't mean it's not Kawasaki's. And then the trick with, or the difficulty with, the incomplete Kawasaki's cohort is exactly that. They look unwell, they have some features of Kawasakis, they may not have any obvious, targeted, specific features or findings to point you towards a different diagnosis. You've got an inconsolable, irritable child with high inflammatory markers, you don't want to miss a possible Kawasakis, so you have to consider whether giving IV Ig and some aspirin for a period of time, as per your cardiologist, would be sensible. And often these children end up getting a dose of IV Ig and then some aspirin. But if they recover incredibly quickly, and there's definitely nothing on the Echo, then those might be the ones that have a slightly shorter course of aspirin, and a slightly shorter period of follow-up surveillance echoes, which all of them will need, than the ones that had clearly clinically-diagnosed Kawasaki's, or they took longer to get better, or they had some form of abnormality on the ECG, on the Echo, or both.
Emma F: And how long are they typically in hospital for, before they start to recover?
Alexis B: So in my experience, you do see improvements within the first 48 to 72 hours I would say. You are probably doing daily blood tests. You are liaising with your cardiologist, and then often in DGHs, the rate-limiting step of being able to discharge somebody is to be able to get an Echo to see whether or not there is a problem, rather than because they are more clinically unwell. Also, you've got to consider that whilst a lot of these children get started on IV antibiotics for query sepsis, query meningitis or such-like, you've then got to wait for the cultures to come back for that, so that can often take you sort of into 36/48 hours before you can start that. So you're probably looking at a few days in hospital on average for these children, and you'd want to be seeing their fever settle, probably 24, if not 48, hours apyrexial ideally. You'd want to see the start of improvement in the inflammatory markers. I don't think I'd keep them until they had a normal crp, but I'd want to see a really good, clear, sustained drop in it. And similarly improvements in the FBC, I'd probably want to see before sending them home. So it's difficult to say an exact period of time or expected length of admission.
Emma F: And what's the long-term prognosis like for these children?
Alexis B: So most of these children do well, but it is still the commonest cause of acquired heart disease. So 25% of them get some form of complication, which means 75% of them don't. They will need long-term or at least medium- to long-term follow up, including Echoes, with cardiology, and it will probably be cardiology that decide at what point they've had enough normal Echoes to discharge them.
Emma F: So just to finish, the questions that we ask at the end of every session. Firstly, are there any classic exam questions that pop up about Kawasaki?
Alexis B: So I think you are most likely to see this in the theory and written papers. Fever, rash, unwell-looking child, nonspecific symptoms. Again, that's just such a core part of acute general pediatrics, being able to make a differential, being able to explain your rationale for making a differential diagnosis, knowing what to look for, in Kawasakis, and then knowing what the initial treatments would be. I think you could definitely see something like that in the theory and written papers.
I do think it would be a reasonable option as a communication station in the clinical part of the exam. For example, an explanation to a parent of the diagnosis, in terms of the features and presentations, it's ticked enough of the boxes or why it hasn't, and you're treating it is incomplete. Why you're gonna treat, what you're gonna treat with. The concerns about morbidity and mortality if a parent, for example, wasn't so keen on IV Ig, and then the need for ongoing follow up. So that could be a diagnosis, explanation, and parental discussion communication station in the clinical exam.
Emma F: And secondly, are there any useful resources that you would recommend?
Alexis B: So everywhere I've worked has had a slightly different guideline for Kawasakis, mostly in relation to what availability they have of imaging, or of IV Ig, or who their local linked-in cardiologists, immunologists, infectious disease specialists are. So I would recommend looking up your local guideline for Kawasaki's disease because you are gonna see a case of possible Kawasaki, or looking enough like incomplete Kawasakis, at some point.
And then there is the Kawasaki Disease Foundation website, that's freely available online, and that's got FAQs and basics both for doctors, in terms what to look for, as well as some information, basic leaflets, for parents that you might be able to give them as part of your explanation of "This is what I think might be going on, this is why I think this might be going on, and this is why I think we need to do this about it".
Emma F: And finally, what are your three takeaway learning points from today?
Alexis B: The first of the key points would be to think Kawasakis if you've got a more than five day consecutive fever. Especially if there's a rash, especially if they look unwell. Most of them won't be, but think about it and be able to then tell your senior and your colleagues, "I thought about it, but I didn't think it was Kawasaki's because they didn't have enough surrounding features, or because they had a clear diagnosis". Know the symptoms and signs. Know what you're looking for in Kawasakis, to be able to say yes or no. So the eyes, the oral mucosa, the extremities, the rash, the lymph nodes, and the fever, so know to look for those.
And, not just for Kawasakis, but the importance of taking a good fever history. So many parents will come in saying that their child has had two weeks of fever, for example. And you say, "Have you ever checked their temperature?", "No. They felt hot every so often". "Have they felt hot every day?" "No." And almost immediately, you can downgrade from two weeks of ongoing fever down to, "Oh, he had a few days on, and then he was better, and then he had a few days of fever, and then he was better. Most of them were 37.5s, one of them was a 38", and you're like, "Okay, so you had one day of fever". So just the importance of taking a good fever history as part of a presentation for half of the children that you're gonna see in A&E, especially over a winter.
Emma F: Thank you. That's been a really useful summary. I really appreciate you joining us today.
Alexis B: A pleasure as always.
Emma F: Thank you for listening to this episode of Master the MRCPCH. We would love to get your feedback about the episode, and get your ideas for future topics that you would like to hear covered. You can find a link to our feedback page in the description for the episode, or email us at
[email protected].
If you want to hear more about the work of the GOSH Learning Academy, you can find us on Twitter, Instagram, and LinkedIn, or visit our website at www.gosh.nhs.uk and search "Learning Academy". We hope you enjoyed this episode and we'll see you next time. Goodbye.
