Once seen, never forgotten – but still a diagnosis that can be easily missed.
Kawasaki disease is reasonably uncommon, and that ‘textbook’ case of a miserable child, five days febrile and bright red all over, ticking all the Kawasaki boxes, is even more uncommon. Here is a short guide to the ins and outs of Kawasaki disease, including some common pitfalls to avoid when diagnosing and managing this potentially devastating condition.
First, some background…
Kawasaki disease is a vasculitis of medium sized arteries (particularly the coronary arteries). It’s the second most common vasculitis in children, after Henoch Schonlein Purpura, and the most common cause of acquired heart disease in childhood.
Its exact cause is unclear, and there are ongoing national research projects investigating this. It most commonly affects children under 6, with a peak at 9-11 months – more severe disease is associated with children under 1 year old at diagnosis. All ethnicities can develop Kawasaki disease, but it is more common in children of Asian origin.
The reason we need to be alert to the possibility of Kawasaki in febrile children is that if not picked up and treated early, it can lead to some horrible complications – the most common of which is coronary artery inflammation leading to aneurysms (more of which later).
How does it present?
Kawasaki disease is a vasculitic disorder, so many of the clinical signs actually show you that the blood vessels are inflamed (most easily seen in mucosal tissue). Not all of the signs need to be there for it to be Kawasaki – as you can see here the onset of each symptom typically happens at a slightly different time:
|Fever (exceeding 38 degrees C)||every day for 5 days
|AND at least 4 of the following 5 features (which may not be present all at once)
|1. Conjunctivitis||bilateral conjunctivitis, without discharge. May be mild.
|2. Cervical lymphadenopathy||(>1.5cm in size) Often one painful, solitary lymph node
|3. Rash||(polymorphous exanthem) May be transient. Check the perineum – can have inflammation and peeling.|
|4. Lips/oral mucosa||erythema, cracked lips, ‘strawberry tongue’, injection of oral and pharyngeal mucosae (NO purulent exudate)|
|5. Fingers / toes||Acute: erythema and/or oedema of palms and soles (usually first week)
Subacute: peeling (14-21 days after onset)
|OR: Positive echocardiogram (at any time) with less than 4 features|
Image: Kawasaki Disease Canada www.kdcanada.org
A full (i.e. ‘textbook’) case of Kawasaki disease is significantly more obvious and easier to diagnose than an incomplete case – these are the ones that get missed, and so here is the really important bit to remember:
INCOMPLETE CASE (especially relevant in infants)
|Fever (exceeding 38 degrees)||for 3 days|
|AND less than 4 features, but diagnosis supported by:|
|1. Lack of alternative diagnosis||(eg failure to respond to antibiotics, no other pathogen) AND/OR|
|2. High CRP and/or ESR or neutrophilia|
|Presence of other clinical features including:|
|1. Irritability||without CNS infection|
|2. BCG scar inflammation
||(document size of scar)|
|3. Other system involvement||CSF pleiocytosis, arthritis, uveitis, pneumonitis, gastroenteritis, myocarditis, gall bladder hydrops, dysuria, sterile pyuria|
A couple of points to reinforce here:
You DO NOT need to wait until day 5 of fever to make the diagnosis. If a child has signs which raise clinical suspicion, a Kawasaki diagnosis can be made on day 3 or 4 of fever. Picking it up early is a winner – treatment can be started to stop the inflammatory process and outcomes are much better (plus you get to look very clever if you’re the one who thought of ‘early’ Kawasaki).
You DO NOT need a ‘full house’ of clinical features. If there is high fever and mucous membrane inflammation – OR persistent fever with raised inflammatory markers – it might be Kawasaki. Discuss with your local Infectious Diseases team if you have a possible case that’s not clear-cut.
What else could it be?
This is a list of differentials (not exhaustive)–
|Infectious||Vascular / inflammatory||Drugs|
|Staphylococcal infection (scalded skin syndrome, toxic shock syndrome)||Systemic onset juvenile idiopathic arthritis||Stevens-Johnson syndrome|
(scarlet fever, toxic shock-like syndrome)
|Polyarteritis nodosum||Drug reaction with Eosinphilia and Systemic Symptoms (DRESS)|
|Measles, herpes simplex, adenovirus, enterovirus||Haemophagocytic lymphohistiocytosis (HLH)|
|Fever in returning traveller – consider ‘imported’ pathogens|
Diagnosing Kawasaki disease depends on excluding other conditions, but it’s important to keep in mind that the presence of infection does NOT rule out Kawasaki disease (which may be a trigger). For example, positive throat swabs with ß-haemolytic strep, or positive ASOT do not rule out Kawasaki. If in doubt, treat for both.
OK, it looks like Kawasaki – what investigations are needed?
There is no diagnostic test for KD – the clinical signs and symptoms are what makes the diagnosis. However, before you start treatment you need at least the following:
- FBC, chemistry (including CRP, LFTs, albumin, U&E) Varicella serology
- Blood culture, ASOT, throat swab, MRSA screen
- Serum save (2 bottles if possible) before IVIG is given
- ECG (changes to look out for – arrhythmias; heart block; signs of pericarditis or myocarditis)
- Consider any other tests that may be needed to rule out differential diagnoses (depending on the clinical picture in each case).
- An ECHOCARDIOGRAM needs to be organised, but this should not delay treatment and can be done after treatment has started.
If the child does have Kawasaki, you might find some of the following abnormalities in the blood results:
- Neutrophils – HIGH
- CRP and ESR – HIGH
- Albumin and haemoglobin – LOW
- Platelets LOW in early disease; HIGH later (second week onwards)
- Liver function deranged
- Conjugated bilirubin – HIGH
- Sodium – LOW
How do we treat Kawasaki Disease?
First of all, if you are suspecting Kawasaki, the child should be admitted to hospital and management should be guided by your local Infectious Diseases team. Treatment should be started as soon as possible – the earlier in the illness the better the outcome. Children with Kawasaki disease are usually given three drugs to start with:
- Aspirin (high dose)
- Intravenous Immunoglobulin (IVIG)
- Proton pump inhibitor
- (Steroids are also sometimes given as part of initial treatment)
Aspirin is given because it is an anti-inflammatory, and prevents platelets from forming blood clots (which can form in small inflamed arteries).
Proton pump inhibitor – for gastric protection, since high-dose aspirin can cause gastritis (as can steroids).
The role IVIG plays is to ‘switch off’ the inflammatory process, minimising damage to the vessels and so hopefully preventing coronary aneurysms from developing. And it’s a really effective drug – reducing the risk of coronary artery aneurysms from 25% to 5%. IVIG does however have some potential risks which you should be aware of. It can trigger haemolysis; so if your patient’s haemoglobin drops after IVIG, check the DAT, blood film and bilirubin, and speak to the ID team.
Some cases fail to respond to first line treatment. Second line options – which would only be used after discussion with an ID consultant – include a second dose of IVIG, adding steroids or infliximab.
What are the possible complications?
Coronary aneurysms – which lead to an increased risk of thrombosis and myocardial ischaemia/infarction. It is also possible for aneurysms to rupture, causing cardiac tamponade. Children who have myocardial ischaemia can present with typical ‘adult’ type MI symptoms – chest pain, shortness of breath or pallor / looking unwell; HOWEVER they can also present with other features such as unexplained crying, tachycardia, abdominal pain or tachypnoea. Any suspicion of MI should be investigated with ECG, troponin, repeat echo (and of course urgent discussion with local paediatric cardiology services). ECG can be normal initially, so repeat if symptoms are continuing.
Myocarditis is common in the acute phase of Kawasaki, and when severe this can cause heart failure, with arrhythmia and ECG changes. Again, involve paediatric cardiology early.
Bleeding – high dose aspirin is a risk for bleeding (especially if excessively anticoagulated). Bleeding can occur after trauma, or even spontaneously. If a patient being treated for Kawasaki develops CNS symptoms, check INR and consider haemorrhagic stroke.
Other possible complications – aseptic meningitis, sterile pyuria, liver dysfunction, hydrops of the gall bladder, arthritis, and pneumonitis.
Once stabilised (and providing there are no acute complications) children are discharged with a weaning course of aspirin. The initial echocardiogram needs to be repeated at two weeks, and again at six weeks, to monitor for coronary aneurysms.
Kawasaki disease won’t always present in the way you expect.
Earlier diagnosis and treatment leads to better outcomes – think about ‘incomplete’ or ‘early’ Kawasaki in your differential diagnosis of a febrile child
Be aware of the possible cardiac complications that may occur acutely and long term.
Disclaimer: This article is not intended to guide clinical management for specific patients – please refer to your local protocols
Dr Katie Knight, with advice kindly given by Dr James Seddon, Paediatric ID Consultant