2022 Corticosteroids for the treatment of Kawasaki disease in children

Cochrane
Library
Cochrane Database of Systematic Reviews
Corticosteroids for the treatment of Kawasaki disease in children

(Review)
Green J, Wardle AJ, Tulloh RMR
Green J, Wardle AJ, Tulloh RMR.

Corticosteroids for the treatment of Kawasaki disease in children.
Cochrane Database of Systematic Reviews 2022, Issue 5. Art. No.: CD011188.
DOI: 10.1002/14651858.CD011188.pub3.
www.cochranelibrary.com
Corticosteroids for the treatment of Kawasaki disease in children (Review)

Copyright © 2022 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Cochrane Trusted evidence.
Informed decisions.
Library Better health. Cochrane Database of Systematic Reviews
TABLE OF CONTENTS
ABSTRACT..................................................................................................................................................................................................... 1
PLAIN LANGUAGE SUMMARY....................................................................................................................................................................... 2
SUMMARY OF FINDINGS.............................................................................................................................................................................. 3
BACKGROUND.............................................................................................................................................................................................. 5
OBJECTIVES.................................................................................................................................................................................................. 6
METHODS..................................................................................................................................................................................................... 6
RESULTS........................................................................................................................................................................................................ 9
Figure 1.................................................................................................................................................................................................. 10
Figure 2.................................................................................................................................................................................................. 12
Figure 3.................................................................................................................................................................................................. 13
DISCUSSION.................................................................................................................................................................................................. 15
AUTHORS' CONCLUSIONS........................................................................................................................................................................... 16
ACKNOWLEDGEMENTS................................................................................................................................................................................ 17
REFERENCES................................................................................................................................................................................................ 18
CHARACTERISTICS OF STUDIES.................................................................................................................................................................. 22
DATA AND ANALYSES.................................................................................................................................................................................... 42
Analysis 1.1. Comparison 1: Corticosteroids versus no corticosteroid use, Outcome 1: Incidence of coronary artery 43
abnormalities........................................................................................................................................................................................
Analysis 1.2. Comparison 1: Corticosteroids versus no corticosteroid use, Outcome 2: Incidence of any serious adverse effects 43
attributable to the administration of corticosteroids.........................................................................................................................
Analysis 1.3. Comparison 1: Corticosteroids versus no corticosteroid use, Outcome 3: Mortality (all-cause)................................. 44
Analysis 1.4. Comparison 1: Corticosteroids versus no corticosteroid use, Outcome 4: Duration of clinical symptoms: fever and 44
rash.........................................................................................................................................................................................................
Analysis 1.5. Comparison 1: Corticosteroids versus no corticosteroid use, Outcome 5: Time for laboratory parameters to 45
normalise: CRP and ESR (days)............................................................................................................................................................
Analysis 1.6. Comparison 1: Corticosteroids versus no corticosteroid use, Outcome 6: Length of hospital stay............................ 45
Analysis 2.1. Comparison 2: Subgroup: single, pulse dose corticosteroid use versus longer course of corticosteroids, Outcome 46
1: Coronary artery abnormalities.........................................................................................................................................................
Analysis 3.1. Comparison 3: Subgroup: geography, Outcome 1: Coronary artery abnormalities.................................................... 47
Analysis 4.1. Comparison 4: Subgroup: high-risk scores versus lower-risk scores or all participants if risk score not calculated 48
in study, Outcome 1: Coronary artery abnormalities.........................................................................................................................
Analysis 4.2. Comparison 4: Subgroup: high-risk scores versus lower-risk scores or all participants if risk score not calculated 49
in study, Outcome 2: Duration of clinical symptoms..........................................................................................................................
ADDITIONAL TABLES.................................................................................................................................................................................... 49
APPENDICES................................................................................................................................................................................................. 49
WHAT'S NEW................................................................................................................................................................................................. 54
HISTORY........................................................................................................................................................................................................ 54
CONTRIBUTIONS OF AUTHORS................................................................................................................................................................... 54
DECLARATIONS OF INTEREST..................................................................................................................................................................... 54
SOURCES OF SUPPORT............................................................................................................................................................................... 54
DIFFERENCES BETWEEN PROTOCOL AND REVIEW.................................................................................................................................... 54
NOTES........................................................................................................................................................................................................... 55
INDEX TERMS............................................................................................................................................................................................... 55
Corticosteroids for the treatment of Kawasaki disease in children (Review) i

Informed decisions.
[Intervention Review]
Corticosteroids for the treatment of Kawasaki disease in children
Jessica Green1, Andrew J Wardle2, Robert MR Tulloh3
1Children's & Adolescent Services, Whiston Hospital, St Helens and Knowsley Teaching Hospitals NHS Trust, Prescot, Merseyside,
UK. 2Cardiology, Hammersmith Hospital, Imperial College London, London, UK. 3Congenital Heart Disease, Bristol Royal Hospital for
Children and Bristol Heart Institute, Bristol, UK
Contact: Andrew J Wardle, [email protected].
Editorial group: Cochrane Vascular Group.

Publication status and date: New search for studies and content updated (no change to conclusions), published in Issue 5, 2022.
Citation: Green J, Wardle AJ, Tulloh RMR.Corticosteroids for the treatment of Kawasaki disease in children. Cochrane Database of
Systematic Reviews 2022, Issue 5. Art. No.: CD011188. DOI: 10.1002/14651858.CD011188.pub3.
ABSTRACT
Background
Kawasaki disease (KD), or mucocutaneous syndrome, is the leading cause of childhood-acquired heart disease in high-income countries.
There is much controversy on how best to treat children with KD and in particular who may benefit from additional treatment beyond
the standard intravenous immunoglobulin (IVIG) and aspirin, such as the addition of corticosteroids. This is an update of the review first
published in 2017.
Objectives
To assess the impact of corticosteroid use on the incidence of coronary artery abnormalities in KD as either first-line or second-line
treatment.
Search methods
The Cochrane Vascular Information Specialist searched the Cochrane Vascular Specialised Register, CENTRAL, MEDLINE, Embase, CINAHL
and two trials registers to 8 February 2021. We searched the reference lists of relevant articles for additional studies.
Selection criteria
We selected randomised controlled trials involving children with all severities of KD who were treated with corticosteroids, including
different types of corticosteroids, different durations of treatment, and where corticosteroids were used alone or in conjunction with other
accepted KD treatments. We included trials using corticosteroids for both first- and second-line treatment.
Data collection and analysis

Two review authors independently selected studies, assessed study quality and extracted data using standard Cochrane methods. We
performed fixed-effect model meta-analyses with odds ratios (ORs) or mean difference (MD) with 95% confidence intervals (CIs). We used
a random-effects model when there was heterogeneity. We assessed the certainty of the evidence using GRADE. The outcomes of interest
were incidence of coronary artery abnormalities, serious adverse events, mortality, duration of acute symptoms (such as fever), time for
laboratory parameters to normalise, length of hospital stay and longer-term coronary morbidity.
Main results
This update identified one new study, therefore the analysis included eight trials consisting of 1877 participants. Seven trials investigated
the use of corticosteroids in first-line treatment and one investigated second-line treatment. The trials were all of good methodological
quality.
Corticosteroids for the treatment of Kawasaki disease in children (Review) 1

Informed decisions.
On pooled analysis, corticosteroid treatment reduced the subsequent occurrence of coronary artery abnormalities (OR 0.32, 95% CI
0.14 to 0.75; 8 studies, 986 participants; moderate-certainty evidence), without resultant serious adverse events (0 events; 6 studies,
737 participants; moderate-certainty) and mortality (0 events; 8 studies, 1075 participants; moderate-certainty evidence). In addition,
corticosteroids reduced the duration of fever (MD −1.34 days, 95% CI −2.24 to -0.45; 3 studies, 290 participants; low-certainty evidence),
time for laboratory parameters (erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP)) to normalise (MD −2.80 days, 95% CI
−4.38 to −1.22; 1 study, 178 participants; moderate-certainty evidence), and length of hospital stay (MD −1.01 days, 95% CI −1.72 to −0.30;
2 studies, 119 participants; moderate-certainty evidence). None of the included studies reported long-term (greater than one year after
disease onset) coronary morbidity.
Authors' conclusions
Moderate-certainty evidence shows that use of steroids in the acute phase of KD can be associated with reduced coronary artery
abnormalities, reduced inflammatory markers and shorter duration of hospital stay when compared to no corticosteroids. There were
no serious adverse events or deaths reported with or without corticosteroid use. Low-certainty evidence shows use of corticosteroids
can reduce duration of clinical symptoms (fever and rash). None of the included studies reported on long-term (greater than one year
after disease onset) coronary morbidity. Evidence presented in this systematic review agrees with current clinical guidelines on the use of
corticosteroids in the first-line treatment in KD.
PLAIN LANGUAGE SUMMARY
Using corticosteroids to treat Kawasaki disease
Key message
Corticosteroids appear to reduce the risk of heart problems after Kawasaki disease without causing any important side effects. They also
reduce the length of symptoms (fever and rash), length of hospital stay, and blood markers associated with being unwell.
What is Kawasaki disease and how is it treated?
Kawasaki disease is an inflammation of the blood vessels. Standard medicines to treat Kawasaki disease are intravenous immunoglobulin
(IVIG) and aspirin. This treatment is usually effective, but it does not work for all children. We currently have a limited understanding of
Kawasaki disease and how best to manage it. This is important as one of the long-term consequences can involve the heart, putting the
children with the disease at higher risk of life-shortening outcomes.
What did we do?
We searched for randomised controlled studies that treated children with Kawasaki disease using medicines known as corticosteroids, to
see if they reduced the chance of future heart problems. We also investigated the effect on the duration of fever, signs of infection in the
blood, and the number of days spent in hospital. In randomised controlled studies, the treatments or tests people receive are decided at
random and these usually give the most reliable evidence about treatment effects.
What did we find?
We found eight studies involving 1877 participants, which compared the use of corticosteroids with no corticosteroids in children with
Kawasaki disease. Combining the results from these studies showed that heart problems were reduced after corticosteroid treatment.
There were no serious side effects or deaths reported in the corticosteroid treatment groups or the no corticosteroid treatment groups.
Blood markers that indicate illness were reduced and children had shorter stays in hospital when treated with corticosteroids. Symptoms
of Kawasaki disease (fever and rash) lasted for a shorter time after corticosteroid treatment. None of the studies reported on long-term
(more than one year after disease onset) heart problems (coronary morbidity), which can be associated with Kawasaki disease.
How certain are we in the evidence?
Evidence was of moderate certainty for serious side effects and deaths, the risk of future heart problems, blood markers and length of
hospital stay. This means that we are moderately confident that the true effect is close to that estimated in this review. Evidence was
considered low certainty for the duration of clinical symptoms (fever and rash). This means our confidence in the effect estimate is limited
and this was because of how these symptoms were measured and reported.
How up-to-date is this evidence?
This Cochrane Review updates our previous evidence. The evidence is current to February 2021.

SUMMARY OF FINDINGS
Summary of findings 1. Corticosteroids compared to no corticosteroid use for the treatment of Kawasaki disease in children
Library
Cochrane
Corticosteroids compared to no corticosteroid use for the treatment of Kawasaki disease in children
Patient or population: children diagnosed with Kawasaki disease

Setting: hospital
Intervention: corticosteroids
Better health.
Informed decisions.
Trusted evidence.
Comparison: no corticosteroid usea
Outcomes Anticipated absolute effects* (95% CI) Relative effect № of partici- Certainty of Comments
(95% CI) pants the evidence
Risk with no corticos- Risk with corticosteroids (studies) (GRADE)
teroid use
Incidence of coronary artery Study population OR 0.32 986 ⊕⊕⊕⊝ —

abnormalities (0.14 to 0.75) (8 RCTs) Moderateb
Follow-up: range 2–6 weeks 167 per 1000 60 per 1000
(27 to 131)
Incidence of any serious ad- Study population — 737 ⊕⊕⊕⊝ 0 cases of serious
verse effects attributable to (6 RCTs) Moderatec adverse events at-
the administration of corti- See comment See comment tributable to cor-
costeroids ticosteroids use
Follow-up: range 2–6 weeks recorded by the in-
cluded studies.
Mortality (all-cause) Study population — 1075 ⊕⊕⊕⊝ 0 deaths record-

follow-up: range 2–6 weeks (8 RCTs) Moderatec ed by the included
See comment See comment studies.

Duration of clinical symp- The mean duration of clin- The mean duration of clin- — 290 ⊕⊕⊝⊝ —
toms: fever and rash ical symptoms (fever and ical symptoms (fever and (3 RCTs) Lowd
rash) across the control rash) was1.34 days lower
(days) groups ranged from 1.5 to (2.24 lower to 0.45 lower).
11.2 days.
Follow-up: range 2–6 weeks
Time for laboratory parame- The mean time for labo- The mean time for labo- — 178 ⊕⊕⊕⊝ —
ters to normalise: CRP, ESR e ratory parameters (CRP, ratory parameters (CRP, (1 RCT) Moderatef
ESR) to normalise was ESR) to normalise was 2.8
(days) 11.2 days. days lower
(4.38 lower to 1.22 lower).
3
Length of hospital stay The mean length of hospi- The mean length of hos- — 119 ⊕⊕⊕⊝ —
tal stay across the control pital stay was1.01 days (2 RCTs) Moderateg
(days) groups ranged from 3.31 lower
Library
Cochrane
to 6.7 days. (1.72 lower to 0.30 lower).
Longer term (> 1 year after Study population — — — 0 studies included

disease onset) coronary mor- data on outcomes
bidity (non-aneurysmal) See comment See comment (including coronary
morbidity (non-
Better health.
Informed decisions.
Trusted evidence.
aneurysmal)) > 1
year after study en-
rolment.
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and
its 95% CI).
CI: confidence interval; CRP: C-reactive protein; ESR: erythrocyte sedimentation rate; MD: mean difference; OR: odds ratio; RCT: randomised controlled trial.
GRADE Working Group grades of evidence

High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.
Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is
substantially different.
Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect.
Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.
aThese comprised alternative treatments such as intravenous immunoglobulin, aspirin and diphenhydramine. See Characteristics of included studies for more details.
bDowngraded one level for inconsistency: large group and effect size; however, subgroup analysis suggests that those with low-risk scores or receiving single-dose treatment
may not benefit. Further investigation to reduce potential confounding is required to explain inconsistencies in data (i.e. geographical variation, first-line versus second-line
treatment). However, overall effect is unlikely to change (likely beneficial).
cDowngraded one level for imprecision: small overall number of participants or events (for very rare events large number of participants needed).
dDowngraded one level for inconsistency: significant heterogeneity within the data, likely due to the subjective nature of the included outcome (i.e. rash). Downgraded one level
for indirectness: some indirectness of treatment effect since one study used corticosteroids as second-line treatment.
eLower numbers and a downward trend are generally better. A low result might be acceptable, rather than a completely negative result. The definitions of normal CRP can range

slightly. The definition of a normal ESR range depends on age and sex.
fDowngraded one level for imprecision due to only one study with a relatively small number of participants.
gDowngraded one level for indirectness of treatment effect (participants in one of two studies received corticosteroids as second-line treatment) and imprecision (small number
of participants).
4
Informed decisions.
BACKGROUND Description of the intervention
Description of the condition It is thought that the prompt and effective treatment of KD can
decrease the incidence of its cardiac sequelae. Accepted and
Kawasaki disease (KD), or mucocutaneous syndrome, is the proven initial pharmacological management involves intravenous
leading cause of childhood-acquired heart disease in high-income immunoglobulin (IVIG) at a dose of 2 g/kg in a single 12-hour
countries (Kato 1996). Originally described by Kawasaki 1967, infusion alongside aspirin 30 mg/kg to 50 mg/kg in four divided
it is a medium vessel vasculitis of unclear aetiology that has doses (Eleftheriou 2013). This has been shown to limit the duration
been linked with an abnormal host response to an infectious of the acute phase of KD as well as reduce the long-term coronary
trigger. Generally affecting children less than five years old, peak sequelae from 25% to 4.7% (Levin 1991). Both medications
onset is between 18 and 24 months. The incidence in those aged have already been subjects of Cochrane Reviews (Baumer 2006;
under five years varies widely throughout the world, including Oates-Whitehead 2003). Plasma exchange is also used in certain
8.4 per 100,000 in the UK, 17.5 per 100,000 to 20.8 per 100,000 institutions (Hokosaki 2012).
in the USA, and 239.6 per 100,000 in Japan (Gardner-Medwin
2002; Harnden 2002; Holman 2003; Nakamura 2012; Singh 2015). Patients do not always respond to the above regimen. A subset of
Rate of recurrence is approximately 3600 per 100,000, while acute people with KD, approximately 20%, have clinical symptoms that
mortality occurred in just one of the 23,730 cases Nakamura and are resistant to the first dose of IVIG and aspirin after 48 hours. This
colleagues analysed in the 2009 to 2010 period (Nakamura 2012). group has been proven to be at higher risk for cardiac sequelae
Such varied epidemiology has strengthened theories linking KD (Brogan 2002). Various systems for identifying this group have
with genetics and one or more infectious agents (Wood 2009). been formulated, including the Kobayashi score; however, while
specific, these have shown poor sensitivity in Western populations
KD is a multisystem vasculitis but its most important complication compared with the Japanese groups in which they were devised
involves the predisposition for coronary artery vasculitis leading to (Kobayashi 2008; Sleeper 2011). Currently, accepted identifiers for
aneurysms in up to 25% of untreated patients (Burns 1996). Such this high-risk group include the following (Eleftheriou 2013).
complications render the coronary vessels vulnerable to stenoses
and thromboses, with subsequent risk of myocardial infarction and 1. Resistance to IVIG.
death (Daniels 2012). Furthermore, these thromboses act as focus 2. Very young age of onset (less than 12 months).
for accelerated KD vasculopathy, increasing cardiovascular risk. 3. Severe inflammatory markers.
There is no diagnostic test for KD but laboratory findings typically 4. Clinical features of shock.
show a raised white cell count, erythrocyte sedimentation rate 5. Existing arterial aneurysms.
(ESR) and C-reactive protein (CRP). Diagnosis is generally based on 6. Kobayashi score of five or greater.
clinical symptoms or features from one of two major sets of criteria.
The Revised Diagnostic Guidelines of the Japan KD Research These high-risk patients have a higher prevalence of coronary
Committee require any five from 1. fever (without requirement aneurysms, especially giant aneurysms (greater than 8 mm),
for a specific duration), 2. bilateral conjunctivitis, 3. changes of and have associated greater long-term cardiovascular morbidity
lips and oral cavity: reddening of lips, strawberry tongue, diffuse and mortality (Tatara 1987). Current consensus on management
injection of oral and pharyngeal mucosa, 4. rash (including redness recommends a repeat dose of IVIG which facilitates disease
at the site of Bacille Calmette Guerin (BCG) inoculation), 5. changes defervescence in approximately half of the patients (Hashino
in the extremities (reddening, swelling and peeling of the skin) 2001a). Measures to improve the success rate have been reviewed
and 6. non-suppurative cervical lymphadenopathy (Kobayashi and the utility of intravenous steroids (IVS), with or without
2020). Complete KD is diagnosed with at least five principal infliximab has shown mixed results. Current data on infliximab are
clinical features, or with four if there is evidence of coronary insufficiently powered to draw conclusions (Davies 2013). These
artery abnormality found on echocardiography or at angiography. studies were subject to significant selection bias, with only the most
Incomplete KD is diagnosed when there are three principal clinical severe cases, bearing the greatest probability of a poor outcome,
features, plus coronary artery abnormality or if there are other given steroids (Kato 1979). Therefore, despite IVS being long used
significant clinical features, which are detailed in the guidelines in vasculitides similar to KD, their use in KD has been subject to
(Kobayashi 2020). long-standing controversy due to these earlier works showing a
deleterious effect (Levin 2013).
The revised diagnostic guidelines in the American Heart
Association KD guideline are similar, but specify the duration of There have been recent gains in knowledge regarding the use
fever; classic (complete) KD is diagnosed in the presence of fever of corticosteroids in KD. The early use of corticosteroids has
for five days along with at least four principal clinical features In the been advocated, but only in high-risk patients (as defined
presence of four or more features, especially if redness and swelling above). This acknowledged, it remains unclear how best to use
of the extremities are present, the diagnosis of KD can be made after corticosteroids (Chen 2016). One editorial highlighted some of
four days of fever (or three days if diagnosed by an experienced the critical issues with current meta-analyses in this area (Levin
clinician)(McCrindle 2017). 2013). Problems include varying inclusion criteria and populations,
differing methodologies of included studies and an overall lack of
Diagnosis is complicated by these symptoms being prevalent in power with respect to data on adverse effects. It is not currently
various common childhood viral exanthems. Symptoms may also known which demographic groups show the greatest benefit with
occur sequentially rather than simultaneously. It is important to respect to coronary sequelae, or if there is the potential for
exclude other febrile illnesses before diagnosing KD. complications with IVS treatment. Furthermore, the most effective
types, frequencies and doses of corticosteroid have not yet been

Informed decisions.
clarified, nor whether corticosteroids should be administered KD diagnostic guidelines include the following principal clinical
alongside IVIG, aspirin or infliximab (Levin 2013). features (Kobayashi 2020).
How the intervention might work 1. Fever (without requirement for a specific duration).
2. Bilateral conjunctivitis.
Corticosteroid treatment is already utilised in a broad range of
vasculitides to great effect. Furthermore, corticosteroids were a 3. Changes of lips and oral cavity: reddening of lips, strawberry
key part of KD treatment prior to the advent of IVIG. Research tongue, diffuse injection of oral and pharyngeal mucosa.
into the exact pathological mechanisms is ongoing and current 4. Rash (including redness at the site of Bacille Calmette Guerin
theories of KD pathogenesis implicate immunological responses to (BCG) inoculation).
infectious agents (Eleftheriou 2013). Such reactions are thought to 5. Changes in the extremities (reddening, swelling and peeling of
be controllable via corticosteroid administration due to a reduction the skin).
in inflammatory mediator transcription. In the context of KD, 6. Non-suppurative cervical lymphadenopathy.
this may mean a reduction in fever as well as lower levels of
inflammation, leading to a reduction in the formation of coronary For the Japanese KD research committee guidelines, complete KD
abnormalities and subsequent incidence of future cardiovascular is diagnosed with at least five principal clinical features, or with
sequelae (Levin 2013). Therefore, it remains important to ascertain four if there is evidence of coronary artery abnormality found on
the utility of IVS in KD. echocardiography or at angiography. Incomplete KD is diagnosed
when there are three principal clinical features plus coronary artery
Why it is important to do this review abnormality, or if there are other significant clinical features, which
There is much controversy regarding how best to identify the are detailed in the guidelines (Kobayashi 2020). In either case, other
high-risk patients who may benefit from additional treatment febrile illnesses should be excluded before diagnosing KD. The
beyond the standard IVIG and aspirin. Furthermore, this situation Japanese guidelines also recommend using Z-scores to evaluate
is complicated by early reports of harmful effects of steroids in KD, the internal coronary artery diameter (Kobayashi 2020).
although these conclusions are now considered to be the result of
The Japanese KD research committee and the American Heart
selection bias as only the most ill patients were investigated (Kato
Association guidelines are similar, but the latter specify the
1979). Current markers used to indicate resistance to IVIG include
duration of fever; classic (complete) KD diagnosed in the presence
very young age of onset (less than 12 months), high inflammatory
of fever for five days along with at least four principal clinical
markers, clinical features of shock, existing arterial aneurysms, or
features (McCrindle 2017). In the presence of four or more features,
a Kobayashi score of five or greater. The utility of corticosteroids
especially if redness and swelling of the extremities are present,
in KD as a whole remains unclear despite several meta-analyses
the diagnosis of KD can be made after four days of fever (or three
and we seek to use this review to clarify the situation through a
days if diagnosed by an experienced clinician). A diagnosis of
more targeted approach to analysis. In addition, it is unclear how
incomplete (or atypical) KD should be considered in a child with
best to define the patients who may benefit from corticosteroids
prolonged unexplained fever, some of the principal clinical features
the most, for example whether ethnic origin, severity of KD or
and compatible laboratory findings (such as raised inflammatory
pre-corticosteroid treatment status may help define the optimum
markers e.g. CRP, ESR) or echocardiographic evidence (McCrindle
population. It is hoped that the whole group and subgroup analyses
2017).
of this review will cast greater light on this issue. This is an update
of the original review published in 2017. Corticosteroids could have been part of the initial treatment for
KD or have formed part of the second-line treatment after failure
OBJECTIVES of first-line treatment that did not include corticosteroids. The
comparison group had to be in parallel. Cross–over trials were not
To assess the impact of corticosteroid use on the incidence of
eligible for inclusion.
coronary artery abnormalities in KD as either first-line or second-
line treatment. We excluded participants with positive blood cultures.
METHODS Types of interventions
Criteria for considering studies for this review All forms of corticosteroid therapy in conjunction with any
combination of no treatment, placebo, immunoglobulin, aspirin or
Types of studies infliximab for the treatment of KD were the intervention of interest.
We searched all randomised controlled trials (RCTs) and quasi-RCTs That stated, the use of corticosteroids had to be the only difference
(RCTs in which allocation methods were not completely random, in management between trial arms.
for example using alternation). We excluded cross-over trials as the
Comparator groups included any of:
response to corticosteroid intervention may depend on timing and
previous treatment state. We excluded all studies not conforming 1. placebo;
to the RCT format.
2. immunoglobulin only;
Types of participants 3. aspirin only;
We included all studies of children (less than 19 years old) 4. immunoglobulin and aspirin;
worldwide diagnosed with KD. The diagnosis of KD should be made 5. infliximab only;
according to existing KD guidelines at the time of the study. The 6. infliximab and immunoglobulin;

Informed decisions.
7. infliximab, immunoglobulin and aspirin. sensitive search strategy designed by Cochrane for identifying
RCTs and controlled clinical trials (as described in the Cochrane
Types of outcome measures Handbook for Systematic Reviews of Interventions Chapter 6,
Primary outcomes Lefebvre 2011). Search strategies for major databases are provided
in Appendix 1.
1. Incidence of coronary artery abnormalities (measured via
diameter or z-scores; Boston scores (de Zorzi 1998)) per study The Information Specialist searched two trials registries on 8
group found at either cardiac angiography or echocardiography February 2021.
within three months of KD diagnosis. Coronary abnormality was
defined using either the de Zorzi criteria (a coronary dimension 1. World Health Organization International Clinical Trials Registry
that was 2.5 standard deviations (SDs) or greater above the Platform (who.int/trialsearch).
mean for body surface area) (de Zorzi 1998); or the Japanese 2. ClinicalTrials.gov (clinicaltrials.gov).
Ministry of Health criteria (Research Committee on Kawasaki
Disease 1984), as follows. Searching other resources
a. Lumen greater than 3 mm in children less than 5 years old; We contacted the authors of trials that met the eligibility criteria
b. Lumen greater than 4 mm in children greater than 5 years old; identified by the searches as ongoing or unpublished trials. We also
c. Internal diameter of a segment measuring 1.5 times or searched reference lists of relevant trials for further publications.
greater that of an adjacent segment.
2. Incidence of any serious adverse effects per study group
Data collection and analysis
that was attributable to the administration of corticosteroids Selection of studies
at any point after treatment initiation. Known side effects
of corticosteroids in other diseases include, for example, For this update, two review authors (JG and RMRT) independently
immunosuppression with resultant opportunistic infection and applied the selection criteria to the studies identified by the
avascular necrosis of the femoral head. search strategy. This included independently assessing whether
the studies fulfilled the inclusion and exclusion criteria. If there was
Secondary outcomes insufficient information to decide whether a study was truly eligible
we contacted the study authors to request further information. A
1. Mortality (all-cause). third review author (AJW) resolved disagreements.
2. Duration of clinical symptoms: fever and rash.
3. Time for laboratory parameters to normalise: C-reactive protein We identified and excluded duplicates and collated multiple
(CRP) and erythrocyte sedimentation rate (ESR). reports of the same study, so that each study rather than each
report was the unit of interest in the review. We recorded the
4. Length of hospital stay.
selection process in sufficient detail to complete a PRISMA flow
5. Longer-term (greater than one year after disease onset) coronary diagram.
morbidity (non-aneurysmal).
Data extraction and management
Search methods for identification of studies
For this update, two review authors (JG and RMRT) independently
We included studies reported as full-text or published as abstract extracted data using a modified version of the Cochrane Vascular
only. standard data extraction form. These data were then brought
together and monitored for discrepancies by a third review author
Electronic searches
(AJW), before being entered into Review Manager 5 (Review
The Cochrane Vascular Information Specialist conducted Manager 2014). We contacted study authors for any required
systematic searches of the following databases for RCTs and information that was not included in the publications. The key
controlled clinical trials without language, publication year or information gathered in the data collection form included the
publication status restrictions. following.
1. Cochrane Vascular Specialised Register via the Cochrane 1. General study information: publication type.
Register of Studies (CRS-Web searched on 8 February 2021). 2. Fulfilment of eligibility criteria: study type, interventions,
2. Cochrane Central Register of Controlled Trials (CENTRAL) outcomes measured and reasons for exclusion.
Cochrane Register of Studies Online (CRSO 2021, Issue 1). 3. Study methods: allocations methods, study dates, duration,
3. MEDLINE (Ovid MEDLINE Epub Ahead of Print, In-Process & Other ethical approval and statistical methods.
Non-Indexed Citations, Ovid MEDLINE Daily and Ovid MEDLINE) 4. Participants: methods of recruitment, consent, total number,
(searched from 25 November 2016 to 8 February 2021). treatment groups, age, sex, race, KD severity, subgroup analyses
4. Embase Ovid (searched from 25 November 2016 to 8 February reported and eligibility criteria.
2021). 5. Intervention (corticosteroids): number of participants, dosing,
5. CINAHL EBSCO (searched from 25 November 2016 to 8 February frequency, duration, delivery method, providers, compliance
2021). and concomitant treatment.
6. Outcomes: coronary diameters (acute), coronary diameter z-
The Information Specialist modelled search strategies for other scores (acute), coronary abnormality (long term), duration of
databases on the search strategy designed for CENTRAL. Where clinical symptoms (e.g. fever), adverse effects, duration of
appropriate, they were combined with adaptations of the highly

Informed decisions.
laboratory parameter abnormality (e.g. CRP) and duration of Assessment of reporting biases
hospital stay.
We planned to screen for publication and reporting bias using
7. Study funding and study author declarations of interest. funnel plot asymmetry and measure using tests if there were
10 or more trials, as outlined in Chapter 10 of the Cochrane
Assessment of risk of bias in included studies
Handbook for Systematic Reviews of Interventions (Higgins 2011).
We assessed the risk of bias using Cochrane's RoB 1 tool as For smaller studies, we considered 'the small-study effect', where
described in section 8.5 of the Cochrane Handbook for Systematic smaller studies can show larger treatment effects due to poor
Reviews of Interventions (Higgins 2011). In this update, two review methodology, heterogeneity, selection bias, chance or artefact. We
authors (JG and RMRT) performed this independently, and resolved also searched for eligible studies that had been registered and
any disagreements by discussion with a third review author (AJW). should have been completed, but were without available published
The domains assessed included: data. It should be noted that the risk of bias assessment has taken
into account selective outcome reporting.
1. sequence generation (selection bias);
2. allocation sequence concealment (selection bias); Data synthesis
3. blinding of participants and personnel (performance bias); We only undertook meta-analyses when it was meaningful to
4. blinding of outcome assessment (detection bias); do so. That is if the treatments, participants and the underlying
5. incomplete outcome data (attrition bias); clinical question were sufficiently similar for pooling to make
6. selective outcome reporting (reporting bias); sense. Statistical analysis took place using a fixed-effect model if
there was low heterogeneity, and a random-effects model if there
7. other bias.
was significant heterogeneity (I2 > 60%). We undertook outcome
Measures of treatment effect analyses using an ITT model. Two-sided P values of 0.05 or less
were considered significant and all analyses were undertaken using
The effect measure of choice for dichotomous data was the odds Review Manager 5 (Review Manager 2014). If meta-analysis was
ratio (OR) with a 95% confidence interval (CI). This is a ratio between not possible, we planned to report the results using a narrative
the corticosteroid intervention group and its parallel comparator synthesis.
group. We reported continuous data, including time-to-event data,
using mean differences (MD) and 95% CIs. If necessary, we planned Subgroup analysis and investigation of heterogeneity
to use a standardised mean difference (SMD) for studies that
We planned to include the following subgroup analyses (data
measured the same outcome but used different methods.
permitting):
Unit of analysis issues
1. first-line versus second-line management;
The unit of analysis was each participant recruited into a trial. 2. type of corticosteroid used;
3. corticosteroid dosing;
Dealing with missing data
4. corticosteroid treatment frequency;
We accounted for all missing data due to dropouts via an intention- 5. total corticosteroid treatment duration;
to-treat (ITT) analysis. We reported if the individual trials conducted
6. corticosteroid route of administration;
ITT analyses. If they did not, then we endeavoured to apply
an ITT analysis. In the event that we were unable to do this, 7. geographical distribution of trial participants, ethnicity;
we utilised a per protocol analysis. We have explained all post- 8. KD severity (non-high risk versus high risk as detailed earlier);
allocation dropouts. We have followed up any data missing from 9. recognised concomitant treatments for KD (as detailed in Types
the published document with the study's original authors. If data of interventions).
still remained absent, we considered this in the risk of bias
assessments. We also employed further subgroup analyses if heterogeneity
remained significant (I2 > 60%) using a Chi2 P value threshold of
Assessment of heterogeneity 0.05.
We measured heterogeneity using the I2 statistic for quantification Sensitivity analysis
of variability (less than 40% = likely low heterogeneity; 40% to 60%
= possible moderate heterogeneity; greater than 60% = possible We planned to perform sensitivity analysis to explore causes of
significant heterogeneity) (Higgins 2011). We also considered the heterogeneity and the robustness of the results if there were
magnitude and direction of effects using the Chi2 test (limit = sufficient data available. We planned to include the following
degrees of freedom) and P values (10% significance threshold) factors in the sensitivity analysis.
(Higgins 2011). Where heterogeneity exceeded generally accepted
limits (greater than 60% heterogeneity), we subgrouped the 1. Type of study design (RCT versus quasi-RCT).
analysis in a logical manner to explain these differences and reduce 2. Low risk of bias trials versus high risk of bias trials. A trial was
remaining heterogeneity. These methods were reinforced by visual defined as high risk if any domain in the trial was assessed at
recognition on forest plots to assess for overlapping CIs (Higgins high risk of bias.
2011). 3. Rates of dropouts for each treatment group. We planned to
perform a sensitivity analysis if the rates of dropout were
considered sufficient to impact results.

Informed decisions.
Summary of findings and assessment of the certainty of the heterogeneity, precision of effects estimates, and risk of publication
evidence bias, according to the GRADE principles as described by Higgins
2011 and Atkins 2004. We used the GRADEpro GDT software
We presented the main findings of the review results concerning
to assist in the preparation of the summary of findings table
the certainty of the evidence, the magnitude of effect of the
(www.guidelinedevelopment.org/).
interventions examined and the sum of available data on the
outcomes (Types of outcome measures) in Summary of findings RESULTS
1. We assessed the certainty of the evidence for each primary
and secondary outcome as high, moderate, low and very low, Description of studies
based on within-study risk of bias, directness of evidence,
See Figure 1.

Informed decisions.
Figure 1. Flow diagram for studies in 2021 updated review

Informed decisions.
Figure 1. (Continued)
Results of the search 2012; Sundel 2003). Cytokine levels was the primary outcome in
one study (Okada 2003). One study did not specify which outcomes
The previous version of this review included seven studies (Wardle
were primary or secondary (Wang 2020).
2017). The search on 8 February 2021 identified 268 records leaving
233 records after removal of duplicates. After assessing titles and Further details including dosing and baseline characteristics, see
abstracts, we excluded 199 records. We assessed the full text of Characteristics of included studies table.
34 articles. For this 2021 update, we identified one new included
study (Wang 2020; see Characteristics of included studies table), Excluded studies
two additional reports to add to studies already included in the
See Characteristics of excluded studies table.
review, one new excluded study (Shiari 2011; see Characteristics
of excluded studies table), four reports as awaiting classification We excluded 15 studies (18 reports) from the review (Asai 1985;
(see Characteristics of studies awaiting classification table), and Hashino 2001b; ISRCTN74427627; Jibiki 2004; Kato 1979; Kusakawa
three new ongoing studies (NCT04078568; EUCTR2019-004433-17- 1983; Miura 2008; Nakamura 1985; Nonaka 1995; Ogata 2009;
GB; IRCT20181202041817N1; see Characteristics of ongoing studies Sekine 2012; Seto 1983; Shiari 2011; Xu 2002; Yuan 2000).
table). See Figure 1.
ISRCTN74427627 was stopped due to lack of funding and there are
Included studies no results. Kato 1979 was not randomised. Shiari 2011 was only
After applying the aforementioned inclusion and exclusion criteria, available as an abstract which did not include enough information
we identified eight trials (16 reports) suitable for inclusion within about methodology or outcome data (we contacted the study
this review (Ikeda 2006; Inoue 2006; Kobayashi 2012; Newburger authors without success). Seto 1983 had inadequate information
2007; Ogata 2012; Okada 2003; Sundel 2003; Wang 2020). Full- on methodology for validation, specifically, no evidence of
text publications were obtained from the websites of the original randomisation. Hashino 2001b had trial design issues. Four
source of publication. When studies had multiple publications, we studies did not meet the criteria for participants described in our
combined data to give as complete an interpretation of available Types of participants section (Miura 2008; Nonaka 1995; Sekine
data as possible, ensuring data were not duplicated. 2012; Yuan 2000). Six studies combined corticosteroids with an
additional intervention meaning corticosteroids were not the only
One trial was conducted in South China (Wang 2020), two trials in differentiating factor between the trial arms (Asai 1985; Jibiki 2004;
North America (Newburger 2007; Sundel 2003), and five in Japan Kusakawa 1983; Nakamura 1985; Ogata 2009; Xu 2002).
(Ikeda 2006; Inoue 2006; Kobayashi 2012; Ogata 2012; Okada 2003).
Four trials were single centre (Ikeda 2006; Ogata 2012; Sundel Studies awaiting classification
2003; Wang 2020), and four trials were multicentre (Inoue 2006; Four trials are awaiting classification and there are currently
Kobayashi 2012; Newburger 2007; Okada 2003). insufficient details to assess for inclusion or exclusion (ChiCTR-
IOR-16007862; ChiCTR1800017994; JPRN-UMIN000005022; JPRN-
The eight trials included 1877 participants, ranging from 32 to
UMIN000009946).
955 participants per trial. Three studies segmented high-risk
participants using criteria as set out in Table 1 (Ikeda 2006; Ongoing studies
Kobayashi 2012; Ogata 2012). One study used corticosteroids
as part of second-line management for KD after treatment Four trials are ongoing and there are currently no
failure (Wang 2020), whereas the other included studies used suitable data available for review (EUCTR2019-004433-17-GB;
corticosteroids as part of first-line management for KD. IRCT20181202041817N1; JPRN-UMIN000009524; NCT04078568).
Coronary artery abnormality was the primary outcome in four Risk of bias in included studies
studies (Ikeda 2006; Inoue 2006; Kobayashi 2012; Newburger 2007).
All included studies were assessed for bias as illustrated in Figure 2,
Duration of fever was the primary outcome in two studies (Ogata
Figure 3, and the Characteristics of included studies table.

Informed decisions.
Figure 2. Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Blinding of participants and personnel (performance bias): All outcomes

Blinding of outcome assessment (detection bias): All outcomes
Incomplete outcome data (attrition bias): All outcomes
Random sequence generation (selection bias)
Allocation concealment (selection bias)
Selective reporting (reporting bias)

Other bias
Ikeda 2006 ? ? ? ? ? + +
Inoue 2006 ? + ? ? + + +
Kobayashi 2012 + + ? + + + +
Newburger 2007 ? ? + + + + +
Ogata 2012 + ? ? + + + +
Okada 2003 + + ? ? + + +
Sundel 2003 ? ? ? + + + +
Wang 2020 + + ? ? + ? +

Informed decisions.
Figure 3. Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages
across all included studies.
Random sequence generation (selection bias)

Allocation concealment (selection bias)
Blinding of participants and personnel (performance bias): All outcomes
Blinding of outcome assessment (detection bias): All outcomes
Incomplete outcome data (attrition bias): All outcomes
Selective reporting (reporting bias)
Other bias
0% 25% 50% 75% 100%
Low risk of bias Unclear risk of bias High risk of bias
Allocation Sundel 2003; Wang 2020). These were all judged at low risk of bias.
One study did not state how data were handled and was at unclear
All studies were RCTs (Ikeda 2006; Inoue 2006; Kobayashi 2012;
risk of bias (Ikeda 2006).
Newburger 2007; Ogata 2012; Okada 2003; Sundel 2003; Wang
2020). All trials except Wang 2020 had a relatively short follow-up period,
up to five weeks, so there was limited loss to follow-up. Wang 2020
Random sequence generation
followed participants up over two years to measure coronary artery
Four studies were at low risk of bias from random sequence lesions (CALs) and by the end of the two years, 47/80 participants
generation due to the use of recognised and accepted approaches had dropped out. However, at one-month follow-up (with the data
to this step (e.g. use of computer generation, random numbers relevant to this review), only 1/80 participants had dropped out.
table) (Kobayashi 2012; Ogata 2012; Okada 2003; Wang 2020). Four The study authors also addressed this and clearly stated that none
studies were at unclear risk of bias because, although they reported of the CAL-positive patients had dropped out. We judged this at low
use of random allocation, there were no details regarding the exact risk of attrition bias.
approach (Ikeda 2006; Inoue 2006; Newburger 2007; Sundel 2003).
Selective reporting
Allocation concealment
One study published the protocol before initiation of the trial
Four studies were at low risk of allocation disclosure due to the (Newburger 2007). The published methodology including outcome
use of centrally held databases (Inoue 2006; Kobayashi 2012; Okada measures reported by the remainder of the trials was consistent
2003) and allocation by a non-participating nurse (Wang 2020). Four with that reported in the respective results section (Ikeda 2006;
studies were at unclear risk of bias for allocation concealment due Inoue 2006; Kobayashi 2012; Ogata 2012; Okada 2003; Sundel 2003).
to inadequate detail being published (Ikeda 2006; Newburger 2007; Therefore, the risk of selective reporting bias for these studies
Ogata 2012; Sundel 2003). was low. One study was at unclear risk of bias as there was
some discrepancy between the numbers in the text compared
Blinding to one of the figures when reporting the results of the third-
One trial was double-blinded and at low risk of performance line treatment (Wang 2020). This raised questions about bias in
bias (Newburger 2007). Six trials were not blinded (Inoue 2006; reporting. However, this third-line treatment was not relevant to
Kobayashi 2012; Ogata 2012; Okada 2003; Sundel 2003; Wang 2020), the outcomes in this review and the data for the main intervention
and one trial did not state whether participants or investigators were fully reported.
were blinded (Ikeda 2006). Therefore, seven studies were at unclear
Other potential sources of bias
risk of performance bias (Ikeda 2006; Inoue 2006; Kobayashi 2012;
Ogata 2012; Okada 2003; Sundel 2003; Wang 2020). All studies reported no potential other biases.
Four studies had blinded outcome reporters and were at low risk Effects of interventions
of bias (Kobayashi 2012; Newburger 2007; Ogata 2012; Sundel
2003); two studies had non-blinded outcome reporters and were at See: Summary of findings 1 Corticosteroids compared to no
unclear risk of detection bias (Inoue 2006; Wang 2020). Two studies corticosteroid use for the treatment of Kawasaki disease in children
did not mention if outcome reporters were blinded and were judged Primary outcomes
at unclear risk of detection bias (Ikeda 2006; Okada 2003).
Incidence of coronary artery abnormalities
Incomplete outcome data
Eight trials with 986 participants assessed incidence of coronary
Two studies mentioned ITT analysis (Kobayashi 2012; Newburger artery abnormalities (Ikeda 2006; Inoue 2006; Kobayashi 2012;
2007). Inoue 2006 reported key outcome variables as per-protocol Newburger 2007; Ogata 2012; Okada 2003; Sundel 2003; Wang
analysis and performed ITT analyses. Four further studies were able 2020). Seven trials used corticosteroids as part of first-line
to account for all or most of the data (Ogata 2012; Okada 2003;
Informed decisions.
treatment and one trial used corticosteroids as part of second-line CI −3.31 to 0; 2 studies, 210 participants; P = 0.05) (Inoue 2006;
treatment of KD (Wang 2020). Okada 2003). Although the second-line management also showed
a possible reduction, this was not clear (MD −0.90, 95% CI −1.84 to
Overall, pooled data of first-line and second-line treatment with 0.04; 1 study, 80 participants; P = 0.06) (Wang 2020). The P value for
corticosteroids showed a reduced incidence of coronary artery the test for subgroup differences was 0.44 (Analysis 1.4).
abnormality favouring corticosteroid use (OR 0.32, 95% CI 0.14
to 0.75; 8 studies, 986 participants; P = 0.009; moderate-certainty Time for laboratory parameters to normalise: C-reactive protein
evidence; Analysis 1.1). Overall, there was heterogeneity (I2 = 63%), and erythrocyte sedimentation rate
so we used a random-effects model.
One study with 178 participants directly measured the time for
In subgroup analysis of corticosteroid use versus no corticosteroid laboratory parameters to normalise (Inoue 2006). Analysis favoured
use for first-line management, there were fewer coronary artery use of corticosteroid (MD −2.80 days, 95% CI −4.38 to −1.22; P =
abnormalities in the corticosteroid group (OR 0.25, 95% CI 0.10 to 0.0005; moderate-certainty evidence; Analysis 1.5).
0.58; 7 studies, 907 participants; P = 0.001; Analysis 1.1). There was
Length of hospital stay
possible moderate heterogeneity of 55%, reduced from overall.
Two studies reported the length of hospital stay and demonstrated
For second-line management, there was no evidence of effect for an overall reduction in days stayed in hospital favouring
coronary artery abnormalities between groups (OR 1.38, 95% CI corticosteroid use (MD −1.01 days, 95% CI −1.72 to −0.30; 2 studies,
0.43 to 4.41; 1 study, 79 participants; P = 0.59; Analysis 1.1). 119 participants; P = 0.006; moderate-certainty evidence; Analysis
1.6) (Sundel 2003; Wang 2020).
Overall, differences were detected between the subgroups by the
test for subgroup differences (P = 0.02). For length of hospital stay, first-line management showed a
reduction in hospital stay (MD −1.41, 95% CI −2.36 to −0.46; 1 study,
Incidence of any serious adverse events attributable to the
39 participants; P = 0.004) (Sundel 2003). However, there was no
administration of corticosteroids
evidence of a difference for second-line management (MD −0.50,
From six trials with 737 participants, there were no serious adverse 95% CI −1.58 to 0.58; 1 study, 80 participants; P = 0.36) (Wang 2020).
events attributable to corticosteroid use and the effects were not The P value for the test for subgroup differences was 0.21 (Analysis
estimable (Analysis 1.2) (Inoue 2006; Kobayashi 2012; Newburger 1.6).
2007; Ogata 2012; Okada 2003; Sundel 2003). Ikeda 2006 and Wang
2020 did not provide information despite attempts to contact the Longer-term (greater than one year after disease onset)
authors. coronary morbidity (non-aneurysmal)
None of the included studies included data on outcomes (including
Secondary outcomes
coronary morbidity (non-aneurysmal) more than one year after
Mortality (all-cause) study enrolment.
The eight trials with 1075 participants reported no deaths within Additional subgroup analysis
the observed study period (Analysis 1.3) (Ikeda 2006; Inoue 2006;
Kobayashi 2012; Newburger 2007; Ogata 2012; Okada 2003; Sundel Planned subgroup analyses are listed below with comments on the
2003; Wang 2020). ability for them to be performed.
Duration of clinical symptoms: fever and rash 1. Type of corticosteroid used: this subgroup analysis was
performed for the outcome coronary artery abnormalities for
Three studies reported data on the duration of clinical symptoms seven studies (Inoue 2006; Kobayashi 2012; Newburger 2007;
(days), showing a reduction in time with corticosteroid treatment Ogata 2012; Okada 2003; Sundel 2003; Wang 2020). When
(MD −1.34 days, 95% CI −2.24 to −0.45; 3 studies, 290 participants; using corticosteroids as a single intravenous (IV) dose there
P = 0.003; low-certainty evidence; Analysis 1.4) (Inoue 2006; Okada was no evidence of an effect with regards to coronary artery
2003; Wang 2020). There was heterogeneity (I2= 76%), so a random- abnormalities (OR 0.70, 95% CI 0.40 to 1.22; 4 studies, 356
effects model was used. The heterogeneity was not explained by participants; I2 = 35%) (Newburger 2007; Okada 2003; Sundel
subgroup analysis for high- and low-risk scores (Analysis 4.2). Two 2003; Wang 2020). When this was followed with an oral course of
studies considered a child afebrile if their temperature was below corticosteroids (longer corticosteroid course) there were fewer
37.5 °C for more than 24 hours (Inoue 2006; Okada 2003). Wang 2020 coronary artery abnormalities in the corticosteroid group (OR
considered a child afebrile if their temperature was below 38 °C 0.13, 95% CI 0.05 to 0.32; 3 studies, 452 participants) (Inoue
for 48 hours after treatment. The differences in how these studies 2006; Kobayashi 2012; Okada 2003). The P value for the test
defined 'fever' may account for the heterogeneity. Furthermore, for subgroup differences was 0.002. Subgroups explained the
Inoue 2006 measured axillary temperature; Okada 2003 and Wang heterogeneity identified for the combined analysis (Analysis
2020 simply measured "body temperature", but did not state the 2.1). We were limited by the small number of studies in each
site. This may account for the heterogeneity as body temperature subgroup so results must be interpreted with caution.
can vary depending on measurement site. 2. Corticosteroid dosing: this subgroup analysis was not applicable
First-line versus second-line management was also performed for as corticosteroid dosing was broadly classed into single, pulse
the duration of clinical symptoms. When considering duration of dose of IV methylprednisolone or a longer tapering course of
fever, the subgroup analysis showed that first-line management prednisolone, and the subgroup analysis for this was performed
of corticosteroids may reduce clinical symptoms (MD −1.65, 95% in subgroup analysis 1 above.

Informed decisions.
3. Corticosteroid treatment frequency: this subgroup analysis was Cochrane Handbook for Systematic Reviews of Interventions (Higgins
not applicable as corticosteroid dosing was broadly classed into 2011).
single, pulse dose IV methylprednisolone or a longer tapering
course of prednisolone, and the subgroup analysis for this was DISCUSSION
performed in subgroup analysis 1 above.
4. Total corticosteroid treatment duration: this subgroup analysis
Summary of main results
was not applicable as corticosteroid dosing was broadly classed Pooled data from eight studies showed that the use of
into single, pulse dose IV methylprednisolone or a longer corticosteroids compared to no corticosteroids in the acute phase
tapering course of prednisolone, and the subgroup analysis for of KD in children can lead to reduced incidence of coronary
this was performed in subgroup analysis 1 above. artery aneurysms (moderate-certainty evidence), reduced time
5. Corticosteroid route of administration: this subgroup analysis for laboratory parameters to normalise (CRP and ESR)(moderate-
was not applicable as corticosteroid dosing was broadly classed certainty evidence), reduced length of hospital stay (moderate-
into single, pulse dose IV methylprednisolone or a longer certainty evidence), and reduced duration of clinical symptoms
tapering course of prednisolone. All corticosteroids in the initial (fever and rash) (low-certainty evidence). There was no evidence
phase were given IV, and the subgroup analysis for this was to suggest that corticosteroids caused adverse effects or increased
performed in subgroup analysis 1 above. mortality. None of the included studies reported long-term (greater
6. Geographical distribution of trial participants, ethnicity: this than one year) coronary morbidity during the study periods
subgroup analysis found that the use of corticosteroids included in this analysis.
appeared to have an increased beneficial effect on coronary
Seven studies investigated corticosteroid use as first-line
artery abnormalities in the studies conducted in Japan (OR
management (Ikeda 2006; Inoue 2006; Kobayashi 2012; Newburger
0.14, 95% CI 0.07 to 0.29; 5 studies, 678 participants; P <
2007; Ogata 2012; Okada 2003; Sundel 2003), and one as second-
0.00001) (Ikeda 2006; Inoue 2006; Kobayashi 2012; Ogata 2012;
line management (Wang 2020). Similar to the overall analysis,
Okada 2003) versus no evidence of benefit in North America
subgroup analysis showed first-line corticosteroid treatment
(OR 0.77, 95% CI 0.37 to 1.59; 2 studies, 229 participants; P
was more effective at reducing coronary artery abnormalities
= 0.48) (Newburger 2007; Sundel 2003), and no evidence of
compared to no corticosteroids, but this effect was not evident
benefit in South China (OR 1.38, 95% CI 0.43 to 4.41; 1 study,
in the second-line treatment subgroup. There were differences
79 participants; P = 0.59) (Wang 2020). The P value for test for
between the subgroups by the test for subgroup differences (P =
subgroup differences was 0.0004; however, it is important to
0.02). The children who received second-line management with
note that this analysis is vulnerable to confounding as most
corticosteroids were IVIG resistant, which may have had an impact
participants who received a single, pulse dose of corticosteroid
on the treatment effect for this study (Wang 2020). There was
were from the North American cohorts. Furthermore, Wang
a similar effect for first-line and second-line treatment for the
2020 (conducted in South China) was the only trial to include
duration of clinical symptoms and length of hospital stay, and no
IVIG-resistant participants and, therefore, the only study to use
differences between the groups (duration of clinical symptoms: P =
corticosteroids as second-line management, which is another
0.44; length of hospital stay: 0.21). It is likely we were limited in our
important and potentially confounding factor (Analysis 3.1).
subgroup analysis due to the small number of studies involved and
We were also limited by the small number of studies in each
these results should be interpreted with caution.
subgroup so results must be interpreted with caution.
7. KD severity (non-high risk versus high risk as detailed earlier): on Additional subgroup analysis demonstrated that with respect to
analysis of high-risk participants only there is strong evidence coronary abnormality:
of an effect (OR 0.13, 95% CI 0.06 to 0.29; 3 studies, 377
participants) (Ikeda 2006; Kobayashi 2012; Ogata 2012). When 1. there may be more benefit to children in Japan versus those
considering low-risk participants alone, the benefit was unclear in North America or China, but the use of different regimens
(OR 0.66, 95% CI 0.38 to 1.13; 6 studies, 609 participants) (Ikeda may have contributed to the different effects seen and we were
2006; Inoue 2006; Newburger 2007; Okada 2003; Sundel 2003; limited by the small number of studies in the subgroups (test for
Wang 2020). There was a difference by the test for subgroup subgroup difference P = 0.0004);
differences (P = 0.001). In both subgroups, heterogeneity was 2. there may be more benefit in children with high-risk scores
reduced to negligible levels (Analysis 4.1). We were limited by versus those with low-risk scores, although both display benefit
the small number of studies in each subgroup so results must be (test for subgroup difference P = 0.001);
interpreted with caution. 3. there may be a benefit of corticosteroids if taken over
8. Recognised concomitant treatments for KD (as detailed earlier in a prolonged course versus the potential for no benefit if
the text): there were insufficient data to perform this subgroup corticosteroids are given as a single, pulse dose (test for
analysis. subgroup difference P = 0.002).
Sensitivity analysis and reporting bias However, potential confounding of the subgroup analyses must be
A sensitivity analysis was not required for this review as all studies noted here – those studies completed in North America were both
were RCTs, no studies had a significant risk of bias and dropout considered part of the lower-risk group and employed single-dose
rates were not significantly different between studies. regimens (Newburger 2007; Sundel 2003).
As there were only eight trials included in this Cochrane Review, we Subgroup analysis demonstrated that with respect to duration
have not screened for publication and reporting bias by assessing of clinical symptoms both groups with high- and low-risk scores
funnel plot asymmetry using tests as outlined in Chapter 10 of the benefit from corticosteroid treatment, with the greatest benefit in

Informed decisions.
those with high-risk scores although the numbers of studies and serious adverse events and mortality for imprecision as no events
participants on which this was based were small. were recorded (small overall number of participants or events-
for very rare events, large number of participants needed). We
Overall completeness and applicability of evidence downgraded duration of clinical symptoms firstly due to the large
heterogeneity and secondly for indirectness due to patients in
All studies in this review collected data on our stated primary
one study (out of the three included) receiving corticosteroids as
outcomes: coronary artery abnormalities and serious adverse
second-line treatment (Wang 2020).
events attributable to corticosteroids use. All relevant participants,
interventions and outcomes have been investigated. Overall, the Overall, this means that we are reasonably confident that the true
evidence collected is highly applicable to this review. The evidence effect is close to that estimated in this review.
demonstrates that corticosteroids have some benefit in the acute
treatment of KD in the populations studied in this review. That Potential biases in the review process
stated, further data looking at different ethnicity subgroups,
disease risk scores, first-line versus second-line management of There are no known biases to disclose in the implementation of this
corticosteroids and duration of corticosteroid use are required review. There were no limitations within the search and this study
for a more complete guide. In particular, an investigation outside followed the predefined protocol following Cochrane guidelines
Japan employing risk stratification and IV corticosteroid treatment (Wardle 2014).
followed by oral doses would be beneficial. Furthermore, it is
worth noting that although the incidence of coronary artery Agreements and disagreements with other studies or
abnormalities was a primary outcome of this review, the severity of reviews
coronary artery abnormalities was not investigated. The results of this review reflect the shifting paradigm regarding
the use of corticosteroids in KD and show that corticosteroids
None of the studies demonstrated any serious adverse events
are beneficial in the treatment of KD, fitting with their use
during their follow-up periods due to the use of corticosteroids.
in other vasculitic diseases. The included studies suggest that
Difficulties remain with the application of the results of this review corticosteroids enhance the resolution of inflammatory markers
to Western populations, where there is no comparable severity risk and are associated with improved coronary artery outcomes.
score. The identification of groups who might gain the greatest This is thought to be due to suppression of the inflammatory
benefit from corticosteroids will remain problematic until a reliable reaction in KD that causes the coronary artery abnormalities (influx
risk stratification score is developed for this group. of neutrophils, large mononuclear cells and lymphocytes which
destroy the internal elastic lamina, followed by myofibroblast
Overall the results of this review are applicable to the majority of proliferation causing a coronary aneurysm (Eleftheriou 2013). In
children worldwide diagnosed with KD, and should aid with clinical addition, other systematic reviews on this question have come
decision-making. to similar conclusions on the efficacy of corticosteroids for KD
(Chen 2016; Yang 2018; Zhu 2012). There has also been some
Quality of the evidence study evidence to suggest that corticosteroids may be effective
in refractory KD as second-line treatment, where people are IVIG
See Summary of findings 1
resistant (Kimura 2017). However, this has not been a clear finding
All studies included in this meta-analysis were randomised trials, in this review (Wang 2020).
with variable incidence in the blinding of participants and outcome
assessors. However, we do not believe the risks of bias identified AUTHORS' CONCLUSIONS
in this review would affect the direction of the reported outcome
Implications for practice
variables. Trials reported data using a mixture of per-protocol
and ITT analyses. The overall certainty of the current evidence Moderate-certainty evidence shows that use of corticosteroids in
can be considered moderate. This is represented by several the acute phase of KD can be associated with reduced coronary
inconsistencies in the different findings incorporated within this artery abnormalities, shorter duration of hospital stay and reduced
review. inflammatory markers when compared to no corticosteroids. Low-
certainty evidence shows that the use of corticosteroids can reduce
We graded evidence according to the GRADE system. We duration of clinical symptoms, namely fever and rash. There were
considered evidence was moderate for incidence of coronary artery no serious adverse events or deaths reported with or without
abnormalities, incidence of serious adverse events, mortality, time corticosteroid use. None of the included studies reported on
for laboratory parameters to normalise and length of hospital long-term (greater than one year) coronary morbidity. Evidence
stay. We considered evidence was low for the duration of clinical presented in this systematic review agrees with current clinical
symptoms. We downgraded coronary artery abnormality evidence guidelines on the use of corticosteroids in the first-line treatment
for inconsistency, as subgroup analysis suggests that those with in KD.
low-risk scores, receiving single-dose treatment or second-line
treatment may not benefit. We downgraded the time for laboratory Implications for research
parameters to normalise for imprecision due to only one study with
a small number of participants being available for this outcome. We This meta-analysis shows the utility of corticosteroids in the
downgraded length of hospital stay as patients in one study (out treatment of the acute phase of KD. However, the follow-up periods
of the two studies included) received corticosteroids as second- were generally short and data are lacking to ascertain the long-term
line treatment, which may have been a confounding factor and benefits.
caused indirectness of evidence. We downgraded incidence of

Informed decisions.
Furthermore, the number of days from onset of treatment was ACKNOWLEDGEMENTS

not evaluated, since it is so variable and there is no protocol
that advises this. Some clinicians use the corticosteroids at first The review authors would like to thank Georgia M Connolly and
diagnosis, some use them in high-risk cases, some use them Matthew J Seager for their contributions to previous versions of this
as rescue therapy. It is hoped that the forthcoming Kawasaki review.
Disease Coronary Artery Aneurysm Prevention randomised trial
of corticosteroids (EUCTR2019-004433-17-GB) will help us answer The review authors and the Cochrane Vascular editorial base
these important questions. are grateful to the following peer reviewers for their time and
comments: Dr Mamoru Ayusawa, Nihon University School of
Overall, further trials and increased follow-up of previous Medicine & Itabashi Hospital, Tokyo, Japan; Prof Kevin Friedman,
trials is required to evaluate the longer-term implications of Boston Children's Hospital, Boston, USA; Prof Michael A Portman,
corticosteroid use in KD, to investigate possible geographical Seattle Children’s Hospital, Seattle, USA.
treatment differences, to investigate possible treatment differences
seen in higher-risk patients and to investigate treatment duration.

Informed decisions.
REFERENCES
References to studies included in this review trial of pulsed corticosteroid therapy for primary treatment
of Kawasaki disease. New England Journal of Medicine
Ikeda 2006 {published data only}
2007;356(7):663-75.
Ikeda K, Kobayashi T, Inoue Y, Okada Y, Tomomasa T,
Shinohara M, et al.Risk stratification and effectiveness of Ogata 2012 {published data only}
intravenous immunoglobulin plus prednisolone as the initial Ishii M, Ogata S, Ogihara Y, Sato K, Ebato T.Clinical utility of a
treatment of Kawasaki disease. European Journal of Pediatrics new strategy for preventing coronary artery lesions in refractory
2006;165(Suppl 1):38-9. Kawasaki disease patients: a randomized prospective study.
Pediatrics International 2012;54(Suppl 1):104.
Inoue 2006 {published data only}
Inoue Y, Okada Y, Shinohara M, Kobayashi T, Kobayashi T, * Ogata S, Ogihara Y, Honda T, Kon S, Akiyama K,
Tomomasa T, et al.A multicentre prospective randomized trial of Ishii M.Cortocosteroid pulse combination therapy for
corticosteroids in primary therapy for Kawasaki disease: clinical refractory Kawasaki disease: a randomized trial. Pediatrics
course and coronary artery outcome. Journal of Pediatrics 2012;129(1):e17-e23.
2006;149(3):336-41.
Okada 2003 {published data only}
Kobayashi 2012 {published data only} Okada Y, Shinohara M, Kobayashi T, Inoue Y, Tomomasa T,
Ikeda K, Saji T, Kobayashi T, Yahata T, Okamoto A, Arakawa H, Kobayashi T, et al, Gunma Kawasaki Disease Study Group.Effect
et al.Optimal timing of initial treatment in severe Kawasaki of corticosteroids in addition to intravenous gamma
disease: a sub-analysis of the RAISE study. Cardiology in the globulin therapy on serum cytokine levels in the acute
Young 2013;23:S144. phase of Kawasaki disease in children. Journal of Pediatrics
2003;143(3):363-7.
Ishii M, Ogata S, Honda T, Ogihatra Y.The efficacy and safety of
new strategy for refractory Kawasaki disease. Cardiology in the Sundel 2003 {published data only}
Young 2011;18(21):-. * Sundel RP, Baker AL, Fulton DR,
Newburger JW.Corticosteroids in the initial treatment of
JPRN-UMIN000000940.Randomized controlled trial to assess
Kawasaki disease: report of a randomised trial. Journal of
immunoglobulin plus steroid efficacy for Kawasaki disease
Pediatrics 2003;142(6):611-6.
(protocol). apps.who.int/trialsearch/Trial2.aspx?TrialID=JPRN-
UMIN000000940 (first received 27 January 2008). Sundel RP, Baker AL, Fulton DR, Newburger JW.Randomized
trial of pulse steroids in the initial treatment of Kawasaki
* Kobayashi T, Saji T, Otani T, Takeuchi K, Nakamura T,
disease (KD). Pediatric Research 2003;53:164.
Arakawa H, et al, RAISE study group investigators.Efficacy of
immunoglobulin plus prednisolone for prevention of coronary Wang 2020 {published data only}
artery abnormalities in severe Kawasaki disease (RAISE study):
Wang Z, Chen F, Wang Y, Li W, Xie X, Liu P, et
a randomised, open-label, blinded-endpoints trial. Lancet
al.Methylprednisolone pulse therapy or additional IVIG
2012;379(9826):1613-20.
for patients with IVIG-resistant Kawasaki disease. Journal
Kobayashi T, Saji T, Otani T, Takeuchi K, Nakamura T, Arakawa H, of Immunology Research 2020;2020:ID 4175821. [DOI:
et al.Efficacy of immunoglobulin plus prednisolone for 10.1155/2020/4175821]
prevention of coronary artery abnormalities in severe Kawasaki
disease. Pediatrics International 2012;54(Suppl 1):53.
References to studies excluded from this review
Kobayashi T, Saji T, Otani T, Takeuchi K, Nakamura T, Arakawa H, Asai 1985 {published data only}
et al.Significance of primary therapy with intravenous
Asai T.Follow up data of prospective controlled study for
immunoglobulin plus prednisolone for severe Kawasaki
the treatment, clinical sign and laboratory data of Kawasaki
disease: result from Japanese multicenter randomized clinical
disease. Shonika. Pediatrics of Japan 1985;26:995-1004.
trial. Circulation 2011;124:A9271.
Hashino 2001b {published data only}
Newburger 2007 {published data only}
Hashino K, Ishii M, Iemura M, Akagi T, Kato H.Re-treatment for
McCrindle BW, Li JS, Minich LL, Colan SD, Atz AM, Takahashi M,
immune globulin-resistant Kawasaki disease: a comparative
et al.Coronary artery involvement in children with Kawasaki
study of additional immune globulin and steroid pulse therapy.
disease: risk factors from analysis of serial normalized
Pediatrics International 2001;43(3):211-7.
measurements. Circulation 2007;116(2):174-9.
ISRCTN74427627 {published data only}
NCT00132080.Trial of pulse steroid therapy in Kawasaki Disease
– Pediatric Heart Network. clinicaltrials.gov/show/NCT00132080 ISRCTN74427627.Multi-centre randomised placebo-controlled
(first received 19 August 2005). trial of corticosteroids in the prevention of coronary artery
abnormalities in acute Kawasaki Disease (KD). www.isrctn.com/
Newburger JW, Sleeper LA, McCrindle BW, Minich LL, Gersony W, ISRCTN74427627 (first received 15 June 2005).
Vetter VL, Paediatric Heart Network Investigators.Randomized

Informed decisions.
Jibiki 2004 {published data only} Shiari 2011 {published data only}
Jibiki T, Terai M, Kurosaki T, Nakajima H, Suzuki K, Inomata H, et Shiari R, Beirami F.The addition of glucocorticoids to initial
al.Efficacy of intravenous immune globulin therapy combined IVIG therapy in children with Kawasaki disease. Clinical and
with dexamethasone for the initial treatment of acute Kawasaki Experimental Rheumatology 2011;29(2):432.
disease. European Journal of Pediatrics 2004;163(4-5):229-33.
Xu 2002 {published data only}
Kato 1979 {published data only} Xu JH, Luo XL.Observation of coronary artery diseases in
Kato H, Koike S, Yokoyama T.Kawasaki disease: effect of Kawasaki diseases with methylprednisolone. Journal of Applied
treatment on coronary artery involvement. Pediatrics Clinical Pediatrics 2002;17(6):680-1.
1979;63(2):175-9.
Yuan 2000 {published data only}
Kusakawa 1983 {published data only} Yuan Y.Analysis of recent effects for corticosteroid in treating
Kusakawa S.A prospective study for treatment of Kawasaki Kawasaki disease. Chinese Journal of Practical Paediatrics
disease with 3 regimes: aspirin, flubiprofen, predonisolone + 2000;15:49.
dipyridamole. Nippon Shonika Gakkai Zasshi 1983;87:2465-91.
Kusakawa S.A randomised controlled study of three different References to studies awaiting assessment
therapies for patients having Kawasaki disease. Nippon Shonika
ChiCTR1800017994 {published data only}
Gakkai Zasshi 1986;90:1844-9.
ChiCTR1800017994.Clinical study of glucocorticoid combined
Kusakawa S.Studies on the treatment of Kawasaki with immunoglobulin in the treatment of Kawasaki disease.
disease during acute stage. Nippon Shonika Gakkai Zasshi trialsearch.who.int/Trial2.aspx?TrialID=ChiCTR1800017994 (first
1985;89:814-8. received 25 August 2018).
Miura 2008 {published data only} ChiCTR-IOR-16007862 {published data only}

Miura M, Kohno K, Ohki H, Yoshiba S, Sugaya A, Satoh M.Effects ChiCTR-IOR-16007862.Research on individualized treatment for
of methylprednisolone pulse on cytokine levels in Kawasaki intravenous immunoglobulin resistant children with Kawasaki
disease patients unresponsive to intravenous immunoglobulin. disease. trialsearch.who.int/Trial2.aspx?TrialID=ChiCTR-
European Journal of Pediatrics 2008;167(10):1119-23. IOR-16007862 (first received 30 January 2016).
Nakamura 1985 {published data only} JPRN-UMIN000005022 {published data only}

Nakamura T.Studies on activated platelets in Kawasaki disease. JPRN-UMIN000005022.A methylprednisolone-plus-intravenous
Nippon Shonika Gakkai Zasshi 1985;89:1845-60. immunoglobulin combination therapy for refractory Kawasaki
disease. trialsearch.who.int/Trial2.aspx?TrialID=JPRN-
Nonaka 1995 {published data only} UMIN000005022 (first posted 1 November 2012).
Nonaka Z, Maekawa K, Okabe T, Eto Y, Kubo M.Randomized
controlled study of intravenous prednisolone and JPRN-UMIN000009946 {published data only}
gammaglobulin treatment in 100 cases with Kawasaki disease. JPRN-UMIN000009946.Prospective multicenter trial to assess
In: KatoH, editors(s). Kawasaki Disease. Amsterdam (the the effectiveness and safety of pulse methylprednisolone
Netherlands): Elsevier Science, 1995:328-31. combined with intravenous immunoglobulin (IVIG) as a
primary treatment in patients with acute Kawasaki disease.
Nonaka Z, Usui N, Mackawa K.The combination therapy center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?
with gamma-globulin and prednisolone is an supportive or recptno=R000011649 (first posted 20 February 2013).
alternative for Kawasaki disease in high risk. Pediatric Research
2000;47(4):565.
References to ongoing studies
Ogata 2009 {published data only}
EUCTR2019-004433-17-GB {published data only}
Ogata S, Ogihara Y, Nomoto K, Akiyama K, Nakahata Y, Sato K,
et al.Clinical score and transcript abundance patterns identify EUCTR2019-004433-17-GB.Kawasaki disease coronary artery
Kawasaki disease patients who may benefit from addition of aneurysm prevention trial. www.clinicaltrialsregister.eu/ctr-
methylprednisolone. Pediatric Research 2009;66(5):577-84. search/trial/2019-004433-17/GB (first received 16 December
2019).
Sekine 2012 {published data only}
ISRCTN71987471.Corticosteroids plus standard of care
Sekine K, Mochizuki H, Inoue Y, Kobayashi T, Suganuma E,
treatment versus standard of care treatment alone to prevent
Matsuda S, et al.Regulation of oxidative stress in patients with
heart complications in Kawasaki disease. trialsearch.who.int/
Kawasaki disease. Inflammation 2012;35(3):952-8.
Trial2.aspx?TrialID=ISRCTN71987471 (first received 31 March
Seto 1983 {published data only} 2020).
Seto S.A controlled trial of steroid pulse therapy for Kawasaki IRCT20181202041817N1 {published data only}
disease. Nippon Shonika Gakkai Zasshi 1983;87:399-406.
IRCT20181202041817N1.Evaluation of primary intravenous
methyl prednisolone pulse treatment in the prevention of

Informed decisions.
coronary artery abnormality in children with Kawasaki disease. Davies 2013

www.irct.ir/trial/35607 (first received 26 August 2019). Davies S, Gold-von Simson G.Should infliximab be used as an
adjuvant to IVIG in the treatment of children with Kawasaki
JPRN-UMIN000009524 {published data only}
disease who are at high risk for resistance to conventional
JPRN-UMIN000009524.A prospective randomized controlled therapy? Pediatric Cardiology 2013;34(7):1756.
trial of immunoglobulin plus prednisolone for Kawasaki
disease patients with high risk for coronary abnormalities. de Zorzi 1998
trialsearch.who.int/Trial2.aspx?TrialID=JPRN-UMIN000009524 de Zorzi A, Colan SD, Gauvreau K, Baker AL, Sundel RP,
(first received 17 December 2012). Newburger JW.Coronary artery dimensions may be
misclassified as normal in Kawasaki disease. Journal of
NCT04078568 {published data only}
Pediatrics 1998;133(2):254-8.
NCT04078568.Addition of methylprednisolone to initial
treatment for Kawasaki disease patients. clinicaltrials.gov/ Eleftheriou 2013
ProvidedDocs/68/NCT04078568/Prot_000.pdf (Date of protocol: Eleftheriou D, Levin M, Shingadia D, Tulloh R, Klein NJ,
25 July 2019). [CLINICALTRIALS.GOV: NCT04078568] Brogan PA.Management of Kawasaki disease. Archives of
Disease in Childhood 2013;99(1):74-83.
Additional references Gardner-Medwin 2002
Atkins 2004 Gardner-Medwin JM, Dolezalova P, Cummins C,
Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S, Southwood TR.Incidence of Henoch-Schönlein purpura,
et al.GRADE Working Group. Grading quality of evidence and Kawasaki disease, and rare vasculitides in children of different
strength of recommendations. BMJ 2004;328(7454):1490-4. ethnic origins. Lancet 2002;360(9341):1197-202.
Ayusawa 2005 Harnden 2002

Ayusawa M, Sonobe T, Uemura S, Ogawa S, Nakamura Y, Harnden A, Alves B, Sheikh A.Rising incidence of Kawasaki
Kiyosawa N, et al.Kawasaki Disease Research Committee. disease in England: analysis of hospital admission data. BMJ
Revision of diagnostic guidelines for Kawasaki disease (the 5th 2002;324(7351):1424-5.
revised edition). Pediatrics International 2005;47(2):232-4.
Hashino 2001a
Baumer 2006 Hashino K, Ishii M, Iemura M, Akagi T, Kato H.Re-treatment for
Baumer JH, Love SJ, Gupta A, Haines LC, Maconochie I, immune globulin-resistant Kawasaki disease: a comparative
Dua JS.Salicylate for the treatment of Kawasaki disease in study of additional immune globulin and steroid pulse therapy.
children. Cochrane Database of Systematic Reviews 2006, Issue Pediatrics International 2001;43(3):211-7.
4. Art. No: CD004175. [DOI: 10.1002/14651858.CD004175.pub2]
Higgins 2011
Brogan 2002 Higgins JP, Green S, editor(s).Cochrane Handbook for
Brogan P, Bose A, Burgner D, Shingadia D, Tulloh R, Michie C, et Systematic Reviews of Interventions Version 5.1 (updated
al.Kawasaki disease: an evidence based approach to diagnosis, March 2011). The Cochrane Collaboration, 2011. Available from
treatment, and proposals for future research. Archives of training.cochrane.org/handbook/archive/v5.1/.
Disease in Childhood 2002;86(4):286-90.
Hokosaki 2012
Burns 1996 Hokosaki T, Mori M, Nishizawa T, Nakamura T, Imagawa T,
Burns JC, Shike H, Gordon JB, Malhotra A, Schoenwetter M, Iwamoto M, et al.Long-term efficacy of plasma exchange
Kawasaki T.Sequelae of Kawasaki disease in adolescents and treatment for refractory Kawasaki disease. Pediatrics
young adults. Journal of the American College of Cardiology International 2012;54(1):99-103.
1996;28(1):253-7.
Holman 2003
Chen 2016 Holman RC, Curns AT, Belay ED, Steiner CA,
Chen S, Dong Y, Kiuchi MG, Wang J, Li R, Ling Z, et al.Coronary Schonberger LB.Kawasaki syndrome hospitalizations in
artery complication in Kawasaki disease and the importance of the United States; 1997 and 2000. Pediatrics 2003;112(3 Pt
early intervention: a systematic review and meta-analysis. JAMA 1):495-501.
Pediatrics 2016;170(12):1156-63.
Jones 2014
Daniels 2012 Jones KS, Baumer JH, Gupta A, Dua JS.Steroid hormone
Daniels LB, Gordon JB, Burns JC.Kawasaki disease: late treatment for Kawasaki disease in children. Cochrane Database
cardiovascular sequelae. Current Opinion in Cardiology of Systematic Reviews 2014, Issue 6. Art. No: CD003284. [DOI:
2012;27(6):572-7. 10.1002/14651858.CD003284.pub3]
Kato 1996
Kato H, Sugimura T, Akagi T, Sato N, Hashino K, Maeno Y,
et al.Long-term consequences of Kawasaki disease. A 10-
Informed decisions.
to 21-year follow-up study of 594 patients. Circulation for the treatment of Kawasaki disease in children. Cochrane
1996;94(6):1379-85. Database of Systematic Reviews 2003, Issue 4. Art. No:
CD004000. [DOI: 10.1002/14651858.CD004000]
Kawasaki 1967
Kawasaki T.Acute febrile mucocutaneous syndrome with Research Committee on Kawasaki Disease 1984
lymphoid involvement with specific desquamation of the Ministry of Health and Welfare Japan.Report of subcommittee
fingers and toes in children. Arerugi 1967;16(3):178-222. on standardisation of diagnostic criteria and reporting
of coronary artery lesions in Kawasaki disease. Research
Kimura 2017 Committee on Kawasaki Disease 1984.
Kimura M, Harazaki M, Fukuoka T, Asakura I, Sakai H,
Kamimaki T, et al.Targeted use of prednisolone with the Review Manager 2014 [Computer program]
second IVIG dose for refractory Kawasaki disease. Pediatrics Review Manager (RevMan).Version 5.3. Copenhagen: The Nordic
International 2017;59(4):397-403. Cochrane Centre, The Cochrane Collaboration, 2014.
Kobayashi 2008 Singh 2015

Kobayashi T, Inoue Y, Morikawa A.Risk stratification and Singh S, Vignesh P, Burgner D.The epidemiology of Kawasaki
prediction of resistance to intravenous immunoglobulin disease: a global update. Archives of Disease in Childhood
in Kawasaki disease. Japanese Journal of Clinical Medicine 2015;100(11):1084-8.
2008;66(2):332-7.
Sleeper 2011
Kobayashi 2020 Sleeper LA, Minich LL, McCrindle BM, Li JS, Mason W, Colan SD,
Kobayashi T, Ayusawa M, Suzuki H, Abe J, Ito S, Kato T, et et al, Pediatric Heart Network Investigators.Evaluation
al.Revision of diagnostic guidelines for Kawasaki disease (6th of Kawasaki disease risk-scoring systems for intravenous
revised edition). Pediatrics International 2020;62(10):1135-8. immunoglobulin resistance. Journal of Pediatrics
2011;158(5):831-5, e3.
Lefebvre 2011
Lefebvre C, Manheimer E, Glanville J.Chapter 6: Searching for Tatara 1987
studies. In: Higgins JP, Green S, editor(s). Cochrane Handbook Tatara K, Kusakawa S.Long-term prognosis of giant coronary
for Systematic Reviews of Interventions Version 5.1.0 (updated aneurysm in Kawasaki disease: an angiographic study. Journal
March 2011). The Cochrane Collaboration, 2011. Available from of Pediatrics 1987;111(5):705-10.
training.cochrane.org/handbook/archive/v5.1/.
Wood 2009
Levin 1991 Wood LE, Tulloh RM.Kawasaki disease in children. Heart
Levin M, Tizard EJ, Dillon MJ.Kawasaki disease: recent 2009;95(10):787-92.
advances. Archives of Disease in Childhood 1991;66(12):1369-72.
Yang 2018
Levin 2013 Yang TJ, Lin MT, Lu CY, Chen JM, Lee PI, Huang LM, et
Levin M.Steroids for Kawasaki disease: the devil is in the detail. al.The prevention of coronary arterial abnormalities in
Heart 2013;99(2):69-70. Kawasaki disease: a meta-analysis of the corticosteroid
effectiveness. Journal of Microbiology, Immunology and Infection
McCrindle 2017 2018;51(3):321-31.
McCrindle BW, Rowley AH, Newburger JW, Burns JC, Bolger AF,
Gewitz M, et al, American Heart Association Rheumatic Zhu 2012
Fever, Endocarditis, and Kawasaki Disease Committee of the Zhu BH, LV HT, Sun L, Zhang JM, Cao L, Jia HL, et al.A meta-
Council on Cardiovascular Disease in the Young, Council on analysis on the effect of corticosteroid therapy in Kawasaki
Cardiovascular and Stroke Nursing, Council on Cardiovascular disease. European Journal of Pediatrics 2012;171(3):571-8.
Surgery and Anesthesia, and Council on Epidemiology and
Prevention.Diagnosis, treatment, and long-term management
of Kawasaki disease: a scientific statement for health References to other published versions of this review
professionals from the American Heart Association. Circulation Wardle 2014
2017;135(17):927-99.
Wardle AJ, Kiddy HC, Seager MJ, Tulloh RM,
Nakamura 2012 Levin M.Corticosteroids for the treatment of Kawasaki disease
in children. Cochrane Database of Systematic Reviews 2014,
Nakamura Y, Yashiro M, Uehara R, Sadakane A, Tsuboi S,
Issue 8. Art. No: CD011188. [DOI: 10.1002/14651858.CD011188]
Aoyama Y, et al.Epidemiologic features of Kawasaki disease in
Japan: results of the 2009–2010 nationwide survey. Journal of Wardle 2017
Epidemiology 2012;22(3):216-21.
Wardle AJ, Connolly GM, Seager MJ, Tulloh RM.Corticosteroids
Oates-Whitehead 2003 for the treatment of Kawasaki disease in children. Cochrane
Database of Systematic Reviews 2017, Issue 1. Art. No:
Oates-Whitehead RM, Baumer JH, Haines L, Love S,
CD011188. [DOI: 10.1002/14651858.CD011188.pub2]
Maconochie IK, Gupta A, et al.Intravenous immunoglobulin

Informed decisions.
* Indicates the major publication for the study
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Ikeda 2006
Study characteristics
Methods Participants "randomly assigned" and stratified into high- or low-risk groups
Unclear how many centres
Participants 178 children with KD were randomly assigned
No statement of baseline characteristics at enrolment between groups
90 assigned to corticosteroid, 88 to control, divided as follows
Low risk: corticosteroid: 45 participants, control: 46 participants
High risk: corticosteroid: 45 participants, control: 42 participants
Interventions Control
1. IVIG
Treatment
1. IVIG
2. Prednisolone
No information on route, duration or dose of corticosteroid used
Outcomes 1. Coronary artery abnormalities
Unclear at what time point outcome measure assessed
Funding No details provided
Declarations of interest No details provided
Notes No statement of blinding participants or researchers
No clear dose, route or duration of corticosteroid use
No clear definition of "coronary artery abnormalities"
No indication of when measurements were taken after treatment
Undertaken at Gunma Children's Medical Centre, Japan
Risk of bias
Bias Authors' judgement Support for judgement
Random sequence genera- Unclear risk Quote: "Randomly assigned".

tion (selection bias)
Comment: unclear how this was done.
Allocation concealment Unclear risk Not stated.

(selection bias)

Informed decisions.
Ikeda 2006 (Continued)

Comment: impossible to know if participants or physicians were blinded to the
allocation.
Blinding of participants Unclear risk Not stated.

and personnel (perfor-
mance bias) Comment: impossible to know if participants or physicians were blinded to the
All outcomes study treatment.
Blinding of outcome as- Unclear risk Not stated.

sessment (detection bias)
All outcomes Comment: impossible to know if there was any blinding of outcome assessors.
Incomplete outcome data Unclear risk All data present.

(attrition bias)
All outcomes
Selective reporting (re- Low risk Published methodology consistent with that reported in published results.
porting bias)
Other bias Low risk None identified.
Inoue 2006
Methods Randomised, prospective, non-blinded, multicentre trial (12 centres in Gunma and Saitama, Japan)
September 2000 to March 2005
Participants 588 children referred to the medical institutions. 410 were ineligible (389: parents decided against par-
ticipation, 17: had previous diagnosis of KD, 4: had a coronary artery abnormality before randomisa-
tion)
178 participants diagnosed with KD (had ≥ 5 of: fever (> 38 °C), non-exudative conjunctival injection,
changes in the oropharynx (including mucosal erythema; dry, cracked lips; and "strawberry tongue"),
changes in the extremities (including palmar and plantar erythema), oedema of the hands and feet or
periungual desquamination, rash and cervical lymphadenopathy
Groups well balanced in baseline demographic and clinical characteristics, and risk score. All partici-
pants followed up for ≥ 2 months (range 2–50 months)
Control: 88
Corticosteroid: 90
1. IVIG 1 g/kg/day for 2 days given over 12 hours

2. Aspirin 30 mg/kg/day decreased to 5 mg/kg/day once CRP resolved
Treatment
1. IVIG 1 g/kg/day for 2 days given over 12 hours

2. Aspirin 30 mg/kg/day decreased to 5 mg/kg/day once CRP resolved
3. Prednisolone sodium succinate 2 mg/kg/day in 3 divided doses, IV until fever resolved, then given
orally until CRP normalised. Once CRP normalised, prednisolone given orally in tapering doses over
15 days in 5-day steps (2 mg/kg/day for 5 days, 1 mg/kg/day for 5 days, then 0.5 mg/kg/day for 5 days).
If there was difficulty with oral delivery, doses were given IV

Informed decisions.
Inoue 2006 (Continued)

Excluding 1 participant whose corticosteroid treatment was discontinued at the discretion of the treat-
ing physician, duration of corticosteroid treatment 18–100 days (median 23 days). Total dose of pred-
nisolone 23.5–90 mg/kg (median 32 mg/kg)
Outcomes Primary outcome
1. Detection of coronary artery abnormality (luminal diameter > 3.0 mm in child < 5 years, or > 4.0 mm in
child > 5 years, when the internal diameter of a segment was ≥ 1.5 times that of an adjacent segment,
or when a luminal contour was clearly irregular, before a 1-month echocardiographic assessment)
Secondary outcomes
1. Duration of fever after initial treatment

2. Time to normalisation of CRP level (< 5 mg/dL)
3. Incidence of treatment failure and recurrence
Declarations of interest Quote: "The study was conducted in accordance with good clinical practice and the Declaration of
Helsinki and was approved by the Gunma University Ethics Committee in May 2000".
No details about declarations of interest provided.
Notes Study terminated on 31 March 2005 by data monitoring committee at the time of the deadline despite a
lower enrolment rate than expected.
Risk of bias
Random sequence genera- Unclear risk Quote: "centre randomly assigned the patient".
Comment: no clear statement of mechanism for random assignment.
Allocation concealment Low risk Quote: "Centrally maintained table of random numbers".
(selection bias)
Comment: likely adequate.
Blinding of participants Unclear risk Non-blinded.

mance bias) Comment: unclear how this would have influenced results.
All outcomes
Blinding of outcome as- Unclear risk Echocardiography operators were non-blinded.

All outcomes Quote: "findings reviewed in a non blinded manner".
Unclear if laboratory team processing blood samples were blinded.
Comment: unclear how this would have influenced results.
Incomplete outcome data Low risk Key outcome variables were reported as per-protocol analysis with inten-
(attrition bias) tion-to-treat analysis variable performed.
All outcomes
porting bias)

Informed decisions.
Kobayashi 2012
Methods Prospective, randomised, multicentre (74 hospitals in Japan) study with blinded endpoints
29 September 2008 to 2 December 2010
Participants Children with diagnosis of KD (Japanese diagnostic guidelines; Ayusawa 2005) and a risk score of ≥ 5
points, which emphasised the positive predictive value of no response to initial treatment with IVIG (2
points for each of the following: blood sodium ≤ 133 mmol/L, ≤ 4 days of illness at diagnosis, AST ≥ 100
units/L, neutrophils ≥ 80% of total WCC; 1 point for each of the following: platelet count ≤ 30 × 104/μL,
CRP > 100 mg/L, age ≤ 12 months)
Excluded: children with history of KD, diagnosed on or after day 9 of fever, with coronary artery abnor-
malities prior to enrolment, afebrile prior to enrolment, had received corticosteroids in the previous 30
days before the study, who had received IVIG in the previous 180 days before the study, with concomi-
tant severe medical disorders, with suspected infectious disease
2014 children assessed for eligibility. 1547 did not meet inclusion criteria (1436 had low-risk scores, 44
had previous diagnosis of KD, 24 had defervescence, 16 had suspected infectious disease, 12 had pres-
ence of coronary artery abnormality at diagnosis, 10 had illness for > 9 days, 3 had pre-existing severe
disease, 1 had a history of IVIG use, 1 had history of corticosteroid use, 219 declined to participate
248 participants randomly assigned
Treatment group: 125 (4 excluded: 1 withdrew consent, 2 had presence of coronary artery abnormality,
1 had misdiagnosis of KD)
Control group: 123 (2 excluded: 1 had presence of coronary artery abnormalities at enrolment, 1 did
not receive IVIG)
No significant differences between groups at baseline
1. IVIG 2 g/kg over 24 hours

2. Aspirin 30 mg/kg/day until afebrile, then 3–5 mg/kg/day for ≥ 28 days from fever onset
Treatment

2. Aspirin 30 mg/kg/day until afebrile, then 3–5 mg/kg/day for ≥ 28 days from fever onset
3. IV prednisolone 2 mg/kg/day in 3 divided doses for 5 days, or until fever resolved, if longer. When CRP
normalised (< 5 mg/L) prednisolone doses were tapered in 5-day steps over 15 days, from 2 mg/kg/
day, to 1 mg/kg/day, to 0.5 mg/kg/day
4. famotidine 0.5 mg/kg/day (histamine 2 blocker) while receiving prednisolone
1. Detection of coronary artery abnormality during study period (luminal diameter > 3.0 mm in a child
aged < 5 years, or > 4.0 mm in a child aged > 5 years, when the internal diameter of a segment was ≥
1.5 times that of an adjacent segment, or when a luminal contour was clearly irregular)
Secondary outcomes
1. Incidence of coronary artery abnormality at week 4 after enrolment

2. Incidence of need for additional rescue treatment
3. CRP concentrations at 1 and 2 weeks after enrolment
4. Incidence of serious adverse events

Informed decisions.
Kobayashi 2012 (Continued)
Funding Funding source: Japanese Ministry of Health, Labour and Welfare, which had no role in study design,
data collection, data analysis, data interpretation or writing of the report.
Declarations of interest Study authors declared that there were no conflicts of interest.
Notes RAISE study. Registered with University Hospital Medical Information Network clinical trials registry,
number UMIN000000940
Preplanned interim analysis after enrolment of the 200th participant in June 2010 demonstrated signif-
icance difference in the incidence of coronary artery abnormalities between the 2 treatment groups (P
< 0.0001), therefore independent data and safety monitoring committee recommended termination of
the study. Study terminated on 2 December 2010
Risk of bias
Random sequence genera- Low risk Quote: "Computer generated randomisation sequence".
Comment: likely adequate and performed.
Allocation concealment Low risk Quote: "Centrally maintained table of random numbers".
(selection bias)
Comment: likely adequate.
Blinding of participants Unclear risk Quote: "Participant and treating physician non-blinded".
All outcomes
Blinding of outcome as- Low risk Quote: "Blinded end point".

All outcomes Echocardiography assessors blinded.
Comment: appeared adequate.
Incomplete outcome data Low risk Quote: "Intention to treat analysis used".
(attrition bias)
All outcomes Comment: appeared adequate.
porting bias)
Newburger 2007
Methods Randomised, double-blind, placebo-controlled multicentre (8 centres in North America) trial
December 2002 to December 2004
Participants All participants between day 4 and day 10 of illness

Informed decisions.
Newburger 2007 (Continued)

Inclusion criteria: child met > 4 of principal clinical criteria; coronary artery z-score of > 2.5 (RCA or LAD)
and met 2 principal clinical criteria if < 6 months, 3 criteria if > 6 months; presence of a coronary artery
aneurysm and met ≥ 1 clinical criterion
Exclusion criteria: previous treatment with IVIG; oral corticosteroid use in preceding 2 weeks; presence
of a disease known to mimic KD; contraindications to corticosteroid use; allergy to aspirin
589 children identified. 276 were ineligible (185 met ≥ 1 exclusion criterion. 95 did not meet inclusion
criteria)
313 considered eligible, of these 199 parents granted consent
199 children randomised
Treatment: 101 participants
Control: 98 participants
No significant differences between groups at baseline randomisation
1. Diphenhydramine 1 mg/kg
3. Aspirin 80–100 mg/kg/day until they were afebrile for 48 hours, then 3–5 mg/kg/day aspirin until study
completion
Treatment
1. Single, pulse dose of IV methylprednisolone 30 mg/kg

2. Diphenhydramine 1 mg/kg
4. Aspirin 80–100 mg/kg/day until they were afebrile for 48 hours, then 3–5 mg/kg/day aspirin until study
completion
If children had recurrent fever > 36 hours after the initial infusion (and no alternative source was found)
then a further dose of IVIG 2 mg/kg, then if they remained pyrexial after a further 36 hours another dose
of IVIG 2 mg/kg was administered
1. Detection of coronary artery abnormality during study period (luminal diameter > 3.0 mm in a child
aged < 5 years, or > 4.0 mm in a child aged > 5 years, when the internal diameter of a segment was ≥
1.5 times that of an adjacent segment, or when a luminal contour was clearly irregular, of a z-score
of > 2.5)
Secondary outcomes
1. Duration of fever
2. Number of days spent in hospital until first discharge
3. Number of total days spent in hospital during study period
4. Number of episodes of retreatment with IVIG
5. Incidence of adverse events
6. Laboratory data at week 1 and 5 after randomisation
Funding Supported by grants from the National Institutes of Health (U01 HL068285 and RR 02172, to Dr New-
burger Ms Baker, and Dr Sundel; U01 HL068270, to Drs Sleeper and Colan, Mr Mitchell and Ms Klein; U01
HL068288, to Dr McCrindle; U01 HL068292, to Dr Minich; U01 HL068290, to Dr Gersony; U01 HL068279,
to Dr Vetter; U01 HL068281, to Dr Atz; and U01 HL068269, to Dr Li) and from the Higgins Family Cardiol-
ogy Research Fund (to Dr Colan).

Informed decisions.
Newburger 2007 (Continued)
Declarations of interest No potential conflict of interest relevant to this article reported.
Notes ClinicalTrials.gov number: NCT00132080
Risk of bias
Random sequence genera- Unclear risk Quote: "Randomly assigned" with use of "dynamic balancing" at each centre.
Comment: no information of what "dynamic balancing" entailed.

(selection bias)
allocation.
Blinding of participants Low risk Quote: "double blind".

mance bias) Comment: appeared adequate.
All outcomes
Blinding of outcome as- Low risk Quote: "double blind".

Incomplete outcome data Low risk Quote: "Interim analysis reviewed by an independent data and safety monitor-
(attrition bias) ing board".
All outcomes
Data excluded for (quote) "6 patients who were discovered to have met an ex-
clusion criteria, for 2 patients who did not receive IV steroid despite randomi-
sation into this group, and 8 patients who were enrolled because they had
coronary abnormalities but did not meet the classic criteria for KD."
Selective reporting (re- Low risk Pretrial protocol published on ClinicalTrials.gov and consistent with method-
porting bias) ology and results published.
Ogata 2012
Methods Single centre in Japan (Kitasato University Hospital), unblinded with stratification of participants into
high- and low-risk groups. Participants were then randomly assigned to each treatment group
April 2007 to November 2010
Participants Inclusion criteria: 5 of 6 clinical criteria for KD
Exclusion criteria: previous diagnosis of KD; presence of a coronary artery abnormality before treat-
ment; use of corticosteroid treatment before recruitment
122 participants enrolled into the study – all Japanese
100% consent rate for parents
High-risk group (Egami score > 3): aspirin + IVIG: 26, aspirin + IVIG + corticosteroid: 22
Informed decisions.
Ogata 2012 (Continued)

Low-risk group (Egami score < 2; Table 1): aspirin alone: 6, aspirin + IVIG: 62
No significant differences at baseline between groups in the high-risk group
Interventions Low-risk group
Control
1. Aspirin 30 mg/kg 3 times daily until afebrile for 36 hours, then 5 mg/kg/day
Treatment
High-risk group
Control
Treatment
3. Single, pulse dose of IV methylprednisolone 30 mg/kg over 2 hours
1. Treatment resistance (temperature > 37.5 °C) at 36 hours
Secondary outcomes
2. Laboratory markers of vasculitic markers
3. Z-scores of coronary arteries
Funding Grant-in-aid for scientific research from Ministry of Education, Culture, Sports, Science and Technolo-
gy; a Parents' Association Grant at Kitasato University School of Medicine, and a grant-in-aid from the
Kawasaki Disease Research Centre in Japan
Declarations of interest No conflict of interest reported.
Notes Trial number: UMIN000005021
Risk of bias
Random sequence genera- Low risk Quote: randomly assigned using a "random number list".

(selection bias)
allocation.

Informed decisions.
Ogata 2012 (Continued)

All outcomes
Blinding of outcome as- Low risk Quote: "double-blinded".

Incomplete outcome data Low risk All data included in analysis.

(attrition bias)
All outcomes
porting bias)
Okada 2003
Methods Prospective, randomised, multicentre (9 centres in Gunma Prefecture, Japan) study. Once participants
identified, randomised into treatment groups
June 2001 to May 2002
Participants Inclusion criteria: ≥ 5 of the following: fever (> 38 °C), non-exudative conjunctival injection, changes in
the oropharynx (including mucosal erythema; dry, cracked lips; and strawberry tongue), changes in ex-
tremities (palmar and plantar erythema, oedema of the hands and feet, or periungual desquamation
in subacute phase of the disease), rash, cervical lymphadenopathy; enrolled within 9 days of onset of
fever
32 participants recruited: corticosteroid treatment group: 14, control group: 18
Age, sex distribution and severity score were similar between groups
1. IVIG 1 g/kg for 2 days

2. Aspirin 30 mg/kg per day throughout study period
3. Dipyridamole 2 mg/kg throughout study period
Treatment
• IVIG 1 g/kg/day for 2 days

• Prednisolone IV 2 mg/kg 3 times daily until resolution of fever, then oral prednisolone 2 mg/kg 3 times
daily until CRP normalised, then a reducing regimen of prednisolone was used, 1 mg/kg twice daily
for 5 days, then 0.5 mg/kg once daily for 5 days
• Aspirin 30 mg/kg per day throughout study period
• Dipyridamole 2 mg/kg throughout study period
1. Cytokine levels pretreatment, post IVIG dose, day 7 and day 14
Secondary outcomes

Informed decisions.
Okada 2003 (Continued)

1. Mean duration of fever
2. Time to normalisation of CRP
3. Coronary artery abnormality (luminal diameter > 3.0 mm in a child aged < 5 years, or > 4.0 mm in a child
aged > 5 years, when the internal diameter of a segment was ≥ 1.5 times that of an adjacent segment,
or when a luminal contour was clearly irregular)
4. Adverse effects
Declarations of interest Study conducted in accordance with good clinical practice and the Declaration of Helsinki and was ap-
proved by the Gunma University Ethics Committee in May 2000.
No details of declarations of interest provided.
Notes
Risk of bias
Random sequence genera- Low risk Quote: "Randomised via a controller in the registration centre".
Allocation concealment Low risk Quote: "Controller in the registration centre".

(selection bias)
Comment: efforts made to separate randomisation from the study trial, reduc-
ing chance of information spread.

All outcomes
Blinding of outcome as- Unclear risk Unclear if blinded.

All outcomes Comment: unclear how this would have influenced results.
Incomplete outcome data Low risk Data for all participants appeared in analysis.
(attrition bias)
All outcomes
porting bias)
Sundel 2003
Methods Prospective randomised single-centre (Boston, North America) trial. Once recruited participants were
randomised to each treatment group
February 1998 to November 2000
Participants Inclusion criteria: fever < 10 days; 4 out of 5 clinical criteria

Informed decisions.
Sundel 2003 (Continued)

Exclusion criteria: previous diagnosis of KD; presence of coronary artery aneurysm; possible infection;
known hypersensitivity to IVIG; contraindications to corticosteroid therapy
71 children met eligibility criteria. 10 declined consent. 20 not enrolled due to insufficient time prior to
IVIG infusion. 2 excluded due to coronary artery abnormalities
39 participants recruited; control: 21, treatment: 18
No significant differences in demographic data at enrolment

2. Aspirin 20–25 mg/kg every 6 hours until afebrile for 48 hours, then 3–5 mg/kg/day for duration of study
3. Diphenhydramine pretreatment prior to IVIG
Treatment

2. Aspirin 20–25 mg/kg every 6 hours until afebrile for 48 hours, then 3–5 mg/kg/day for the duration of
the study
3. Diphenhydramine pretreatment prior to IVIG
4. Single, pulse dose of IV methyl prednisolone 30 mg/kg (maximum dose 1.5 g) 3 hours before IVIG
Secondary outcomes
1. Length of hospital stay

2. Inflammatory markers at 6 weeks
3. Coronary artery abnormalities
4. Incidence of adverse events
Funding From the Departments of Medicine and Cardiology, Children's Hospital, and the Department of Pedi-
atrics at Harvard Medical School, Boston, Massachusetts, USA. Supported in part by the Kobren Fund.
Declarations of interest Study sponsors had no role in study design; in the collection, analysis or interpretation of data; in the
writing of the report; or in the decision to submit the paper for publication
Notes
Risk of bias
Random sequence genera- Unclear risk Quote: "Randomly assigned".

Comment: unclear how this was done.

(selection bias)
allocation.
Blinding of participants Unclear risk Quote: "Participants and investigators were not blinded to treatment assign-
and personnel (perfor- ment".
mance bias)
All outcomes Comment: unclear risk.

Informed decisions.
Sundel 2003 (Continued)
Blinding of outcome as- Low risk Quote: "Outcome markers were generally measured objectively (e.g. temp,
sessment (detection bias) laboratory values), or were read by blinded observers unaware of patients'
All outcomes treatment status (echocardiograms)".
Incomplete outcome data Low risk 95% of children had echocardiography at 2 weeks.
(attrition bias)
All outcomes 93% of children had echocardiography at 6 weeks (1 in treatment group, 2 in
control group).
porting bias)
Wang 2020
Methods Prospective, single-centre, non-blinded randomised controlled trial in children with initial IVIG-resis-
tant KD
Participants randomised using a random numbers table
January 2018 to June 2020 (30 months)
Participants 955 children with KD at the Department of Cardiology of Guangzhou Women and Children's Medical
Center in China from January 2018 to June 2019 were selected and initially treated with IVIG.
80 cases of these who were assessed as IVIG resistant were randomly divided into 2 groups. 40 per
group
Inclusion criteria: diagnosed and hospitalised in study hospital who met the diagnostic criteria for KD,
established by American Heart Association in 2017.
Exclusion criteria: presence of other diseases that affect the temperature change in the course of the
KD, such as sepsis, influenza and juvenile idiopathic arthritis; absence of detailed information about
their initial treatment outside the hospital; history of KD; treatment with hormone or immunosuppres-
sant therapy in preceding 30 days; presence of a severe immune disease, such as immunodeficiency or
chromosomal abnormality; refusal to sign informed consent; inability to follow up for ≥ 6 months
1. IVIG 2 g/kg single intravenous infusion

2. Aspirin 30–50 mg/kg/day orally
Treatment
1. Methylprednisolone 15 mg/kg/day IV for 3 days, without a subsequent course and taper of oral pred-
nisone
Outcomes Outcomes (not stated in the study whether primary/secondary)
1. The whole fever time (total days from the onset of fever (> 38 °C) to stable temperature for 48 hours
after treatment)
2. Duration of fever after retreatment (time required (in hours) to stabilise body temperature for 48 hours
from the start of retreatment.)
3. Hospital days (from admission date to a discharge date after retreatment)
4. Medical costs (Yuan)

Informed decisions.
Wang 2020 (Continued)

5. Readmission rate (i.e. admitted to hospital within 3 days of discharge)
6. Laboratory examination difference (calculated using the equation values 7 days after treatment minus
values 1 day before treatment)
a. Difference in white blood cells
b. Difference in neutrophils
c. Difference in CRP
d. Difference in haemoglobin
e. Difference in platelets
f. Difference in albumin
g. Difference in sodium
7. CAL outcomes were followed up > 2 years. Diagnostic criteria of CALs: the z-scores from echocardio-
graphic assessment of luminal dimensions are classified into 5 categories as follows: 1. no involve-
ment: z-scores always < 2; 2. dilation only: z-score ≥ 2 but < 2.5; 3. small aneurysm: z-score ≥ 2.5 to < 5;
4. medium aneurysm: z-score ≥ 5 to < 10 and absolute dimension < 8 mm; 5. giant aneurysm: z-score
≥ 10, or absolute dimension ≥ 8 mm. Z-scores measured at 7 days, 1 month, 3 months, 6 months, 12
months and 2 years.
Funding Study supported by the Guangdong Natural Science Fund, China (grant number 2016A030313836), and
the Guangdong Science and Technology Project of China (grant number 2017A030223003).
Declarations of interest Authors declared no competing financial interests.
Notes
Risk of bias
Random sequence genera- Low risk Random number table used to allocate the participants to group.
Allocation concealment Low risk Non-participating nurses generated the allocation.

(selection bias)
Blinding of participants Unclear risk Non-blinding of participants and personnel; however, unlikely to have affected
and personnel (perfor- objectively measured outcomes.
mance bias)
All outcomes
Blinding of outcome as- Unclear risk Unclear how the non-blinding of outcome assessors would have affected
sessment (detection bias) those measuring echocardiography or duration of clinical symptoms/hospital
All outcomes days.
Incomplete outcome data Low risk All or most of data accounted for.
(attrition bias)
All outcomes
Selective reporting (re- Unclear risk Numbers receiving third-line treatment were reported differently in the figures
porting bias) compared to the text description, which raises questions about bias. However,
this was not relevant for the data analysed and all the figures were present for
the main intervention.
Other bias Low risk No other sources of bias found.
AST: aspartate aminotransferase; CRP: C-reactive protein; IV: intravenous; IVIG: intravenous immunoglobulin; KD: Kawasaki disease; LAD:
left anterior descending; RCA: right coronary artery; WCC: white cell count.

Informed decisions.
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Asai 1985 Intervention included dipyridamole as well as prednisolone. Therefore, effect of corticosteroids not
ascertainable. Primary outcome of arterial changes was poorly defined.
Hashino 2001b Corticosteroids only used if children were non-responders to 2 treatments with IVIG. These IVIG
non-responsive children were then divided into groups allocated to receiving corticosteroid (9 chil-
dren) and those who received another dose of IVIG (8 children). This caused a significant delay in
the administration of any corticosteroids.
ISRCTN74427627 Study stopped early due to lack of funding/sponsorship.
Jibiki 2004 Investigators assessed the laboratory mechanism of action of IVIG and corticosteroid therapy in
KD.
Kato 1979 Not randomised. Reported different protocols used at different hospitals.
Kusakawa 1983 Intervention included dipyridamole as well as prednisolone. Therefore, effect of corticosteroids not
ascertainable.
Miura 2008 Other than age, this study did not specify inclusion or exclusion criteria for participants.
Nakamura 1985 Study reported 3 arms. Prednisolone given with dipyridamole. Therefore, effect of corticosteroids
not ascertainable. The outcome of this study was regarding platelet laboratory processes.
Nonaka 1995 Other than age, this study did not specify any inclusion or exclusion criteria for participants.
Ogata 2009 Investigators assessed the molecular mechanism of action of IVIG and corticosteroid therapy in KD.
Therefore, effect of corticosteroids not ascertainable.
Sekine 2012 There were 6 healthy children without KD and other diseases who participated in this study, which
did not match the inclusion criteria for this review.
Seto 1983 Inadequate information on methodology for validation, specifically no evidence of randomisation.
Shiari 2011 Full-text version of the study not available despite attempts to contact author. Abstract did not
include enough information on methods or outcome measures. Inadequate details on exact ap-
proach.
Xu 2002 Interventions included immunoglobulin versus corticosteroids, meaning corticosteroids were not
the only differential factor between the groups and, therefore, their effect could not be clearly de-
fined. Inadequate information on methodology, specifically around randomisation, despite con-
tacting authors.
Yuan 2000 No exclusion criteria for participants. Therefore, it was unclear whether the participants had posi-
tive blood cultures, which is an exclusion criterion for this review. Study did not clearly define coro-
nary aneurysm or how this was measured.
IVIG: intravenous immunoglobulin; KD: Kawasaki disease.
Characteristics of studies awaiting classification [ordered by study ID]
ChiCTR1800017994
Methods Unknown

Informed decisions.
ChiCTR1800017994 (Continued)
Participants Inclusion criteria: 2017 AHA KD diagnostic standard:
1. Fever for > 5 days and is accompanied by ≥ 4 of the following 5 main characteristics:
a. changes in limbs: in the acute phase, erythema and oedema of the hands and feet, during the
recovery period, membranous desquamation of the fingertips;
b. pleomorphic rash;
c. bilateral painless conjunctival hyperaemia without exudate;
d. lip and mouth changes: lip flushing and cleft palate, strawberry tongue, oral and pharyngeal
mucosa diffuse redness;
e. swelling of the cervical lymph nodes (diameter = 1.5 cm), usually unilateral.
2. Fever with the above-mentioned main clinical manifestations of < 4, if 2-dimensional echocardio-
graphy or coronary angiography found coronary artery disease, can be diagnosed as KD.
3. Has only 2 or 3 main features, and excludes febrile diseases such as scarlet fever, drug allergy
syndrome, Stevens-Johnson syndrome, toxic.
Interventions Treatment: glucocorticoid + immunoglobulin
Control: immunoglobulin
Outcomes Primary outcome: drug resistance rate
Secondary outcomes: incidence of coronary arterial lesions; hospitalisation expenses; time for
eliminating the clinical symptoms; incidence of glucocorticoid adverse effects; NT-proBNP; blood
pressure; alanine transaminase
Notes Unable to obtain further details on methods to allow assessment for inclusion.
ChiCTR-IOR-16007862
Methods Unknown
Participants Inclusion criteria: KD diagnosis based on 5th edition diagnostic criteria revised by the Japanese KD
Research Committee:
1. fever lasting > 5 days (including cases who were responsive to treatment with a fever lasting < 5
days);
2. binocular conjunctival congestion (no exudation);
3. changes in the lip and oral cavity, flushed lips and rhagades, strawberry-like tongue, oral and pha-
ryngeal mucosal diffuse hyperaemia;
4. polymorphous rash;
5. changes in the tail end of limbs, palmoplantar redness, hard swelling in the finger (toe) tip;
6. acute non-suppurative cervical lymphadenopathy that was often unilateral, diameter > 1.5 cm.
KD diagnosed if patients presented with ≥ 5 of the criteria. KD also diagnosed in patients who met
4 criteria and had a coronary artery aneurysms or coronary artery ectasia observable by 2-dimen-
sional echocardiogram or coronary arteriography.
Interventions Control: IVIG 2 g/kg/day + aspirin 30–50 mg/kg/day
Treatment: IVIG + aspirin + methylprednisolone
Outcomes Primary outcomes: white blood cell; haemoglobin; platelet count; percent of neutrophile granulo-
cyte; alanine transaminase; aspartate aminotransferase; gamma-glutamyl transpeptidase; sodium;
erythrocyte sedimentation rate; CRP; albumin; gene

Informed decisions.
ChiCTR-IOR-16007862 (Continued)
Notes Unable to obtain further details on methods to allow assessment for inclusion.
JPRN-UMIN000005022
Methods Unknown
Participants All participants were Japanese and fulfilled the Criteria for Diagnostic Guidelines for KD (5th revi-
sion) published by the KD Research Committee in Japan
Interventions IVIG therapy
Outcomes Unknown
Notes Unable to obtain further details on methods to allow assessment for inclusion
JPRN-UMIN000009946
Methods Unknown
Participants Inclusion criteria: complete KD diagnosed by the Japanese diagnostic guidelines for KD; deferves-
cence period < 5 days; age > 3 months; no previous treatments for KD; written informed consent
obtained
Interventions Group 1: 2 doses of intravenous methylprednisolone pulse + IVIG
Group 2: 1 dose of intravenous methylprednisolone pulse + IVIG
Outcomes Primary outcomes: reduction of defervesence period; improvement of laboratory tests; frequency
of adverse events
Secondary outcome: frequency and severity of coronary arterial lesions in acute phase and long-
term prognosis
Notes Unable to obtain further details on methods to allow decision on inclusion.
CRP: C-reactive protein; IVIG: intravenous immunoglobulin; KD: Kawasaki disease; NT-proBNP: N-terminal pro-brain natriuretic peptide.
Characteristics of ongoing studies [ordered by study ID]
EUCTR2019-004433-17-GB
Study name Multicentre, randomised, open-label, blinded endpoint assessed, trial of corticosteroids plus intra-
venous immunoglobulin (IVIG) and aspirin, versus IVIG and aspirin for prevention of coronary artery
aneurysms in Kawasaki disease
Methods Controlled, randomised, parallel, open-label trial
Participants Aged 30 days (postnatal age) to 15 years inclusive, and below the country-specific age of consent
for the duration of the trial
KD defined in ≥ 1 of 3 following methods: as per AHA criteria:
1. fever for ≥ 5 days in addition to 4 of the following 5 clinical criteria:

a. bilateral non-purulent conjunctivitis
Informed decisions.
EUCTR2019-004433-17-GB (Continued)
b. cervical lymphadenopathy
c. polymorphous skin rash
d. changes in lips or mucosa (strawberry tongue, red cracked lips, diffuse erythematous orophar-
ynx)
e. extremity changes (erythema, oedema of palms and soles in initial phase, and at convalescent
stage skin peeling);
2. OR < 5 days of fever but all 5 clinical criteria above;
3. OR incomplete KD cases, as per a modified AHA definition, namely:
a. children/adolescents (aged > 1 year) with fever ≥ 5 days AND ≥ 2 other compatible clinical cri-
teria as listed above; OR infants aged ≤ 1 year with fever ≥ 7 days without other explanation;
AND for both age groups
b. CRP ≥ 30 mg/L or ESR ≥ 40 mm/hour (or both) AND for both age groups
c. EITHER the presence of any ≥ 3: anaemia for age (haemoglobin < lower limit of normal refer-
ence range for local laboratory) platelet count ≥ 450 × 109/L or < 140 × 109/L; albumin < 30 g/L;
elevated ALT (> upper limit of normal reference range for local laboratory); white cell count ≥
15 × 109/L; urine ≥ 10 white blood cells per high power field
d. OR abnormal echocardiogram compatible with KD but without established CAA, with ≥ 3 of the
following suggestive features: decreased left ventricular function, mitral regurgitation, peri-
cardial effusion or dilated but non-aneurysmal coronary arteries (internal diameter 2 ≤ z < 2.5;
and not meeting the exclusion criteria for aneurysmal change as defined below).
Written informed consent from appropriate legal representative(s), and assent from patients who
have not reached the age of consent and will not reach the age of consent for the duration of the
trial in the participating country, but are judged to have capacity for this (depending on both age
and acuity of illness).
Disease-related exclusions: diagnosis is a second or further episode of KD; already established CAA
at screening; severe congestive heart failure or cardiogenic shock defined as the presence of hy-
potension and shock requiring the initiation of volume expanders; congenital coronary artery ab-
normality that would impair assessment of the primary endpoint; suspected macrophage activa-
tion syndrome.
Exclusions related to medications: started IVIG > 24 hours prior to randomisation; known hypersen-
sitivity to prednisolone or methylprednisolone; current oral, intravenous or intramuscular corticos-
teroid treatment for > 3 days in previous 7 days prior to randomisation; history of previous severe
reaction to any human immunoglobulin preparation.
Exclusions related to general health or other issues: active varicella zoster virus infection, or known
exposure to a case of varicella within the previous 21 days prior to randomisation if known to be
non-immune; co-enrolment in another study/trial of an investigative medicinal product.
Interventions Group 1: corticosteroid + IVIG + aspirin
Group 2: IVIG + aspirin
Outcomes Primary outcomes: any CAA documented within the 12 weeks of trial follow-up; mean estimate
across weeks 1, 2 and 6 of the maximum of the z-score of the internal diameters of the proximal
right coronary artery or left anterior descending coronary artery, adjusting for rescue treatment
Secondary outcomes: maximum coronary z-score individually obtained from echocardiograms at

weeks 1, 2, 6 and 12; CAA defined solely by a luminal internal diameter z-score of ≥ 2.5 identified
within the 12 weeks of trial follow-up from echocardiograms completed during this time; receipt
of rescue treatment within the 12 weeks of trial follow-up; receipt of second dose of IVIG within the
12 weeks of trial follow-up; duration of fever after enrolment (time to temperature < 38 °C), temper-
ature collected from the axilla measured daily while child/adolescent is febrile; daily serum con-
centrations of CRP from days 1–5, and at 1 and 2 weeks after enrolment, and time to normalisation
of CRP (time from randomisation to first measurement ≤ 10 mg/L); duration of hospitalisation cal-
culated from date of admission to date of discharge for the KD admission; serious adverse events
including deaths occurring at any time during the 12 weeks of trial follow-up; grade 3 or 4 adverse
events occurring at any time during the 12 weeks of trial follow-up; clinical adverse events of any

Informed decisions.
EUCTR2019-004433-17-GB (Continued)
grade judged related to IVIG, aspirin or corticosteroids occurring at any time during the 12 weeks of
trial follow-up.
Starting date 16 December 2019 (date on which this record first entered in the EudraCT database)
Contact information Cara Purvis, Clinical Trial Manager; [email protected]
Notes ISRCTN71987471
KD-CAAP: Kawasaki Disease Coronary Artery Aneurysm Prevention trial
IRCT20181202041817N1
Study name Evaluation of primary intravenous methyl prednisolone pulse treatment in the prevention of coro-
nary artery abnormality in children with Kawasaki disease (The effect of corticosteroid on the treat-
ment of Kawasaki)
Methods Single-blind, parallel-group, randomised trial
Participants Inclusion criteria: age of 6 month to 5 years; complete and incomplete KD definition according to
AHA
Exclusion criteria: complicated cases; recurrent KD; previous coronary artery abnormality; conges-
tive heart failure; chronic renal failure; prednisolone sensitisation; active viral infection of zoster
or exposure to varicella in the past 21 days if not immunised; injecting oral, intravenous or muscu-
lar corticosteroid current > 3 days in the last 7 days; history of severe reaction to the preparation of
any human globulin product; registration in another study possible it will affect the effects of treat-
ment, effectiveness or follow-up
Interventions Quote: "After entering the study, patients were randomly divided into two groups: IVIG 2 g / kg, as-
pirin and 3-day intravenous pulsomethyl prednisolone at a dose of 30 mg / kg, and continued treat-
ment with oral prednisolone 1mg / kg for 3 days"
Quote: "Control group: Treatment with IVIG at 2 g / kg and aspirin at 30 mg / kg"
Quote: "Intervention group: Aspirin 30 mg / kg and intravenous methyl prednisolone pulse at 30

mg / kg in three days. Within 12 hours after the first dose of the steroid pulse, if the temperature
return to normal and CRP drops to less than half before treatment, it is considered as response
to treatment and treatment will continue with oral prednisolone at a dose of 1 mg / kg and if the
temperature do not return to normal and CRP drops to less than half before treatment, it is consid-
ered as failure of treatment and will be treated with standard IVIG treatment similar to the control
group."
Outcomes Primary outcomes: coronary artery abnormality (echocardiography performed ≥ 3 times (immedi-
ately at diagnosis, 2 weeks and 2 months after onset of disease))
Secondary outcomes: severity of coronary artery disease ≥ 3 times (immediately at diagnosis, 2

weeks and 2 months after onset of disease)
Starting date Expected recruitment start date: 19 February 2019
Expected recruitment end date: 19 February 2020
Actual recruitment start date: not reported
Contact information Nahid Aslani; email: [email protected]
Notes Only protocol available

Informed decisions.
JPRN-UMIN000009524
Study name A prospective randomised controlled trial of immunoglobulin plus prednisolone for KD patients
with high risk for coronary abnormalities
Methods Parallel, assessor-blinded study
Participants Target sample size: 240
Inclusion criteria: people with severe KD with risk score ≥ 4 points; written informed consent ob-
tained from patients or their parents; Streptococcus, Epstein-Barr virus, adenovirus or yersinia in-
fection, or measles, or Stevens-Johnson syndrome ruled out
Exclusion criteria: history of KD; diagnosed as KD on the 9th day of illness or later; presence of coro-
nary artery lesions before treatment; defervescence before treatment; received corticosteroids
within 28 days before treatment; received IVIG within 180 days before treatment; severe underlin-
ing diseases; concurrent infection
Interventions Control group: IVIG 2 g/kg over 24 hours with aspirin 30 mg/kg/day. Dose of aspirin can be reduced
to 5 mg/kg/day after resolution of fever
Treatment group: IVIG 2 g/kg over 24 hours plus intravenous prednisolone 2 mg/kg/day with as-
pirin 30 mg/kg/day. Dose of aspirin can be reduced to 5 mg/kg/day after resolution of fever
Prednisolone will be given intravenously for ≥ 3 days and then can be given orally after resolution
of fever. When concentration of CRP is ≤ 1.0 mg/dL, the dose of prednisolone will be tapered over 9
days
Outcomes Primary outcome: incidence of CALs within 4 weeks after primary treatment
Secondary outcomes: incidence of CALs at 4 weeks after primary treatment; z-score of coronary
artery diameters; incidence of resistance to primary treatment or relapse; duration of fever after
primary treatment; serum CRP concentrations at 1 week and 2 weeks after primary treatment; inci-
dence of adverse events
Starting date 17 December 2012
Contact information Taichi Kato, Nagoya University Graduate School of Medicine, Department of Pediatrics, Japan;
[email protected]
Notes
NCT04078568
Study name Addition of methylprednisolone to initial treatment for Kawasaki disease patients
Methods Multicentre, open-label, blind-endpoints, RCT at > 10 hospitals in China
Participants and physicians will not be blinded to the assignment
Paediatric cardiologists who assess CAL by echocardiography will be blinded to the allocation
Participants Participants will be screened for eligibility if diagnosed with Kawasaki disease (KD) according to
the American Heart Association diagnostic guidelines for KD, including incomplete KD.
Inclusion criteria: meeting the above diagnostic criteria for KD; diagnosed before the tenth day of
onset (day of onset defined as the first day of fever); not treated with intravenous immunoglobulin
(IVIG) yet; age ≥ 1 month to < 12 years old; body weight ≤ 30 kg.

Informed decisions.
NCT04078568 (Continued)
Exclusion criteria: z score of any coronary artery before initial treatment ≥ 10; receiving steroids
or other immunosuppressive agents in the previous 30 days; previous history of KD; afebrile be-
fore enrolment; suspected infectious diseases including sepsis, septic meningitis, peritonitis, bac-
terial pneumonia, varicella and influenza; serious immune diseases such as immunodeficiency or
chromosomal abnormalities; refusing to sign the informed consent; unable to be followed up for at
least 3 months.
Interventions Active comparator
• IVIG 2 g/kg once, given within 12–24 hours

• Aspirin 30 mg/kg orally per day (given in 3 divided doses), then 3–5 mg/kg per day when fever
subsides for 3 days and CRP is normal. Aspirin will be continued for ≥ 6 weeks after onset of illness
Experimental
• IVIG 2 g/kg once, given within 12 to 24 hours

• Aspirin 30 mg/kg orally per day (given in 3 divided doses), then 3–5 mg/kg per day when fever
subsides for 3 days and CRP is normal. Aspirin will be continued for ≥ 6 weeks after onset of illness
• Intravenous methylprednisolone 1.6 mg/kg per day (given in 2 divided doses) for 3 days, then
changed to oral prednisolone 2 mg/kg when fever subsides for 3 days. If CRP is normal, the oral
dose will be reduced every 5 days from 2 mg/kg to 1 mg/kg to 0.5 mg/kg (tapered over 15 days).
Then prednisolone will be discontinued
Outcomes Primary outcome: CAL at 1 month of illness measured by 2-dimensional echocardiography. Mea-
surement included diameter of left main coronary artery, left anterior descending artery, left cir-
cumflex coronary artery, and proximal and middle segments of the right coronary artery. Z-score of
each coronary artery will be calculated. CAL defined as z-score ≥ 2 of any coronary artery
Secondary outcomes: need for additional treatment (about 2 weeks of illness). Axillary tempera-
ture (or rectal temperature) will be measured every 6 hours during hospitalisation. Participants
who have recurrent or persistent fever (axillary temperature ≥ 37.5 °C or rectal temperature ≥ 38
°C) after 36 hours of completion of initial IVIG infusion will be given additional treatment. Duration
of fever (hours) after initiation of initial IVIG infusion: from initiation of initial IVIG infusion to the
first record of being afebrile (defined as an axillary temperature < 37.5 °C for > 24 hours). Axillary
temperature (or rectal temperature) will be measured every 6 hours during hospitalisation. Par-
ticipants with an axillary temperature < 37.5 °C (or rectal temperature < 38 °C) for > 24 hours are
considered afebrile. Changes in z-scores throughout study. This is a repeated measurement. Inter-
nal diameter of left main coronary artery, left anterior descending artery, left circumflex coronary
artery, proximal and middle segment of right coronary artery will be measured by echocardiogra-
phy at 6 time points: enrolment, 2 weeks, 1 month, 3 months, 6 months and 12 months of illness.
Z-score will be calculated based on the height, weight and coronary artery diameter. Change in
serum CRP concentration (measured before initial IVIG infusion and 72 hours after completion of
initial IVIG infusion). Number of participants with serious adverse events (composite outcome, in-
cluding death, hypertension, severe infection, allergic reactions, heart failure, thrombosis, etc.)
Starting date 15 January 2020
Contact information Fang Liu, MD; [email protected]
Lan He, MD; [email protected]
Notes
AHA: American Heart Association; ALT: alanine aminotransferase; CAA: coronary artery abnormality; CAL: coronary artery lesion; CRP:
C-reactive protein; ESR: erythrocyte sedimentation rate; IVIG: intravenous immunoglobulin; KD: Kawasaki disease; RCT: randomised
controlled trial.

Informed decisions.
DATA AND ANALYSES
Comparison 1. Corticosteroids versus no corticosteroid use
Outcome or subgroup title No. of studies No. of partici- Statistical method Effect size
pants
1.1 Incidence of coronary artery ab- 8 986 Odds Ratio (M-H, Random, 0.32 [0.14, 0.75]
normalities 95% CI)
1.1.1 First-line treatment 7 907 Odds Ratio (M-H, Random, 0.25 [0.10, 0.58]
95% CI)
1.1.2 Second-line treatment 1 79 Odds Ratio (M-H, Random, 1.38 [0.43, 4.41]
95% CI)
1.2 Incidence of any serious adverse 6 737 Odds Ratio (M-H, Fixed, 95% Not estimable
effects attributable to the adminis- CI)
tration of corticosteroids
1.2.1 First-line treatment 6 737 Odds Ratio (M-H, Fixed, 95% Not estimable
CI)
1.3 Mortality (all-cause) 8 995 Odds Ratio (M-H, Fixed, 95% Not estimable
CI)
1.3.1 First-line treatment 7 915 Odds Ratio (M-H, Fixed, 95% Not estimable
CI)
1.3.2 Second-line treatment 1 80 Odds Ratio (M-H, Fixed, 95% Not estimable
CI)
1.4 Duration of clinical symptoms: 3 290 Mean Difference (IV, Random, -1.34 [-2.24, -0.45]
fever and rash 95% CI)
1.4.1 First-line treatment 2 210 Mean Difference (IV, Random, -1.65 [-3.31, 0.00]
95% CI)
1.4.2 Second-line treatment 1 80 Mean Difference (IV, Random, -0.90 [-1.84, 0.04]
95% CI)
1.5 Time for laboratory parameters 1 178 Mean Difference (IV, Fixed, 95% -2.80 [-4.38, -1.22]
to normalise: CRP and ESR (days) CI)
1.6 Length of hospital stay 2 119 Mean Difference (IV, Fixed, 95% -1.01 [-1.72, -0.30]
CI)
1.6.1 First-line treatment 1 39 Mean Difference (IV, Fixed, 95% -1.41 [-2.36, -0.46]
CI)
1.6.2 Second-line treatment 1 80 Mean Difference (IV, Fixed, 95% -0.50 [-1.58, 0.58]
CI)

Informed decisions.
Analysis 1.1. Comparison 1: Corticosteroids versus no corticosteroid

use, Outcome 1: Incidence of coronary artery abnormalities
Corticosteroid Control Odds Ratio Odds Ratio

Study or Subgroup Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI
1.1.1 First-line treatment

Ikeda 2006 2 90 10 88 13.5% 0.18 [0.04 , 0.83]
Inoue 2006 2 90 10 88 13.5% 0.18 [0.04 , 0.83]
Kobayashi 2012 4 121 28 121 17.6% 0.11 [0.04 , 0.34]
Newburger 2007 15 95 18 95 20.7% 0.80 [0.38 , 1.70]
Ogata 2012 2 22 10 26 12.7% 0.16 [0.03 , 0.84]
Okada 2003 0 14 0 18 Not estimable
Sundel 2003 0 18 1 21 5.3% 0.37 [0.01 , 9.64]
Subtotal (95% CI) 450 457 83.2% 0.25 [0.10 , 0.58]
Total events: 25 77
Heterogeneity: Tau² = 0.58; Chi² = 11.22, df = 5 (P = 0.05); I² = 55%
Test for overall effect: Z = 3.22 (P = 0.001)
1.1.2 Second-line treatment

Wang 2020 8 40 6 39 16.8% 1.38 [0.43 , 4.41]
Subtotal (95% CI) 40 39 16.8% 1.38 [0.43 , 4.41]
Total events: 8 6
Heterogeneity: Not applicable
Total (95% CI) 490 496 100.0% 0.32 [0.14 , 0.75]

Total events: 33 83
Heterogeneity: Tau² = 0.75; Chi² = 16.37, df = 6 (P = 0.01); I² = 63% 0.001 0.1 1 10 1000
Test for overall effect: Z = 2.62 (P = 0.009) Favours corticosteroid Favours control
Test for subgroup differences: Chi² = 5.45, df = 1 (P = 0.02), I² = 81.7%
Analysis 1.2. Comparison 1: Corticosteroids versus no corticosteroid use, Outcome 2:

Incidence of any serious adverse effects attributable to the administration of corticosteroids

Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI

Inoue 2006 0 90 0 88 Not estimable
Kobayashi 2012 0 121 0 121 Not estimable
Newburger 2007 0 101 0 97 Not estimable
Ogata 2012 0 22 0 26 Not estimable
Sundel 2003 0 18 0 21 Not estimable
Subtotal (95% CI) 366 371 Not estimable
Total events: 0 0
Test for overall effect: Not applicable
Total (95% CI) 366 371 Not estimable

Total events: 0 0
Heterogeneity: Not applicable 0.01 0.1 1 10 100
Test for overall effect: Not applicable Favours corticosteroid Favours control
Test for subgroup differences: Not applicable

Informed decisions.
Analysis 1.3. Comparison 1: Corticosteroids versus no corticosteroid use, Outcome 3: Mortality (all-cause)


Ikeda 2006 0 90 0 88 Not estimable
Inoue 2006 0 90 0 88 Not estimable
Kobayashi 2012 0 121 0 121 Not estimable
Newburger 2007 0 101 0 97 Not estimable
Ogata 2012 0 22 0 26 Not estimable
Sundel 2003 0 18 0 21 Not estimable
Total events: 0 0

Wang 2020 0 40 0 40 Not estimable
Total events: 0 0
Total (95% CI) 496 499 Not estimable

Total events: 0 0
Heterogeneity: Not applicable 0.01 0.1 1 10 100
Test for overall effect: Not applicable Favours corticosteroid Favours control
Test for subgroup differences: Not applicable
Analysis 1.4. Comparison 1: Corticosteroids versus no corticosteroid

use, Outcome 4: Duration of clinical symptoms: fever and rash
Corticosteroid Control Mean Difference Mean Difference
Study or Subgroup Mean [days] SD [days] Total Mean [days] SD [days] Total Weight IV, Random, 95% CI [days] IV, Random, 95% CI [days]

Inoue 2006 0.6 0.5 90 1.5 1 88 43.9% -0.90 [-1.13 , -0.67]
Okada 2003 0.3 0.5 14 2.9 2.4 18 26.1% -2.60 [-3.74 , -1.46]
Subtotal (95% CI) 104 106 69.9% -1.65 [-3.31 , 0.00]

Wang 2020 (1) 10.3 2 40 11.2 2.3 40 30.1% -0.90 [-1.84 , 0.04]
Subtotal (95% CI) 40 40 30.1% -0.90 [-1.84 , 0.04]
Total (95% CI) 144 146 100.0% -1.34 [-2.24 , -0.45]

Test for overall effect: Z = 2.94 (P = 0.003) -10 -5 0 5 10
Test for subgroup differences: Chi² = 0.60, df = 1 (P = 0.44), I² = 0% Favours corticosteroid Favours control
Footnotes
(1) Only duration of fever measured

Informed decisions.
Analysis 1.5. Comparison 1: Corticosteroids versus no corticosteroid use,

Outcome 5: Time for laboratory parameters to normalise: CRP and ESR (days)
Corticosteroid Control Mean Difference Mean Difference Risk of Bias
Study or Subgroup Mean [days] SD [days] Total Mean [days] SD [days] Total Weight IV, Fixed, 95% CI [days] IV, Fixed, 95% CI [days] A B C D E F G
Inoue 2006 8.4 3.7 90 11.2 6.6 88 100.0% -2.80 [-4.38 , -1.22] ? + ? ? + + +
Total (95% CI) 90 88 100.0% -2.80 [-4.38 , -1.22]

Test for subgroup differences: Not applicable Favours corticosteroid Favours control
Risk of bias legend

(A) Random sequence generation (selection bias)
(B) Allocation concealment (selection bias)
(C) Blinding of participants and personnel (performance bias)
(D) Blinding of outcome assessment (detection bias)
(E) Incomplete outcome data (attrition bias)
(F) Selective reporting (reporting bias)
(G) Other bias
Analysis 1.6. Comparison 1: Corticosteroids versus no corticosteroid use, Outcome 6: Length of hospital stay
Study or Subgroup Mean [days] SD [days] Total Mean [days] SD [days] Total Weight IV, Fixed, 95% CI [days] IV, Fixed, 95% CI [days]

Sundel 2003 1.9 0.7 18 3.31 2.1 21 55.9% -1.41 [-2.36 , -0.46]
Subtotal (95% CI) 18 21 55.9% -1.41 [-2.36 , -0.46]

Wang 2020 6.2 2.3 40 6.7 2.6 40 44.1% -0.50 [-1.58 , 0.58]
Subtotal (95% CI) 40 40 44.1% -0.50 [-1.58 , 0.58]
Total (95% CI) 58 61 100.0% -1.01 [-1.72 , -0.30]

Heterogeneity: Chi² = 1.54, df = 1 (P = 0.21); I² = 35%
Test for subgroup differences: Chi² = 1.54, df = 1 (P = 0.21), I² = 35.0% Favours corticosteroid Favours control
Comparison 2. Subgroup: single, pulse dose corticosteroid use versus longer course of corticosteroids
pants
2.1 Coronary artery abnormalities 7 808 Odds Ratio (M-H, Fixed, 95% 0.38 [0.24, 0.60]
CI)
2.1.1 Single, pulse dose corticos- 4 356 Odds Ratio (M-H, Fixed, 95% 0.70 [0.40, 1.22]
teroid use CI)
2.1.2 Longer course of corticosteroids 3 452 Odds Ratio (M-H, Fixed, 95% 0.13 [0.05, 0.32]
CI)

Informed decisions.
Analysis 2.1. Comparison 2: Subgroup: single, pulse dose corticosteroid use

versus longer course of corticosteroids, Outcome 1: Coronary artery abnormalities

2.1.1 Single, pulse dose corticosteroid use

Newburger 2007 15 95 18 95 22.7% 0.80 [0.38 , 1.70]
Ogata 2012 2 22 10 26 12.5% 0.16 [0.03 , 0.84]
Sundel 2003 0 18 1 21 2.0% 0.37 [0.01 , 9.64]
Wang 2020 8 40 6 39 7.3% 1.38 [0.43 , 4.41]
Subtotal (95% CI) 175 181 44.6% 0.70 [0.40 , 1.22]
Total events: 25 35
2.1.2 Longer course of corticosteroids

Inoue 2006 2 90 10 88 14.8% 0.18 [0.04 , 0.83]
Kobayashi 2012 4 121 28 121 40.6% 0.11 [0.04 , 0.34]
Subtotal (95% CI) 225 227 55.4% 0.13 [0.05 , 0.32]
Total events: 6 38
Test for overall effect: Z = 4.51 (P < 0.00001)
Total (95% CI) 400 408 100.0% 0.38 [0.24 , 0.60]

Total events: 31 73
Heterogeneity: Chi² = 15.20, df = 5 (P = 0.010); I² = 67% 0.001 0.1 1 10 1000
Test for overall effect: Z = 4.21 (P < 0.0001) Favours corticosteroid Favours control
Comparison 3. Subgroup: geography
pants
3.1 Coronary artery abnormal- 8 986 Odds Ratio (M-H, Fixed, 95% CI) 0.36 [0.23, 0.55]
ities
3.1.1 Centres in Japan 5 678 Odds Ratio (M-H, Fixed, 95% CI) 0.14 [0.07, 0.29]
3.1.2 Centres in North America 2 229 Odds Ratio (M-H, Fixed, 95% CI) 0.77 [0.37, 1.59]
3.1.3 Centres in China 1 79 Odds Ratio (M-H, Fixed, 95% CI) 1.38 [0.43, 4.41]

Informed decisions.
Analysis 3.1. Comparison 3: Subgroup: geography, Outcome 1: Coronary artery abnormalities

3.1.1 Centres in Japan

Ikeda 2006 2 90 10 88 12.9% 0.18 [0.04 , 0.83]
Inoue 2006 2 90 10 88 12.9% 0.18 [0.04 , 0.83]
Kobayashi 2012 4 121 28 121 35.4% 0.11 [0.04 , 0.34]
Ogata 2012 2 22 10 26 10.9% 0.16 [0.03 , 0.84]
Subtotal (95% CI) 337 341 72.1% 0.14 [0.07 , 0.29]
Total events: 10 58
3.1.2 Centres in North America

Newburger 2007 15 95 18 95 19.8% 0.80 [0.38 , 1.70]
Sundel 2003 0 18 1 21 1.8% 0.37 [0.01 , 9.64]
Subtotal (95% CI) 113 116 21.6% 0.77 [0.37 , 1.59]
Total events: 15 19
3.1.3 Centres in China

Wang 2020 8 40 6 39 6.3% 1.38 [0.43 , 4.41]
Subtotal (95% CI) 40 39 6.3% 1.38 [0.43 , 4.41]
Total events: 8 6
Total (95% CI) 490 496 100.0% 0.36 [0.23 , 0.55]

Total events: 33 83
Comparison 4. Subgroup: high-risk scores versus lower-risk scores or all participants if risk score not calculated in
study
pants
4.1 Coronary artery abnormalities 8 986 Odds Ratio (M-H, Fixed, 95% 0.36 [0.23, 0.55]
CI)
4.1.1 High-risk scores 3 377 Odds Ratio (M-H, Fixed, 95% 0.13 [0.06, 0.29]
CI)
4.1.2 Lower-risk scores or all partici- 6 609 Odds Ratio (M-H, Fixed, 95% 0.66 [0.38, 1.13]
pants in study if risk score not calcu- CI)
lated
4.2 Duration of clinical symptoms 3 290 Mean Difference (IV, Random, -1.34 [-2.24, -0.45]
95% CI)

Informed decisions.
pants
4.2.1 High-risk scores 1 32 Mean Difference (IV, Random, -2.60 [-3.74, -1.46]
95% CI)
4.2.2 Lower-risk scores or all partici- 2 258 Mean Difference (IV, Random, -0.90 [-1.13, -0.67]
pants in study if risk score not calcu- 95% CI)
lated
Analysis 4.1. Comparison 4: Subgroup: high-risk scores versus lower-risk scores or all
participants if risk score not calculated in study, Outcome 1: Coronary artery abnormalities

4.1.1 High-risk scores

Ikeda 2006 2 45 9 42 11.5% 0.17 [0.03 , 0.84]
Kobayashi 2012 4 121 28 121 35.1% 0.11 [0.04 , 0.34]
Ogata 2012 2 22 10 26 10.8% 0.16 [0.03 , 0.84]
Subtotal (95% CI) 188 189 57.5% 0.13 [0.06 , 0.29]
Total events: 8 47
4.1.2 Lower-risk scores or all participants in study if risk score not calculated
Ikeda 2006 0 45 1 46 1.9% 0.33 [0.01 , 8.40]
Inoue 2006 2 90 10 88 12.8% 0.18 [0.04 , 0.83]
Newburger 2007 15 95 18 95 19.7% 0.80 [0.38 , 1.70]
Sundel 2003 0 18 1 21 1.8% 0.37 [0.01 , 9.64]
Wang 2020 8 40 6 39 6.3% 1.38 [0.43 , 4.41]
Subtotal (95% CI) 302 307 42.5% 0.66 [0.38 , 1.13]
Total events: 25 36
Total (95% CI) 490 496 100.0% 0.36 [0.23 , 0.55]

Total events: 33 83

Informed decisions.
Analysis 4.2. Comparison 4: Subgroup: high-risk scores versus lower-risk scores or all
participants if risk score not calculated in study, Outcome 2: Duration of clinical symptoms
Study or Subgroup Mean [days] SD [days] Total Mean [days] SD [days] Total Weight IV, Random, 95% CI [days] IV, Random, 95% CI [days]
4.2.1 High-risk scores

Okada 2003 0.3 0.5 14 2.9 2.4 18 26.1% -2.60 [-3.74 , -1.46]
Subtotal (95% CI) 14 18 26.1% -2.60 [-3.74 , -1.46]
4.2.2 Lower-risk scores or all participants in study if risk score not calculated
Inoue 2006 0.6 0.5 90 1.5 1 88 43.9% -0.90 [-1.13 , -0.67]
Wang 2020 10.3 2 40 11.2 2.3 40 30.1% -0.90 [-1.84 , 0.04]
Subtotal (95% CI) 130 128 73.9% -0.90 [-1.13 , -0.67]
Total (95% CI) 144 146 100.0% -1.34 [-2.24 , -0.45]

Test for subgroup differences: Chi² = 8.23, df = 1 (P = 0.004), I² = 87.8% Favours corticosteroid Favours control
ADDITIONAL TABLES
Table 1. Criteria of high-risk participants

Study High-risk criteria
Ikeda 2006 Developed own risk score based upon IVIG unresponsiveness in a multiple logistic regression analy-
sis. 42/178 randomly identified KD participants were deemed high risk.
Kobayashi 2012 Kobayashi risk score of ≥ 5 (≤ 4 days fever prediagnosis, aged ≤ 12 years, CRP ≥ 100 mg/L, ≤ 300 ×
103/μL platelets, ALT ≥ 100 units/L, sodium ≤ 133 mmol/L, neutrophils ≥ 80%)
Ogata 2012 Egami score ≥ 3 (aged ≤ 6 months, ≤ 4 days fever prediagnosis, ≤ 300 × 103/μL platelets, CRP ≥ 7 mg/
dL, ALT ≥ 80 units/L)
ALT: alanine transaminase; CRP: C-reactive protein; IVIG: intravenous immunoglobulin; KD: Kawasaki disease
APPENDICES
Appendix 1. Database search strategies
Source Search strategy Hits retrieved
CENTRAL via CRSO #1 MESH DESCRIPTOR Mucocutaneous Lymph Node Syndrome EXPLODE ALL Feb 2021: 43
TREES 104
#2 kawasaki*:TI,AB,KY 313
#3 (mucocutaneous near5 syndrome):TI,AB,KY 189
#4 #1 OR #2 OR #3 331
#5 MESH DESCRIPTOR Glucocorticoids EXPLODE ALL TREES 19175
#6 steroid*:TI,AB,KY 29613
Informed decisions.
(Continued)
#7 (corticosteroid* or corticoid* or glucocorticoid*):TI,AB,KY 27108
#8 dexamethasone:TI,AB,KY 11715
#9 methylprednis*:TI,AB,KY 5330
#10 prednisone:TI,AB,KY 9443
#11 prednisolone:TI,AB,KY 7039
#12 hydroxycorticosteroid*:TI,AB,KY 112
#13 corticosterone:TI,AB,KY 136
#14 #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 73042
#15 #4 AND #14 83
#16 01/01/2016 TO 08/02/2021:CD 858409
#17 #15 AND #16 43
Clinicaltrials.gov Mucocutaneous Lymph Node Syndrome OR kawasaki Feb 2021: 18
ICTRP Search Portal Mucocutaneous Lymph Node Syndrome OR kawasaki Feb 2021: 0
MEDLINE (Ovid 1 exp Mucocutaneous Lymph Node Syndrome/ Feb 2021: 85

MEDLINE Epub Ahead
of Print, In-Process 2 kawasaki*.ti,ab.
& Other Non-In-
3 (mucocutaneous adj5 syndrome).ti,ab.
dexed Citations, Ovid
MEDLINE Daily and 4 or/1-3
Ovid MEDLINE) 1946 to
present 5 exp Glucocorticoids/
6 steroid*.ti,ab.
7 (corticosteroid* or corticoid* or glucocorticoid*).ti,ab.
8 dexamethasone.ti,ab.
9 methylprednis*.ti,ab.
10 prednisone.ti,ab.
11 prednisolone.ti,ab.
12 hydroxycorticosteroid*.ti,ab.
13 corticosterone.ti,ab.
14 or/5-13
15 4 and 14
16 randomized controlled trial.pt.
17 controlled clinical trial.pt.
18 randomized.ab.
19 placebo.ab.
20 drug therapy.fs.
21 randomly.ab.
Informed decisions.
(Continued)
22 trial.ab.
23 groups.ab.
24 or/16-23
25 exp animals/ not humans.sh.
26 24 not 25
27 15 and 26
28 (2016* or 2017* or 2018* or 2019* or 2020* or 2021*).ed.
29 27 and 28
Embase (Ovid) 1 exp mucocutaneous lymph node syndrome/ Feb 2021: 75
2 kawasaki*.ti,ab.
3 (mucocutaneous adj5 syndrome).ti,ab.
4 or/1-3
5 exp glucocorticoid/
6 steroid*.ti,ab.
7 (corticosteroid* or corticoid* or glucocorticoid*).ti,ab.
8 dexamethasone.ti,ab.
9 methylprednis*.ti,ab.
10 prednisone.ti,ab.
11 prednisolone.ti,ab.
12 hydroxycorticosteroid*.ti,ab.
13 corticosterone.ti,ab.
14 or/5-13
15 4 and 14
16 randomized controlled trial/
17 controlled clinical trial/
18 random$.ti,ab.
19 randomization/
20 intermethod comparison/
21 placebo.ti,ab.
22 (compare or compared or comparison).ti.
23 ((evaluated or evaluate or evaluating or assessed or assess) and (compare

or compared or comparing or comparison)).ab.
24 (open adj label).ti,ab.
25 ((double or single or doubly or singly) adj (blind or blinded or blindly)).ti,ab.

Informed decisions.
(Continued)
26 double blind procedure/
27 parallel group$1.ti,ab.
28 (crossover or cross over).ti,ab.
29 ((assign$ or match or matched or allocation) adj5 (alternate or group$1 or

intervention$1 or patient$1 or subject$1 or participant$1)).ti,ab.
30 (assigned or allocated).ti,ab.
31 (controlled adj7 (study or design or trial)).ti,ab.
32 (volunteer or volunteers).ti,ab.
33 trial.ti.
34 or/16-33
35 15 and 34
36 (2016* or 2017* or 2018* or 2019* or 2020* or 2021*).dc.
37 35 and 36
CINAHL (EBSCO) S33 S31 AND S32 Feb 2021: 24
S32 EM 2016 OR EM 2017 OR EM 2018 OR EM 2019 OR EM 2020 OR EM 2021
S31 S15 AND S30
S30 S16 OR S17 OR S18 OR S19 OR S20 OR S21 OR S22 OR S23 OR S24 OR S25
OR S26 OR S27 OR S28 OR S29
S29 MH "Random Assignment"
S28 MH "Triple-Blind Studies"
S27 MH "Double-Blind Studies"
S26 MH "Single-Blind Studies"
S25 MH "Crossover Design"
S24 MH "Factorial Design"
S23 MH "Placebos"
S22 MH "Clinical Trials"
S21 TX "multi-centre study" OR "multi-center study" OR "multicentre study"

OR "multicenter study" OR "multi-site study"
S20 TX crossover OR "cross-over"
S19 AB placebo*
S18 TX random*
S17 TX trial*
S16 TX "latin square"
S15 S4 AND S14
S14 S5 OR S6 OR S7 OR S8 OR S9 OR S10 OR S11 OR S12 OR S13

Informed decisions.
(Continued)
S13 corticosterone
S12 TX hydroxycorticosteroid*
S11 TX prednisolone
S10 TX prednisone
S9 TX methylprednis*
S8 TX dexamethasone
S7 TX corticosteroid* or corticoid* or glucocorticoid*
S6 TX steroid*
S5 (MH "Glucocorticoids+")
S4 S1 OR S2 OR S3
S3 TX mucocutaneous N5 syndrome
S2 TX kawasaki*.
S1 (MH "Mucocutaneous Lymph Node Syndrome")
Appendix 2. Glossary of terms
Term Definition
Aetiology Cause of a condition
Defervesence The abatement of a fever
Deleterious Negative effect
Desquamation Loss of the outermost layer of a surface
Erythematous Reddening - a term generally reserved for the skin
Exanthem A virus known to be associated with a rash
Mucocutanoeous Type of bodily surface, such as that inside the mouth
Oropharyngeal Area of the body encompassing the mouth and throat
Polymorphous rash Rash of varying appearance
Sequelae Consequence
Stenosis Restriction in the diameter of a vessel
Thrombosis The formation of a blood clot within a blood vessel

Informed decisions.
WHAT'S NEW
Date Event Description
8 February 2021 New search has been performed New search run. One new study included. One new study exclud-
ed and three new ongoing studies identified.
8 February 2021 New citation required but conclusions New search run. One new study included. One new study exclud-
have not changed ed and three new ongoing studies identified. New author joined
team. Text updated to reflect current Cochrane guidelines.
HISTORY
Protocol first published: Issue 8, 2014
Review first published: Issue 1, 2017
CONTRIBUTIONS OF AUTHORS
JG: study selection, assessing risk of bias, data extraction and analysis, updating review.
AJW: final reviewing and drafting of the review update, guarantor of the review.
RMRT: study selection, assessing risk of bias, data extraction and analysis, updating review.
DECLARATIONS OF INTEREST
JG: none.
AJW: none.
RMRT: none.
SOURCES OF SUPPORT
Internal sources
• No sources of support provided
External sources
• Chief Scientist Office, Scottish Government Health Directorates, The Scottish Government, UK
The Cochrane Vascular editorial base is supported by the Scientist Office.
DIFFERENCES BETWEEN PROTOCOL AND REVIEW
2021
We carried out additional subgroup analyses based on first- or second-line corticosteroid treatment to investigate if this had any impact
on the results. The previous version of the review only included studies using first-line treatment.
2016
In conducting this study it was decided that we would refer to coronary 'aneurysms' in the context of coronary 'abnormalities'. This is due
to a wide variation in the literature, including trials included in this review, as to what the definition of an aneurysm is.
We used odds ratio instead of risk ratio to report dichotomous data, in line with standard statistical analysis.
We renamed the outcome 'incidence of any adverse effects' to 'incidence of serious adverse events' to reflect more accurately the effects
we intended to study.

Informed decisions.
NOTES
This review replaced the withdrawn protocol 'Steroid hormone treatment for Kawasaki disease in children' (Jones 2014).
INDEX TERMS
Medical Subject Headings (MeSH)

Adrenal Cortex Hormones [adverse effects]; *Coronary Artery Disease; Fever [drug therapy]; Immunoglobulins, Intravenous [adverse
effects]; Length of Stay; *Mucocutaneous Lymph Node Syndrome [drug therapy]
MeSH check words

Child; Humans


2022 Corticosteroids for the treatment of Kawasaki disease in children

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

2022 Corticosteroids for the treatment of Kawasaki disease in children

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2022 Corticosteroids for the treatment of Kawasaki disease in children

Uploaded by

Copyright:

Available Formats

Cochrane

Corticosteroids for the treatment of Kawasaki disease in children

Green J, Wardle AJ, Tulloh RMR

Green J, Wardle AJ, Tulloh RMR.

Corticosteroids for the treatment of Kawasaki disease in children (Review)

Corticosteroids for the treatment of Kawasaki disease in children (Review) i

Corticosteroids for the treatment of Kawasaki disease in children

Jessica Green1, Andrew J Wardle2, Robert MR Tulloh3

Contact: Andrew J Wardle, [email protected].

Editorial group: Cochrane Vascular Group.

Data collection and analysis

Corticosteroids for the treatment of Kawasaki disease in children (Review) 1

PLAIN LANGUAGE SUMMARY

Using corticosteroids to treat Kawasaki disease

What is Kawasaki disease and how is it treated?

What did we do?

What did we find?

How certain are we in the evidence?

How up-to-date is this evidence?

Corticosteroids for the treatment of Kawasaki disease in children (Review) 2

Patient or population: children diagnosed with Kawasaki disease

Incidence of coronary artery Study population OR 0.32 986 ⊕⊕⊕⊝ —

Mortality (all-cause) Study population — 1075 ⊕⊕⊕⊝ 0 deaths record-

Cochrane Database of Systematic Reviews

Longer term (> 1 year after Study population — — — 0 studies included

GRADE Working Group grades of evidence

Cochrane Database of Systematic Reviews

BACKGROUND Description of the intervention

Corticosteroids for the treatment of Kawasaki disease in children (Review) 5

Corticosteroids for the treatment of Kawasaki disease in children (Review) 6

Corticosteroids for the treatment of Kawasaki disease in children (Review) 7

Corticosteroids for the treatment of Kawasaki disease in children (Review) 8

Corticosteroids for the treatment of Kawasaki disease in children (Review) 9

Figure 1. Flow diagram for studies in 2021 updated review

Corticosteroids for the treatment of Kawasaki disease in children (Review) 10

Corticosteroids for the treatment of Kawasaki disease in children (Review) 11

Blinding of participants and personnel (performance bias): All outcomes

Selective reporting (reporting bias)

Corticosteroids for the treatment of Kawasaki disease in children (Review) 12

Random sequence generation (selection bias)

0% 25% 50% 75% 100%

Low risk of bias Unclear risk of bias High risk of bias

Corticosteroids for the treatment of Kawasaki disease in children (Review) 14

Corticosteroids for the treatment of Kawasaki disease in children (Review) 15

Corticosteroids for the treatment of Kawasaki disease in children (Review) 16

Furthermore, the number of days from onset of treatment was ACKNOWLEDGEMENTS

Corticosteroids for the treatment of Kawasaki disease in children (Review) 17

Corticosteroids for the treatment of Kawasaki disease in children (Review) 18

Miura 2008 {published data only} ChiCTR-IOR-16007862 {published data only}

Nakamura 1985 {published data only} JPRN-UMIN000005022 {published data only}

Corticosteroids for the treatment of Kawasaki disease in children (Review) 19

coronary artery abnormality in children with Kawasaki disease. Davies 2013

Ayusawa 2005 Harnden 2002

Kobayashi 2008 Singh 2015

Corticosteroids for the treatment of Kawasaki disease in children (Review) 21

* Indicates the major publication for the study

Characteristics of included studies [ordered by study ID]

Unclear how many centres