Study designs of evaluations included in the review
Studies needed to include a control group for the entire study, or subsets of patients whose treatment differed only by the inclusion of a corticosteroid.
Specific interventions included in the review
Studies where a corticosteroid was part of the initial management of disease were eligible for inclusion. The comparator was either control or a drug regimen identical to the treatment group but without a corticosteroid (standard therapy). Additional drugs were allowed so long as both groups received them. Prednisolone or methylprednisolone were assessed in the included studies. Standard therapy was aspirin or intravenous immunoglobulin (IVIG), alone or in combination. Additional therapies included cephalexin, dipyridamole and propranolol, or combinations of these. Where stated, the length of treatment ranged from one dose to 48 days. Further details of drug doses were given.
Participants included in the review
Studies on children, where all participants had a diagnosis of Kawasaki disease, were eligible for inclusion. Diagnosis of Kawasaki disease was defined as meeting the American Heart Association's or the Japanese Ministry of Health's diagnostic criteria, or fever for 5 days plus any four of the following: conjunctivitis, mucositis of the oral pharynx, peripheral oedema, unilateral lymphadenopathy and rash. Where stated, the mean ages in the included studies ranged from 1.5 to 4.4 years and the maximum period of symptoms before enrolment were 7 to 10 days.
Outcomes assessed in the review
The primary outcome was the development of coronary aneurysms. Only studies where a 2-dimensional echocardiograph or coronary artery catheterisation were performed at least 2 weeks after treatment were eligible for inclusion. Where stated, coronary imaging follow-up was performed at between 3 weeks and 3 months.
How were decisions on the relevance of primary studies made?
One author assessed studies for inclusion, blinded to author or location of study. A second author made the selection unblinded. Any conflicts were resolved by discussion.