Objective To compare the efficacy of low-dose or no aspirin with conventional high-dose aspirin for the initial treatment in the acute-phase of Kawasaki disease (KD).

Design A meta-analysis and systematic review of randomised control trials and cohort studies.

Methods All available articles that compared different dosage of aspirin in the acute-phase of KD published until 20 September 2019 were included from the databases of PubMed, Embase and Cochrane Central Register of Controlled Trials Central without language restrictions. Extracted data from eligible studies were reviewed by two authors independently and analysed by using RStudio software.

Results Nine cohorts with a total of 12 182 children were enrolled. We found that low-dose (3–5 mg/kg/day) or no aspirin in the acute-phase KD was associated with reducing the risk of coronary artery lesions (CALs, OR=0.81, 95% CI 0.69 to 0.95). No differences were observed in intravenous immunoglobulin resistance, length of hospital stay and fever days after admission (OR=1.35, 95% CI 0.91 to 1.98; standard mean difference (SMD)=0.17, 95% CI −1.07 to 1.4; SMD=0.3, 95% CI −1.51 to 2.11) in the low-dose/no aspirin subgroup compared with the high-dose (≥30 mg/kg/day) aspirin subgroup. We did not identify any potential factors affecting the homogeneity of CAL risk as well as clinical important effects in all included studies.

Conclusions Prescribing low-dose or no aspirin in the acute-phase of KD might be associated with a decreased incidence of CAL. However, additional well-designed prospective trials are required to support the theory.