Meta-analysis of the results of intravenous gamma globulin treatment of coronary artery lesions in Kawasaki disease.

The objective of this study was to evaluate the efficacy of intravenous gamma globulin (IVGG) therapy on the prevention of coronary artery lesions (CALs) in patients with Kawasaki disease (KD), with reference to the literature on meta-analyses in randomized controlled studies. Studies from 1984 to the end of 2000 obtained from the National Library of Medicine or from the bibliographies of these articles were used in the analysis. The total number of patients with KD covered in 17 articles was 4020. All the articles were examined for the number of doses per day, the duration of administration, and the total number of IVGG doses. The number of patients in each group was counted, and the incidence of CALs was evaluated at 30 or 60 days after onset. The results of these searches were further analyzed by meta-analytical methods. The administration of IVGG significantly decreased the incidence of CALs in a dose-dependent manner: at 30 days after onset the incidence of CALs was 29.4% without IVGG but 21.6% with a total dosage of IVGG < 1000 mg/kg, 10.8% with a total dosage of 1000-2000 mg/kg, and 10.2% with a total dosage of > or =2000 mg/kg. Compared with the incidence of CALs without IVGG, the odds ratio (OR) was 0.662 [95% confidence interval (CI) 0.519-0.815)] at <1000 mg/kg, 0.292 (95% CI 0.222-0.371) with 1000-2000 mg/kg, and 0.274 (95% CI 0.207-0.349) with > or =2000 mg/kg. At 60 days, the values had decreased to 17.3%, 13.8% [OR 0.767 (95% CI 0.585-1.005)], 5.8% [OR 0.296 (95% CI 0.200-0.436)], and 4.9% [OR 0.244 (95% CI 0.170-0.349)], respectively. The meta-analyses also indicated that high doses of IVGG (> or =2000 mg/kg per day) given in a single dose prevented CALs more effectively than the same dosages divided into five daily doses in the patients with KD: The incidence of CALs at 30 days after disease onset was 2.4% with a single dose versus 12.9% with divided doses. Compared with divided doses, the OR with a single dose was 0.164 (95% CI 0.064-0.393) and 2.8% versus 6.1% at 60 days [OR 0.450 (95% CI 0.206-0.956)]. We clearly confirmed that higher doses of IVGG (> or =2000 mg/kg per day) administered in a single infusion were more effective for preventing CALs, as evaluated during both the subacute and convalescent phases of KD.