BACKGROUND: Although idiopathic cardiomyopathies(i-CMP) are very important in all age groups, the epidemiology of i-CMP in children has not been well defined. A retrospective study in Korean children was performed in 1998 to obtain basic data on i-CMP. MATERIAL AND METHOD: The medical records of all patients aged birth to 15 years from the hospitals where pediatric cardiologists worked were reviewed to obtain information on i-CMP. Pediatric cardiologists from a total of 22 hospitals were participated in reviewing the medical records of their patients and filling up the protocol. The data were pooled to the study committee and reviewed. RESULTS: Of the 278 cases with i-CMP, there were dilated cardiomyopathy (d-CMP) in 182 (65.4%): hypertrophic cardiomyopathy (h-CMP) in 74 (26.6%): restrictive cardiomyopathy (r-CMP) and unclassified in 17 (6.1%) and 5 (1.9%) each. The average annual occurrence of new cases as a whole was 2.65 per 100,000 (95% CI: 1.5-3.7): d-CMP, 1.73/100,000/year (95% CI: 0.73-2.73): h-CMP, 0.71/100,000/year (95% CI: 0.35-1.07): r-CMP, 0.16/100,000/year (95% CI: 0.02-0.3). The median age at the time of diagnosis was 11 months in d-CMP: 3.0 years in h-CMP: 6.9 years in r-CMP. The survival rate in d-CMP was 76% at 1 year, 72.5% at 2 year, 70% at 5 year. There was no difference in survival rate according to age (in d-CMP, between children less than 2 years of age and over 2 years of age (74% vs. 79% at 1 year: 67% vs. 76% at 5 year, p=NS): in h-CMP, between children less than 1 year of age and over 1 year of age (84% vs. 96% at 1 year: 63% vs. 81% at 5 year, p=NS)). R-CMP showed the worst survival rate (72% at 1 year, 30.2% at 5 year). CONCLUSION: In spite of the inherent defects of retrospective analysis, this study provides the useful epidemiological data in children with i-CMP. However, more systemic approach is needed to define the nature of the i-CMP in children.