The aims of the study were to describe Candida species in children with candidemia, to determine the changing epidemiology of candidemia over time in our tertiary care hospital, and to examine the demographic and clinical characteristics of patients with candidemia caused by parapsilosis and nonparapsilosis Candida spp. From 2012 to 2018, we identified a total of 126 cases of candidemia. The most commonly isolated Candida sp. was C. parapsilosis (n = 71, 56.3%), followed by C. albicans (n = 34, 26.9%). A total of 21 candidemia episodes (16.6%) were caused by other Candida species. Patients were divided into two groups (parapsilosis and nonparapsilosis) to identify any potential differences between the groups in terms of risk factors, mortality, and antifungal resistance. The median age of the patients, the median durations of the hospital and pediatric intensive care unit stay, receipt of immunosuppressive therapy within 2 weeks of developing candidemia, the rate of using total parenteral nutrition, need for mechanical ventilation, and receipt of carbapenems were statistically significantly higher in the parapsilosis group than in the nonparapsilosis group (P = 0.020, P = 0.001, P = 0.011, P = 0.036, P = 0.002, P = 0.038, and P = 0.004, respectively). The overall 30-day mortality rates (4.2% versus 3.6%) and resistance to fluconazole (33.8% versus 32.7%) were similar between the groups (P = 0.790 and P = 0.860, respectively). The distribution of Candida strains isolated in this study was consistent with the global trend, with C. parapsilosis being the most commonly identified species. Determining local epidemiologic data at regular intervals in candidemia cases is important in terms of determining both the changing epidemiology and empirical antifungal agents. IMPORTANCE In our study, the changing epidemiology of Candida species in candidemia in children was evaluated. The dominance of Candida parapsilosis species in the changing epidemiology was remarkable. We found that fluconazole resistance was high in both parapsilosis and nonparapsilosis groups. Updating local epidemiologic data at certain intervals in candidemia cases is important in determining both the changing epidemiology and empirical antifungal agents.