Frequency of Fungemia in Hospitalized Pediatric Inpatients Over 11 Years at a Tertiary Care Institution

Abstract

To determine the frequency of bloodstream fungal infections in children who were admitted to our tertiary institution over an 11-year period. We conducted a retrospective cohort study of patients who were aged 0 to 21 years, had bloodstream fungal infections, and were admitted to the University of California, Los Angeles from 1991 through 2001. Patients were identified through the microbiology laboratory database. All positive fungal cultures for pediatric inpatients were reviewed. For each fungemic patient, a review of clinical course, cause, and outcome was performed. Over 11 years, 1124 pediatric inpatients with 3633 positive cultures had evidence of fungal colonization or infection. The mean incidence of positive fungal cultures increased from 105 between 1991 and 1996 to 129 patients per year between 1997 and 2001. Fungal isolates were mainly Candida species (85%) obtained primarily from respiratory (41%) and urine (27%) cultures. Only 7.5% of positive fungal cultures were from blood, although 24490 pediatric admissions prompted 72960 bacterial and fungal blood cultures, at charges of 2.52 million dollars. Of 14592 fungal blood cultures, <2% (n = 272) were positive, involving <1% (n = 97) of patients. The mean rise in number of children with fungemia was significant, from 6.8 between 1991 and 1996 to 13.0 patients per year between 1997 and 2001. Fungemia was associated with a high all-cause mortality rate (46%), particularly in immunocompromised patients (57%). Organisms recovered were primarily Candida species (91%). There was a decline in C albicans and C glabrata fungemia and an increase in C parapsilosis organisms. In 84% of patients, fungal organisms were isolated from both bacterial and fungal blood cultures, and in 74%, the same organism was isolated from additional body sites. Episodes of fungemia increased significantly over 11 years as compared with a moderate increase in positive fungal cultures and were associated with high all-cause mortality rates. More sensitive assays for early identification of fungal bloodstream infections are warranted.