To determine the risk factors for disseminated infection in hospitalized children with candidemia.
We performed a nested case-control study within a cohort of hospitalized children with candidemia. The cohort was defined by all patients with positive blood cultures for Candida species in a large, urban, academic, tertiary care children's hospital from 1998 to 2001. Cases were patients with clinical, microbiologic or radiographic evidence of disseminated candidiasis. Controls were patients with no evidence of disseminated candidiasis.
Among 168 total children with candidemia, the median age was 3.5 years (interquartile range, 0.6–14.3). There were 189 episodes of candidemia. Candida species included:Candida albicans (41%), Candida parapsilosis (24%), Candida glabrata (13%) and Candida tropicalis (9%). The most common underlying diagnoses were oncologic (24%), gastrointestinal (15%) and cardiac (10%) diseases. Eighty-nine patients (53%) were admitted to an intensive care unit, 46 (27%) to a general pediatric or surgical ward and 33 (20%) to the oncology ward.
Of the 168 patients with candidemia, 153 were included in the analysis of risk factors for disseminated candidiasis. Of 153 (17%) patients, 26 had evidence of organ dissemination. Organ involvement was most commonly identified in the lung (58%), followed by the liver (23%), kidney (16%), brain (12%), spleen (8%), eye (8%) and heart (8%). Eight of the 26 patients had evidence of dissemination to more than 1 organ. Independent risk factors for disseminated candidiasis were persistently positive blood cultures for Candida (>3 days) with a central venous catheter in place (odds ratio, 3.0; 95% confidence interval, 1.2, 7.8; P = 0.02) and immunosuppression (odds ratio, 2.9; 95% confidence interval, 1.2, 7.0; P = 0.02).
Prolonged duration of candidemia with a central venous catheter in place and immunosuppression were independent risk factors for disseminated candidiasis in children with candidemia. Furthermore review of the epidemiology of candidemia at our institution revealed a heterogeneous population of children at risk for candidemia and a predominance of non-albicans species as the cause of these infections. Future studies are needed to determine the extent of evaluation needed for detecting dissemination among children with candidemia and to explore interventions for its prevention.
From the *Division of Infectious Diseases, The Children's Hospital of Philadelphia, Department of Pediatrics, University of Pennsylvania School of Medicine, Philadelphia, PA; the †Department of Pediatrics, Children's Hospital and Regional Medical Center, University of Washington School of Medicine, Seattle, WA; ‡Division of Infectious Diseases, Department of Medicine, the §Center for Clinical Epidemiology and Biostatistics, Center for Education and Research on Therapeutics, and the ¶Department of Biostatistics and Epidemiology, University of Pennsylvania School of Medicine, Philadelphia, PA
Accepted for publication March 8, 2004.
Supported in part by National Research Service Award T32-HD430210-01 from the National Institutes of Health and by Centers for Education and Research on Therapeutics grant U18-HS10399 from the Agency for Healthcare Research and Quality.
Presented in part at the Pediatric Academic Societies Meeting, Seattle, WA, 2003 and at the 13th Annual Meeting of the Society for Healthcare Epidemiology, Arlington, VA, April 5–8, 2003.
Address for reprints: Theoklis E. Zaoutis, MD, Division of Infectious Diseases, The Children's Hospital of Philadelphia, 34th and Civic Center Boulevard, Philadelphia, PA 19104. Fax 215-590-2025; E-mail email@example.com