Epidemiology of candidemia in neonatal intensive care units: a persistent public health problem.

BACKGROUND Candidemia has become an increasingly important problem in infants hospitalized in the Neonatal Intensive Care Units (NICUs). Candida species are the third most common agents of late-onset infections in critically ill neonates and they are associated with high morbidity and mortality rates. In this study we evaluated the epidemiology of Candida bloodstream infections in the NICU of an Italian university hospital during a 15-year period. Our specific aims were to analyze the change in species distribution and the vitro susceptibility of these yeasts to fluconazole (FCZ) and amphotericin B (AmB). METHODS A retrospective study of candidemia in the NICU of a university hospital in southern Italy, covering the years 2000-2014 was carried out. The isolates were identified using the VITEK2 yeast identification system and antifungal susceptibility was determined using the E-test method. RESULTS Among the 57 patients with confirmed candidemia, 60% were males (n = 34 cases) and 82% (n = 47) had a gestational age of 24-32 weeks. Twenty-seven neonates (47%) had a very low birth weight (<1500 g), 20 (35%) an extremely low birth weight (<1000 g), and 10 (18%) a low birth weight (<2500 g). The most important potential risk factors were the placement of a central venous catheter, total parenteral nutrition, and endotracheal intubation (100%, each). Candida albicans was the most frequent yeast (47%), followed by Candida parapsilosis (44%). The proportion of Candida non-albicans increased slightly, from 46% in 2000-2004 to 71% in 2010-2014 (χ2 test for trend, p = 0.030). All isolates were susceptible to FCZ and AmB. CONCLUSIONS The detection in this epidemiologic study of an increase in Candida non-albicans highlights the importance of correct species-level identification in the rapid diagnosis for an efficient treatment of candidemia. Knowledge of the local epidemiological trends in Candida species isolated in blood cultures will facilitate therapeutic decision-making.