Eur J Pediatr (2006) 165: 422–423 DOI 10.1007/s00431-006-0088-1 SHORT REPORT Renato S. Procianoy . Mariluce V. Enéas . Rita C. Silveira Empiric guidelines for treatment of Candida infection in high-risk neonates Received: 28 November 2005 / Accepted: 12 January 2006 / Published online: 14 March 2006 # Springer-Verlag 2006 Keywords Invasive candidiasis . Premature infants . Risk factors . Fungal infection The incidence of neonatal Candida infection is steadily increasing and has been reported to be 1.6% in infants admitted to neonatal intensive care units (NICUs) [3].The use of therapeutic guidelines based on risk factors for fungal infection is important for achieving an early diagnosis of invasive Candida infection and for obtaining a decrease of its morbidity and mortality [1]. We performed this study in order to evaluate the efficacy of empiric guidelines in reducing proven mortality resulting from Candida infection. Patients and methods The study was approved by the Ethics Committee at the Hospital de Clínicas de Porto Alegre. There is no conflict of interest. Mortality by candidemia was analyzed during a 5-year period: (1) period 1 (January 2000 to June 2002), during which each neonatologist initiated antifungal therapy according to his own criteria based on clinical suspicion or presence of positive blood culture; (2) period 2 (July 2002 to December 2004), during which antifungal therapy was carried out according to the guidelines of the institute. The same neonatologists worked in the NICU during the study period, and all adhered to the new policy during period 2. Institutional guidelines indicated therapy in patients that fulfilled all the following criteria: (1) birth weight <1500 g or a very ill newborn; (2) clinical signs of infection and/or neutropenia; (3) use of broad-spectrum antibiotics (vancomycin and/or third-generation cephalo- sporin) for 7 days or more in association with one of the following total parenteral nutrition, mechanical ventilation, postnatal corticosteroid therapy, use of H2 blocker or mucocutaneous candidiase. Blood and cerebral spinal fluid (CSF) cultures were obtained prior to therapy in both periods. Candida infection was defined by at least one positive blood culture or by positive CSF culture. Mortality caused by fungal infection was defined by death during antifungal therapy with persistent positive blood culture. The statistical analyses included: Student’ s t-test, Mann-Whitney test, chi-square test or Fischer ’ s exact test. Significance was established at p<0.05. Results and discussion A total of 3178 newborn infants were admitted to the Hospital de Clínicas de Porto Alegre during the study period: 1607 in period 1 and 1571 in period 2. Antifungal therapy was used in 149 newborns: 72 (4.5%) in period 1 and 77 (4.9%) in period 2 (p=0.63). All 149 had central vascular catheter and received broad-spectrum antibiotic therapy (vancomycin plus third-generation cephalosporin) before antifungal therapy. There was no significant differ- ences between periods 1 and 2 with respect to the number of infants with a very low birth weight, those requiring mechanical ventilation and those requiring total parenteral nutrition. There was a significant difference between the two periods with respect to the number of patients with positive blood and/or CSF cultures for Candida spp.: in period 1, 18 (1.1%); in period 2, 6 (0.4%) (p=0.027). M. V. Enéas . R. C. Silveira Department of Pediatrics, Newborn Section, Universidade Federal do Rio Grande do Sul and Hospital de Clínicas de Porto Alegre, Porto Alegre, RS Brazil R. S. Procianoy (*) Department of Pediatrics, Newborn Section, Universidade Federal do Rio Grande do Sul and Hospital de Clínicas de Porto Alegre, Rua Tobias da Silva 99, #302, Porto Alegre, RS 90570-020, Brazil Tel.: +55-51-32227889 Fax: +55-51-33312738 e-mail: renatosp@terra.com.br