áñ"ã To the Editor, We read with interest the article by McCrossan et al on selective fluconazole prophylaxis. Reviewing the data presented in Table 2, in 3 out of 4 infants with positive blood culture, cephalosporin was involved. A practical conclusion would be to eliminate the use of cephalosporin in VLBW infants. Another measure would be to shorten empiric antibiotic use to 3-4 days (if cultures prove to be negative) or to 7 days if positive cultures for bacteria or if the infant's clinical condition improves under antibiotic coverage. Prevention rather than prophylaxis could be the corner stone for minimizing invasive fungal infections.

As to the study methodology, prophylaxis was given at the discretion of the caring neonatologist and thus only 14 of the 31 eligible infants received prophylaxis. The right way to do it would have been to give prophylaxis to all the 31 eligible infants, according to the eligibility and risk factors set by the authors. In the pre-prophylaxis period, a total of 6 of the 33 eligible infants developed proven fungal infection, but the authors do not suggest which of these 33 infants (all with risk factors) would have received prophylaxis. In the post-prophylaxis period, 17 eligible infants did not receive prophylaxis! Reading the manuscript, one can not sure that these 17 infants did not have fungi in blood, urine or other sites. In the "Results" section, second paragraph, row 4, 6% should read 18% (6/33).

The study by McCrossan et al showed that fluconazole prophylaxis in a selected group of preterm infants with birth weight less than 1,500 grams decreases invasive fungal infection [1]. We agree that a carefully delineated risk-factor approach to the prevention of Candida infection in neonates may be a useful alternative to continuous fluconazole prophylaxis and may decrease the risk of the emergence of fluconazole-resistant species.

We recently reported our data on 3178 newborns, using a protocol based on more restricted risk factors than that used by McCrossan et al. We studied a 5 year period, comparing two and half years before (period 1) and after implementation of the restricted risk factor guideline (period 2) to start anti fungal therapy. We showed a significant decrease in culture proven Candida sepsis (1.1% and 0.4% in periods 1 and 2 respectively, p=0.027), and also in mortality due to Candida infection (61% and 0% in periods 1 and 2 respectively, p=0.016), with no increase in the use of anti fungal therapy (4.5% and 4.9% in periods 1 and 2 respectively, p=0.63) [2,3]. Feja et al showed that infants with birth weight higher than 1,500 grams with risk factors for fungal infection are in risk for Candida infection [4].

With the use of our protocol fewer infants will use fluconazole, and it will also cover those with birth weight higher than 1,500 grams that have risk factors for Candida sepsis.

References.

1. McCrossan BA, McHenry E, O'Neill F, et al . Selective fluconazole prophylaxis in high-risk babies to reduce invasive fungal infection. Arch Dis Child Fetal Neonatal Ed. 2007;92;F454-458

2.Procianoy RS, Enéas MV, Silveira RC. Empiric guidelines for treatment of Candida infection in high-risk neonates. Eur J Pediatr 2006;165:422-3.

3.Procianoy RS, Silveira RC. Prophylactic Fluconazole in Preterm Neonates. New Engl J Med 2007;357:1349

4. Feja KN, Wu F, Roberts K, et al. Risk factors for candidemia in critically ill infants: a matched case-control study. J Pediatr 2005; 147:156–161