Clinical characteristics and response to prophylactic fluconazole of preterm VLBW neonates with baseline and acquired fungal colonisation in NICU: data from a multicentre RCT

Abstract

Background: Fungal colonisation by Candida spp. affects a high proportion of VLBW neonates in NICU. However, few data are available on the clinical characteristics of colonisation in preterm infants who are colonised at baseline via vertical transmission, compared to preterms who become colonised during their stay in NICU via horizontal transmission.

Material and methods: We reviewed the database of a multicentre, randomised trial of prophylactic fluconazole in VLBW neonates conducted in 8 Italian NICUs in the years 2004 and 2005 (Manzoni et al., NEJM 2007;356(24):2483-95). Per the protocol, all enrolled infants underwent weekly surveillance cultures from birth till discharge. We investigated the frequency of the two different modalities of Candida colonisation in this population, as well as the clinical and outcome characteristics possibly related to them.

Results: Overall, Candida colonisation affected 54 of 336 infants (16.1%). Baseline (i.e., detected <3(rd) day of life) colonisation affected 16 (4.7%), and acquired 38 (11.4%), of the 54 colonised preterms. Infants with baseline colonisation had significantly higher birth weight (1229 ± 28 g vs. 1047 g ± 29, p = 0.01) and gestational age (30.2 wks ± 2.7 vs. 28.5 wks ± 2.6, p = 0.01), and were significantly more likely to limit progression from colonisation to invasive Candida infection when fluconazole prophylaxis was instituted (21.6% vs. 42.7%, p = 0.009). Isolation of C. parapsilosis was significantly more frequent in infants with acquired colonisation.

Conclusions: Infants with baseline and acquired colonisation differ for demographics characteristics and for their response to fluconazole prophylaxis. This information may be useful for targeting more accurate management strategies for these two different groups of colonised preterms in NICU.