Oral nystatin prophylaxis and neonatal fungal infections

Alison J Howell ¹, David Isaacs ^2* and Robert Halliday ²

¹ John Radcliffe Hospital, Oxford, United Kingdom
² Children's Hospital at Westmead, Australia

^* To whom correspondence should be addressed. E-mail: davidi{at}chw.edu.au.

Accepted 2 March 2009

Abstract

Background: The value of antifungal prophylaxis depends partly on the incidence of neonatal fungal infection. We compared the incidence of fungal infection in babies in neonatal units which do and do not give antifungal prophylaxis using oral nystatin.

Methods: Prospective, multi-centre surveillance study from 1993-2006 of invasive fungal infection, defined as positive blood or CSF culture, in babies <1500g birth weight in neonatal units in Australia and New Zealand.

Results: There were 118 episodes of invasive fungal infection (IFI) in 14,778 babies <1500g, an incidence of 0.80% (95% CI 0.66-0.94%). All infections were due to Candida species, mostly C. albicans (74, 62.7%) and C. parapsilosis (39, 33.1%). The mortality was 16.5%. The incidence was 0.54% (0.38-0.70%) for babies <1500g in units using selective or universal oral nystatin prophylaxis and 1.23% (0.84-1.62%) in units using no prophylaxis (P<0.0001). The incidence of infection in babies <1000g was 1.78% (106/5948) (95% CI 1.44-2.12%). The incidence was 1.23% (0.92-1.54%) for babies <1000g in units using nystatin prophylaxis and 2.67% (1.97-3.37%) in units using no prophylaxis (P<0.0001).

Conclusions: The incidence of neonatal fungal infection was low in Australia and New Zealand, even without antifungal prophylaxis. Antifungal prophylaxis with oral nystatin was associated with a significantly lower incidence of fungal infection compared with no prophylaxis.