Candida species are the third most frequent cause of septicaemia after 72 hours of life among pattern neonates with an estimated incidence of 1.6-15% in neonatal intensive care units. Candida colonization is the most important predictor for invasive candidiasis among preterm neonates. Of all colonised preterm neonates, 7-25% develop invasive candidiasis which is associated with mortality rate approaching 50%. Prophylactic antifungal agents effectively reduce fungal colonisation and invasive disease.
This study was conducted to determine the prevalence of early candida colonization, associated factors and early outcome among preterm neonates admitted in to the special care unit of Mulago National Referral Hospital in Uganda.
A prospective observational study was conducted in the special care unit of Mulago Hospital between November 2008 and February 2009. Consecutively 213 preterm neonates were enrolled, candida colonisation of groin, rectum and throat were determined by taking swabs for culture on CHROMagar after 72 hours and within first week of life. Colonised neonates were followed up for 2 weeks and for those who developed features of septicaemia by WHO definition; blood was drawn for bacterial and candida isolation.
Data was entered into the computer using EPIDATA package 3.1 and analysed using STATA version 10. The fishers exact tests were performed for bivariate analysis and logistic regression was conducted to assess the factors independently associated with candida colonisation. P values of <0.05 were considered significant and confidence interval of 95% was used.
Of the 213 enrolled preterm neonates, 50(23.5%) had early candida colonisation. The main common species was C. albicans 28(40%). Gestational age <_ 30 weeks was the only factor independently associated with candida colonisation p=0.005. Forty six (46) candida colonised preterm neonates were followed up of whom 13(28%) developed mucocutaneous candidiasis and 14(30%) died. Six (43%) deaths followed bacterial septicaemia, two followed blood culture negative septicaemia, four followed necrotizing enterocolitis and two followed aspirations. All the neonates who died following necrotising enterocolis had candida rectal colonisation and three of them had multiple body sites colonised by candida. Mortality among colonised neonates was significantly associated with colonisation of more than one body site p=0.035 and gestation age <_ 30 weeks =p=0.044. Candida was not found in this study.
Early candida colonisation was highly prevalent (23.5%) among preterm neonates admitted to the special care unit of Mulago Hospital. Irrespective of mode or place of delivery, preterm neonates of gestational age ≤30 weeks are more predisposed to early candida colonisation. Candida albicans was the commonest species that colonised these neonates with the rectum and groin being the most frequently colonised sites. In this study candidemia was not found, although mortality among candida colonised preterm neonates was high (30%) especially among those with multiple body sites colonised. Most of the deaths were associated with bacterial septicaemia.
In a view of the high early candida colonisation and high mortality observed among candida colonised preterm neonates in this study specially among preterm neonates of gestation age ≤ 30 weeks, a study to elucidate to the possible sources of candida that colonise
Preterm neonates in the special care unit are very necessary in order to minimize candida colonisation.
The use of antifungal prophylaxis among neonates for prevention of candidemia cannot be recommended in special care unit since candidimia was not found in this study. A large study should be conducted to compare outcome of preterm neonates colonised and those not colonised by candida, since a high mortality was observed among candida colonised preterm neonates.