The present prospective cross-sectional study was conducted in NICU of a tertiary care teaching hospital at Mysore between August 2018-December 2018 after taking approval from Institutional Research and Ethics Committee (EC REG: ECR/134/Inst/KA/2013/RR-16).

Inclusion criteria: The subjects included for the study were neonates admitted in NICU for more than 48 hours.

Exclusion criteria: Neonates with an already established invasive fungal infection were excluded.

Sample size calculation: Sample size was calculated using the below formula:

Where n is the sample size, Z is the statistic corresponding to level of confidence, Z value for 95% confidence is 1.96. P is expected prevalence (taken as 32.5% from the previous study done by Nazir A and Masoodi T [7]), and d is precision, taken as 5%.

Sample size was calculated as 100 (rounded-off), 150 subjects were included.

The detailed history was noted down on the proforma after taking consent from the parent. The samples (swabs) were collected from 150 neonates from different sites like perineal area, ear canal, throat and umbilicus with the help of sterile cotton tipped swabs pre-moistened with sterile saline. Different swabs were used for collection from different sites. The swabs were rubbed on the selected site with gentle pressure.

All swabs were inoculated immediately on chocolate agar and SDA, incubated at 37°C for seven days. Culture tubes and plates were examined daily for their morphological and colonial characters. Growth observed on chocolate agar and SDA was creamy, smooth, pasty colonies after 24-48 hours or at the end of seven days [Table/Fig-1] which was confirmed morphologically, gram positive budding yeast cells by gram’s stain were seen [Table/Fig-2]. In addition, blood culture records were looked if any blood sample from the study subjects i.e., of the neonates were received and checked for Candida growth and correlated with Candida colonisation.

[Table/Fig-1]:

Growth on SDA- Candida spp grow as white creamy mucoid colonies.

[Table/Fig-2]:

Gram stain of Candida albicans showing typical oval budding yeast cells (100x oil immersion).

Identification of Candida species was done by its ability to form germ tube by Reynolds Braude phenomenon, hyphal, chlamydospore and blastospore formation by inoculating on to cornmeal agar and incubating for 24-48 hours and also growth on CHROM agar. Confirmed Candida colonies were inoculated on CHROM agar plates, incubated at 37°C and the plates were examined at 24 hours, 48 hours and 72 hours after incubation. After 24 hours of incubation majority of the Candida colonies had grown well but the accurate colour of the different species developed only after 48 hours of incubation [Table/Fig-3].

[Table/Fig-3]:

Colony morphology of Candida species on CHROMagar: (a) Candidaparapsilosis: cream to white smooth colonies; (b) Candida tropicalis: blue to metallic blue coloured raised colonies; (c) Candida glabrata: pink smooth colonies; (d) Candida albicans: light green coloured smooth colonies.

Statistical analysis was done using SPSS version 11.0 and the prevalence of organisms was determined and expressed in percentage.

In the present study Candida colonisation was observed in 32 (21.3%) out of 150 neonates. Colonisation was found predominantly in perineal region in 17 (53%) out of 32 isolates, the colonisation in other sites is shown in [Table/Fig-4]. Among the various species isolated in colonised neonates Candida tropicalis was the common species followed by C.parapsilosisC.glabrata and C.albicans [Table/Fig-5].

[Table/Fig-4]:

Distribution of Candida colonisation in various sites.

Age distribution among the colonised neonates, 2-10 days was predominant with 18 (56%) out of 32, total 26 (81%) of the affected neonates were below 30 days of age and 06 (19%) were above 30 days. The low birth weight 22 (55%) was most predominant risk factor, distribution of other risk factors in Candida colonised neonates are shown in [Table/Fig-6].

Blood culture data of neonates colonised with Candida was collected from the departmental blood culture register for correlation. It was found that Candida spp was isolated from blood culture among 14 out of the 32 Candida colonised neonates and same Candida spp was isolated from blood and swab in those neonates.

Neonates represent a highly vulnerable patient population. It is estimated that invasive infections cause >1.4 million neonatal deaths worldwide annually [11]. There has been a significant increase in the incidence of invasive fungal infections worldwide especially in patients admitted in intensive care units. Among different fungal pathogens, colonisation by Candida species is very frequent in NICU and a necessary first step in the pathogenesis of systemic infection. Candida species are a common cause of neonatal nosocomial BSIs in premature infants and are a leading cause of infectious-related mortality in the NICU. The overall prevalence of Candida colonisation was 21.3% in neonates admitted to the NICU. Prevalence reported by other studies over the last decade point towards an increasing Candida prevalence, findings of these studies have been summarised in [Table/Fig-7] [7-10,12-18].

In this study, most of the neonates had Candida colonisation by day ten and maximum colonisation was seen in perineum region. Several risk factors have been associated with Candida colonisation, major risk factors identified in this study were low birth weight, prematurity and use of antibiotic therapy. High percentage of the neonates with colonisation with Candida in present study had Low Birth Weight (LBW) (68%) and Extremely Low Birth Weight (ELBW) (13%) and 75% of them were born preterm, which is consistent with observations of other studies that the incidence of Candida colonisation was inversely proportional to birth weight and gestational age. This association of Candida colonisation and lower gestational age can be due to prematurity being associated with compromised immunity such as decreased skin and mucosal integrity, the relative quantitative deficiency of protective maternal IgA to Candida, also excessive handling of preterm neonates in neonatal units promotes horizontal acquisition [5]. Prolonged PROM (≥24hrs), especially in vaginally colonised mothers has been associated with neonatal septicaemia [19]. In present study, there was no association between Premature Rupture Of Membranes (PROM) and Candida colonisation was found, also the mothers with vaginal colonisation did not checked.

Historically, Candida albicans was the most frequently isolated species, but recently non-albicans Candida (NAC) species have emerged as important opportunistic pathogens [20]. In this study, the NAC species were the predominant organism responsible for colonisation in neonates and the remaining were of C. albicans. C.tropicalis was the most common species isolated.

The study had certain limitations. Blood samples did not collect from all the subjects to correlate the isolated coloniser with systemic infection. In addition, follow-up of candidemic neonates could not be done.

In conclusion, in the present study it was observed that the neonates with risk factors have increased chance of being colonised with Candida in turn they may progress to candidaemia, which may increase the hospital stay, add socio-economic burden and increased morbidity and mortality.

Hence, focus should be shifted on prevention of candidaemia in these high-risk neonates by screening them for Candida colonisation, selectively keeping under observation and if suspected candidaemia they can start with antifungal agents instead of giving empirical treatment to all. Also, preventive measures like screening of NICU staff for carriers, restricted antibiotic use can decrease Candida colonisation and hence its infection.