This randomized controlled trial demonstrates that oral nystatin significantly reduced the incidence of fungal colonization in our neonatal unit. Although no cases of SFI occurred in the nystatin group, the difference in SFI rates between the two groups did not reach statistical significance.
A number of risk factors have been reported to be associated with an increased risk of developing SFI. Preceding colonization, particularly in GI tract and skin, is consistently recognized as the most important predictor of SFI (23, 24, 36-38). High rates of colonization (ranging between 22-87%) have been noted among preterm and VLBW infants who do not receive any antifungal prophylaxis (21, 26, 28, 30, 37, 39). A similar result has been noted in our study, with the colonization rate in the control group being 56.3%, as compared to a 29.8% rate in the nystatin group. Density and number of colonization sites reported have a positive correlation to the risk of subsequent SFI development (24, 40), thereby increasing the risk of fungal translocation and dissemination (24, 41). Other studies have reported that high fungal densities and multiple fungal colonization sites are associated with greater SFI risk. Kaufman, et al demonstrated that the risk of SFI increased with each additional site colonized. The same fungal species are mostly documented in both the infection and colonization sites (37, 42, 43). These results are similar to those reported in our study. Four of five cases in the control group were preceded by either heavy or multiple site colonization with similar fungal species. The exception was the one SFI that occurred without evidence of prior colonization.
Nystatin is a well-studied polyene antifungal that has a comparable efficacy with fluconazole for SFI prophylaxis amongst preterm and/or VLBW infants (21, 25, 44). Previous studies have proven that nystatin can prevent the development of colonization and SFI (21, 28, 30). A meta-analysis demonstrated that one in every 4-9 infants was prevented from developing SFI after receiving either fluconazole or nystatin prophylaxis (32). Our study demonstrated a statistically significant reduction of fungal colonization by 26% in the nystatin group as compared to the control group. All five SFI cases in our study were in control group and no cases of SFI were found in nystatin group. Although the difference in SFI incidence between both groups was not statistically significant, this result has suggested a declining trend of SFI risk and potential preventive effect of nystatin prophylaxis against SFI. We observed that incidence of SFI in our current study (5.2%) was much lower than our previous rate in the epoch of 2005-2008 (26%) (8). This large difference in incidence of SFI may have been influenced by a recent change in our clinical practice with the implementation of restrictive of antibiotic guidelines (narrow spectrum of antibiotics) since 2008.
The most common fungal organism cultured in this study was C. albicans, a finding similar to that of previous studies (26, 28, 45). Although C. albicans was still the most frequently encountered fungal organism, the occurrence of NAC species (consisting of C. parapsilosis, C. tropicalis, C. glabrata, C. krusei, and C. kefyr) continues to increase. The shift from C. albicans to NAC species in our unit has been previously reported by Wahyuningsih et al (33). This effect may be due to long-term extensive use of fluconazole when Amphotericin B was not available during that epoch. The notable increase in the incidence of NAC may become an important determinant in selecting a prophylactic drug because some NAC species, particularly C. glabrata and C. krusei, have been reported to be natively resistant to fluconazole (24, 46, 47). In contrast, nystatin is efficacious against all Candida species and no resistance has been documented with its use. Malassezia spp. was also found in our study. Prematurity, low birth weight, use of parenteral nutrition with lipid emulsion, use of central venous catheters, and contact with healthcare staff with contaminated hands were related to Malassezia spp. colonization in infants. There were some studies which have reported that this organism is an important cause of SFI and have demonstrated reduced susceptibility to fluconazole and flucytosine. To date, there are no reports of Malassezia spp. resistance to nystatin (48-51).
Based on the findings of the current study and previous studies, it is more difficult to reduce the risk of SFI once colonization has occurred (15, 26, 42, 46, 52). Currently, there is no established guidelines regarding the most effective timing and duration in using nystatin prophylaxis among high-risk infants. Several previous studies suggested that colonization may start within 3-6 days of life (15, 26, 38, 53). Therefore, to be effective in preventing fungal overgrowth and infection, oral nystatin should be commenced before the onset of colonization, particularly within the first 72 hours of life (26, 52). In contrast, the mean age of first colonization in our study was 10-11 days, longer than reported in the other previous studies. Most studies administered antifungal prophylaxis until patients were not in intensive care or no longer experiencing any risk factors for fungal infections. Although risk factors in every unit may be different, previous studies mostly agreed that antifungal prophylaxis administration for four to six weeks was safe and resulted in lower risk of colonization (5, 38, 41).
No mortality attributable to fungal infection was reported in any of our SFI cases. However, further study is still required to specifically evaluate the effect of nystatin on fungal-related mortality and long-term outcome. Oral nystatin is not absorbed, hence it is unlikely that systemic adverse reactions would be expected from this drug. The few adverse effects that were previously reported among children and adults, such as vomiting, diarrhea, and allergic reaction (5, 24, 54) were not found in our study.
Study Limitations and Further Suggestions
To date, this study is the first randomized controlled trial in our nation assessing the effectiveness of nystatin prophylaxis in preterm VLBW infants. In our study, no newborn infants under 28 weeks’ gestational age were recruited due to low survival related to limited resources in our NICU during the study period. Further study is required to assess the effect of nystatin prophylaxis in this particular gestational age group. It is notable that this study was conducted 8 years ago. Since the time of recruitment for this trial, our NICU has implemented a number of changes including improvements in management of respiratory care, hemodynamic instability and a comprehensive infection control program including rationalization of use of antibiotics. We acknowledge that these changes may have significant impact on the current validity of the results of this trial in the context of current practices within our NICU.
In the setting of a unit with high SFI incidence, this study has shown that nystatin prophylaxis is effective in preventing colonization and can potentially decrease the risk of SFI. Since some neonatal centers in Indonesia still face financial limitations and difficulties in maintaining intravenous access for systemic antifungal prophylaxis, the outcomes of this study have potential significance in enhancing clinicians understanding about the possible benefits of using nystatin as an alternative agent for fungal prophylaxis in a neonatal intensive care setting in Indonesia.