Assessment of skin yeast colonization and its predisposing factors among neonates admitted to the neonatal intensive care unit
摘要
Studies indicate that yeast colonization is a key factor in invasive infections among high-risk neonates. Understanding how colonization develops and what worsens it is vital for improving early intervention and infection control in neonatal intensive care units (NICUs). This study assessed the extent of yeast colonization in NICU admissions, identified clinical predictors of colonization severity using the colonization index (CI), and performed antifungal susceptibility testing (AST) on the isolates.
MethodsFrom January 2020 to June 2021, all neonates (≤ 28 days old) admitted to a referral NICU in Sari, Iran, were included. Skin swabs were collected from various anatomical sites three times per week. Yeasts were identified by standard culture and molecular methods. Colonization was graded using the CI. In-vitro AST was performed using the Clinical Laboratory Standards Institute (CLSI) protocol.
ResultsA total of 1,026 skin samples collected from 78 neonates, 213 yeasts were isolated of which 165 (77.5%) Candida, 17.4% Cryptococcus, and 5.2% Rhodotorula were the most common. Among Candida species, C. albicans and C. parapsilosis were the most prevalent. Of 78 neonates, 35 (44.9%) met the colonization criteria. CI was high in 31.8%, mild in 29.6%, and low in 38.6% of cases. Cesarean delivery and prolonged antibiotic were associated with colonization. Whereas male gender and longer NICU stay correlated with lower odds of high CI. AST indicated notable fluconazole resistance in subsets of C. albicans and C. parapsilosis and reduced caspofungin activity across several species.
ConclusionOur findings indicate that gestational age, mode of delivery, antibiotic exposure, phototherapy, and gender were associated with yeast colonization. Our results also revealed the frequent occurrence of Cryptococcus (Naganishia) diffluens and high incidence of caspofungin resistance in Candida albicans isolates, supporting species-level identification.