To investigate the clinical characteristics of late-onset septicemia caused by Candida parapsilosis (CPLOS) in preterm neonates in NICU.

From February 2014 to February 2019, 19 preterm neonates diagnosed as CPLOS in NICU of Beijing Obstetrics and Gynecology Hospital, Capital Medical University, were selected as observation group, and 19 preterm neonates with laboratory diagnosis of late-onset bacterial septicemia (BLOS) and with the same gestational age, birth weight and gender were selected as control group. The clinical data of the two groups were analyzed, retrospectively, including neonatal baseline, perinatal factors, clinical manifestations and infection related indicators. CPLOS and BLOS susceptibility factors were analyzed.

The mean gestational ages and birth weights of premature neonates with CPLOS were (30.21 ± 2.66) weeks and [1 250 (1 133, 1 408)] g, respectively. Compared with BLOS group, the hospitalization time of cases in CPLOS group was significantly shorter [20 (13, 29) d vs. 29 (21, 42) d; Z =-2.032, P = 0.021], the incidence of thrombocytopenia (< 100 × 10⁹/L) was significantly higher [52.63% (10/19) vs. 15.79% (3/19); χ² = 5.729, P = 0.021], but the incidence of encephalitis was significantly lower [10.53% (2/19) vs. 42.11% (8/19); χ² = 4.886, P = 0.027], all with significant differences. All 19 (100%) preterm neonates were sensitive to fluconazole and amphotericin B. The median treatment time of fluconazole and amphotericin were 28 days and 12 days, respectively. There were no adverse reactions related to anti-fungal therapy. Compared with preterm neonates with BLOS, the incidence of premature rupture of membranes was significantly higher [47.37% (9/19) vs. 0% (0/19); χ² = 15.360, P = 0.001], the proportion of cefepime application was significantly higher [ (73.68% (14/19) vs. 31.58% (6/19); χ² = 6.756, P = 0.009], the days of penicillin sodium application was significantly longer [16 (13, 22) d vs. 1 (1, 13) d; Z =-2.825, P = 0.003], and the catheterization time of umbilical vein was significantly longer [11 (6, 13) d vs. 7 (2, 11) d; Z =-2.042, P = 0.042] in preterm neonates with CPLOS before infection.

Compared with preterm neonates with BLOS, preterm neonates with CPLOS had different clinical characteristics, including earlier infection, less incidences of thrombocytopenia and encephalitis. Preterm neonates with CPLOS were sensitive to fluconazole. Preterm neonates with long-term umbilical vein catheterization, intravenous infusion of penicillin sodium, and with application of cefepime and other restricted antibiotics before infection, were more likely to develop to CPLOS.