[Analysis of risk factors for onset of acquired Serratia marcescens infection in neonatal intensive care unit].

Objective: To analyze the clinical characteristics of critically ill neonates in the neonatal intensive care unit (NICU) who acquired Serratia marcescens infection for onset or colonization, and to explore the risk factors contributing to the onset of Serratia marcescens infection.

Methods: A retrospective case-control study was conducted by collecting clinical data from NICU neonates at the Women and Children's Hospital of Ningbo University between January 2017 and December 2023. Forty-four neonates with clinical signs and/or symptoms consistent with Serratia marcescens infection, and with Serratia marcescens isolated from specimens, would be enrolled as the infection onset group, while 45 neonates who tested positive for Serratia marcescens in rectal and/or pharyngeal cultures during the same period, but had no clinical signs or infection symptoms, were enrolled as the colonization control group. The distribution of bacteria in the neonates infected with Serratia marcescens was observed, and clinical data were subjected to univariate and binary multivariate Logistic regression analyses for screening the independent risk factors for onset of acquired Serratia marcescens infection in NICU.

Results: During the 7-year period, 5 972 neonates were admitted to the NICU, of which 297 developed hospital-acquired infections. Among these, 44 neonates were identified with Serratia marcescens infection, accounting for 14.8% of hospital-acquired infections. During the same period, a total of 45 neonates were diagnosed with the colonization of Serratia marcescens, but did not develop any symptoms. The primary infection sites of the neonates in both the colonization control group and infection onset group were respiratory tract, accounting for 86.7% (39/45) and 43.2% (19/44), respectively. The highest rate of infection in the infection onset group was respiratory tract (43.2%), followed by bloodstream infection [29.6% (13/44)], intracranial infection [15.9%, (7/44)], intestinal infection [6.8% (3/44)], and urinary tract infection [4.5% (2/44)]; no deaths were reported. In addition to respiratory tract infection, 13.3% (6/45) of the neonates in the colonization control group had intestinal infection, and no pathogenic bacteria was detected in their blood, cerebrospinal fluid, or urine. Univariate analysis showed that compared with the colonization control group, the neonates in the infection onset group had lower gestational ages [days: 28.5 (26.9, 30.0) vs. 32.0 (30.1, 34.6), P < 0.01], lower birth weights and proportion of probiotic usage [birth weights (kg): 1.20 (0.96, 1.44) vs. 1.75 (1.45, 2.23), probiotic usage: 29.5% (13/44) vs. 57.8% (26/45), both P < 0.01], longer length of NICU stay and duration of antibiotic usage [length of NICU stay (days): 65.11±23.00 vs. 40.31±20.04, duration of antibiotic usage (days): 23.09±9.57 vs. 11.80±7.19, both P < 0.01], and higher proportions of invasive procedures such as mechanical ventilation > 3 days and central venous catheterization > 7 days [mechanical ventilation > 3 days: 61.4% (27/44) vs. 20.0% (9/45), central venous catheterization > 7 days: 81.8% (36/44) vs. 28.9% (13/45), both P < 0.01], indicating that these factors were associated with Serratia marcescens infection onset acquired in NICU. Binary multivariate Logistic regression analysis showed that a birth weight of ≤ 1.5 kg [odds ratio (OR) = 5.745, 95% confidence interval (95%CI) was 1.345-24.549, P = 0.018], a length of NICU stay > 45 days (OR = 3.642, 95%CI was 1.102-12.041, P = 0.034), duration of antibiotic usage (OR = 0.871, 95%CI was 0.799-0.949, P = 0.002), non-usage of probiotics (OR = 3.191, 95%CI was 1.058-9.627, P = 0.039), and invasive procedures such as mechanical ventilation > 3 days (OR = 5.302, 95%CI was 1.510-18.619, P = 0.009), and central venous catheterization > 7 days (OR = 3.818, 95%CI was 1.103-13.212, P = 0.034) were independent risk factors for the onset of NICU-acquired Serratia marcescens infection.

Conclusions: The incidence of NICU-acquired Serratia marcescens infection is high. Low birth weight, prolonged length of NICU stay, long-term antibiotic usage, and invasive treatments are independent risk factors for the onset of NICU-acquired Serratia marcescens infection. Oral probiotics may be a new method for preventing onset of NICU-acquired Serratia marcescens infection.