563 - Late-onset Carbapenem-resistant Enterobacteriaceae Bacteremia among Very Preterm Infants: A Multicenter Study in China

Publication Number: 563.5743

Shujuan Li, Children's Hospital of Fudan University, Shanghai, Shanghai, China (People's Republic); Weiyin Yu, Children's Hospital of Fudan University, Shanghai, Shanghai, China (People's Republic); Jinglin Chen, Children‘s hospital of fudan university, 上海市 / 市辖区 / 杨浦区, Shanghai, China (People's Republic); Jie Yang, Fudan Children's Hospital, Shanghai, Shanghai, China (People's Republic); Shoo K Lee, Mount Sinai Hospital, Toronto, ON, Canada; Yun Cao, Children's Hospital, Fudan University, Shanghai, PRC, Shanghai, Shanghai, China (People's Republic); Siyuan Jiang, Children's Hospital of Fudan University, Shanghai, Shanghai, China (People's Republic)

Background: Data on CRE infections in very preterm infants are limited

Objective: To determine the prevalence of CRE bacteremia in very preterm infants in China and to evaluate the risk factors for, the outcome and treatment of, very preterm infants with CRE bacteremia.

Design/Methods: This cohort study included all infants born at 24-31 weeks of gestation or birth weight < 1500g admitted to hospital participating in the Chinese Neonatal Network from January to December, 2022, with 1 episode of bacteremia caused by Enterobacteriaceae. Infants who had missing data on antimicrobial susceptibility testing and bacteremia occurring within 72 hours after birth were excluded. Demographics, clinical characteristics, and outcomes were compared between infants with CRE and non-CRE bacteremia.

Results: Among 11447 preterm infants, 205 infants were included in the study presenting 207 episodes of Enterobacteriaceae. A total of 27 (13.0%) episodes of bacteremia were caused by CRE and the other 180 (87.0%) were caused by non-CRE Enterobacteriaceae. The most frequently isolated CREs were Klebsiella app. (66.7%, 18/27), followed by Enterobacter spp. (29.6%, 8/27). In a multivariate model, birth weight and previous antibiotic exposure to carbapenem were the independently associated with CRE bacteremia. Six (22.2%) patients in CRE group and 18(10.1%) patients in non-CRE group died due to Enterobacteriaceae bacteremia (p =0.07). The overall mortality during hospitalization was 29.6% in the CRE group and 13.5% in the non-CRE group (p=0.04). Infants with CRE bacteremia had a significantly higher rate of concurrent meningitis than those with non-CRE bacteremia (adjusted OR 4.54; 95% CI 1.09-18.9).
For empirical antibiotic treatment within 48 hours of CRE bacteremia onset, 21/27 (77.8%) infants were treated with meropenem solely for Gram-negative bacteria coverage, and 6/21 (28.6%) of these infants died within 7 days of bacteremia onset. After antimicrobial susceptibility test reports were available at sixth day of blood culture, 22 infants were survival. Among these infants, 17 (77.3 %) infants received monotherapy and 5 (22.7%) received combination therapy. The most frequent antibiotic given in monotherapy was meropenem (12/17, 70.1%). The median duration of antibiotic course for CRE bacteremia was 22.0 days (IQR 8.0-24.8).

Conclusion(s): A significant proportion of late-onset Enterobacteriaceae bacteremia in very preterm infants in Chinese NICUs is caused by CRE, resulting in a high case fatality rate. Meropenem-based regimens remain the primary treatment strategy for CRE bacteremia, while the treatment failure rate is considerable.

Baseline and demographic characteristics of all episodes of Enterobacterales bacteremia.
Table 1.pdf

Illness severity and outcomes of infants with CRE compared with non CRE bacteremia.
Table 2.pdf

Table 3 Antimicrobial regimen of infants with CRE bacteremia
Table 3.pdf