PC1129: Association of Gestational Age with Central Line and Non-central Line-associated Bloodstream Infections in the Neonatal ICU

Background: The risk of central line-associated bloodstream infections (CLABSIs) is inversely associated with gestational age (GA) for infants in the NICU. Less is known about the relationship between GA and non-CLABSIs, the risk of CLABSIs and non-CLABSIs in late preterm and full term infants, and pathogens associated with CLABSIs and non-CLABSIs in different GA groups. Our objectives were to determine the incidence of CLABSIs and non-CLABSIs among infants of different GA groups ( < 28, 28-33, 34-37, and ≥38 weeks) and assess the pathogens causing CLABSIs and non-CLABSIs in those GA strata.

Methods: From 2013-2019, we performed a retrospective study in an academically-affiliated, Level IV NICU that cares for preterm infants and infants with complex congenital anomalies. Eligible infants were admitted during the first week of life and had ≥1 positive blood culture obtained at ≥3 days of life. BSIs were classified using CDC’s National Healthcare Safety Network criteria as CLABSIs, primary BSIs, and BSIs secondary to infections at another site. Data collected included GA, pathogens (including polymicrobial infections), and symptoms, laboratory results, and radiologic studies required to identify BSI type. Data analysis included descriptive statistics, incidence rates per 1000 admissions, and differences in proportions analyzed by chi-squared tests.

Results: Among 6708 admissions, 214 infants (3.2%) experienced 255 BSIs at ≥3 days of life. Of these infants, 36 (17%) experienced >1 BSI. Incidence of CLABSIs and non-CLABSIs were highest in infants < 28 weeks gestation (Table 1) and non-CLABSI episodes were more common than CLABSI episodes in all GA strata. Coagulase-negative staphylococci (CoNS) were more likely to cause BSIs in infants < 34 weeks compared to infants ≥34 weeks gestation (Table 2; p=0.008). In contrast, gram positive cocci (excluding CoNS and Staphylococcus aureus) caused a higher proportion of BSIs in infants ≥34 weeks compared to infants < 34 weeks gestation (p=0.01). S. aureus, gram negative rods (GNRs), and Candida species caused a similar proportion of BSIs in all GA groups, although Candida spp. were least common. CoNS and S. aureus caused 42% and 32% of CLABSIs, respectively. GNRs caused 18% and 24% of CLABSIs and non-CLABSIs, respectively.

Conclusion: Similar to prior studies, our population demonstrated an inverse relationship between CLABSI rates and GA. The same relationship was observed for non-CLABSIs, which were also more common than CLABSIs in all GA strata. While CoNS were the most common cause of both CLABSIs and non-CLABSIs overall, late preterm and full term infants were more likely to have BSIs caused by other pathogens. These data suggest the need to increase infection prevention measures for non-CLABSIs and to consider tailoring infection prevention strategies by GA, particularly CoNS in infants < 34 weeks gestation and gram positive cocci (excluding CoNS and S. aureus) in late preterm and full term infants.