Major article
Neonatal gram-negative bacillary late-onset sepsis: A case-control-control study on a prospectively collected database of 5,233 admissions

https://doi.org/10.1016/j.ajic.2015.09.009Get rights and content

Highlights

  • Gram-negative bacillary late-onset sepsis results in greater mortality and morbidity in extremely preterm infants and late-preterm and term-born neonates.

  • Prolonged use of total parenteral nutrition is the most important risk factor for acquisition of gram-negative bacillary late-onset sepsis in the neonatal intensive care unit.

  • Independent predictors of in-hospital mortality in neonates with gram-negative bacillary late-onset sepsis were Pseudomonas aeruginosa etiology and underlying secondary pulmonary hypertension, renal disease, and neuromuscular comorbidities.

Background

Gram-negative bacillary (GNB) bloodstream infections account for 20%-30% of neonatal late-onset sepsis (LOS). We aimed to identify the incidence, clinical characteristics, and risk factors for adverse outcomes in neonates with GNB LOS.

Methods

All patients with GNB LOS admitted to the neonatal intensive care units (NICUs) of a university-affiliated teaching hospital in Taiwan from January 1, 2004-December 31, 2011, were enrolled. A case-control-control study was performed to evaluate risk factors for acquisition of neonatal GNB LOS.

Results

Of the 5,010 neonates, 290 (5.8%) had a total of 346 episodes of GNB LOS (36.7% of total LOS), with an incidence rate of 13.6 per 10,000 neonate hospital days. The overall mortality rate was 17.6% (51/290), and the sepsis attributable mortality rate was 9.8% (34/346 episodes). After multivariate logistic regression analysis, neonates with prolonged use of total parenteral nutrition (adjusted odds ratio [OR] = 1.53; 95% confidence interval [CI], 1.02-2.29; P = .041) were independently associated with acquisition of GNB LOS. The independent predictors of in-hospital mortality were Pseudomonas aeruginosa etiology (OR = 11.45; 95% CI, 2.83-46.24) and underlying secondary pulmonary hypertension (OR = 18.02; 95% CI, 3.28-98.89), renal disease (OR = 17.16; 95% CI, 2.96-99.38), and neuromuscular comorbidities (OR = 2.72; 95% CI, 1.06-7.00).

Conclusion

Given the higher illness severity and sepsis-attributable mortality rate of neonatal GNB LOS in the NICU, strategies to reduce the incidence need to be addressed urgently.

Section snippets

Study population and design

This study was conducted in the NICU of Chang Gung Memorial Hospital (CGMH), which provides care from primary to tertiary levels in a university-affiliated teaching hospital in Northern Taiwan. The NICU of the CGMH includes 3 units and has a total capacity of 49 beds equipped with mechanical ventilators and 28 beds with special care nurseries. We made use of the prospectively collected neonatal database of our NICU, which have been kept by research nurses for >10 years.5, 19 Between January

Incidence of gram-negative LOS

During the study period, a total of 5,233 neonates were hospitalized in our NICUs, and a total of 223 neonates, including 181 neonates who died within the first 3 days of life and 42 neonates who were transferred to other hospitals, were excluded from the analysis. Of 5,010 neonates enrolled, 713 (14.2%) neonates had a total of 942 episodes of LOS; of them, 290 (5.8%) neonates had a total of 346 episodes of GNB LOS (36.7% of all LOS). The incidence rate of GNB LOS in the entire cohort was 13.6

Discussion

Results from this study showed that more than one-third of NICU GNB LOS occurred in neonates with a BW of >1,500 g or GA ≥33 weeks, and they accounted for 29.4% (15/51) of all fatal cases; however, VLBW infants or extremely preterm infants were more likely to have GNB LOS than late-preterm or term-born infants. These suggest that not only VLBW infants but also non-VLBW infants encounter GNB LOS and the non-VLBW infant population were ignored previously. The GNB LOS-attributable mortality rate

Conclusions

GNB LOS deserves greater concern because it is associated with a significantly higher severity of illness and sepsis-attributable mortality rate. The risk factor identified in this study, longer duration of total parenteral nutrition, is consistent with risk factors identified by previous investigators. Patients with certain underlying chronic diseases or patients infected with P aeruginosa should be targeted and assessed for higher risk of in-hospital mortality when deciding the treatment

Acknowledgments

We thank Chiao-Ching Chiang for keeping the database of our NICU, and we thank all nursing staff working in our NICUs for keeping extremely detailed patient records, which contributed greatly to the completion of this research. We also thank Chun-Chun Cheng and Yu-Jr Lin for statistical consultation, who was supported by grants from the Biostatistical Center for Clinical Research, Chang Gung Memorial Hospital (grant no. CLRPG340599).

References (40)

  • B.J. Stoll et al.

    Neonatal outcomes of extremely preterm infants from the NICHD Neonatal Research Network

    Pediatrics

    (2010)
  • H. Aly et al.

    Is bloodstream infection preventable among premature infants? A tale of two cities

    Pediatrics

    (2005)
  • C. Auriti et al.

    Determinants of nosocomial infection in 6 neonatal intensive care units: an Italian multicenter prospective cohort study

    Infect Control Hosp Epidemiol

    (2010)
  • M.H. Tsai et al.

    Incidence, clinical characteristics, and risk factors for adverse outcome in neonates with late onset sepsis

    Pediatr Infect Dis J

    (2014)
  • M.J. Bizzarro et al.

    Seventy-five years of neonatal sepsis at Yale: 1928-2003

    Pediatrics

    (2005)
  • F.C. Lessa et al.

    Trends in incidence of late-onset methicillin-resistant Staphylococcus aureus infection in neonatal intensive care units: data from the National Nosocomial Infections Surveillance System, 1995-2004

    Pediatr Infect Dis J

    (2009)
  • M. Cohen-Wolkowiez et al.

    Early and late onset sepsis in late preterm infants

    Pediatr Infect Dis J

    (2009)
  • M.Y. Lai et al.

    Characteristics of neonates with culture-proven bloodstream infection who have low levels of C-reactive protein (<=10 mg/L)

    BMC Infect Dis

    (2015)
  • P.L. Graham et al.

    Risk factors for late onset gram-negative sepsis in low birth weight infants hospitalized in the neonatal intensive care unit

    Pediatr Infect Dis J

    (2006)
  • S. Samanta et al.

    Risk factors for late onset gram-negative infections: a case-control study

    Arch Dis Child Fetal Neonatal Ed

    (2011)
  • Cited by (15)

    • Healthcare-Associated Infections

      2023, Avery's Diseases of the Newborn
    • Epidemiology and microbiology of late-onset sepsis among preterm infants in China, 2015–2018: A cohort study

      2020, International Journal of Infectious Diseases
      Citation Excerpt :

      Gram-negative bacteria LOS was related with the most significantly increased risk of death, PVL, BPD, and NEC. This is consistent with previous reports that infants with Gram-negative bacteremia have a higher severity of illness (Tsai et al., 2016). No association was found between Gram-positive bacteremia and adverse neonatal outcomes except NEC; however, the study might have been underpowered to detect the association because of the relatively low event rates of adverse outcomes.

    • Health Care-Associated Infections

      2018, Avery's Diseases of the Newborn: Tenth Edition
    • Risk factors and clinical outcomes for carbapenem-resistant Gram-negative late-onset sepsis in a neonatal intensive care unit

      2017, Journal of Hospital Infection
      Citation Excerpt :

      The incidence of nosocomial infections caused by Gram-negative (GN) bacteria has been mounting over recent decades [3]. GN late-onset sepsis (LOS) is often associated with a higher mortality rates than is LOS with Gram-positive (GP) bacteria [4,5]. The emergence of carbapenem-resistant (CR) strains of GN bacteria in neonatal intensive care units (NICUs) has been reported by several surveys [6–8].

    • Health Care-Associated Infections

      2017, Avery's Diseases of the Newborn, Tenth Edition
    View all citing articles on Scopus

    Conflicts of Interest: None to report.

    1

    Contributed equally to this article.

    View full text