Article Text

Download PDFPDF

Question 2: Does coagulase negative staphylococcal sepsis cause neurodevelopmental delay in preterm infants?
  1. Matthew McGovern1,2,
  2. Lisa Flynn3,
  3. Sheena Coyne3,
  4. Eleanor J Molloy1,2,3,4
  1. 1 Department of Paediatrics, Academic Centre, Tallaght Hospital, Trinity College, University of Dublin, Dublin, Ireland
  2. 2 Trinity Translational Medicine Institute, St James’s Hospital, Dublin, Ireland
  3. 3 Department of Neonatology, Coombe Women and Infants University Hospital, Dublin, Ireland
  4. 4 Deparment of Neonatology, Our Lady’s Children’s Hospital, Dublin, Ireland
  1. Correspondence to Professor Eleanor J Molloy, Department of Paediatrics, Trinity College Dublin, Dublin, Ireland; eleanor.molloy{at}

Statistics from


A preterm male neonate who was born at 24+2 weeks’ gestation is now 3 weeks old, on full enteral feeds and nasal continuous positive airway pressure in room air. Following multiple episodes of oxygen desaturation and an increase in the oxygen requirement a sepsis evaluation is performed and empiric intravenous antibiotic therapy is commenced according to local guidelines.

Coagulase negative Staphylococcus (CONS) is subsequently isolated on blood culture. Your registrar has enquired about the possible neurodevelopmental implications of this episode of CONS sepsis.

Structured clinical question

Do preterm neonates (patient) requiring treatment for CONS sepsis (intervention) have poorer neurodevelopmental outcomes (outcome) compared with neonates without sepsis (comparison)?


Secondary sources: The Cochrane library and DARE were searched and revealed no relevant systematic reviews.

PubMed: PubMed was searched using the following search strategy: (((neurodevelopment*[Title/Abstract] OR development*[Title/Abstract])) AND (vlbw[Title/abstract] OR Low-birth-weight[Title/Abstract] OR low birth weight[Title/Abstract] OR preterm[Title/Abstract] OR premature[Title/Abstract])) AND (CONS[Title/Abstract] OR sepsis[Title/Abstract]). The search was limited to English publications in humans, within the infant age group (birth 23 months). The search revealed 573 publications, the abstracts of which were reviewed by three authors (MMG, LF, SC). Our literature search identified seven relevant studies: one systematic review and meta-analysis,1 four prospective cohort studies,2–5 one retrospective cohort study6 and one case–control study.7 All studies included looked at preterm neonates <37 weeks with confirmed diagnoses of CONS sepsis in whom the outcomes of interest included either neuroimaging or detailed developmental follow-up. See table 1 for included studies.

Table 1

CONS and neurodevelopmental outcome.


Late-onset sepsis (LOS) is a commonly encountered problem in the neonatal intensive care unit (NICU) which is associated with increased mortality,8 and prolonged hospital stay in the paediatric population.8 9 CONS organisms, which are normally commensals on the skin of healthy adults, are recognised as one of the most important causes of LOS among preterm infants.10–18 A variety of factors predispose the preterm neonate to sepsis with CONS organisms including underdeveloped mucosal barriers, immature neutrophils and impaired pattern recognition receptor signalling.19 CONS organisms have often been viewed as more indolent pathogens as they are less often associated with focal complications such as osteomyelitis or meningitis; however, they are more likely to cause persistent infection in the very low birthweight (VLBW) population.20 CONS infections have also been associated with an attenuated neonatal inflammatory response21 and are the most commonly noted organism in cases of polymicrobial bloodstream infection where they have increased mortality and complications compared with single-organism infections.22 23 In addition, it is now recognised that each episode of bacteraemia within the preterm population can cause cerebral injury even in the absence of meningitis,24 and evidence from animal models supports the role of CONS in contributing to this inflammatory brain injury.25

As the most important nosocomial infection in the preterm neonate, prevention of CONS sepsis is given high priority in neonatal care.26 A variety of quality improvement interventions including improved hand hygiene, vascular device management and more accurate diagnosis of nosocomial infection are known to reduce the incidence of CONS sepsis within the NICU.12 27 28 Despite this, CONS sepsis remains an important cause of morbidity, and neonates diagnosed with CONS sepsis tend to be a higher risk population who are medically complex, of lower gestation and of lower birth weight.29–31 Thus, while CONS has historically been viewed as less invasive infective organism, CONS sepsis occupies an important place in models of preterm sepsis as it often affects those patients who are more medically complex, and therefore at higher risk of long-term neurodevelopmental complications.

The only systematic review and meta-analysis identified1 included observational studies comparing neurodevelopmental outcomes in VLBW infants with culture-proven sepsis and non-septic VLBW controls. In a subgroup analysis on studies of preterm infants with CONS sepsis, three prospective cohort studies were identified3–5 with similar methodology and all used the same developmental assessment scale (Bayley Scales of Infant Development, BSID) to measure endpoints between 18 and 24 months’ corrected age. Inclusion criteria differed between studies (based on either weight or gestation) and it is unclear whether developmental assessors were blinded to previous sepsis episodes. Subsequent pooled analysis found significant associations between CONS sepsis and adverse neurodevelopmental outcome as measured by neurodevelopmental index and diagnosis of cerebral palsy (CP).

A retrospective cohort study of preterm infants <29 weeks with confirmed CONS sepsis from a single centre evaluated neurodevelopment as the primary outcome.6 Neurodevelopmental outcome was measured using Wechsler Preschool and Primary Scale of Intelligence-Revised, BSID II or Stanford-Binet IV scales. This study used a clear definition of cases, included all cases of confirmed CONS sepsis in patients <29 weeks, had an appropriate control population and had relevant follow-up data on the majority of patients. This study found that CONS sepsis in those <29 weeks was associated with cognitive delay but not major disability after adjusting for potential confounders such as gestational age and intraventricular haemorrhage.

Of the four prospective cohort studies identified from our search three were included for analysis in the systematic review on developmental outcomes discussed above. The remaining prospective cohort study by Hemels et al examined MRI findings and neurodevelopmental outcomes in preterm infants aged <32 weeks with CONS sepsis and compared these with preterm infants without sepsis episodes. The study had a clear case definition and follow-up data were available at 15 months’ corrected age in 93% of patients enrolled and at 24 corrected months in 90%. The study detected no difference in MRI findings of white matter injury on a term-corrected MRI brain between groups and no differences in neurodevelopmental outcome at 24 months using the Griffiths Mental Development Scales, though results were not corrected for potential confounders.2

A case–control study was identified which attempted to identify the microbiological antecedents of CP by studying the blood culture results of patients with CP who were previously enrolled in a study of the effects of magnesium sulfate on women in preterm labour at <34 weeks.7 Case definition for CONS sepsis was based on microbiology results only, with no data on clinical assessment or other laboratory parameters. This study found no association between CONS sepsis and subsequent CP at 18 months after controlling for potential confounders, though notably only five patients with CP were included and more detailed data on developmental scoring are not provided. Interestingly the same study purported an association between CONS cultured from maternal swabs and subsequent diagnosis of CP.7 While on the surface this may seem like and interesting observation, the authors do not prove causation and do not present mechanistic data to support the association observed. In addition, the study includes only five cases of CP and most notably the findings have not since been replicated in similar studies.32


The only systematic review identified after literature search showed that CONS sepsis is associated with adverse neurodevelopmental index and CP. While the inclusion criteria differed somewhat between studies included, the methodology and endpoints were otherwise similar, and the pooled data included 4431 infants. The remaining observational studies identified on literature review were smaller and of poorer design. The findings of these studies were variable with regard to neurodevelopmental outcome; however, several did not adjust for potential confounders. Also notable is that none of the studies included investigated the potential effects of differing antibiotic regimens on the developmental outcomes on those enrolled. Based on the literature search performed, best evidence suggests that CONS sepsis is associated with adverse developmental outcome, though all studies identified from literature search were observational in nature and their causal inference is limited by the inability to show a dose–response relationship (grade B).

Clinical bottom line

  • Coagulase negative Staphylococcus (CONS) is the most common organism in late-onset sepsis among preterm infants and is associated with adverse outcome (grade B).

  • Similar to other episodes of neonatal sepsis, CONS sepsis is associated with developmental impairment in the preterm neonate (grade B).

  • Due to the potential association with developmental impairment, episodes of CONS sepsis may be important considerations during the subsequent routine developmental follow-up of affected preterm infants (grade B).



  • Contributors MMG performed the literature review, reviewed the resultant articles and wrote the manuscript. LF and SC performed the literature review, reviewed the resultant articles and contributed to the writing of the manuscript. EJM designed the research question, aided in the review of articles and edited the manuscript before submission.

  • Funding This research was funded by the National Children’s Research Centre, Crumlin, Dublin, Ireland.

  • Competing interests None declared.

  • Patient consent Not required.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles