Cerebrospinal fluid shunt placement for the treatment of hydrocephalus is the most common pediatric neurosurgery procedure performed in the United States (Hanak et al., 2017). Unfortunately, infants under the age of one year of age are at the highest risk for shunt infection following this life-saving procedure and may experience many shunt infections over the course of their childhood (Vinchon and Dhellemmes, 2006; Hommelstad et al., 2013; Simon et al., 2014). Shunt infections are most commonly caused by Staphylococcus epidermidis (S. epidermidis), which forms biofilms on the catheter that are recalcitrant to antibiotic therapy and immune-mediated clearance (Kockro et al., 2000; Fux et al., 2006; Vinchon and Dhellemmes, 2006; Gutierrez-Murgas and Snowden, 2014). Thus, infants with shunt infections undergo many surgical procedures to replace infected shunts (Kockro et al., 2000; Fux et al., 2006; Gutierrez-Murgas and Snowden, 2014). Additionally, shunt infections are associated with long-term increases in seizure risk, decreases in intellectual ability, and overall mortality (Fux et al., 2006; Vinchon et al., 2012). The mechanisms for the increased risk of shunt infections in infants are unknown. Studies designed to advance our understanding of the immune response to shunt infections in infants could significantly improve their life-long outcomes through development of diagnostic, treatment and prevention strategies for these serious infections.

Our laboratory previously developed adult murine models of S. epidermidis and Staphylococcus aureus (S. aureus) CNS catheter infection, mimicking what is seen in humans with ventricular shunt infections (Snowden et al., 2012; Snowden et al., 2013; Gutierrez-Murgas et al., 2016). Establishing a catheter model within the CNS was necessary because the immune response in this compartment often differs from that seen in the periphery due to the inherent anti-inflammatory environment of the normal CNS (Hampl et al., 2003). However, in response to injury or infection, the CNS is capable of mounting a robust immune response (Hampl et al., 2003; Lucas et al., 2006; Kielian et al., 2007). We have observed increases in pro-inflammatory cytokines IL6, IL1β, CXCL1, and CCL2 in this model, driven primarily by bacterial dissemination to surrounding parenchymal tissue, with less inflammation attributable to the catheter-associated infection (i.e. biofilm) (Snowden et al., 2013; Gutierrez-Murgas et al., 2016). This demonstrates that catheter-associated infections trigger an immune response distinct from parenchymal infections. It is important to consider the differences between biofilm and parenchymal inflammatory responses when attempting to understand the pathophysiology of device associated infections such as CNS catheter infections. In this report, we characterize our newly adapted infant mouse model of CNS catheter infection, which will serve as a powerful tool to identify key factors in the host immune response to staphylococcal infections in the CNS. Our results show that the inflammatory response to S. epidermidis is not markedly attenuated, which may have significant implications for diagnosis of shunt infection in this population and may contribute to their increased morbidity.

Infants are at higher risk of S. epidermidis infection following CNS catheter placement in the treatment of hydrocephalus, but the mechanisms of this increased morbidity are not well-defined. Our initial characterization of an infant model of S. epidermidis CNS catheter shows that the early immune response to CNS S. epidermidis catheter infection is notably attenuated in young mice, which is surprising given the significant bacterial burden in the brain tissue ( Figure 1 ) and speaks to a distinct difference in the early immune responses of infants as opposed to adults. Specifically, we have observed that infant mice do not mount the early pro-inflammatory response observed in the previously reported adult model of this infection (Gutierrez-Murgas et al., 2016). This may have significant clinical implications, as diagnosis of shunt infections frequently relies on the detection of systemic and CNS markers of inflammation, which are variable in young infants with central nervous infections and are often absent until late in illness (Gutierrez-Murgas and Snowden, 2014). The relatively lower levels of CNS inflammation observed in our studies may explain the common lack of traditional measures of infection like fever in children with shunt infections and suggests that we may need an increased sensitivity for diagnosing infection in these children, as they may not mount the robust inflammatory response we expect in older children with meningitis or device associated infection.

Human studies have demonstrated a similar attenuation in inflammatory responses to S. epidermidis in infants, with age-dependent decreases in cytokineproduction following exposure of cord blood monocytes to S. epidermidis in vitro and decreased systemic levels of C-reactive protein in infants with S. epidermidis device-associated infections (Klingenberg et al., 2005; Strunk et al., 2012). However, an in vivo model of S. epidermidis bacteremia reported increased CXCL1, IL6, and TNFα in the bloodstream of neonatal mice, suggesting that the immune response to S. epidermidis may be tailored to the site of infection (Kronforst et al., 2012). Systemic measures of inflammation will be evaluated in future studies of infant CNS catheter-associated infection to assess parallels between human and murine illness, as well to identify potential novel biomarkers. Microglia and astrocytes also appear to have altered responses in infancy, which likely plays a key role in the phenotype observed in this infant model of infection and will be examined in future studies directly comparing adult and infant responses (Floden and Combs, 2006; Kurronen et al., 2012; Christensen et al., 2014). This highlights the importance of longitudinal mechanistic studies to increase our understanding of the immune response to S. epidermidis in the CNS, as this infection is clinically and immunologically distinct from peripheral bacterial infection.

Interestingly, a greater proportion of bacteria spread from the catheter to the parenchyma in young mouse brains ( Figure 1 ). In contrast, in adult animals, significantly more bacteria are adherent to the catheter, with a lesser degree of parenchymal spread (Gutierrez-Murgas et al., 2016). Importantly, this increase in parenchymal bacterial spread in the presence of a hypo-inflammatory immune response in young hosts may play a role in the increased susceptibility of human infants to S. epidermidis disease. The mechanisms responsible for differential bacterial localization in young vs. adult brains will be pursued in other studies.

There is increasing evidence that inflammation in early development, from intrauterine, peripartum and early childhood insults, may have lifelong impacts on neurologic function. For example, elevated pro-inflammatory mediators in cord blood have been associated with an increased risk of cerebral palsy in several case series although it is difficult to determine how much of this effect is due to inflammation vs. other confounders (Armstrong-Wells et al., 2015; Bi et al., 2014; Varner et al., 2015). Animal studies attempting to better define the impact of specific inflammatory processes on neurologic function have demonstrated that cytokine exposure during infancy increases the risk for changes in memory or neuropsychiatric function later in life following a second immune challenge (Bilbo et al., 2005; Bilbo et al., 2006; Bilbo et al., 2007; Williamson et al., 2011). Additionally, following E. coli sepsis, rodents are less responsive to cognitive enrichment strategies than those without a history of infection (Bilbo et al., 2007). Interestingly, these effects are only observed in rodents less than 30 days of age, suggesting an early period of vulnerability in the developing brain (Bilbo et al., 2006; Williamson et al., 2011). Therefore, the attenuated inflammatory response observed in this infant model of S. epidermidis CNS catheter infection may be beneficial to the developing brain. The impact of CNS S. epidermidis infection on neurodevelopmental outcomes has not been previously explored, despite the frequency of this pathogen in human infants. The trend we observed in nest building studies ( Figure 5 ) suggest that early CNS catheter infection may impact long-term neurologic outcomes; however, there was variability in the phenotypes observed. This variation parallels variations in human outcomes and suggests the need for more study of neurologic changes in this setting. Future studies will also need to include other biofilm-forming pathogens reported in CNS catheter infection, such as S. aureus, E. coli and other Gram-negative bacteria, and fungal pathogens such as Candida albicans (Gutierrez-Murgas and Snowden, 2014; Simon et al., 2014; Hanak et al., 2017).

The development of this infant murine model of central nervous system catheter infection is an important advancement as it now allows an investigation of the immunologic and neurologic responses to these infections in infants. With the identification of inflammatory responses that are skewed in young animals, we can design future studies to evaluate key signaling decision points, which will provide more specific targets for potential immunomodulatory therapy and screening for high-risk patients. New treatment and diagnostic modalities are imperative given the high costs, in healthcare expense and lifelong patient morbidity, associated with CNS catheter infections in infancy.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was approved by University of Nebraska Medical Center IACUC; University of Arkansas for Medical Sciences IACUC. The study was conducted in accordance with the local legislation and institutional requirements.

GS: investigation, methodology, writing – review & editing. KE: investigation, methodology, writing – review & editing. MB: investigation, methodology, writing – review & editing. BW: investigation, methodology, writing – review & editing. MH: investigation, methodology, writing – review & editing. JS: conceptualization, formal analysis, investigation, methodology, project administration, supervision, writing – original draft, writing – review & editing.