Early-onset sepsis is generally acquired via maternal ascending vaginal infection (8) (Table 1). Untreated genital tract colonization with group B streptococcus (GBS), prolonged rupture of membranes, and chorioamnionitis are known risk factors for the development of EOS (9–12). Importantly, two large, retrospective studies of very low birth weight (VLBW) neonates failed to identify an association between the development of EOS and the risk for subsequent or late-onset sepsis (LOS; defined as sepsis occurring after 72 h of life) (13, 14). Interestingly, neonates who were born at <25 weeks gestation and survived EOS showed a significant reduction in risk of LOS or death by 120 days (14). Extremely preterm neonates that are able to survive EOS may simply have a more robust immune response and thus, have a bias to equally fair well with subsequent infections. An alternative explanation offered by the authors is that the early immune stimulus may transform the preterm neonatal immune system from a relative state of tolerance to a level of competence that is better suited to defend against pathogens in the extrauterine environment (14).

Chorioamnionitis likewise has been shown to be associated with reductions in respiratory distress syndrome, chronic lung disease, and mortality in preterm neonates (16–19) (Figure 1). The mechanism by which chorioamnionitis decreases the incidence of pulmonary disease in preterm neonates is likely via increased levels of interleukin (IL)-1 and IL-6, which stimulate pulmonary surfactant production and promote fetal lung maturation (20–23). Meanwhile, the observed difference in mortality among neonates born to mothers with chorioamnionitis may be explained in part by the fact that these neonates develop significantly fewer cases of LOS, which has been associated with prolonged hospital stays and death (15, 24).

The presence of early exposure to inflammation, bacteria, and infections may have lasting beneficial effects. Sepsis among preterm neonates appears protective to the development of childhood asthma (25). Birth cohort studies investigating the impact of endotoxin [bacterial lipopolysaccharide (LPS)] exposure during infancy on the risk of later wheezing, atopy, and asthma have had largely mixed results (26–32). More recently, Lynch et al. demonstrated that children who had reduced bacterial exposure in the first year of life were more likely to develop atopy at age 3 years (33). In the September 2015 issue of Science, Schuijs et al. published their results utilizing a murine model of asthma to investigate the impact of chronic exposure to low-dose endotoxin and concluded that endotoxin protects mice from asthma development by increasing the synthesis of the enzyme A20 (a nuclear factor-κB attenuator) in airway epithelial cells (34).

Early exposure to inflammation and infection is not without harmful and devastating effects. VLBW neonates with EOS are about threefold more likely to die than those without EOS with an overall mortality of 35–37% by 120 days (3, 35). Moreover, a meta-analysis of 17 studies demonstrated that sepsis was associated with poor long-term neurodevelopment among VLBW neonates including cerebral palsy (36). Likewise, chorioamnionitis appears to be associated with cystic periventricular leukomalacia in preterm neonates, encephalopathy in term neonates, and cerebral palsy in both preterm and term neonates (37–41). Injury to the preterm brain is believed to result from a multi-hit mechanism in which the neonate is first exposed to inflammation and cytokine release in utero, leading to increased susceptibility to subsequent perinatal and postnatal insults (42). This model is supported by the work of Korzeniewski et al. who demonstrated the cumulative contributions of chronic placental inflammation, acute fetal inflammation, and postnatal inflammatory events on neonatal white matter injury (43). Microglial activation has a central role in this process via excitotoxic, inflammatory, and free radical injury to the developing central nervous system (44). The aforementioned hypothesis has been further validated in rat models of the generation of neuroinflammation in which rat pups subjected to both endotoxin and hypoxic ischemia demonstrated white water injury but sham and endotoxin alone groups did not (45). The group that received both endotoxin and hypoxic ischemia notably had an increase in activated microglia and tumor necrosis factor (TNF)-alpha expression compared to the other groups. The investigation of potential therapies to treat theses prenatal insults by targeting activated microglia and astrocytes is ongoing and includes the administration of dendrimer-based N-acetyl-l-cysteine treatment in the postnatal period, which has been shown to suppress neuroinflammation and improve motor function in newborn rabbits with cerebral palsy (46).

The first exposure to antibiotics often occurs prior to birth in the form of intrapartum antibiotic prophylaxis against GBS, treatment of suspected chorioamnionitis, or antibiotic prophylaxis for women undergoing elective or emergency Cesarean sections. The use of intrapartum antibiotics is steadily increasing; for example, Van Dyke et al. demonstrated that the percentage of pregnant women receiving intrapartum increased from 26.8% in 1998–1999 to 31.7% in 2003–2004 (47). In an era of widespread prophylactic treatment of GBS for colonized pregnant women, the incidence of invasive early-onset GBS disease has decreased by more than 80%; however, the incidence of late-onset invasive GBS disease has remained unchanged (48). The use of maternal antibiotic prophylaxis is not without risks including the emergence of antimicrobial resistance invasive GBS and other neonatal pathogens. Between 1996 and 2003, clindamycin and erythromycin resistance has significantly increased in invasive GBS isolates (49). The incidence of early-onset Escherichia coli sepsis has also significantly increased in VLBW neonates with 64–85% of recent cases having resistance to ampicillin (35, 50). Not unexpectedly, neonates with ampicillin-resistant E. coli infections were more likely to be born from mothers who received intrapartum ampicillin (35, 50–52). Moreover, a multicenter case–control study during 1995–1996 demonstrated that cases of resistance E. coli infection were more often preterm (91 vs 20%, p < 0.001) and had significantly greater morality (40.9 vs 0%, p = 0.017), compared to cases of susceptible E. coli infections (51).

Among underweight and preterm neonates, the use of empiric antibiotics has essentially become standard of practice with antibiotics being the most prescribed medications in the neonatal intensive care unit (53). This phenomenon is largely due to the difficulty of accurately diagnosing neonatal sepsis in symptomatic neonates with developmental immaturity. A retrospective cohort analysis of 5,693 extremely low birth weight (ELBW; <1,000 g) neonates demonstrated that 98% of neonates received antibiotic treatment in the first three postnatal days, while <2% of neonates had positive blood cultures and clinical symptoms of EOS (7). The majority of neonates in the cohort received >5 days of empiric antibiotics despite having negative cultures; each additional day of empiric treatment was associated with a 4% increase in the odds of necrotizing enterocolitis (NEC) and a 16% increase in the odds of death (7). Similarly, Kuppala et al. demonstrated prolonged administration of empirical antibiotics was associated with increased LOS and the composite outcome of LOS, NEC, or death (54). These short-term deleterious outcomes, as well as an increased incidence of invasive candidiasis, may be the result of intestinal microbiome modification, including decreased microbial diversity, which is associated with broad-spectrum antibiotic use in ELBW and preterm neonates (55–57). The disruption of the microbiome may lead to long-term health consequences including decreased absorption of nutrients and vitamin production, as well as increased risk of infections, asthma, diabetes, and obesity. Further discussion on the effect of antibiotics on the microbiome and the role of dysbiosis in pediatric disease is beyond the scope of this mini review, but the interested reader is directed to a number of outstanding recent reviews (58–60).

Vaccines provide a unique opportunity to prevent infection-associated disease. Hepatitis B vaccine is the only vaccine currently recommended in the first month of life by the United States Department of Health and Human Services, Centers for Disease Control and Prevention and is often administered during the birth hospitalization for healthy, term neonates (61, 62). Essentiality, all infants administrated hepatitis B vaccine respond with hepatitis B surface antigen-specific humoral and cell-mediated immunity following completion of the primary vaccine series (63). Although antibody titers decrease over time, immunological memory persists with vaccinated responders mounting a rapid anti-hepatitis B surface antibody response to a vaccine challenge (63). This immunological memory has had a dramatic impact on reducing hepatitis B infection and disease worldwide. After the implementation of universal hepatitis B vaccination program in Taiwan, the seroprevalence rate of hepatitis B surface antigen in children decreased from 10 to 0.7% (64). Likewise, universal vaccination significantly reduced the incidence of pediatric fulminant hepatitis and hepatocellular carcinoma (64). In addition to its clear beneficial effects, early vaccination for hepatitis B remains remarkably safe. Over one billion doses of hepatitis B vaccine have been administered worldwide with few true adverse reactions, and no evidence of an association with sudden infant death syndrome, multiple sclerosis, or chronic fatigue syndrome (65).

Contrasting with the success of the hepatitis B vaccine, the use of other vaccines early in life has been more challenging and frequently less successful. The administration of vaccines against influenza, measles, and mumps during infancy has been unsuccessful given the poor generation of host antibodies (66, 67). Likewise, infants demonstrate decreased cell proliferation and IFN-γ production in response to the polio vaccine, compared to adults (68). This relative resistance to the development of life-long adaptive immunity early in life has impeded the use of many current vaccines in neonates. Generally, this has been attributed to the absence of a strong type 1 T helper cellular response to the antigen. The use of immune adjuvants appears to be one of the best methods to elicit a stronger immune response and overcome this limitation. Currently, aluminum salts, oil-in-water emulsions (MF59, AS03, AF03), virosomes, and AS04 [monophosphoryl lipid A (MPLA) preparation with aluminum salt] are being used as adjuvants in vaccines approved for use in the United States and/or Europe (69).