This year, The Journal of Perinatal & Neonatal Nursing (JPNN) marks 30 years of publication-a terrific achievement to be celebrated by all who have contributed to the journal over the past 3 decades. Congratulations and thank you to the editorial team and authors who have made an important contribution to the knowledge of the JPNN readership and, as a result, have improved the nursing care of women and children during the perinatal and neonatal periods. Given the 30-year milestone and focus of this edition on infection, it seemed appropriate to write about a clinical issue in perinatal and neonatal health that has an important history with many lessons learned to guide our future practice. The wealth of knowledge relative to the discovery and administration of antibiotics during the perinatal and neonatal periods and our increasing understanding of the importance of optimizing microbial health early in life make this a critical issue in nursing practice.

Antibiotics are the most commonly prescribed medication in the neonatal intensive care unit (NICU). Overuse and misuse of these drugs occur frequently, with studies showing a 40-fold variation in prescribing practices in NICUs within the United States.1 In this column, an overview of the history of antibiotic administration is followed by a review of the consequences that have been associated with routine antibiotic administration to preterm infants following birth. The culmination of our collective knowledge of these consequences has resulted in multiple approaches to foster antimicrobial stewardship in the newborn intensive care unit. A summary of best practices relative to antimicrobial stewardship is provided, along with suggestions for future research on antibiotic administration and microbial health of the newborn.

HISTORY OF ANTIBIOTIC ADMINISTRATION DURING THE PERINATAL AND NEONATAL PERIODS

One of the most serendipitous moments in the recent history of science was Alexander Fleming's discovery of the first antibiotic, penicillin. Fleming made his discovery in 1928. However, penicillin was not used to treat infections in humans until 1942 and was not widely available for use in medicine until a large-scale production process was developed in 1945. As a result of Fleming's discovery of penicillin and the generations of antimicrobial medications that followed, infectious disease as a cause of death has fallen from the first 3 causes of death (pneumonia, tuberculosis, and diarrhea and enteritis) in 1900 to the eighth cause of death (influenza and pneumonia) in 2013.2

Pregnant women and infants have been significantly influenced by Fleming's discovery and the widespread use of antibiotics. Prior to antibiotics, infection was one of the leading causes of pediatric death. Today, bacterial sepsis has been reduced to the seventh cause of death in children.2 Before widespread use of antibiotics, the most common bacterium to be associated with a serious infection in pregnant women and infants was the group A [beta]-hemolytic Streptococcus.3 Group B [beta]-hemolytic Streptococcus (GBS) emerged later as an important pathogenic organism among pregnant women and infants. By the late 1950s, at least 6 case reports of infants with GBS septicemia were documented in the literature. Of these 6 cases, 4 had meningitis and 2 had pneumonia. None of the infants survived the GBS infection.4

During the next 2 decades, concern regarding infection associated with GBS mounted. Hood and colleagues5 first reported GBS to be a pathogenic organism for pregnant women and infants in 1961, and this was soon followed by a report of early-onset meningitis caused by GBS.6 The mechanisms associated with GBS and infection during infancy were highlighted in a report by Eickhoff and colleagues7 in 1964, and within 10 years, GBS was thought to be one of the most important challenges facing the neonatal care team, as it had been associated with a variety of lethal infectious diseases. This concern was coupled with other insights into infectious disease during infancy, namely, the concepts of "early" versus "late" onset infection8 and nosocomial transmission of infection in the newborn nursery.9

At this time, the emerging knowledge of neonatal infectious disease, along with the high case fatality rate for all infants with infection, especially those born preterm, resulted in a significant amount of apprehension among the neonatal care team. This was magnified by the challenges associated with differentiating the radiographic findings of preterm infants with respiratory distress syndrome (RDS) from those with early-onset GBS pneumonia, both of which are characterized by the presence of hyaline membranes.10,11 It was this clinical diagnostic challenge of differentiating RDS from GBS and other types of pneumonia that resulted in the routine, prophylactic administration of antibiotics to preterm infants that became common in the late 1960s and early 1970s. While this practice was widely adopted as a prevention strategy against infection in the preterm infant, it was not without concern for the resistant organisms that were likely to emerge. For this reason, many neonatal clinicians at that time and over the intervening years have worked to develop a predictive diagnostic test for neonatal infection to better guide therapy. Unfortunately, one has yet to be translated for use in clinical practice.

The past several decades have been characterized by routine administration of antibiotics to preterm infants in the early postnatal course as a preventative strategy against infection. While this practice has reduced deaths associated with neonatal infectious disease, it has not prevented all neonatal mortality associated with GBS sepsis, as some infants, regardless of antibiotic therapy, do not survive their infection. Differentiating the infants who respond to postpartum antibiotic therapy is difficult based on maternal factors, and only gestational age and birth weight have been associated with mortality. Thus, attempts to prevent GBS colonization in pregnant women prior to delivery were adopted along with postpartum prophylaxis in the late 1970s and early 1980s. This approach was effective in the prevention of early-onset GBS infection.12,13 However, this intervention, like the indiscriminate administration of antibiotics to preterm infants, was not without concern for the development of resistant organisms and other unintended clinical consequences. Concerns at this time were magnified by the findings reported in 1983 from a large randomized clinical trial conducted on preterm infants born less than 2 kg that showed no difference in early-onset GBS infection between those receiving postpartum prophylactic penicillin and their untreated controls.14 The angst over these findings was somewhat offset by a study reported in 1986 showing that a more selective use of antibiotics (ampicillin) in high-risk situations could interrupt mother-to-infant transmission of GBS. However, this also required the administration of antibiotics to the infant during the postpartum course,15 increasing the overall exposure of antimicrobial medications.

Since 1990, the rate of early-onset GBS has been dramatically reduced as a result of guidelines for the prevention of GBS infection published by the American College of Obstetricians and Gynecologists (ACOG), the American Academy of Pediatrics (AAP), and the Centers for Disease Control and Prevention (CDC).16 In 1990, there were approximately 1.7 cases of GBS sepsis per 1000 live births. In 2008, this had been reduced to approximately 0.3 cases per 1000 live births. It should be noted that although only 20% to 30% of GBS infections occur in preterm infants, they are 7.7 times more likely to die from early-onset disease compared with infants who are born at term (95% confidence interval [CI], 4.9-12.3).17 While adherence to the ACOG, AAP, and CDC guidelines has reduced the incidence of early-onset disease, these guidelines do not appear to have influenced the rate of late-onset disease, which has not changed over the past 30 to 40 years.18 Unlike early-onset sepsis that can be influenced by guidelines that are focused on perinatal management and targeted toward maternal risk factors, late-onset sepsis is multifactorial in nature and requires a different approach to prevention. Future research aimed at the development and implementation of a GBS vaccine may prove to be an effective strategy in preventing late-onset infection.

CONSEQUENCES OF ANTIBIOTIC ADMINISTRATION ON INFANT HEALTH OUTCOMES

The discovery and administration of antibiotics to newborns have reduced the morbidity and mortality associated with infection. However, studies have shown that the prolonged routine prophylactic use of antibiotics in preterm infants is associated with some unintended consequences. The most significant of these discussed in the literature include death, necrotizing enterocolitis (NEC), and late-onset sepsis.

In a study involving data from 5693 extremely low-birth-weight infants (birth weight <1000 g) cared for at 19 different NICUs, the median duration of initial antibiotic administration was 5 days (range, 1-36 days).19 Interestingly, more than half (53%) of these infants received a prolonged course of antibiotics, defined as 5 days or more in the setting of sterile culture results. The infants who received a prolonged duration of initial antibiotics were less mature at birth (25.6 weeks vs 26.3 weeks in the not prolonged group), has an Apgar score of less than 5 (19% vs 11% in the not prolonged group), and were black (48% vs 42% in the not prolonged group). When controlling for these factors, as well as the NICU center where they received care, this study found that a prolonged course of antibiotics was associated with increased odds of NEC or death (odds ratio [OR] = 1.30; 95% CI, 1.10-1.54], NEC alone (OR = 1.21; 95% CI, 0.98-1.51), and death alone (OR = 1.46; 95% CI, 1.19-1.78). Each additional day of antibiotic exposure in this population of infants was associated with increased odds of death, NEC, and the composite measure of NEC or death. This study reported approximately 4% increase in the odds of an infant in this population having NEC or dying with each additional day of initial empirical antibiotic treatment. When measuring the outcome of NEC alone, the study reported approximately 7% increase in the odds of and infant having NEC for each additional day of antibiotics. When measuring the outcome of mortality alone, the study reported approximately 16% increase in the odds of an infant dying for each additional day of initial empirical antibiotic exposure.19 This study highlighted that while antibiotics are the most commonly administered medication in the NICU, prolonged duration when negative blood cultures are likely to be present early in the postnatal course is associated with both mortality and NEC among low-birth-weight infants.

Additional studies have explored the influence of prolonged initial antibiotic administration with adverse outcomes in preterm infants.20,21 In a large study of 365 preterm infants born 32 weeks' gestation or less and weighing 1500 g or less, infants who received a prolonged course of initial antibiotics (>=5 days) in the setting of sterile blood cultures were at significantly greater risk of a composite outcome of NEC, late-onset sepsis, and death after 7 days (OR = 2.66; 95% CI, 1.12-6.30).20 This study was not powered to evaluate the risk of NEC or death alone but did find increased odds of late-onset sepsis after 7 days in infants with prolonged exposure to initial antibiotics (OR = 2.45; 95% CI, 1.28-4.67).20 Late-onset sepsis was also shown to be associated with prolonged exposure to antibiotics in a study conducted with more than 200 preterm infants born less than 28 weeks' gestation in Australia.21 Here, the authors reported that a course of initial antibiotics for 4 days or more for suspected early-onset sepsis was associated with increased odds of proven late-onset sepsis (OR = 2.1; 95% CI, 1.2-3.7) but not NEC or death.21 While there is a need for antibiotic administration to the preterm infant in the setting of positive blood cultures, mounting evidence strongly suggests against prolonged use of these medications soon after birth in the setting of negative blood cultures.

The mechanisms underpinning the associations that have been shown between early antibiotic exposure, late-onset sepsis, NEC, and death are likely to be related to the influence of antibiotics on early intestinal colonization in the preterm infant. Antibiotics are known to alter the intestinal microbiome, putting the infant host at risk of infection from pathogenic bacteria and antimicrobial resistance. During infancy when the initial intestinal microbiome is being acquired and commensal bacteria are activating toll-like receptors to protect against gut injury and associated death, the unintended risks of antibiotic administration must be carefully considered. For example, the lack of bacterial diversity and overwhelming abundance of Proteobacteria associated with broad-spectrum antibiotic exposure may predispose the infant to inflammatory diseases such as late-onset sepsis and NEC.22-26 All of this evidence taken together has resulted in a growing body of work focused on antimicrobial stewardship and a more careful consideration of how antibiotics are administered to the preterm infant in the NICU.

ANTIMICROBIAL STEWARDSHIP IN THE NICU

The current practice patterns and clinical outcomes associated with antibiotic administration have contributed to the CDC calling for a need to improve how antibiotics are used as 1 of 4 key strategies required to address the problem of antibiotic resistance in the United States.27 Many hospitals have developed and implemented antimicrobial stewardship programs focused on adult patient populations. These programs have decreased the overall use of broad-spectrum antibiotics and, in turn, have decreased adverse events such as opportunistic infections associated with antibiotic administration.27 The CDC has outlined 7 core elements of successful Hospital Antibiotic Stewardship Programs:

1. Leadership commitment: Dedicating necessary human, financial, and information technology resources.

2. Accountability: Appointing a single leader responsible for program outcomes. Experience with successful programs show that a physician leader is effective.

3. Drug expertise: Appointing a single pharmacist leader responsible for working to improve antibiotic use.

4. Action: Implementing at least 1 recommended action, such as systemic evaluation of ongoing treatment need after a set period of initial treatment (ie, "antibiotic time out" after 48 hours).

5. Tracking: Monitoring antibiotic prescribing and resistance patterns.

6. Reporting: Regular reporting information on antibiotic use and resistance to doctors, nurses, and relevant staff.

7. Education: Educating clinicians about resistance and optimal prescribing.

To successfully ensure that antimicrobial stewardship is achieved in the NICU setting, the CDC and the Vermont Oxford Network have developed a collaboration focused on the development of iNICQ 2016. iNICQ 2016 will

focus specifically on key components of antibiotic stewardship programs within NICUs. The iNICQ Collaborative center-level improvement teams will pioneer evidence-based best practices that will reduce the overuse of antibiotics, while bringing much-needed standardized protocols and baseline data for hospitals to share.28

The literature suggests that the development and implementation of guidelines to reduce the use of specific antibiotics in the NICU (ie, vancomycin) can be effective.29 For example, 2 large hospitals (Brigham and Women's Hospital [BWH] and Massachusetts General Hospital [MGH], both in Boston, Massachusetts) significantly reduced their start rate of vancomycin by 35% and 62%, respectively, after the implementation of a guideline. In addition, the use of a vancomycin administration guideline resulted in a reduction in the number of infants exposed to vancomycin from 5.2 to 3.1 per 1000 patient-days (40% reduction, P = .008) at BWH and 10.8 to 5.5 per 1000 patient-days (49% reduction, P = .009) at MGH without compromising patient safety.29 Given this example of success with antibiotic guidelines and the resources being invested in iNICQ 2016, antimicrobial stewardship practices in the NICU are likely to rapidly evolve with a positive impact on preterm infant health outcomes.

Some of the best practices that have been reported relative to antimicrobial stewardship in the NICU include (1) assembly of a multidisciplinary antimicrobial stewardship team and (2) taking a comprehensive approach to antibiotic management strategies.30

The Infectious Diseases Society of America and the Pediatric Infectious Diseases Society of America recommend the creation of a multidisciplinary and interprofessional antimicrobial stewardship team to lead healthcare organizations in the development and implementation of interventions that will optimize microbial heath among infants.31 Recommended members of this team include representatives from neonatology, infectious disease, pharmacy, infection control and prevention, bioinformatics, and neonatal nursing. It is important to underscore the important role that neonatal nurses play on antimicrobial stewardship teams. Neonatal nurses are often the first to identify changes in vital signs or subtle symptoms that may indicate a new infection or unexpected problems associated with current antibiotic therapy. In addition, hospital policies and procedures that are important to antimicrobial stewardship efforts often depend on nursing insights and input. For example, any changes in how diagnostic blood cultures or other blood samples are collected should involve neonatal nursing input.30

A comprehensive approach to antibiotic management in the NICU requires accurate measurement of antibiotic administration, improvement in diagnostic techniques, rational selection of empirical therapy, and ongoing reevaluation and decision making regarding the potential discontinuation of antibiotic administration.30 Of these, accurately measuring antibiotic administration so that data are readily available to the team may be most important. Without accurately measuring the utilization of antibiotics in the NICU, it is challenging, if not impossible, to benchmark against other institutions or set improvement targets that will reflect antimicrobial stewardship efforts. One commonly used measure of antibiotic utilization in the NICU is days of therapy (DOT). DOT reflects total days of antibiotics administered, regardless of dosing by weight or renal function. This measure can then be adjusted for 1000 patient-days of hospitalization, which is a typical benchmark among pediatric hospitals. Other useful antimicrobial stewardship metrics for the NICU include DOT for concurrent use of medications, DOT for broad-spectrum antibiotic use, DOT of perioperative prophylaxis of more than 48 hours, and reduction in DOT of inadequate therapy per 100 late-onset sepsis evaluations.30 These measures have the ability to reflect stewardship efforts focused on reduction in use of concurrent antibiotics, broad-spectrum antibiotics, prolonged prophylaxis, and ineffective antibiotic administration. Time and resources spent by hospitals in developing systems that generate these measurement data are an investment in antimicrobial stewardship efforts in the NICU.

CONCLUSION

Antibiotics remain the most commonly prescribed medication type in the NICU. The administration of antibiotics to preterm infants has greatly improved morbidity and mortality associated with infection over the past several decades. However, the prolonged use of these medications early in life has been shown to be associated with some significant unintended consequences. The risks of late-onset sepsis, NEC, and death have all been shown to be increased in preterm infants who had sterile blood cultures following birth but received antibiotics for a longer duration. Current and future efforts focused on antimicrobial stewardship in the NICU, as well as ongoing research in the area of the preterm infant microbiome, are likely to further enhance best practices associated with antibiotic administration and, in turn, optimize microbial health of the preterm infant.

-Katherine E. Gregory, PhD, RN

Senior Nurse Scientist

Department of Pediatric Newborn Medicine

Department of Nursing

Brigham and Women's Hospital

Boston, Massachusetts

References