Seminars in Fetal & Neonatal Medicine 22 (2017) 278e283

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine journal homepage: www.elsevier.com/locate/siny

Antibiotic stewardship in perinatal and neonatal care Jayashree Ramasethu a, *, Tetsuya Kawakita b a b

Division of Neonatal Perinatal Medicine, MedStar Georgetown University Hospital, Washington DC, USA Department of Obstetrics and Gynecology, MedStar Washington Hospital Center, Washington DC, USA

a b s t r a c t Keywords: Antibiotics Stewardship Newborn Neonatal intensive care Sepsis

The spread of antibiotic resistance due to the use and misuse of antibiotics around the world is now a major health crisis. Neonates are exposed to antibiotics both before and after birth, often empirically because of risk factors for infection, or for non-specific signs which may or may not indicate sepsis. There is increasing evidence that, apart from antibiotic resistance, the use of antibiotics in pregnancy and in the neonatal period alters the microbiome in the fetus and neonate with an increased risk of immediate and long-term adverse effects. Antibiotic stewardship is a co-ordinated program that promotes the appropriate use of antibiotics, improves patient outcomes, reduces microbial resistance, and decreases the spread of infections caused by multidrug-resistant organisms. This review addresses some of the controversies in antibiotic use in the perinatal period, examines opportunities for reduction of unnecessary antibiotic exposure in neonates, and provides a framework for antibiotic stewardship in neonatal care. © 2017 Elsevier Ltd. All rights reserved.

1. Introduction When considering the use and misuse of drugs in perinatal care, the topic of antibiotic stewardship is of prime importance. Antimicrobial resistance has emerged as one of the most serious health threats in the world, generating calls for action from the World Health Organization [1] and the Centers for Disease Control and Prevention [2]. Antimicrobial resistance has been postulated to be responsible for about 30% of deaths from neonatal sepsis worldwide, primarily-in low and middle-income countries [3], but there is growing realization that developed countries are not immune from this crisis [2]. When standard antibiotics are no longer effective, second- and third-line antibiotics are required, with potential delays in effective therapy, increased morbidity and mortality, adverse effects and costs. Antimicrobial stewardship is a co-ordinated program that promotes the appropriate use of antimicrobials (including antibiotics), improves patient outcomes, reduces microbial resistance, and decreases the spread of infections caused by multidrug-resistant organisms [4]. The Infectious Diseases Society of America, the Pediatric Infectious Disease Society, and the Society for Healthcare

Epidemiology of America published guidelines in 2007 for developing an institutional program to enhance antimicrobial stewardship, endorsed by the American Academy of Pediatrics [4]. A small number of studies evaluating formal antibiotic stewardship programs (ASPs) in pediatric patients have demonstrated reductions in antibiotic utilization, cost, and prescribing errors [5]. Even though antibiotics are among the most widely prescribed medications in hospitalized neonates in industrialized nations [6,7], there is little published data on formal ASPs specifically addressing the neonatal population [8]. Many pregnant women are also exposed to antibiotics [9]. Evidence is mounting that antibiotic treatment of mothers in pregnancy may have immediate and long-term consequences for neonates. This article reviews some of the controversies and opportunities for antibiotic stewardship in the neonataleperinatal period and suggests a framework for setting up neonate-specific ASPs. This review does not address antibiotic stewardship related to neonatal sepsis in resource-constrained nations, which is also critical, but the epidemiology, resistance patterns and clinical conditions pose a very different set of challenges [3]. 2. Antibiotic use in pregnancy

* Corresponding author. Division of Neonatal Perinatal Medicine, MedStar Georgetown University Hospital, 3800 Reservoir Rd, NW, Suite M3400, Washington DC 20007, USA. E-mail address: [email protected] (J. Ramasethu). http://dx.doi.org/10.1016/j.siny.2017.07.001 1744-165X/© 2017 Elsevier Ltd. All rights reserved.

It is estimated that 40% of pregnant women in the USA receive antibiotics prior to delivery to prevent infectious complications in both the mother and fetus [9]. Given the 10e30% incidence of maternal colonization with group B streptococcus (GBS) and with

J. Ramasethu, T. Kawakita / Seminars in Fetal & Neonatal Medicine 22 (2017) 278e283

33% of women undergoing cesarean delivery, between 1.5 and 2.0 million women and their infants are exposed to antibiotics annually in the USA. In addition, antibiotics are prescribed for the treatment of asymptomatic bacteriuria, which complicates 2e10% of pregnancies, pyelonephritis, in the context of preterm premature rupture of membranes (PPROM) [10], and for the treatment of chorioamnionitis in labor [11]. The Centers for Disease Control and Prevention (CDC) 2010 revised guidelines recommend universal screening for GBS between 35 and 37 weeks gestation [12]. Intrapartum antibiotic prophylaxis is indicated when culture-based screening is positive for GBS, with a history of a previous infant with invasive GBS disease, maternal GBS bacteriuria (104 colony-forming units), or other clinical factors such as preterm delivery (<37 weeks gestation), rupture of membranes 18 h, intrapartum fever (temperature 38.0  C), and intrapartum nucleic acid amplification test positive for GBS. Implementation of national guidelines for intrapartum antibiotic prophylaxis in the early 1990s has resulted in an 80% reduction of early-onset GBS neonatal sepsis, from 1.7 cases to fewer than 0.4 cases per 1000 live births in the early 2000s [12]. Penicillin is the agent of choice for intrapartum prophylaxis, though ampicillin is an acceptable alternative. Specific antibiotic recommendations have been made for patients with penicillin allergies, depending on the severity of the allergy and isolate susceptibility, but the alternative agents given to penicillin-allergic women often do not comply with CDC recommendations [13]. Unfortunately, the use of medications other than penicillin, ampicillin or cefazolin is not considered to be adequate prophylaxis, and this may in turn lead to additional evaluation and intervention in the newborn. Women who have cesarean deliveries, particularly those who have been in labor, or after membrane rupture, are at increased risk of postpartum infection, including wound infection and endometritis, compared to women delivering vaginally. Antibiotic prophylaxis given after cord clamping, to avoid neonatal exposure to antibiotics has been supplanted by pre-incision prophylaxis, similar to prophylaxis for other surgical procedures, again increasing the exposure to antibiotics to the neonate, albeit briefly [14]. Chorioamnionitis or acute inflammation of the chorion and amnion layers of the membranes is estimated to occur in 3e10% of deliveries at term but clinical diagnosis is subjective and variable, leading to antibiotic treatment in the mother, and, invariably, evaluation and antibiotic treatment of the newborn [11].

3. Antibiotic use in neonates Suspected sepsis is the most usual working diagnosis for infants admitted to neonatal intensive care units (NICUs) [15]. Ampicillin and gentamicin have led the list of 10 drugs most widely used in neonatal units in the USA since 1996 [6,7]. Symptoms of neonatal sepsis range from subtle to severe, but are non-specific and may overlap with many non-infectious clinical conditions in neonates. Given the high risk of morbidity and mortality with neonatal sepsis, and the poor positive predictive value of ancillary laboratory tests [16,17], empirical antibiotic treatment is often initiated, targeting the most likely micro-organisms, based on the clinical situation, but antibiotic use is extremely variable across institutions. A recent retrospective cohort study of more than 50,000 infants in 127 NICUs across California in 2013 showed that overall antibiotic use varied 40-fold, from 2.4% to 97.1% of patient days [18]. Antibiotic use was independent of rates of proven infection, surgical volume, necrotizing enterocolitis or mortality. Most of the intermediate-

279

level NICUs reported zero rates of culture-proven infection, and yet 50% were in the highest antibiotic use quartile.

4. Adverse effects of antibiotics in neonates Whereas the benefits of antibiotics cannot be denied, there are real and potential risks of the prevalent exposure of the maternalefetal dyad to antibiotics. The rate of GBS sepsis has decreased from 1.7 to 0.4 per 1000 births with universal screening and GBS prophylaxis, but this results in almost 700 women (and their unborn infants) being exposed to antibiotics to prevent one case of early onset GBS disease. Concerns have been raised about the potential for development of antibiotic-resistant neonatal sepsis with widespread maternal GBS prophylaxis, but in the absence of an effective vaccine to prevent GBS disease or a rapid, sensitive diagnostic test to detect GBS colonization when women are admitted in labor or with rupture of membranes, antibiotic prophylaxis is likely to be continued. No increase has been noted so far in rates of ampicillin-resistant infections in term infants, but in very low birth weight infants (VLBW, birth weight <1500 g), some studies have noted an increase in rates of E. coli infections [19], whereas others have shown no change in the rate [20]. However, an association has been noted between intrapartum ampicillin exposure and ampicillin-resistant E. coli infection at birth, ranging from 55% to 85% [19,20]. Antibiotics are used to prolong latency and reduce infectious complications in women with PPROM, and a combination of intravenous ampicillin and erythromycin is usually recommended, in order to cover GBS, Gram-negative infections, mycoplasma, and ureaplasma [10]. In the ORACLE 1 study, exposure to co-amoxiclav (a combination of amoxicillin and clavulinic acid) during pregnancy was associated with a four-fold risk of necrotizing enterocolitis in the newborn compared to placebo (relative risk: 4.72; 95% confidence interval: 1.57e14.23) [10]; the pathogenesis of this is uncertain but could be related to abnormal colonization of the intestinal tract. There is now evidence that the intrauterine environment is not sterile, and there is an active maternalefetal exchange of commensal micro-organisms, establishing the fetal intestinal microbiome well before delivery [21,22]. Antibiotics administered before delivery may disrupt the normal colonization of the developing fetal intestinal microbiome with long-lasting effects. Exposure of mothers to antibiotics in the second and third trimester has been associated with increases in childhood obesity [23], and in asthma [24]. Although transfer of macrolides across the placenta is estimated to be small, Cho et al. showed that administration of subtherapeutic antibiotic therapy increased adiposity in young mice with changes in the microbiome and increased hormones related to metabolism, which may explain some of these associations [25]. Perinatal and early empiric antibiotic use has been associated with lower bacterial diversity in the developing microbiome of the neonate, and an increase in colonization with potentially pathogenic Enterobacteriaceae, which may precede bloodstream infection in preterm infants [26]. Prolonged initial antibiotic therapy in VLBW infants has been associated with increased risks of necrotizing enterocolitis (NEC) and death [27,28], with each additional empirical day of treatment associated with measurable increase in risk [27]. In a large study of 11,669 VLBW infants without cultureproven sepsis or NEC in the Canadian Neonatal network, a 10% increase in antibiotic use rate (AUR) was associated with increased odds of mortality, and major morbidity (chronic lung disease, persistent periventricular echogenicity or echolucency, or stage 3

280

J. Ramasethu, T. Kawakita / Seminars in Fetal & Neonatal Medicine 22 (2017) 278e283

retinopathy of prematurity) [28]. It is unclear whether these outcomes were simply related to severity of illness or secondary to antibiotic therapy. In addition, prolonged treatment with broadspectrum antibiotics in the neonatal period has been associated with colonization of infants with antimicrobial-resistant Gramnegative bacteria at the time of discharge from the NICU, with the potential for dissemination of resistant Gram-negative bacteria from colonized infants to other medical facilities or to the community [29]. 5. Opportunities for antibiotic stewardship in neonatal care There are certain clinical paradigms where the routine use of antibiotics and choice of antibiotics is being re-examined: (i) empirical antibiotics for suspected sepsis in term and late preterm infants born to mothers with chorioamnionitis; (ii) continued antibiotic use in the NICU for culture-negative sepsis; (iii) choice of antibiotics for late onset sepsis. 6. Chorioamnionitis and empiric therapy for suspected sepsis Chorioamnionitis has long been recognized as a significant risk factor for early onset sepsis (EOS). The Committee of Fetus and Newborn (COFN) of the American Academy of Pediatrics (AAP) published guidelines in 2012 for “Management of neonates with suspected or proven EOS” [30]. This included recommendations for laboratory evaluation and treatment for babies born to mothers with suspected chorioamnionitis, including empiric treatment with broad-spectrum antibiotics even in asymptomatic infants, with the direction to discontinue antibiotics after 48 h if the infant remains asymptomatic, blood culture is negative and the laboratory results (complete blood count ± C-reactive protein) are normal. The COFN report also recommended treating asymptomatic neonates with prolonged antibiotics despite a sterile blood culture if the mother received intrapartum antibiotics and the laboratory tests were abnormal. A recent survey of 81 nurseries in 31 states across the USA confirmed that obstetric diagnosis of chorioamnionitis was the factor most frequently used to identify risk for EOS [31]. Among well-appearing infants born to mothers with concern for chorioamnionitis, 65% of sites used the AAP or CDC guidelines to determine clinical care, 14% used a published clinical risk calculator, and two sites used clinical observation alone. The complete blood count and C-reactive protein were the most frequently obtained laboratory tests (94.8% and 36.4% respectively), and influenced duration of empirical antibiotics in 13% of sites. Other industrialized nations also have published national guidelines which vary in their recommendations for the initiation and continuation of antibiotics for term and late preterm infants at risk for EOS, again with considerable variation in practice among clinicians surveyed [32]. The recommendation for laboratory tests, blood cultures and empirical antibiotic therapy in this population of newborns is being questioned, particularly since laboratory tests such as the complete blood count and C-reactive protein concentrations have poor positive predictive value for the diagnosis of early neonatal sepsis [16,17]. Kiser et al. showed that adherence to guidelines similar to that proposed by the COFN resulted in 24% of infants evaluated for EOS being treated with prolonged antibiotics, with 20% treated solely on the basis of abnormal laboratory data [33]. Rather than relying on a broad and non-specific diagnosis of chorioamnionitis, Puopolo et al. developed a prediction model for EOS using details of the baby's gestational age, duration of rupture of membranes, GBS carriage status, highest maternal intrapartum

temperature and the nature and timing of intrapartum antibiotic administration to determine the risk of EOS in term and late preterm infants at the time of birth [34]. Using a Bayesian approach, the model was refined to include the infant's clinical examination to further categorize the risk, and an online calculator was developed and validated. It was estimated that judicious use of the calculator could reduce antibiotic use in 80,000 to 240,000 neonates in the USA each year [35]. Use of the EOS calculator in a large integrated healthcare system in California showed that laboratory evaluation for sepsis (blood culture use) decreased from 14.5% to 4.9%, and empirical antibiotic use in the first 24 h decreased from 5.0% to 2.6%, without increasing between 24 and 72 h of age, and without increases in readmissions for EOS within 7 days of birth [36]. The authors emphasized that “no antibiotic treatment is not the same as no care,” and that close clinical observation, increased monitoring of at-risk infants and parental education is required to ensure success in decreasing empiric antibiotic use without adverse effects. 7. Asymptomatic infant with EOS Concerns have been raised about delayed antibiotic treatment in infants who are initially asymptomatic, since this may result in poor outcomes. The incidence of bacteremia in asymptomatic infants is extremely low but not absent. Although the majority of infants with culture-confirmed EOS born to mothers with chorioamnionitis are usually symptomatic within 6 h of birth, there are reports of infants who were initially asymptomatic, including some who remained asymptomatic up to 72 h, despite positive blood cultures. Organisms isolated from these infants included GBS, Enterococcus faecalis, Hemophilus influenza, Staphylococcus aureus, and E. coli [36,37]. Some of these infants may have had ‘transient bacteremia’ since they remained asymptomatic and repeat blood cultures prior to the initiation of antibiotic therapy were negative [36]. It is estimated that if the overall prevalence of chorioamnionitis is 3e10% in term pregnancies, 450e1500-well appearing infants would be evaluated and treated for each initially asymptomatic infant with early onset infection [37]. Given the widespread exposure to antibiotics in essentially uninfected neonates, the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), American College of Obstetricians and Gynecologists (ACOG), Society for MaternaleFetal Medicine (SMFM), and AAP organized a workshop in 2015 to review the data and provide updated guidelines for the diagnosis and management of women and newborns following a maternal diagnosis of chorioamnionitis [11]. The workshop proposed changing the term chorioamnionitis to “intrauterine inflammation or infection, or both,” abbreviated as “Triple I,” in order to better characterize the heterogeneous conditions. Revised terminology and diagnostic criteria were proposed as well as changes in the management of newborns of mothers with suspected intrauterine infection, most notably a proposal to consider close observation of asymptomatic infants greater than 34 weeks gestation born to mothers with suspected (not confirmed) intraamniotic infection rather than initiation of empiric antibiotic therapy [11,38]. 8. Decreasing antibiotic use in the NICU The other group of infants that receives a large proportion of antibiotics are those admitted to the NICU, particularly preterm or VLBW infants. Empiric antibiotic treatment within the first 3 days

J. Ramasethu, T. Kawakita / Seminars in Fetal & Neonatal Medicine 22 (2017) 278e283

of life was administered to 96% of 4039 infants weighing between 401 and 1000 g at birth, cared for in 19 neonatal centers in the USA between 1998 and 2001 [27]. In the Canadian Neonatal Network, 85% of 13,738 VLBW infants admitted to NICUs between 2010 and 2014 received antibiotics during their hospital course [28]. Given the higher prevalence of early and late onset sepsis in preterm/ VLBW infants [19,20] associated with increased mortality and morbidity, and the difficulty in differentiating clinical signs of sepsis from non-infectious symptoms, empirical antibiotic initiation in this population is explicable, but the variation in duration of treatment appears to be arbitrary and institution-based, since there appears to be no correlation between duration of antibiotic use and degree of illness or positive cultures [18,39]. It has been a standard practice to treat all neonates with respiratory distress at birth with antibiotics for suspected GBS sepsis/pneumonia. Some studies have indicated that it may be possible to avoid antibiotics in term/late preterm infants with transient tachypnea of the newborn by paying attention to the risk factors for infection [40]. Similarly, preterm VLBW infants born by cesarean section solely for maternal indications such as pre-eclampsia, without preterm labor or premature rupture of membranes, are at markedly reduced risk of EOS, and consideration may be given to not initiating antibiotics or for discontinuing antibiotics early [41]. Reasons for extending antibiotic treatment include concerns about the reliability of blood cultures when blood volumes are small, or due to intrapartum antibiotics. Automated blood culture systems with optimized enriched broths identify most (>94%) micro-organisms within 48 h of incubation, even with low colony counts [42,43], and maternal intrapartum antibiotic treatment does not appear to prolong the time to positivity of neonatal blood cultures in infants with EOS [41]. Nevertheless, the clinician faced with a neonate with signs suggestive of sepsis is generally reluctant to discontinue antibiotics without some degree of certainty, which can only be done if evaluation is thorough and complete. Diagnostic evaluation of infants with sepsis should include obtaining adequate volume of blood for cultures, and from two separate sites if late onset sepsis (LOS) is suspected, to ensure that the bacterial growth represents a true infection rather than a contaminant. Ruling out meningitis by performing a lumbar puncture before initiating antibiotics would help in reducing the duration of antibiotic therapy in both early and late onset sepsis, but is not always done. Blood and cerebrospinal fluid cultures may not be positive in infants who have urinary tract infections or in infants with pneumonia. The appropriate duration of treatment for neonates with suspected pneumonia and negative blood cultures is unclear, although some authors have suggested a shortened five-day course of treatment [8].

9. Choice of antibiotics Standard empiric treatment for suspected EOS in the USA has been a combination of ampicillin and gentamicin, to cover the most common organisms, GBS and E. coli, acquired vertically from the mother in labor. E. coli is now the most usual cause of early onset sepsis in preterm neonates, and E. coli infection is associated with high mortality and increased morbidity among the survivors [19,20]. Although increasing ampicillin resistance has been noted in E. coli infections, in a study of 258 neonatal E. coli bloodstream infections, including 123 ampicillin resistant isolates, the unadjusted 30-day mortality rate did not differ between infants with ampicillin-resistant and ampicillin-susceptible E. coli infections; moreover, among a subset of infants with ampicillin-resistant

281

E. coli, appropriate empirical antibiotic therapy was not associated with lower mortality [44]. This suggests that changing empiric therapy to broad-spectrum antibiotic treatment to cover possible resistant E. coli infection may simply incur the risk of developing more drug-resistant organisms, without immediate benefit. Vancomycin and gentamicin have been the drugs of choice for suspected LOS in the USA, since coagulase-negative staphylococci (CONS) accounted for the majority (48e55%) of LOS in VLBW infants for more than two decades, and Gram-negative bacilli were responsible for 16e18% of infections [45]. The use of broadspectrum cephalosporins has decreased in the last decade, with the realization that they contribute to antibiotic resistance and invasive candidiasis [7]. NICUs have attempted decreasing empiric vancomycin use, by using empirical oxacillin instead of vancomycin for suspected LOS and by vancomycin audits and restriction guidelines [8,46]. There is no evidence that a delay in vancomycin therapy increases mortality in infants with CONS sepsis, but inadequate empirical therapy for methicillin-resistant Staphylococcus aureus infection has been associated with increased mortality, so the judicious selection of initial antibiotics remains critical [47]. Decreasing rates of central-line-associated bloodstream infections with infection prevention measures will result in lower rates of CONS infections, and may have a further impact on antibiotic utilization rates overall, and on vancomycin use in particular [48]. A combination of antibiotics is normally used in critically ill infants with necrotizing enterocolitis (NEC) or complicated intraabdominal infections, where polymicrobial infection with aerobic and anaerobic micro-organisms is probable, but again there is significant variability in the choice of antibiotics and duration of treatment. Results of the ongoing Phase 2/3 study (SCAMP study, NCT 01994993) of different antibiotic regimens for complicated intra-abdominal infections in infants may help guide future therapy. 10. Antibiotic stewardship program for neonates Box 1 shows a framework for implementing antibiotic stewardship for neonates, based on the recommendations of the CDC and the guidelines published by the Infectious Diseases Society of America, the Pediatric Infectious Disease Society, and the Society for Healthcare Epidemiology of America [2,4]. In some hospitals, babies born to mothers with chorioamnionitis may not be admitted to the NICU, but receive antibiotics in the postnatal wards, so this framework is not restricted to NICU patients. Composition of the team may vary between institutions. A physician lead is essential to spearhead the program, but this is an interdisciplinary and collaborative effort. Pediatric infectious disease specialists provide vital consultations in determining appropriate therapy, particularly when cultures are positive and antibiotic resistance is detected. Pediatric pharmacologists consider pharmacokinetics to ensure that drug dosage is appropriate for optimum therapy without adverse effects. Support from hospital administration is necessary to provide human, financial and information technology resources. Strategies for antibiotic stewardship may also vary between institutions. Evaluation of antibiotic practices within a NICU can inform antibiotic stewardship targets tailored to the individual center [4]. Rather than an all-or-none approach, implementation of specific action items, such as systematic evaluation of the need for continued treatment after 48 h, would help to focus efforts to improve compliance and gradually bring about “institutional culture change.” Electronic hard stops may be set to automatically

282

J. Ramasethu, T. Kawakita / Seminars in Fetal & Neonatal Medicine 22 (2017) 278e283

Box 1 Antibiotic stewardship program for neonates.

The Team Neonatologist Pediatric infectious diseases physician Pediatric pharmacist Infection preventionist Neonatal nurse educator Microbiologist Informatics specialist The Task Optimize antibiotic selection Decrease unnecessary antibiotic use Improve patient safety and outcomes Decrease cost The Strategies Infection prevention measures Discussion and dissemination of best practices/ guidelines e Indication for antibiotics e Choice of antibiotics for empiric therapy e Choice of antibiotics for confirmed infection e Duration of antibiotic treatment for culture-negative and -positive sepsis e Interpretation of colonization versus true infection Formulary restriction/pre-authorization for selected antimicrobials Prospective audit and feedback Therapeutic monitoring The Measurements Compliance with guidelines e Right drug, right dose, right duration Tracking trends e Days of therapy: total days of antibiotics adjusted to 1000 patient-days e Antibiotic-free days Microbiological data e Antimicrobial resistance patterns Incidence of late onset sepsis/necrotizing enterocolitis/ mortality Cost analysis

Antibiotic use rate (AUR) is the total number of patient-days on which infants are exposed to antibiotics (or antifungals) administered intravenously or intramuscularly per 100 patient-days, expressed as a percentage. In NICUs with a significant population of small premature infants with prolonged lengths of stay, AUR may be low, but measuring usage trends indexed to the patient population would still provide valuable information. It is also important to ensure the safety of reduction of antibiotic usage by monitoring relapse or readmission rates. The importance of infection control measures to prevent CLABSI and other LOS cannot be over-emphasized [49]. Reduction of LOS would be associated with commensurate reduction in antibiotic use, but only if empiric antibiotics are discontinued in a timely fashion when cultures are negative. It is interesting to note that a prospective longitudinal study of neonatal infections and antibiotic use throughout 25 years in a tertiary NICU showed that emergence of cephalosporin-resistant Gram-negative bacterial infection was not prevented by responsible antibiotic use, indicating that the relationship between antimicrobial use and drug resistance is complex and that other factors may be involved [50]. Nevertheless, antibiotic stewardship is vital to prevent the catastrophic consequences of resistant infections now being experienced in many countries around the world [3]. Practice points  Widespread antibiotic use in neonates may be associated with development of antibiotic resistance, and through the effect on the microbiome, long-standing immunological and metabolic consequences.  Antibiotic use in neonates should be based on evidence-based guidelines, and monitored, with focus on reducing indiscriminate use.  Prevention of infection is vital to reduction of antibiotic use.

Research directions  An effective vaccine for GBS in pregnant women will reduce the use of antibiotic prophylaxis in 30% of women at delivery.  Improved diagnostic tests to rule out bacterial infections or confirm viral infections will reduce antibiotic use in culturenegative sepsis. Conflict of interest statement

discontinue empirical antibiotic therapy after 48 h unless reinstituted by the physician, but run the risk of inadvertent discontinuation of necessary treatment in some cases [8]. Antibiotic restriction guidelines, requiring pre-authorization for select antibiotics, have been used to prevent indiscriminate use of broadspectrum antibiotics. Education, dissemination of evidence-based guidelines, audits, regular reporting of antibiotic use metrics, and resistance patterns maintain the momentum [46]. Additionally, participation in a multicenter collaborative such as the Vermont Oxford Network “iNICQ 2017: Choosing Antibiotics Wisely” offers the opportunity to share tools, standardized protocols, progress, and lessons learned [48]. There are standard metrics used to measure antibiotic usage, but some have shortcomings when applied to the neonatal population. Days of therapy (DOT) is calculated as the aggregate sum of antibiotics used per patient per day, per 1000 patient-days. A neonate receiving ampicillin and gentamicin daily for 2 days would be measured as 4 DOT (2 DOT for each day). DOT may be applied to all or to select antibiotics, and measured monthly to monitor trends.

None declared. Funding sources None. Acknowledgment We would like to thank Dr Helain Landy, Chair, Department of Obstetrics and Gynecology, MedStar Georgetown University Hospital, for her assistance with the section on antibiotic use in pregnancy. References [1] World Health Organization. Global action plan on antimicrobial resistance. 2015. http://www.who.int/antimicrobial-resistance/global-action-plan/en/. [2] Centers for Disease Control [https://www.cdc.gov/drugresistance/]. [3] Laxminarayan R, Matsoso P, Pant S. Access to effective antimicrobials: a worldwide challenge. Lancet 2016;387:168e75.

J. Ramasethu, T. Kawakita / Seminars in Fetal & Neonatal Medicine 22 (2017) 278e283 [4] * Dellit TH, Owens RC, McGowan Jr JE, et al. Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America guidelines for developing an institutional program to enhance antimicrobial stewardship Clin Infect Dis 2007;44:159e77. [5] Smith MJ, Gerber JS, Hersh AL. Inpatient antimicrobial stewardship in pediatrics: a systematic review. J Pediatr Infect Dis Soc 2015;4:e127e35. [6] Clark RH, Bloom BT, Spitzer AR, Gerstmann DR. Reported medication use in the neonatal intensive care unit: data from a large national data set. Pediatrics 2006;117:1979e87. [7] Hsieh EM, Hornik CP, Clark RH, et al. Medication use in the neonatal intensive care unit. Am J Perinatol 2014;31:811e21. [8] Cantey JB, Wozniak PS, Pruszynski JE, Sanchez PJ. Reducing unnecessary antibiotic use in the neonatal intensive care unit (SCOUT): a prospective interrupted time-series study. Lancet Infect Dis 2016;16:1178e84. [9] * Ledger WJ, Blaser MJ. Are we using too many antibiotics during pregnancy?. Br J Obstet Gynaecol 2013;120:1450e2. [10] Kenyon S, Boulvain M, Neilson JP. Antibiotics for preterm rupture of membranes. Cochrane Database Syst Rev 2013;12:CD001058. [11] * Higgins RD, Saade G, Polin RA, et al. Chorioamnionitis Workshop Participants. Evaluation and management of women and newborns with a maternal diagnosis of chorioamnionitis: summary of a workshop. Obstet Gynecol 2016;127:426e36. [12] Verani JR, McGee L, Schrag SJ. Division of bacterial diseases, national center for immunization and respiratory diseases, centers for disease control and prevention (CDC). Prevention of perinatal group B streptococcal disease e revised guidelines from CDC. MMWR Recomm Rep 2010;2010(59):1e36. [13] Briody VA, Albright CM, Has P, Hughes BL. Use of cefazolin for group B streptococci prophylaxis in women reporting a penicillin allergy without anaphylaxis. Obstet Gynecol 2016;127:577e83. [14] American College of Obstetrics and Gynecology Committee opinion no. 465: antimicrobial prophylaxis for cesarean delivery: timing of administration. Obstet Gynecol 2010;116:791e2. [15] Spitzer AR, Kirkby S, Kornhauser M. Practice variation in suspected neonatal sepsis: a costly problem in neonatal intensive care. J Perinatol 2005;25: 265e9. [16] Newman TB, Puopolo KM, Wi S, Draper D, Escobar GJ. Interpreting complete blood counts soon after birth in newborns at risk for sepsis. Pediatrics 2010;126:903e9. [17] Benitz WE, Han MY, Madan A, Ramachandra P. Serial serum C-reactive protein levels in the diagnosis of neonatal infection. Pediatrics 1998;102:E41. [18] Schulman J, Dimand RJ, Lee HC, et al. Neonatal intensive care unit antibiotic use. Pediatrics 2015;135:826e33. [19] Stoll BJ, Hansen N, Fanaroff AA, et al. Changes in pathogens causing earlyonset sepsis in very-low-birth-weight infants. N Engl J Med 2002;347:240e7. [20] Schrag SJ, Farley MM, Petit S, et al. Epidemiology of invasive early-onset neonatal sepsis, 2005 to 2014. Pediatrics 2016;138:e20162013. [21] Romano-Keeler J, Weitkamp JH. Maternal influences on fetal microbial colonization and immune development. Pediatr Res 2015;77:189e95. [22] * Neu J. The microbiome during pregnancy and early postnatal life. Semin Fetal Neonatal Med 2016;21:373e9. [23] Mueller NT, Whyatt R, Hoepner L, et al. Prenatal exposure to antibiotics, cesarean section and risk of childhood obesity. Int J Obes (Lond) 2015;39: 665e70. [24] Stensballe LG, Simonsen J, Jensen SM, Bønnelykke K, Bisgaard H. Use of antibiotics during pregnancy increases the risk of asthma in early childhood. J Pediatr 2013;162:832e8. [25] Cho I, Yamanishi S, Cox L, et al. Antibiotics in early life alter the murine colonic microbiome and adiposity. Nature 2012;488:621e6. [26] Greenwood C, Morrow AL, Lagomarcino AJ, et al. Early empiric antibiotic use in preterm infants is associated with lower bacterial diversity and higher relative abundance of enterobacter. J Pediatr 2014;165:23e9. [27] * Cotten CM, Taylor S, Stoll B, et al. NICHD Neonatal Research Network. Prolonged duration of initial empirical antibiotic treatment is associated with increased rates of necrotizing enterocolitis and death for extremely low birth weight infants. Pediatrics 2009;123:58e66. [28] * Ting JY, Synnes A, Roberts A, et al. Canadian Neonatal Network

*

References marked with an asterisk are important references.

[29]

[30]

[31] [32]

[33] [34]

[35] [36]

[37] [38] [39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48] [49]

[50]

283

Investigators. Association between antibiotic use and neonatal mortality and morbidities in very low-birth-weight infants without culture-proven sepsis or necrotizing enterocolitis. JAMA Pediatr 2016;170:1181e7. Clock SA, Ferng YH, Tabibi S, et al. Colonization with antimicrobial-resistant Gram-negative bacilli at neonatal intensive care unit discharge. J Pediatr Infect Dis Soc 2016 Mar 28. pii: piw014 [Epub ahead of print]. Polin RA. Committee on Fetus and Newborn. Management of neonates with suspected or proven early-onset bacterial sepsis. Pediatrics 2012;129: 1006e15. Mukhopadhyay S, Taylor JA, Van Kohorn I, et al. Variation in sepsis evaluation across a national network of nurseries. Pediatrics 2017;139:e20162845. Van Herk W, el Halou S, Janota J, et al. Variation in current management of term and late preterm neonates at risk for early onset sepsis. Pediatr Infect Dis J 2016;35:494e500. Kiser C, Nawab U, McKenna K, Aghai ZH. Role of guidelines on length of therapy in chorioamnionitis and neonatal sepsis. Pediatrics 2014;133:992e8. Puopolo KM, Draper D, Wi S, et al. Estimating the probability of neonatal early-onset infection on the basis of maternal risk factors. Pediatrics 2011;128:e1155e63. * Escobar GJ, Puopolo KM, Wi S, et al. Stratification of risk of early-onset sepsis in newborns at 34 weeks' gestation. Pediatrics 2014;133:30e6. * Kuzniewicz MW, Puopolo KM, Fischer A, et al. A quantitative, risk-based approach to the management of neonatal early-onset sepsis. JAMA Pediatr 2017;171:365e71. Wortham JM, Hansen NI, Schrag SJ, et al. Chorioamnionitis and cultureconfirmed, early-onset neonatal infections. Pediatrics 2016;137:e20152323. * Randis TM, Polin RA, Saade G. Chorioamnionitis: time for a new approach. Curr Opin Pediatr 2017;29:159e64. Cordero L, Ayers LW. Duration of empiric antibiotics for suspected earlyonset sepsis in extremely low birth weight infants. Infect Control Hosp Epidemiol 2003;24:662e6. Weintraub AS, Cadet CT, Perez R, DeLorenzo E, Holzman IR, Stroustrup A. Antibiotic use in newborns with transient tachypnea of the newborn. Neonatology 2013;103:235e40. Mukhopadhyay S, Puopolo KM. Clinical and microbiologic characteristics of early-onset sepsis among VLBW infants: opportunities for antibiotic stewardship. Pediatr Infect Dis J 2017;36:477e81. Garcia-Prats JA, Cooper TR, Schneider VF, Stager CE, Hansen TN. Rapid detection of microorganisms in blood cultures utilizing an automated blood culture system. Pediatrics 2000;105:523e7. Lancaster DP, Friedman DF, Chiotos K, Sullivan KV. Blood volume required for detection of low levels and ultralow levels of organisms responsible for neonatal bacteremia by use of Bactec Peds Plus/F, Plus Aerobic/F medium, and the BD Bactec FX system: an in vitro study. J Clin Microbiol 2015;53: 3609e13. Bergin SP, Thaden JT, Ericson JE, et al. Neonatal Escherichia coli bloodstream infections: clinical outcomes and impact of initial antibiotic therapy. Pediatr Infect Dis J 2015;34:933e6. Boghossian NS, Page GP, Bell EF, et al. Late-onset sepsis in very low birth weight infants from singleton and multiple gestation births. J Pediatr 2013;162:1120e4. Holzmann-Pazgal G, Khan AM, Northrup TF, et al. Decreasing vancomycin utilization in a neonatal intensive care unit. Am J Infect Control 2015;43: 1255e7. Thaden JT, Ericson JE, Cross H, et al. Survival benefit of empirical therapy for staphylococcus aureus bloodstream infections in infants. Pediatr Infect Dis J 2015;34:1175e9. Vermont Oxford Network [https://public.vtoxford.org/quality-education/ inicq-2017-choosing-antibiotics-wisely/]. Bizzarro MJ, Shabanova V, Baltimore RS, Dembry LM, Ehrenkrantz RA, Gallagher PG. Neonatal sepsis 2014e2013: the rise and fall of coagulasenegative staphylococci. J Pediatr 2015;166:1193e9. Carr D, Barnes EH, Gordon A, Isaacs D. Effect of antibiotic use on antimicrobial antibiotic resistance and late-onset neonatal infections over 25 years in an Australian tertiary neonatal unit. Archs Dis Childh Fetal Neonatal Ed 2017;102:F244e50.