Chorioamnionitis

Maternal fever during labor, and perhaps other signs and symptoms of chorioamnionitis, often results in a call to the family practitioner, pediatrician, or neonatologist related to concern for the neonate. This communication often causes an evaluation to rule out early-onset neonatal sepsis.[15] Because of a concern for early-onset sepsis (EOS) when signs and symptoms of maternal chorioamnionitis occur, 18-38 newborns are evaluated and treated with antibiotics for every infant with proven bacteremia. The reason for this clinical phenomenon is that newborns who develop EOS, defined as proven infection (positive culture from a normally sterile site like blood, tracheal aspirate, cerebrospinal fluid) at less than 72 hours of life, have a high mortality rate. A strong association is observed between very preterm infants dying when younger than 24 hours and chorioamnionitis.[16, 17]

Heightened clinical evaluations for EOS began in the 1970s because group B streptococcal (GBS) infections resulted in a neonatal mortality of about 50%.[18] Over the past 50 years, awareness of GBS-related neonatal morbidity and mortality resulted in the widespread implementation of intrapartum chemoprophylaxis with antibiotics to reduce the risk of GBS disease, which led to an 85% reduction in the rate of culture-proven early-onset GBS sepsis, from approximately1.8 per 1000 live births in the early 1990s to fewer than 0.26 per 1000 live births in 2010.[19]

Early-onset bacterial infections in the newborn may occur when the mother has abnormal bacterial colonization of the urogenital tract, an ascending but silent amniotic fluid infection, or symptomatic chorioamnionitis. Thus, the physician cannot assume that maternal signs and symptoms alone will identify all infected infants.

GBS infections continues to be the major cause of EOS in term neonates; however, Escherichia coli has surpassed GBS as the most significant pathogen in preterm infants for over 10 years.[20] Intrapartum ampicillin exposure (as part of GBS prophylaxis as used at some institutions) was identified as an independent risk factor for ampicillin-resistant E coli EOS, as well as for a significant increase in E coli late-onset sepsis.[21]

Additionally, methicillin-resistant Staphylococcus aureus (MRSA), already a common cause of nosocomial infection in maternity and neonatal units, looms as a potential cause of EOS.[22] So far, maternal colonization during pregnancy with MRSA has not translated into an increase in MRSA-associated EOS, but close monitoring for this infection is warranted.[23]

This article discusses intraamniotic infection during pregnancy and its effects on the fetus and newborn, as well as summarizes the history, physical examination, and laboratory findings in both mother and infant to provide appropriate decision-making tools for cost-effective management of the neonate. The subject is expansive in scope, and readers are encouraged to seek more information from other sources. Other Medscape Drugs and Diseases articles of interest include Congenital Pneumonia; Meningitis, Bacterial; and Neonatal Sepsis.

An entire 2016 issue of the Journal of Perinatal Medicine was devoted to clinical chorioamnionitis.[24]  Several chapters in the monograph by Romero et al contain information on the intraamniotic inflammatory response in women with clinical chorioamnionitis, molecular mechanisms to identify infecting microorganisms, and the cytokine profiles of the mother and the newborn infant.[25, 26, 27, 28, 29]

Readers are also referred to the 2017 Committee Opinion Number 712 by the American College of Obstetrics and Gynecology (ACOG) on intrapartum management of intraamniotic infection,[30] as well as an excellent 2016 review article about clinical chorioamnionitis by Kim et al in the American Journal of Obstetrics and Gynecology that discusses the definition, pathogenesis, grading, staging, and clinical significance of the most common lesions in placental disease, accompanied by illustrations of the lesions as well as diagrams.[31]

Abnormal bacterial colonization of the distal colon during pregnancy may create an abnormal vaginal and cervical microbial environment.[32] Ascending of cervical and vaginal flora through the cervical canal is the most common pathway to chorioamnionitis. Uncommonly, chorioamnionitis may occur via hematogenous spread as a result of maternal bacteremia (eg, Listeria monocytogenes), or via contamination of the amniotic cavity as a result of an invasive procedure (eg, amniocentesis, fetoscopy). Although spread of peritoneal infection to the amniotic cavity via the fallopian tubes has also been suggested, it is very unlikely.[31] Subsequent activation of the maternal and fetal inflammatory response systems generally lead to labor and/or rupture of membranes.[33] More than 3 decades ago, rectovaginal colonization with group B Streptococcus (GBS) during pregnancy was found to be associated with GBS-related infection of the fetus or newborn.[18] Studies have demonstrated that other types of bacteria residing in the vagina, cervix, or both ascend through intact or ruptured fetal membranes and initiate amniotic fluid infection, chorioamnionitis (inflammation), or both.[34]

Urinary tract infection during pregnancy can bathe the vagina with bacterial pathogens and is a recognized risk factor for neonatal sepsis.[35]  This observation is particularly true for untreated asymptomatic GBS-related bacteriuria.[36]

Bacterial vaginosis is associated with premature labor, although overt infection of the neonate with microbes causing bacterial vaginosis is uncommon. Screening for and treatment of bacterial vaginosis and other genital infections may prevent preterm birth, especially if initiated before 20 weeks' gestation.[37]

Many associations related to infection and preterm birth have been made; however, the mechanisms of these relationships are not necessarily understood. These associations include periodontitis,[38]  blood types A and O,[39] alcoholism,[39]  and obesity during pregnancy.[40]

In the mid-trimester of pregnancy (14-24 weeks), ultrasonographic evidence of a short cervix may be the only clinical finding in intraamniotic fluid infection.[11] Cervical insufficiency, regardless of bacterial culture results from amniotic fluid, is associated with intraamniotic inflammation, preterm birth, and other adverse outcomes of pregnancy.[41, 42] Related issues to cervical insufficiency are mechanical methods of cervical ripening that are also suspected of increasing maternal and neonatal infections.[43] A Cochrane review stated that vaginal prostaglandin to initiate labor after premature rupture of membranes may increase maternal and fetal infection and warrants more research.[44] Each of these factors may be associated with altered host defenses that allow ascending infection from the urogenital tract to placental tissues and amniotic fluid.[45]

Maternal chorioamnionitis occurs when protective mechanisms of the urogenital tract and/or uterus fail during pregnancy or when increased numbers of microbial flora or highly pathogenic microorganisms are introduced into the urogenital environment.[45, 46, 47, 48]

Ascending infection into the vagina, then the cervix, and finally into the uterine cavity, fetal membranes, and placenta is the consequence of many factors (ie, innate host defenses, disrupted healthy bacterial flora, pathologic bacterial load, bacterial virulence factors, and toxin production). A short cervix has been recognized as either a risk factor or a surrogate for microbial invasion of the amniotic fluid.[11, 42, 49]

Urogenital hygiene is obviously important in establishing healthy bacterial flora. Healthy bacteria (ie, lactobacilli)[50]  and natural peptide antibiotics in the vagina and cervix may have roles in preventing infections during pregnancy.[51]  Mucus, phagocytes, and natural antibiotic proteins (ie, lactoferrin, lysozyme, beta defensins) in the cervicovaginal secretions attempt to maintain a normal bacterial flora.[46]  Bacterial interference, mainly produced via lactobacilli living in an acidic vaginal environment and producing bacteriocins, may help to keep pathogenic bacteria from gaining a foothold in the cervicovaginal secretions.[52] These mechanisms of host protection may be altered in a significant number of pregnant women who develop chorioamnionitis. The use of oral probiotics to alter vaginal flora and potentially reduce morbidities associated with intraamniotic infection has been studied extensively, but no clear cut benefits were realized.[53]

Oral hygiene may influence rectal and urogenital bacterial flora during pregnancy. Although the theory is controversial, intense interest has focused on a connection among periodontitis, abnormal rectal colonization, and preterm delivery,[54, 55] as well as whether treatment for periodontal disease during pregnancy decrease the incidence of preterm birth.[56] Orogenital contact may also alter either colonic or urogenital microbial flora and ultimately cause ascending infection and chorioamnionitis, as seen in some case reports.[57, 58]  

Currently, researchers are trying to understand how host defense mechanisms prevent urogenital infection during pregnancy. An intense area of research is the concept of bacterial communities living in the cervicovaginal area (microbiome) that are metabolically active to produce biochemicals (metalobome) that support their existence as well as prevent pathogenic bacteria from gaining access to the amniotic cavity and subsequently cause chorioamnionitis.[59, 60, 61] The prevalence and diversity of bacterial species in fetal membranes during preterm labor emphasizes that further research on this topic is needed.[62, 63] Metagenomics uses nonculture, molecular methods to delineate all microbes inhabiting an environment. Thus, the cervicovaginal and intestinal microbiome are under intense scrutiny to understand preterm labor, preterm premature rupture of membranes (PPROM), and chorioamnionitis relative to the mother, and necrotizing enterocolitis, sepsis, and neurologic injury relative to the newborn. Several published reports exist regarding using molecular methods to understand intrauterine infection, fetal inflammation, and preterm delivery.[61, 62, 64]

Clinical events associated with chorioamnionitis include the following:

• History of premature birth (with increasing risk at earlier gestational age)
• Presence of premature labor
• Preterm premature rupture of fetal membranes (before labor onset)
• Prolonged rupture of the fetal membranes (>18 hours)

A retrospective study (2012-2015) suggests that prolonged spontaneous active labor beyond the median not only significantly raises the risk of chorioamnionitis but also increases the odds of cesarean delivery.[65]

In a report of patients with clinical signs and symptoms of chorioamnionitis at term, and using both cultivation and molecular techniques of amniotic fluid, investigators noted almost 40% of women clinically diagnosed with chorioamnionitis did not have any evidence of bacteria in the amniotic cavity.[66] Additionally, nearly 50% did not have evidence of acute inflammatory lesions of the placenta (ie, histologic chorioamnionitis). Thus, other causes of signs and symptoms that resemble maternal chorioamnionitis must be sought.

Epidural anesthesia during labor is associated with maternal fever[67] and fetal tachycardia (see Special Concerns in the Diagnostic Considerations section). A sterile inflammatory response in the placenta and the fetus has been shown to be associated with epidural-related maternal fever.[68] Other conditions, such as dehydration or maternal exhaustion during labor, may result in maternal fever and must also be considered as causes of the febrile state.

United States data

The prevalence of maternal chorioamnionitis in the United States varies with different publications, but it appears to be inversely correlated with gestational age at birth. In a 2014 study that assessed the entire US population and linked infant birth and death certificate files for the year 2008, the prevalence of chorioamnionitis was 9.7 per 1000 live births.[69] Studies that looked at placentas found histologic chorioamnionitis present in 3%-5% of term placentas and in 94% of placentas delivered at 21-24 weeks of gestation.[31]

The risk of chorioamnionitis increases based on health conditions and behaviors, as outlined in the Pathophysiology section. Furthermore, factors such as gestational age, economic conditions, and ethnic differences influence the incidence. Histopathology of the placenta suggests inflammation may occur in the normal course of parturition at term gestation, thus complicating the definition of chorioamnionitis. An increase in histopathologic chorioamnionitis is noted in cases of preterm birth as compared with delivery of the healthy term infant. Signs of placental inflammation are present in 42% of extremely low birth weight infants.[70] Most investigators agree that infection is directly or indirectly associated with 40%-60% of all preterm births.[71]

Infants exposed to maternal acute chorioamnionitis are at increased risk for early-onset sepsis (EOS). The risk is modified by gestational age and maternal treatment with intrapartum antibiotics. Data from the 1980s and 1990s showed that asymptomatic infants born at term gestation to mothers who received intrapartum treatment for clinical chorioamnionitis have a 1.5% incidence rate of positive blood cultures, whereas symptomatic term infants with chorioamnionitis born to mothers who received intrapartum treatment have a 13% incidence rate of positive cultures 13%.[72]

More recent reports continue to indicate that the risk of EOS in infants born to women with chorioamnionitis remains strongly dependent on gestational age, but this risk is much lower compared to old data. In three reports including 1892 infants born at 35 weeks or more of gestation to mothers with clinical chorioamnionitis,[73, 74, 75] the rates of EOS (positive blood culture at < 72 hours of age) were only 0.47%, 1.24%, and 0.72% (number needed to treat [NNT] to prevent one infection: 80-210). In contrast, 4.8%-16.9% of preterm infants exposed to chorioamnionitis develop EOS (NNT: 6-21).[4, 76]  None of these studies stratified risk according to presence or absence of clinical signs of illness; however, more recent data from the National Institute of Child Health and Human Development (NICHD) Neonatal Research Network suggest the risk to be very low in asymptomatic late preterm and term neonates.[77]

International data

Developed countries (eg, Canada, Western Europe, Australia) probably have an incidence equal to, or perhaps even less than, the rate of chorioamnionitis observed in the United States. In underdeveloped countries, premature rupture of membranes has a strong association with chorioamnionitis, and chorioamnionitis in this setting results in preterm birth with a high mortality rate.[78] Classic studies by Naeye et al demonstrated that malnourished pregnant women in Africa had a higher risk of ascending urogenital infection with subsequent amniotic fluid infection.[79]  Infection in these malnourished women in Africa was attributed to a decrease in host defense factors in amniotic fluid that regularly prevents disease in this liquor.[80] In developed countries where women receive suboptimal care and have poor nutrition during pregnancy, a higher incidence of infection can be expected because of altered immune defenses.[81]

The bacterial pathogens that cause EOS in developing countries differ from the microbes that cause disease in the United States and other more developed countries, with Klebsiella pneumonia and Pseudomonas aeruginosa being the most common organisms in two reports from India and Pakistan.[82, 83]  For ill-defined reasons, the prevalence of group B streptococcal (GBS) disease is lower in developing countries. It is speculated that as developing countries sustain economic development, the prevalence of different bacterial pathogens assumes a profile closer to that of developed countries.

Race-, sex-, and age-related demographics

In select populations, race may increase the risk of maternal chorioamnionitis and preterm delivery.[84] Studying histologic chorioamnionitis and preterm birth, Holzman et al observed evidence of inflammatory pathology in 12% of placentas from white women and women of other races compared to 55% in black women.[85]  However, it is difficult to separate race form other hostile environmental circumstances (eg, violence, inadequate prenatal care, malnutrition) that could lead to chorioamnionitis and adverse maternal and neonatal outcomes.

Existing data on the role of sex in EOS are conflicting. Although some researchers identified male sex as a risk factor for EOS,[17] others failed to demonstrate this association.[86, 21] Advanced maternal age alone, defined as being older than 35 years, has not been identified as a risk factor for chorioamnionitis. However, teenage pregnancy raises the risk of chorioamnionitis.[87, 88, 89]

Maternal sequelae

Acute chorioamnionitis may result in labor abnormalities (dysfunctional labor) that increase the risk for cesarean delivery, uterine atony, and postpartum bleeding, as well as the need for blood transfusion.[2, 90] These complications are likely to occur more often when the amniotic fluid is infected with invasive organisms (eg, E coli and group B Streptococcus [GBS]) as compared with low-virulence organisms (eg, Ureaplasma urealyticum).[91] Chorioamnionitis may also lead to the development of other infectious complications, including endometritis, localized pelvic infections requiring drainage, septic pelvic thrombophlebitis, and intraabdominal infections.[92] More serious sequelae such as sepsis, coagulopathy, and adult respiratory distress syndrome are rare, especially when treatment with broad-spectrum antibiotics is initiated. Additionally, chorioamnionitis may initiate uteroplacental bleeding or a placental abruption.[93]  The risk of intrauterine infection is increased in placenta previa and may manifest with vaginal bleeding.[94]

Neonatal sequelae

The most serious risks of neonatal exposure to chorioamnionitis are preterm delivery[95] and early-onset neonatal infections (especially sepsis and pneumonia). Other adverse outcomes include perinatal death, asphyxia, intraventricular hemorrhage (IVH), cerebral white matter damage, and long-term disability (including cerebral palsy), as well as other morbidities related to preterm birth.[96, 97] The outcome of neonatal infections depends on the causative organism, the nature of the infection, the time of infection onset to time of administration of appropriate therapy, the symptoms at time of birth, and the gestational age of the infant. Prematurity and birth defects are confounding factors that must be considered when a prognosis is offered to parents or caregivers of an infected newborn. Outcomes may not be evident during the neonatal period, and long-term follow-up care is indicated in these infected neonates.

Neonatal mortality and morbidity

In a study that evaluated the whole US population and linked infant birth and death certificate files for the year 2008, the neonatal mortality rate for infants exposed to chorioamnionitis was 1.40 per1000 live births (LB) versus 0.81 per 1000 LB for infants without chorioamnionitis, with an odds ratio (OR) of 1.72 and a 95% confidence interval (CI) 1.20-2.45.[69] The OR for neonatal death for infants with chorioamnionitis exposure who received antibiotics versus those who did not was 0.69 (95% CI = 0.21-2.26).[69]  In another study of infants born at 23-32 weeks' gestation with evidence of intrauterine infection and inflammation, the neonatal death rate was 9.9%-11.1%.[98]

Preterm infants born to mothers with chorioamnionitis have unfavorable short-term (meningitis and intraventricular hemorrhage and periventricular leukomalacia) and long-term (cerebral palsy and neurodevelopmental impairment) neurologic outcomes.[99, 100]  Cerebral palsy (CP)[101]  and cognitive impairment without CP[102]  have been linked to exposure to maternal chorioamnionitis. In particular, funisitis and the fetal inflammatory response syndrome have been associated with white matter brain injury or periventricular leukomalacia that is linked to activation of cytokine networks.[103, 104]  Interleukin (IL)-1beta, IL-6, IL-8, IL-17, IL-18, and tumor necrosis factor (TNF)-alpha are among the cytokines identified as agents related to the fetal inflammatory response syndrome (FIRS) that results in brain injury.[105, 106, 107] However, more recent systematic reviews suggest that the evidence for a causal or associative role of chorioamnionitis in CP is weak[108] and that improvements in neonatal intensive care may have attenuated the impact of chorioamnionitis on brain health outcomes.[109]

The relationship of chorioamnionitis and neonatal cardiopulmonary morbidity is conflicting. Different studies have evaluated the risk of respiratory distress syndrome (RDS), bronchopulmonary dysplasia (BPD), and childhood asthma after fetal exposure to chorioamnionitis. Although some studies showed chorioamnionitis to be associated with lower risk of RDS,[110, 111] other studies found an increased risk of RDS[111, 112] or no association after adjusting to other variables.[99]  Similar conflicting data exist for the link of chorioamnionitis and BPD; however, a 2017 French national prospective, population-based, cohort study that included 2513 live-born singletons delivered at 24-31 weeks of gestation and 1731 placentas concluded that histologic chorioamnionitis is not associated with BPD.[113]

Chorioamnionitis caused by Ureaplasma has been studied extensively[114] (including in animal models) and has been linked to congenital pneumonia, prolonged mechanical ventilation, and cytokine release in the neonatal lungs with subsequent development of BPD.[115] However, studies that looked at antibiotic therapy with erythromycin to reduce the incidence BPD when the neonatal lungs are colonized or infected with Ureaplasma have been disappointing. More recent studies with azithromycin are encouraging.[116, 117]

The link between fetal exposure to chorioamnionitis and the future development of childhood asthma was implied by a systematic review but there was much variation in the included studies with regard to the type of maternal infection, age of the children, and methods of exposure ascertainment that made the conclusion less certain.[118] Lastly, with regard to the association between chorioamnionitis and patent ductus arteriosus, two meta-analyses reached opposing conclusions about the association.[119, 120]

Parents or other caregivers of infected neonates need specific instructions about the subsequent care of these infants. This is particularly true for secondary complications associated with such infections. For example, parents/caregivers of an infant with meningitis that has postinfectious hydrocephalus requiring ventriculoperitoneal shunt placement need to have specific instructions about shunt-related malfunction or shunt-related infection. Education of the parents/caregivers related to the recognition and management of seizures should be mandatory before discharge.

Similarly, parents/caregivers of patients with long-term pulmonary complications of congenital pneumonia may require specific education (eg, administration of oxygen or use of bronchodilators at home). Parental education in neonatal resuscitation is indicated for many graduates of the neonatal intensive care unit (NICU).

The time-honored clinical signs and symptoms of chorioamnionitis include the following:

• Intrapartum fever: The National Institute of Child Health and Human Development (NICHD) workshop report defined maternal fever as temperature of at least 39.0°C [102.2°F] or 38.0°C [100.4°F] to 38.9°C [102.02°F] on two occasions 30 minutes apart, ^[1] without another clear source. ^[2]
• Significant maternal tachycardia (>120 beats per minute [bpm])
• Base line fetal tachycardia (>160 bpm for 10 min or longer, excluding accelerations, decelerations, and periods of marked variability)
• Maternal leukocytosis (total blood leukocyte count >15,000 cells/μL) in the absence of corticosteroids
• Purulent-appearing fluid coming from the cervical os, as visualized by speculum examination
• Other nonspecific signs such as maternal tachycardia and uterine tenderness are deemphasized by the NICHD. ^[2]

Of these criteria, intrapartum maternal fever appears to be the most frequent and necessary to make the diagnosis of intrauterine infection according to the workshop conducted by NICHD[2] and later endorsed by the American College of Obstetricians and Gynecologists (ACOG).[30] According to this report, women who have fever without other chorioamnionitis symptoms or signs are regarded as having “isolated fever” and should not be diagnosed as having chorioamnionitis or “triple I.” Additionally, the NICHD report deemphasized nonspecific signs like uterine tenderness, maternal tachycardia, and vaginal discharge because some of these signs occur during the normal course of physiologic labor or can be masked by neuraxial anesthesia. Of note, for treatment purposes, ACOG further suggests that patients with isolated fever of at least 39.0°C (102.2°F) should be managed as having suspected intraamniotic infection.[30]

When at least two of the aforementioned criteria are present, the risk of neonatal sepsis is increased. Each clinical sign and symptom of chorioamnionitis, however, is by itself of low predictive value. The signs and symptoms of maternal chorioamnionitis are so subjective and the term “chorioamnionitis,” which should be reserved to describe histologic changes in the placenta and membranes, has been used widely by clinicians, sometimes to describe isolated intrapartum low-grade fever. This has led to too many asymptomatic newborns being evaluated for early-onset sepsis (EOS) and being unnecessarily exposed to antibiotics, with significant variations in practice among neonatal intensive care unitis (NICUs).[121] The NICHD workshop expert panel report, even though not considered a formal consensus recommendation by the NICHD, is a step in the right direction to define the problem and recommend a unified approach for diagnosis and management of chorioamnionitis, both for obstetric and neonatal care providers.

The NICHD workshop recommended using the term “triple I” to address the heterogeneity of this disorder. The term "triple I" refers to intrauterine infection or inflammation or both, and it is defined by strict diagnostic criteria (see Diagnostic Considerations); however, this terminology has not been universally accepted.[3] It is important to differentiate between clinical and histologic chorioamnionitis; the latter tend to be “silent” and present only with preterm labor or preterm premature rupture of membranes (PPROM). The risk of neonatal sepsis is increased when chorioamnionitis is diagnosed in the laboring mother; however, the risk is much lower than anticipated based on historical figures when widespread use of intrapartum antibiotics was not a common practice.[4]

An increasing or decreasing total leukocyte count may be more important than a single determination. Abnormalities in either umbilical vein interleukin (IL)-6 levels or an increasing neonatal immature-to-total neutrophil ratio, along with clinical criteria associated with chorioamnionitis, improve the sensitivity and predictive accuracy of identifying the septic neonate.

Although many cytokines and inflammatory markers have been proposed as diagnostic tests for EOS in the newborn, none has been of sufficient predictive value to gain wide acceptance.[10, 122, 123] Some cytokines are only secreted over a limited time frame during the start of infection, and inflammatory markers may be more sustained in their presence. Levels of C-reactive proteins (CRPs) are useful to exclude infection and discontinue antibiotics. Some investigators suggest that using a combination of markers like CRP, procalcitonin, IL-6, presepsin (soluble CD14 subtype), in conjunction with other cell-surface antigens such as CD11b, CD64, and human leukocyte antigen-antigen D related (HLA-DR) may enhance the laboratory diagnosis of EOS.[124, 125, 126]

The diagnosis of clinical chorioamnionitis in pregnancy is commonly made based on clinical findings of fever plus fetal tachycardia, maternal leukocytosis, or purulent fluid coming from the cervical os. Additionally, the pregnant woman with chorioamnionitis may appear ill, even toxic, and she may exhibit hypotension, diaphoresis, and/or cool or clammy skin. However, especially when dealing with histologic chorioamnionitis, maternal clinical signs or symptoms of infection may be absent (silent chorioamnionitis).[5]

Furthermore, clinical signs and symptoms of chorioamnionitis are not always associated with placental evidence of inflammation.[6] This is particularly true if maternal fever is the sole criterion for the diagnosis.

Examination for suspected sepsis in a woman with chorioamnionitis may include the following findings:

• Fever
• Tachycardia (>120 bpm)
• Hypotension
• Diaphoresis
• Cool or clammy skin
• Uterine tenderness
• Foul-smelling or purulent vaginal discharge

The presence of intraamniotic amniotic fluid "sludge," a free-floating hyperechogenic material within the amniotic fluid in close proximity to the uterine cervix, reflects intraamniotic inflammation with or without microorganisms.[127]  On uterine ultrasonography, this finding has been seen in asymptomatic women at risk for preterm delivery.[12, 128]  More recent studies confirm that amniotic fluid sludge is a useful marker of microbial invasion of the amniotic cavity, histologic chorioamnionitis, and funisitis—conditions that increase the risk for preterm delivery at an extreme gestational age.[129]  Aseptic aspiration of the "sludge" may show the material to have a low glucose content, many neutrophils, and organisms such as gram-positive cocci or Candida.[130, 128]

Clinical signs and symptoms of chorioamnionitis are not always associated with placental evidence of inflammation.[66] This is particularly true if maternal fever is the sole criterion for the diagnosis.

Epidural anesthesia during the intrapartum period has been associated with fever in the mother and the neonate[131, 132] ; a primary sterile inflammatory response either in the placenta or in the epidural space (with secondary inflammation in the placenta and chorioamniotic membranes) is the most likely etiology.[67, 68, 133]  Other etiologic factors in epidural anesthesia-induced fever include nulliparity, dysfunctional labor, prolonged labor, maternal exhaustion, dehydration, and/or prolonged rupture of membranes.[134]

Because it is hard to differentiate epidural anesthesia-induced maternal fever from intraamniotic infection, maternal and/or neonatal fevers result in more evaluations for sepsis and antibiotic treatment in neonates.[135] In addition to the risk for unnecessary exposure to antibiotics, fever itself may be damaging to the newborn, especially in the setting of hypoxia-ischemia.[136]

In the setting of epidural anesthesia during labor, the following clinical course has been observed. The fetus usually has tachycardia when the mother is febrile during labor. At birth, the newborn may also have a fever (temperature >38.0ºC [100.4ºF]). If the neonate is not septic, the temperature elevation dissipates rapidly following birth, and the newborn subsequently exhibits normal behavior. Usually a temperature elevation in the neonate has returned to normal within 30-60 minutes after birth. Furthermore, these noninfected, febrile neonates have normal Apgar scores and appear remarkably well following birth. Such newborns can be observed for illness rather than undergo a septic workup and antibiotic therapy. However, clinical judgment must be based on many factors, including the intrapartum administration of broad-spectrum antibiotics to the mother.

Examination for suspected sepsis in the neonate of a mother with chorioamnionitis often yields nonspecific and subtle findings, which may include the following:

• Behavioral abnormalities (eg, lethargy, hypotonia, weak cry, poor suck)
• Pulmonary: Tachypnea, respiratory distress, cyanosis, pulmonary hemorrhage, and/or apnea
• Cardiovascular: Tachycardia, hypotension, prolonged capillary refill time, cool and clammy skin, pale or mottled appearance, and/or oliguria
• Gastrointestinal: Abdominal distention, vomiting, diarrhea, and/or bloody stools
• Central nervous system (CNS): Thermal regulatory abnormalities, behavioral abnormalities, apnea, and/or seizures
• Hematologic and/or hepatic: Pallor, petechiae or purpura, and overt bleeding

The fetus may have tachycardia (>160 bpm) or decreased variability.[137]  A biophysical profile (BPP) performed on the fetus using ultrasonography may reveal a lower than normal score, but ultrasonic biophysical profile assessment has not been predictive of clinical chorioamnionitis.[138]  Lack of fetal breathing has been associated with fetal infection.[139, 140]

Tachypnea, respiratory distress (eg, expiratory grunt, retractions), cyanosis, pulmonary hemorrhage, and/or apnea (ie, pulmonary manifestations of pneumonia, sepsis, or both), must be immediately appreciated by caregivers. Nursery personnel must be aware that a neonate who is born without respiratory distress but who develops signs and symptoms of pulmonary disease in the first 6-12 hours of life has a heightened risk for congenital (infectious) pneumonia.

Tachycardia, hypotension, prolonged capillary refill time, cool and clammy skin, pale or mottled appearance, oliguria (ie, cardiovascular manifestations of sepsis), or a combination of these may be observed, typically occurring late in the course of early-onset sepsis (EOS). Caregivers must also consider other explanations for these physical findings, such as developmental defects in the cardiovascular system (ie, cardiovascular malformations with abnormalities of aortic blood flow) or inborn errors of metabolism.

Gastrointestinal symptoms, as outline above, may be nonspecific in patients with EOS.

Thermal regulatory abnormalities (ie, hypothermia or hyperthermia), behavioral abnormalities, apnea, seizures (ie, CNS manifestations), or a combination may be seen. A bulging fontanel or nuchal rigidity is not a reliable sign of meningitis in a neonate.

Pallor, petechiae or purpura, and overt bleeding (ie, hematopoietic involvement, liver involvement, or both) may be seen and requires immediate diagnostic and therapeutic attention.

Because one physiologic system may affect another, signs and symptoms may originate from more than one dysfunctional organ. However, many neonatal conditions resemble neonatal sepsis; thus, physician caregivers must have an open mind regarding other clinical conditions that may involve signs and symptoms resembling those of neonatal sepsis. Those conditions include, but are not limited to, the following:

• Cardiovascular malformations, especially left-sided obstructive lesions causing poor systemic cardiac output
• Endocrine disorders that may also cause shocklike states, such as different types of congenital adrenal insufficiency or hypoglycemia associated with hyperinsulinemia
• Serious CNS trauma or dysfunction from any cause and subtle seizures with systemic manifestations
• Anemia caused by unrecognized isoimmunization or blood loss from conditions such as fetomaternal transfusion syndrome

During the intrapartum period, diagnosis of chorioamnionitis is usually based on clinical criteria. This is particularly true for pregnancies at term. Chorioamnionitis or intraamniotic infection, as etiologies for preterm labor and preterm premature rupture of the membranes (PPROM), should always be considered. Silent chorioamnionitis is recognized as an important cause of premature labor and PPROM.[5, 158]

To diagnose silent or obvious amniotic fluid infection or chorioamnionitis, the physician often uses laboratory examinations of the amniotic fluid, maternal blood, maternal urine, or a combination of these.

Bacteriologic cultures of amniotic fluid and urogenital discharge may be diagnostic for causative pathogens. Investigators suggest that obtaining cervical cultures or performing frequent digital examination increases the risk of initiating amniotic fluid infection in the presence or absence of ruptured membranes.

Maternal laboratory studies

Examination of amniotic fluid and urogenital secretions

Culture of amniotic fluid remains the "gold standard" and most specific test for documentation of intraamniotic infection, but this study is limited by the fact that it may take days to obtain definitive results. More rapid results can be obtained from several other tests, including Gram stain, glucose concentration, white blood cell (WBC) concentration, and leukocyte esterase level.[7] Amniotic fluid, obtained with amniocentesis, may be screened for leukocyte count; Gram stain; pH; glucose concentration; and levels of endotoxin, lactoferrin, and cytokines (eg, interleukin [IL]-6, IL-8, or tumor necrosis factor [TNF]), or a combination of these markers.

Maternal endotoxin activity appears to show promise as a marker in pregnancies complicated by PPROM; however, more data and larger studies are needed to evaluate its potential in predicting the clinical evolution of preterm birth.[159]  Similarly, findings from a study of 47 women with PPROM suggested that maternal levels of IL-6 in combination with maternal characteristics hold the potential to be good noninvasive predictors of histologic chorioamnionitis.[160]

Cytokines commonly quantified in either amniotic fluid or blood include IL-6, TNF-alpha, IL-1, and IL-8.[31, 105, 161] No consensus has been reached regarding which cytokine offers the best sensitivity, specificity, and positive versus negative predictive accuracy. However, IL-6, a key mediator of the acute phase response to infection and tissue injury, is one of the most studied markers and a bedside, point-of-care (POC) testing has been developed to test for IL-6 in amniotic fluids and vaginal secretions.[136, 162, 31, 163] Elevated IL-6 levels in cord blood and amniotic fluid have been related to adverse long-term neurologic outcomes in the neonate.[163, 164, 165]  This testing has not become routine yet. However, Chaemsaithong et al reported the potential utility of a rapid IL-6 bedside test (20 minutes) (lateral flow-based immunoassay, or POC test) for measuring IL-6 concentrations in amniotic fluid. Their goal was to identify women with intraamniotic inflammation and/or infection and those who might deliver spontaneously before 34 weeks' gestation among women with preterm labor and intact membranes.[165]

Data from 136 women with singleton pregnancies who presented with symptoms of preterm labor and underwent amniocentesis showed that the POC test for amniotic fluid IL-6 concentrations had a 93% sensitivity, 91% specificity, and a positive likelihood ratio of 10 for the identification of intraamniotic inflammation by using a threshold of 745 pg/mL.[165] Moreover, the POC test performed similarly to enzyme-linked immunosorbent assay (ELISA) for IL-6 levels and identification of microbial invasion of the amniotic cavity (MIAC), acute inflammatory lesions of the placenta, and patients at risk of impending spontaneous preterm delivery.[165] These investigators found similar results in the setting of PPROM.[163] Other investigators reported similar findings.[162] More recently, matrix metalloproteinase (MMP)-8, a neutrophil collagenase enzyme, has been shown to be a sensitive marker for intraamniotic inflammation that compares well to IL-6 and was developed into rapid POC test as well.[166, 167]

The rapid development of polymerase chain reaction (PCR) as a diagnostic aid has allowed its use in identifying microbes such as human immunodeficiency virus, cytomegalovirus, herpes simplex, parvovirus, toxoplasmosis, and bacterial DNA in amniotic and other body fluids. PCR has been used for the diagnosis of amniotic fluid infection caused by bacterial pathogens[168] ; however, only university or major academic centers have this relatively expensive technology available to caregivers.

Amniocentesis to obtain amniotic fluid carries the risk of rupturing the fetal membranes and initiating preterm labor. For this reason, screening tests that use cervicovaginal secretions to indicate chorioamnionitis have been reported. Potential markers of cervical or chorionic inflammation include cervical or vaginal concentrations of fetal fibronectin, insulinlike growth factor binding protein-1, and sialidase. Significant association is noted among levels of cervical IL-6, fetal fibronectin, and amnionitis. Conversely, a positive midgestational fetal fibronectin assay was not associated with acute histologic placental inflammation at birth.[169] Proteomic profiling of amniotic fluid detects intrauterine inflammation and/or infection and predicts subsequent neonatal sepsis.[8] Caregivers should follow this research, because in the next 5 to 10 years, proteomic profiling for inflammation or nonculture-based molecular detection of microbes may become routine in obstetric samples.

Antenatal screening uses rectovaginal specimens to detect the presence of maternal group B streptococcal (GBS) colonization at 35-37 weeks' gestation. Using these specimens, the Centers for Disease Control and Prevention (CDC) recommends selective growth of GBS in broth followed by cultivation using the plate method.[170, 171]  This is the criterion standard assay.​

The CDC does not recommend direct PCR detection of GBS in rectovaginal samples. Rather, a rectovaginal sample should undergo enhanced growth in selective broth before performing PCR.[170, 171] Maternal colonization with rectovaginal GBS increases the risk of chorioamnionitis, and intrapartum prophylaxis with antibiotics reduces the incidence of neonatal infection from GBS.[172, 173]

Missed screening and the failure to give intrapartum antibiotics is responsible for the persistence of neonatal GBS infection.[174] Therefore, for mothers that missed GBS screening at 35-37 weeks' gestation, intrapartum testing for GBS using rapid detection methods on vaginal secretions is an option recommended by some authorities. Intrapartum real-time PCR, performed on vaginal swabs, have been shown to be accurate by several investigators, and it performs as well or better than the antepartum culture for identification of GBS vaginal carriers during labor.[175, 176, 177]

Examinations of maternal blood

WBC counts or C-reactive protein (CRP) levels in maternal blood have been used to predict acute chorioamnionitis when maternal fever is present. Different studies have supported or refuted the use of CRP to diagnose chorioamnionitis.[178, 179] The CRP level may be a better predictor of the risk of chorioamnionitis than peripheral WBC counts, especially if the mother has received corticosteroids, which may falsely increase the total WBC count.

Other investigators have suggested that the alpha1-proteinase inhibitor (A1PI) complex in maternal blood is a better predictor of amniotic fluid infection than either CRP levels or WBC count. Analysis of maternal serum for either IL-6 or ferritin content may also be helpful, because elevations in these mediators are associated with maternal or neonatal infection. Serum IL-6 levels may be more predictive of infection than CRP concentrations in maternal blood. Levels of A1PI complex, cytokines, and ferritin in maternal blood have not gained widespread use as markers of acute chorioamnionitis.

Laboratory studies in newborn infants

The criterion standard for making a diagnosis of early-onset bacteremia, pneumonia, or meningitis in neonates is the growth of bacteria in an appropriate specimen (ie, blood, tracheal secretions, cerebrospinal fluid [CSF]). Urinary tract infection is an infrequent cause of early-onset bacterial disease in the neonate; thus, bladder catheterization or suprapubic bladder taps are not usually required as part of an evaluation for early-onset sepsis (EOS).

Controversy has arisen regarding the inclusion of the lumbar puncture as part of the evaluation for EOS. Some clinicians have argued that meningitis is rarely seen as a manifestation of EOS, the neonate with meningitis has obvious manifestations, and the asymptomatic term neonate does not require a lumbar puncture as part of the evaluation for EOS. Furthermore, other caregivers argue that a lumbar puncture can only be performed safely when life-threatening pulmonary dysfunction or hypertension resolves. Alternatively, other investigators have stressed that cases of meningitis are missed with this approach.[180]

The medical literature contains good evidence that meningitis may occur in association with sterile blood cultures. Because meningitis is a devastating neonatal infection, no lumbar puncture may result in inadequate antibiotic therapy. Thus, we recommend that a lumbar puncture be performed selectively; when the newborn is symptomatic, especially with central nervous system (CNS) symptoms/signs such as lethargy, irritability, apneas, and seizures; when inflammatory markers are severely deranged, and also when the blood culture is positive.

Studies that are also considered specific for infection include positive findings on Gram stains of tracheal secretions or CSF.[181] The tracheal secretions must be obtained shortly after birth (< 4-6 hours). The reason is that colonization of the airways may occur from the neonatal intensive care unit (NICU) environment during this time frame. Both tracheal fluids and CSF should be sterile at birth. The presence of bacteria on microscopic analysis (ie, Gram stain) indicates that more than 10,000 colony-forming units (CFUs) of bacteria are present per milliliter of specimen (body fluid). However, the absence of bacteria in either CSF or tracheal secretions does not exclude infection. A final diagnosis should be based on culture results; this testing takes 24-48 hours.

An absence of neutrophils in CSF or tracheal secretions is expected. The presence of neutrophils in tracheal aspirates obtained after birth indicates that the fetus has mounted an inflammatory response to infection in the environment. Studies by the original author and separate studies by pathologists indicate that neutrophils present in tracheal secretions shortly after birth originate from the fetus or neonate and do not represent aspirated maternal neutrophils found in infected amniotic fluid. This conclusion is based on examining Y-body fluorescence in neutrophils present in the tracheal secretions of infected male neonates. In some studies, 50% of the neutrophils present in tracheal secretions of infants with suspected congenital pneumonia had Y-chromosome fluorescence, indicating a fetal origin. Maternal neutrophils can gain access to the fetal lung only when gasping occurs during fetal asphyxia.

Bacterial antigen detection in CSF (especially for GBS) may be a useful indicator of bacterial infection; however, false-positive tests have been reported and the test is rarely done in the present day. Bacterial antigen detection in the urine should not be used in a neonate's evaluation for sepsis.

All other tests used to diagnose early-onset bacterial infection in the neonate should be considered screening tests. The most common laboratory studies used to screen for neonatal sepsis are WBC profiles and CRP determinations. These tests, at best, are presumptive indicators of infection.

WBC profiles (leukopenia [< 5000/µL], leukocytosis [>30,000/µL], a markedly diminished absolute neutrophil count [ANC] [< 500-1500/µL], an immature-to-total neutrophil ratio [>0.3-0.4]) are commonly used screening tests for the septic neonate. Note that the immature-to-total neutrophil ratio of 0.3-0.4 is higher than the previous value of 0.2 reported in the classic studies of Manroe (1977 and 1979).[182, 183] Clinical pathologists have been less accepting of the immature-to-total neutrophil ratio as a diagnostic aid in neonatal sepsis,[184]  and studies have reexamined the WBC counts and the leukocyte profiles present in extremely preterm infants[185] and at high altitude.[186] Other diagnostic tests (eg, inflammatory factors, adhesion molecules, cytokines, neutrophil surface antigens, even bacterial DNA) may be superior alternatives to this test. To date, these markers of neonatal inflammation/infection have not replaced leukocyte counts as diagnostic methods.

WBC profiles and kinetics are influenced by the genetic make-up of the patient, the gestational age, maternal noninfectious disorders such as pregnancy-induced hypertension (PIH), medications administered to the mother, fetal disease, and other factors. Reference range WBC counts in the neonate do not exclude infection, and serial studies of WBC indices at approximately 6- to 12-hour intervals may be more useful in detecting sepsis.[187] A continued assessment of WBC kinetics offers more information regarding decision making. For example, a physician should be particularly concerned with a falling total WBC count, a declining absolute mature neutrophil count, and a rising immature-to-total neutrophil ratio. These findings, taken together, indicate depletion in the bone marrow–related storage pool of neutrophils.[188]

The predictive accuracy of WBC indices for the diagnosis of the EOS is poor. Likewise, the accuracy of CRP determinations to predict neonatal infections shortly after birth is low. However, a negative CRP (especially if done serially, 12-24 hours apart) is a reason to stop antibiotic therapy after 48 hours.[189, 10]

Akin to maternal diagnostic studies for infection, levels of A1PI complex and cytokines (eg, IL-1 and IL-6; in particular, IL-1 receptor antagonist), as well as the detection of bacterial products in neonatal blood, have not gained widespread use as markers of neonatal sepsis. However, these effectors of inflammation may prove to have better predictive accuracy than WBC tests or the CRP level. Procalcitonin level may have better sensitivity, specificity, and positive and negative predictive value than CRP in the diagnosis of early-onset neonatal sepsis and is increasingly being used.[9, 10] Cell-surface markers of inflammation on leukocytes remain under investigation as potential markers to detect EOS.[190, 191, 192]  Serum amyloid A levels appear to be have high sensitivity at the onset of symptoms and 2 days after, although existing data show a variable positive predictive value with a high negative predictive value.[10]

Molecular methods that use real-time PCR and DNA sequencing for amplification and detection of 16S rRNA of pathogenic bacteria in neonatal blood have created enormous interest because a rapid diagnosis is possible.[193, 194, 195]  However, despite showing that 16S rRNA PCR increased the sensitivity in detecting bacterial DNA in newborns with signs of sepsis and allowing shortening of antibiotic courses, a report concluded that uncertainty about the bacterial cause of sepsis was not reduced by this test, and that blood culture remains currently irreplaceable.[196] As technology advances, caregivers should pay close attention to this rapidly advancing field.

Ultrasonography may be used to ascertain fetal well-being. A biophysical profile (BPP) provides information about the status of the fetus. A low BPP score, and especially the loss of fetal breathing movements, has been associated with fetal bacterial infection after premature rupture of membranes (PROM).[139, 140] Other investigations have not confirmed the importance of a low BPP score. Specifically, the absence of fetal breathing, may not be a reliable test for amnionitis prior to 32 weeks' gestation.[197, 198]

Before the fetus is viable, vaginal ultrasonography can be used to identify women with a shortened cervical canal, which is associated with a higher risk of preterm delivery.[11, 12, 13] Researchers suggest a shortened cervical canal or cervical insufficiency are linked to ascending urogenital infection that initiates premature labor, PROM, or both.

Needle aspiration and analysis of amniotic fluid (amniocentesis) is the only invasive procedure used to confirm the diagnosis of acute chorioamnionitis. This procedure can be risky with intact fetal membranes, because the fetal membranes can rupture during or after the procedure. Bleeding or placental abruption can also be a consequence of amniocentesis. The procedure should be performed using ultrasonographic guidance to avoid fetal injury. For these reasons, amniocentesis to diagnose maternal chorioamnionitis has had limited application in obstetric practice.

Histologic findings

Gross and microscopic examinations of the placenta, fetal membranes, and umbilical cord for evidence of inflammation and infection are crucial to make a definitive diagnosis of chorioamnionitis.[14]  Histologic chorioamnionitis is a reliable indicator of infection whether or not it is clinically apparent.[199]  Nevertheless, anatomic studies should be correlated with a culture aseptically obtained from the fetal surface of the placenta.

The microbiologic cultures should include an attempt to isolate aerobic and anaerobic bacteria. Special microbiologic techniques may be required for certain microorganisms such as Listeria monocytogenes. Only with these methods can the pathologist help the bedside clinician delineate the cause of maternal chorioamnionitis and neonatal sepsis. Clinicians are encouraged to ask pathologists for help in their search for infections causing disease in the pregnant woman, fetus, and newborn. Obstetricians must also obtain the placenta, fetal membranes, and umbilical cord samples for analytical studies when suspicious clinical circumstances are noted.

Staging

Redline and colleagues proposed a scoring system for placental examination that promotes consistency when pathologists judge the severity of chorioamnionitis.[200]

Several physiologic scores have also been proposed for neonates who have life-threatening illness, but a report by Lim et al could not conclude that these scores accurately predicted neonatal morbidity and mortality during infection.[201]

Transfer, hospitalization, and discharge considerations

Infected neonates born at hospitals with level 1 (normal) or level 2 (special care) nurseries may require transfer to a level 3 or 4 neonatal intensive care unit (NICU). Transfer depends on the circumstances of the neonatal infection, degree of prematurity, presence of anomalies, and other pathophysiologic states. Reasons for transfer of the neonate from a level 1 or 2 nursery to a higher-level facility are outlined in Consultations.

Transfer requirements such as oxygen or assisted ventilation, mode of transportation (eg, ambulance, helicopter, fixed wing aircraft), and healthcare personnel to transport the patient are beyond the scope of this article.

Both the mother with suspected chorioamnionitis and her newborn with suspected sepsis require frequent assessments over the first 48 hours following birth. Mothers with chorioamnionitis who appear well after a brief intravenous course of antibiotics may be discharged on oral antibiotic therapy, but comprehensive outpatient follow-up care is required. General and gynecologic health is usually normal after maternal chorioamnionitis.

When the newborn exposed to chorioamnionitis is preterm (< 35 weeks' gestation) or symptomatic term, the decision to obtain laboratory evaluation (complete blood cell [CBC] count and blood culture) and treat with antibiotics is relatively straightforward. However, when late preterm or term neonates are asymptomatic and well appearing, the decision to obtain laboratory evaluation, start antibiotics, and separate infants from their mothers is difficult and has generated much debate in recent years.[4, 77, 202, 203, 204]

One approach to limiting the unnecessary use of antimicrobials is to use the “sepsis calculator” developed by Puopolo et al[205] to estimate the probability of early-onset sepsis (EOS) using maternal risk factors in neonates born at 34 weeks of gestation or Later. Utilizing data from more than 600,000 infants at at least 34 weeks’ gestation at birth, the investigators developed a model for EOS risk prediction based on objective maternal factors, then combined that model with findings from examination of the infants.[206] The model uses three categorical variables: group B Streptococcus (GBS) status (positive, negative, uncertain), maternal intrapartum antimicrobial treatment (GBS-specific or broad spectrum), and intrapartum prophylaxis or treatment given 4 hours or longer before delivery (yes, no) in addition to the following continuous variables: highest maternal intrapartum temperature (centigrade or Fahrenheit), gestational age (weeks and days), and duration of rupture of membranes (hours). A predicted probability per 1,000 live births can be estimated using the calculator (http://newbornsepsiscalculator.org). Several retrospective studies demonstrated that the use of the sepsis calculator in a population of well-appearing neonates (≥34 weeks' gestation) exposed to the clinical maternal diagnosis of chorioamnionitis would have substantially reduced the proportion of neonates undergoing laboratory tests and receiving antimicrobial agents.[202, 207, 208, 209]

Term neonates undergoing an evaluation to exclude sepsis who consistently appear well can probably go home with their mothers within 48 hours after birth. Septic-appearing neonates usually receive antibiotic therapy via the parenteral route until treatment is deemed complete and the infant is well. Depending on the nature of the infection and other risk factors associated with the hospitalization (eg, extreme prematurity, need for home oxygen), an outpatient follow-up visit may be scheduled from 1 day to 2 weeks after discharge. Home healthcare follow-up visits by a reliable and well-trained nursing service may also be indicated.

Antimicrobial therapy

Outpatient antibiotics used to treat a term neonate with rule-out sepsis have not been evaluated. Some managed care plans have discharged neonates with proven infection who appear well after antibiotic therapy. These newborns complete a course of intravenous antibiotics at home. The intravenous antibiotics are often administered via a percutaneous venous line placed before hospital discharge. A visiting nurse comes to the home to administer the antibiotics twice daily.

Depending on the type of infection found in the neonate, the duration of intravenous therapy with antibiotics ranges from 7 days (eg, perhaps pneumonia with rapid improvement or “culture-negative sepsis” based on a mildly ill or well-appearing infant with abnormal inflammatory markers like high levels of C-reactive protein [CRP] and/or procalcitonin, white blood cells, or immature-to-total neutrophil ratio) to 4-6 weeks (eg, osteomyelitis). The actual duration of treatment for different types of neonatal infections has not been studied; it is often experience based rather than evidence based.

Surgical intervention

Surgical interventions are infrequently required in early-onset bacterial infections of the neonate. The conditions that may require intervention include epidural or brain abscess, subcutaneous abscesses, infections localized to the pleural space, certain intraabdominal infections (especially if intestinal perforation is present), and bone or joint infections.

This section addresses two topics. The first topic includes maternal interventions to treat suspected chorioamnionitis and protect the fetus from infection. The second topic includes the diagnostic approach and the appropriate treatment of neonates born to mothers with suspected chorioamnionitis.

The observation that epidural anesthesia during labor may create findings suggestive of maternal chorioamnionitis is discussed. A maternal fever that occurs when epidural anesthesia is administered during the intrapartum period has often been interpreted as chorioamnionitis. This may not be the case, and the neonate is often needlessly treated after birth.

Using ampicillin as the chemoprophylactic agent to prevent group B streptococcal (GBS) disease in the neonate is associated with other issues and should be discouraged. Ampicillin-resistant E coli infections in the mother and her infant are reported as an increasing problem, possibly due to this prophylactic practice.[210, 211]  However, the use of penicillin rather than ampicillin to prevent GBS infections of the newborn is encouraged and should be the standard of care.[212] When the mother is allergic to penicillin, she is given clindamycin if her GBS isolate is documented to be susceptible to clindamycin. Approximately 30% of GBS isolates in the United States were clindamycin resistant in 2010, and the proportion varies by country. If, however, clindamycin susceptibility testing has not been performed, vancomycin should be administered instead.[170]

Obstetric management influencing neonatal outcome

When acute chorioamnionitis is evident, delivery must be expedited. Upon signs of serious fetal distress, delivery must be emergent. Withholding maternal antibiotics to obtain postnatal cultures from the neonate is no longer appropriate. This strategy was once an accepted practice based on the assumption that waiting to obtain cultures from the newborn helps to determine the cause of infection. The morbidity and mortality in the mother and newborn may actually increase because of a delay in administering antibiotics.

The neonatal care provider (neonatologist, pediatrician, or family medicine physician) must decide whether the fetus was infected and whether antibiotics given before birth should be continued in the neonate. Those antibiotics may differ from those administered to the mother. The history, physical findings, and results of certain laboratory studies can assist the physician in deciding whether to continue antibiotics started during the intrapartum period. Because antibiotic chemoprophylaxis reduces the risk of GBS infection in neonates, the obstetrician must always consider beginning penicillin during the intrapartum period when a mother has defined risk factors for GBS disease.[212, 213] The neonatal care provider must judge whether the chemoprophylaxis was sufficient to prevent infection (especially in a healthy, full-term neonate) or whether the infant must continue antibiotic therapy after birth. The US Centers for Disease Control and Prevention (CDC) issued guidelines that outline the strategies for screening and treatment to prevent neonatal disease caused by GBS, the most recent guidelines were published in 2010,[170]  with an update in 2012.[171]

A retrospective study evaluating daily gentamicin for the treatment of intrapartum chorioamnionitis in 500 women found that daily gentamicin dosing using ideal body weight compared with traditional 8-hour dosing regimens was associated with a 64% lower risk of postpartum endometritis and a 5% higher chance of successful outcome.[214] These results were adjusted for maternal factors such as race, parity, advanced maternal age (>34 years), body mass index, diabetes mellitus, gestational hypertension (>140/90 mmHg), and GBS status.

Determining the appropriate procedures to prevent fetal infection in the setting of premature, prelabor, rupture of membranes is more complex. The mother who has preterm labor or premature rupture of membranes at less than 34 weeks' gestation and no clinical signs or symptoms of chorioamnionitis should receive corticosteroid therapy.[98]

Attitudes toward antibiotic use have changed over time (see the discussion about the sepsis calculator under Treatment: Approach Considerations. If GBS colonization of the mother is not present, and signs and symptoms of chorioamnionitis are absent, pregnant women with preterm labor or premature rupture of membranes (PROM) at more than 36 weeks' gestation should be observed for infection. Thus, prophylactic antibiotics are not given in these circumstances. Mothers at term gestation with accepted risk factors for GBS infection in their fetus should receive chemoprophylaxis. Mothers at risk of preterm birth, and in whom GBS status is unknown, receive antibiotics during latency until the GBS screening is completed. A period of observation for maternal and/or fetal infection is also required after admission, although signs and symptoms may not be evident (ie, silent disease).

A systematic review and meta-analysis of planned early birth versus expectant management for women with prelabor PROM (PPROM) prior to 37 weeks' gestation found no clinically important difference in the incidence of neonatal sepsis between women who birthed immediately and those managed expectantly.[215] Early planned birth was associated with an increase in the incidence of neonatal respiratory distress syndrome (RDS), need for ventilation, neonatal mortality, endometritis, admission to the neonatal intensive care unit, and the likelihood of birth by cesarean section; however, there was a decreased incidence of chorioamnionitis. Women randomized to early birth also had an increased risk of labor induction but a decreased length of hospital stay.[215] Babies of women randomized to early birth were more likely to be born at a lower gestational age. In women with PPROM before 37 weeks' gestation with no contraindications to continuing the pregnancy, a policy of expectant management with careful monitoring was associated with better outcomes for the mother and baby.[215, 216]

Relatively recent randomized controlled studies[217] and a meta-analysis[218] have shown that antenatal corticosteroids benefits on fetal lung maturity extend beyond 32 weeks' gestation (up to 38 weeks); this led the American College of Obstetricians and Gynecologists and the Society for Maternal-Fetal Medicine to issue guidance statements about extending antenatal steroid use to selected late preterm singleton pregnancies.[219, 220] Studies have not clearly demonstrated that the use of corticosteroids increases the risk of bacterial infection in the fetus.[98]

Magnesium sulfate (MgSO4) is typically given as an obstetric tocolytic; however, its administration also appears to act differentially to modulate infection-associated inflammation in fetal membranes. A 2018 study of human fetal membrane explants indicates that MgSO4 differentially modulates lipopolysaccharide-induced fetal membrane inflammation in a time-dependent manner, partially via modulation of caspase-1 activity.[221] The investigators suggest that MgSO4 may also have utility in prevention of fetal membrane inflammation caused by polymicrobial infection.

Neonatal immunology and the risks created by maternal chorioamnionitis

Newborns are vulnerable to infection because of an immature immune system.[222] Factors that render neonates susceptible to bacterial infections include reduced numbers and/or function of macrophages and dendritic cells in peripheral tissues (eg, lung); lower numbers of neutrophils in the bone marrow storage pool[188] ; decreased immunoglobulin G (IgG) and complement levels, especially in prematurely born infants; an inability to respond to bacterial carbohydrate antigens; an increased percentage of T cells bearing naïve cell surfaces and correspondingly underdeveloped functional behaviors related to foreign antigens; and anatomic and biochemical immaturity of skin and mucosal barriers (eg, lung and gut epithelia) as they relate to local host defenses.

Emerging treatments, such as the use of intravenous immunoglobulins and hematopoietic growth factors, may correct deficiencies of the neonatal immune system.[223]  However, the use of immunotherapy still requires more investigation before these treatments become a standard of care.[224] Specifically, the routine use of intravenous immunoglobulins to treat neonatal sepsis is not established,[225] and the use of granulocyte colony stimulating factor (G-CSF) and granulocyte/macrophage colony stimulating factor (GM-CSF) may have a limited role in managing infected preterm infants with neutropenia.[226] The mainstays of current neonatal intensive care for bacterial sepsis in neonates are prompt recognition of bacterial infection, antimicrobial therapy, and supportive care. (In this review, supportive care is only briefly discussed below. See the Medscape Drugs and Disease article Neonatal Sepsis for a more in-depth care of these critically-ill neonates.)

Treatment of the neonate

Communication between obstetric and pediatric caregivers is essential to recognize neonatal infection. Recognition or suspicion of maternal chorioamnionitis is essential to reducing neonatal morbidity and mortality caused by early-onset bacterial infections in the neonate. Nurses and physicians who care for the mother must communicate their concerns about maternal infection to the nurses and physicians who care for the newborn after birth. Caregivers in the nursery must be critically aware of a neonate's signs and symptoms in relationship to the antepartum and intrapartum history.

Signs and symptoms in the mother that suggest chorioamnionitis and increase the risk of fetal or neonatal infection are described under Physical Examination. Although numerous ways to approach the diagnosis and treatment of neonatal sepsis are recognized, a hands-on assessment is the main key to recognition. The experienced physician or nurse in the nursery may indicate to fellow caregivers that the newborn has a septic appearance.

Depending on the hospital setting and the status of the neonate, a family physician may seek pediatric consultation. Depending on the severity or nature of infection in the neonate in a hospital setting, the pediatrician may seek consultation with a neonatologist, a pediatric infectious disease subspecialist, or both. If organ system failure is present or impending organ system failure (eg, respiratory, cardiovascular, renal) secondary to infection is a concern, the infant should be transferred to an appropriate level 3 or level 4 neonatal intensive care unit (NICU). Transportation to a level 3 or 4 NICU is clearly indicated for extremely premature infants requiring high-frequency oscillatory ventilation or near-term or term neonates (≥ 35 weeks' gestation) who are close to meeting criteria for extracorporeal membrane oxygenation (ECMO).

Diet

Seriously or critically-ill newborns with early-onset bacterial infections often require parenteral nutrition until their condition improves. The use of intravenous lipids during proven bacteremia is the subject of controversy. The fear is that lipid inclusions may interfere with phagocytosis of microbes by hepatic, splenic, or pulmonary macrophages. Infections involving the gastrointestinal tract may need a special approach to enteral nutrition when the feedings are reinstituted.

Activity

Activity and illness is generally related to adults because neonates are typically at rest and are not stressed when seriously or critically ill.

Maternal antibiotic chemoprophylaxis is related to urogenital colonization with group B Streptococcus (GBS). Mothers are screened for GBS-related colonization at 35-37 weeks' gestation. Preterm labor before 35 weeks' gestation indicates that knowledge of GBS-related colonization of the urogenital tract is not immediately available.

The Centers for Disease Control and Prevention (CDC) continues to publish updated guidelines for prevention of perinatal GBS disease.[170, 171]  In 2011, the American Academy of Pediatrics (AAP)[227] and the American College of Obstetricians and Gynecologists (ACOG)[228]  released their recommendations regarding prevention of early-onset GBS disease in newborns on the basis of the 2010 CDC guidelines. The Society of Obstetricians and Gynaecologists of Canada (SOGC) published their recommendations for prevention of early-onset neonatal GBS disease in 2013.[229]

The Centers for Disease Control and Prevention (CDC) continues to publish updated guidelines for prevention of perinatal GBS disease; the most recent one was published in 2010,[170]  with an update in 2012.[171]

Although it was not considered a consensus development conference, in 2016, the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) convened a panel of experts from the Society for Maternal-Fetal Medicine (SMFM), the American College of Obstetricians and Gynecologists (ACOG), and the American Academy of Pediatrics (AAP) for a workshop that issued a report on the evaluation and management of women and newborns with a maternal diagnosis of chorioamnionitis.[2] The panel noted that the term "chorioamnionitis" has been used to label a heterogeneous array of conditions characterized by intrauterine infection, inflammation, or both, with a consequent great variation in clinical practice for mothers and their newborns. Therefore, the panel proposed to replace the term "chorioamnionitis" with a more general, descriptive term: “intrauterine inflammation or infection or both,” abbreviated as “triple I.” The panel also proposed a classification (see the table under Diagnostic Considerations) for triple I as well as recommended approaches for the evaluation and management of pregnant women and their newborns with a diagnosis of triple I. Furthermore, the panel indicated the importance of recognizing that an isolated maternal fever is not synonymous with chorioamnionitis.

Early delivery, supportive care, and antibiotic administration for the mother with chorioamnionitis are discussed in Medical Care. The antibiotics used most often to treat mothers with acute chorioamnionitis are also discussed here.[1]

The treatment of the potentially septic neonate is complex. An overview of the treatment for early-onset neonatal infection is summarized in Medical Care.

Maternal antibiotics for chorioamnionitis

The standard drug treatment in the mother with chorioamnionitis includes ampicillin and an aminoglycoside (ie, usually gentamicin), although clindamycin may be added for anaerobic pathogens.[30] Cefazolin may be used instead of ampicillin for mothers with mild penicillin allergy and clindamycin or vancomycin may be used when infected mothers may be suffering from severe penicillin allergy. In cases involving premature labor or prelabor premature rupture of membranes (PPROM), penicillin or ampicillin is frequently administered as a chemotherapeutic agent to prevent group B streptococcal (GBS) colonization of the fetus. The use of penicillin alone is suggested for GBS chemoprophylaxis during the intrapartum period. Using penicillin rather than ampicillin may avoid colonization of the fetus with ampicillin-resistant E coli. The rationale for ampicillin use when maternal chorioamnionitis is suspected is that ampicillin would treat GBS, Haemophilus species, many enterococci strains, and Lmonocytogenes.

For more information on intrapartum antibiotic use to prevent GBS, see the Medscape Drugs and Diseases topic Bacterial Infections and Pregnancy.

Clindamycin may be used to treat S aureus and anaerobes. Gentamicin provides broad-spectrum coverage against gram-negative bacteria. These antibiotics should be given intravenously. The drugs mentioned above are generally safe for the mother and fetus. An absolute contraindication to using these antibiotics is a known allergic reaction to them. Renal function must always be considered when using antibiotics, especially aminoglycosides. With acute chorioamnionitis, these antibiotics should be used only during the intrapartum period when the mother is febrile. No additional doses are required after vaginal delivery; however, one additional dose of the chosen regimen is indicated following cesarean delivery.[30] The addition of anaerobic coverage after cesarean delivery may be considered (clindamycin or metronidazole) to decrease the risk of endometritis. Alternative regimens for acute chorioamnionitis include monotherapy with ampicillin–sulbactam, piperacillin–tazobactam, cefotetan, cefoxitin, or ertapenem.

If a urinary tract infection is present, the appropriate antibiotic or combination of antibiotics should be used to treat the specific bacterium isolated from the urine.

Erythromycin is infrequently used in women allergic to penicillin. It is not recommended for GBS prophylaxis because significant numbers of GBS strains are erythromycin resistant. Its ability to enter urogenital secretions has been questioned, especially in the treatment of U urealyticum-related or Mycoplasma hominis-related colonization in pregnant women. Of the invasive GBS strains that were isolated in one study, resistance to either clindamycin or erythromycin was in excess of 20%, whereas colonizing isolates of GBS had resistance in more than 40% of cases.[230] A report from the Centers of Disease Control and Prevention (CDC) noted that, of 4882 GBS isolates, 15% and 32% were resistant to clindamycin and erythromycin, respectively.[231] The GBS resistance to erythromycin and clindamycin is global in nature, with several studies reporting this problem from different parts of the world, including China,[232] Brazil,[233] Spain,[234] and Italy,[235] among others. These reports suggest that erythromycin or clindamycin used as chemoprophylaxis to prevent GBS infection in neonates is problematic in women with penicillin allergy. Per CDC guidelines,[170] clindamycin may be used intrapartum only when the isolated GBS is documented to be susceptible to clindamycin; otherwise, vancomycin would be the drug of choice.

Supportive, immune, and antibiotic therapy of early-onset bacterial infection

An extensive discussion of the management of septic neonates is not possible in this article, but it is available in other Medscape Drugs and Diseases chapters (see Neonatal Sepsis). Critical points to ensure intact survival of the neonate are mentioned for completeness. For example, ventilator management and surfactant replacement therapy can be used to treat the neonate with congenital bacterial pneumonia, but a complete discussion of the techniques involved in this therapy are covered in other articles. Physicians and nurses attending the delivery of a newborn whose mother is suspected of having chorioamnionitis should be ready to perform a full resuscitation, including intubation; providing positive-pressure ventilation; and treatment of hypovolemia, shock, and respiratory and/or metabolic acidosis. Low Apgar scores may be another indicator of sepsis.

After initial stabilization of a neonate with potential infection in the delivery room, attention is directed toward the following variables that influence survival:

• Warmth, monitoring of vital signs, maintenance of fluid and electrolyte balance, and correction of significant metabolic acidosis
• Management of the circulation, including correction of hypovolemia, and enhancement of cardiac performance with inotropic drugs if sepsis-related myocardial dysfunction is noted
• Glucose homeostasis
• Treatment of respiratory distress that may entail surfactant replacement (for pneumonia, respiratory distress syndrome) and different modes of assisted ventilation (Inhaled nitric oxide may be considered as a therapy in the presence of pulmonary hypertension.)
• Assessment and treatment of thrombocytopenia and coagulopathy, if present

The aforementioned elements of supportive care are essential to reducing morbidity and mortality. Pulmonary hypertension can complicate the management of neonatal sepsis, and inhaled nitric oxide may reverse this complication. The use of inhaled nitric oxide is an established, FDA-approved therapy for hypoxic respiratory failure in late preterm and term infants; it has been used successfully in many level 3 and level 4 units including use during transport. When myocardial dysfunction, cardiovascular collapse, and severe pulmonary hypertension are not reversible, extracorporeal membrane oxygenation (ECMO) may be a life-saving intervention. In critically ill, septic neonates that are near term or term, the importance of early referral for ECMO cannot be overstated.

Guidelines for immunotherapy in early-onset sepsis (EOS) are not well established. Treatments used include administration of granulocyte colony stimulating factor (G-CSF) or granulocyte/macrophage colony stimulating factor (GM-CSF) (eg, filgrastim, sargramostim),[226] and intravenous administration of immunoglobulin G (IgG),[225] particularly if a high-titer IgG antibody is specifically directed against the bacterial pathogen. Despite extensive research on neonatal immunotherapies, no agreement regarding their use has been reached. A future area of intensive research is maternal immunomodulation therapy for prevention of preterm birth, prematurity-related morbidity, and neonatal infections.[236] An example of such maternal immunomodulation would be the development of a maternal GBS vaccine to prevent invasive GBS disease in the newborn, an issue that is under intensive investigation.[237]

Antibiotic therapy for early-onset bacterial infection of the neonate usually includes the administration of a penicillin (ie, ampicillin is most often used for additional coverage against Haemophilus species, enterococci, and listeriosis) and an aminoglycoside (ie, usually gentamicin). Generally, gentamicin provides ample coverage against gram-negative bacteria that cause EOS. The third-generation cephalosporins should be used as part of the antibiotic regimen if resistant E coli is suspected based on the maternal history, amniotic fluid culture results, and the clinical picture. Cefotaxime has been advocated by many experts when meningitis is suspected or when an asphyxiated infant or an extremely preterm infant is being treated and severe renal dysfunction may occur.

Antibiotic administration in newborns is based on birth-weight criteria and gestational age at birth. Doses of antibiotics change as the postnatal age increases and renal function improves. Administration of aminoglycosides should include changes in dosing based on pharmacokinetics.

Final decisions about antibiotics should be based on positive culture results from appropriate anatomic sites. If renal dysfunction is present, antibiotic dosages should be adjusted during the course of their administration. This is particularly true for aminoglycoside administration in extremely premature newborns and in newborns with urogenital anomalies.

Recommendations on the appropriate antibiotic dose can be found in soft-cover neonatology textbooks (ie, Neonatology: Management, Procedures, On-Call Problems, Diseases, and Drugs or Manual of Neonatal Care) and classic textbooks of neonatal-perinatal medicine. Specific textbooks about antibiotic use in pediatric patients, including neonates (ie, Nelson's Pocketbook of Pediatric Antimicrobial Therapy), have also been written. For this article, we referred to the Lexicomp Pediatric & Neonatal Dosage Handbook, 24th edition.[141] Another source for neonatal drug information is NeoFax, which is no longer available in the print form but is available online through subscription with Micromedex or via an application (app) through a tablet computer or smart phone. The review on antibiotics to treat EOS is not exhaustive.

Lastly, the physician must consider the duration of antibiotic therapy. This is particularly true when deciding the duration of antibiotic treatment for well-appearing term neonates. In the era of managed care, in which cost reductions are typical, discontinuing antibiotics in healthy term neonates within 36-48 hours of initiating therapy is probably safe. With current bacteriologic techniques, more than 90%-95% of neonatal blood cultures become positive within 48 hours of the time they are cultivated. A negative C-reactive protein (CRP) result when reviewed at 48 hours after birth suggests antibiotic treatment can be stopped.

In neonates with proven infection, the well-being of the infected newborn should guide the duration of antibiotic therapy. The bacterium causing the infection and the site of the infection also influence the duration of antibiotic therapy. For example, bacterial pneumonia is often treated for 7-10 days with antibiotics. Bacteremia is often treated with antibiotics for 10-14 days. This duration is based on the potential for recurrence with shorter courses of treatment (ie, 10 days of antibiotics is often considered a minimum for GBS-associated bacteremia).

Cerebrospinal fluid (CSF) infections may require antibiotic therapy for 2-4 weeks based on the bacterium responsible for the infection, findings on an analysis of CSF indicating resolution of the infection, and the presence of complications associated with meningitis. For uncomplicated GBS-related infections of the CSF, 2 weeks may be sufficient; infection with other gram-positive pathogens and all gram-negative bacteria often require 3-4 weeks of antibiotic therapy. Surgical interventions for localized central nervous system infections (eg, an infectious epidural collection, brain abscess) or the presence of postinfectious hydrocephalus may indicate that antibiotic therapy needs to be provided for as long as 4-6 weeks.

Information in the following section reviews the antibiotics that are commonly used to treat early-onset bacterial infections in the neonate. The antibiotics covered are not exhaustive. For example, the use of azithromycin to treat congenital pneumonia caused by U urealyticum or Mycoplasma is not reviewed.

Class Summary

Empiric antimicrobial therapy must be comprehensive and should cover all likely pathogens that cause early-onset sepsis (EOS). Antibiotic combinations are usually recommended for group B Streptococcus (GBS), listeria and serious gram-negative bacillary infections. This approach ensures coverage for a broad range of organisms and polymicrobial infections. In addition, it prevents resistance in bacterial subpopulations and provides additive or synergistic effects. Once organisms and sensitivities are known, the use of antibiotic monotherapy is then recommended. The exception would be bacteria that could be more susceptible to killing with the synergistic action of two antibiotics (eg, enterococcus).

Information about antimicrobials used to treat neonates and the source for this review is the Lexicomp Pediatric & Neonatal Dosage Handbook, 24th edition.[141] Another source for neonatal drug information is NeoFax, which is no longer available in the print form but is available online through subscription with Micromedex or via an application (app) through a tablet computer or smart phone.

Aqueous crystalline penicillin G is considered the first-line agent for GBS. Ampicillin may be used; however, there is concern surrounding the emergence of ampicillin-resistant E coli infections. Other modified penicillins such as oxacillin or nafcillin (antistaphylococcal), netilmicin (antipseudomonal or other gram-negative enteric bacteria), and piperacillin (antipseudomonal) are not typically used as first-line antibiotics for treatment of early-onset neonatal infections. The aforementioned modified penicillins are designed to treat infections caused by penicillin-resistant bacteria that can express beta-lactamase. These modified penicillins are usually reserved for the treatment of postnatally acquired infections in hospitalized neonates. Methicillin-resistant staphylococcal infections are uncommon causes of EOS; neonates with EOS who have these staphylococci reported should be treated with vancomycin.

Penicillin G, aqueous crystalline (Pfizerpen)

Aqueous crystalline penicillin G (pen G) administered IV is the drug of choice for GBS bacteremia or meningitis. Pen G is also known as benzylpenicillin. Do not confuse pen G with benzathine or procaine penicillin used only for IM injections; pen G is the original antibiotic in the penicillin class and inhibits synthesis of the bacterial cell wall. Pen G may provide adequate coverage for S pneumoniae when it is a cause of early-onset bacterial infection in neonates (infrequent) but this bacterium can also have resistance to pen G.

Ampicillin

A more broad-spectrum aminopenicillin used for many years as either a definitive or a prophylactic therapy for early-onset bacterial infection of neonates (ie, GBS, Listeria monocytogenes and susceptible E coli). Ampicillin may provide additional coverage against Haemophilus species, many enterococci, other streptococci, and a limited number of susceptible gram-negative enteric bacteria. It is indicated for neonatal bacteremia or meningitis due to GBS (pen G is the drug of choice). 

Cefotaxime (Claforan)

Cefotaxime is a third-generation cephalosporin with enhanced potency against many gram-negative bacteria. It is generally considered inactive against enterococci, Listeria, and most strains of pseudomonads and bacteroides. Some experts consider this antibiotic the preferred therapy for neonatal meningitis caused by gram-negative bacteria if the bacterium is sensitive to it (and in conjunction with an aminoglycoside). This preference is based on more effective CNS penetration of cefotaxime. Cefotaxime is indicated when aminoglycosides may be contraindicated (eg, significant renal failure) or when aminoglycosides may have enhanced toxicity. 

Gentamicin

Gentamicin is one of the aminoglycoside antibiotics (others include amikacin, netilmicin, kanamycin and tobramycin). Generally, gentamicin has activity against Pseudomonas aeruginosa, whereas kanamycin does not. Gentamicin is the first choice for prophylactic or definitive therapy of early-onset bacterial infections in neonates because it has broad activity against many gram-negative bacilli. Amikacin and tobramycin are usually reserved to treat nosocomial infections caused by gram-negative bacteria that are resistant to gentamicin.

Aminoglycosides should not be used alone to treat infections potentially caused by gram-positive bacteria. Thus, ampicillin is always included in the treatment of early-onset bacterial infections in neonates. Furthermore, to prevent the emergence of highly antibiotic-resistant gram-negative bacteria, nosocomial infections in hospitalized neonates should never be treated with an aminoglycoside alone. A second antibiotic should be administered in addition to the aminoglycoside, and its mechanism of action that causes microbial death should be different from that of the aminoglycoside.

This antibiotic has a black box warning. Elevated blood concentrations of aminoglycosides may cause significant injury to the kidney and the vestibular/auditory nerve. Concurrent use of furosemide or other loop diuretics and use of vancomycin can increase nephrotoxicity. Thus, trough levels of aminoglycosides in neonatal sera must be measured if their use is going to exceed an initial period of prophylaxis (ie, 48 h after birth) to exclude sepsis.

Aminoglycosides demonstrate concentration-dependent killing of bacteria, suggesting a benefit related to higher serum concentrations that are achieved with less-frequent dosing (eg, once daily administration is now the standard of care in term or late-preterm infants).