Bacterial meningitis in the neonate: Treatment and outcome

INTRODUCTION — Bacterial meningitis is more common in the first month than at any other time of life [1]. Despite advances in neonatal intensive care, meningitis in the neonate remains a devastating disease.

The treatment and outcome of bacterial meningitis in the neonate (age <1 month) will be discussed here. The clinical features, diagnosis, and complications of bacterial meningitis are discussed separately, as is bacterial meningitis in older children:

●(See "Bacterial meningitis in the neonate: Clinical features and diagnosis".)

●(See "Bacterial meningitis in the neonate: Neurologic complications".)

●(See "Bacterial meningitis in children older than one month: Clinical features and diagnosis".)

●(See "Bacterial meningitis in children older than one month: Treatment and prognosis".)

SUPPORTIVE CARE — Initial care for all neonates with meningitis should be provided in an intensive care unit setting. Supportive care measures are a crucial part of the management of neonates with bacterial meningitis [2]. These may include:

●Management of cardiovascular instability or shock (algorithm 1) (see "Neonatal shock: Management")

●Provision of oxygen and additional respiratory support as needed (see "Respiratory support, oxygen delivery, and oxygen monitoring in the newborn" and "Overview of mechanical ventilation in neonates")

●Careful fluid therapy, avoiding both hypo- and hypervolemia (see "Fluid and electrolyte therapy in newborns")

●Prevention and management of hypoglycemia (see "Management and outcome of neonatal hypoglycemia")

●Control of seizures (see "Treatment of neonatal seizures")

●Nutritional support (see "Approach to enteral nutrition in the premature infant" and "Parenteral nutrition in infants and children")

ANTIMICROBIAL THERAPY — For neonates whose clinical and initial cerebrospinal fluid (CSF) findings are suggestive of bacterial meningitis (eg, CSF pleocytosis, increased CSF protein and/or decreased CSF glucose, organism present on Gram stain), broad-spectrum antimicrobial therapy should be initiated as soon as possible. An appropriate regimen includes agents that have adequate CSF penetration at appropriate doses to achieve adequate levels in the CSF.

When the results of the CSF and blood cultures, including antimicrobial susceptibilities, are available, antimicrobial therapy is tailored to the specific pathogen. (See 'Definitive therapy' below.)

Empiric therapy — The initial choice of antimicrobials for suspected bacterial meningitis in the neonate is based on the timing of onset (ie, early versus late onset), likely pathogens (table 1), and local susceptibility patterns within the nursery or neonatal intensive care unit and within the community. (See "Bacterial meningitis in the neonate: Clinical features and diagnosis", section on 'Etiology'.)

Often, at the first sign of illness, empiric therapy for neonatal sepsis is initiated as summarized in the table (table 2) and discussed separately. (See "Management and outcome of sepsis in term and late preterm neonates", section on 'Initial empiric therapy'.)

When there is clinical concern for meningitis (eg, critical illness, CSF pleocytosis, organism present on CSF Gram stain, or other suggestive CSF parameters), the empiric regimen should be modified as follows:

●Discontinue the aminoglycoside.

●Add an extended-spectrum cephalosporin (eg, cefotaxime [where available], ceftazidime, or cefepime); ceftriaxone should not be used in neonates, because it displaces bilirubin from albumin binding sites, which might contribute to kernicterus and may precipitate if used with intravenous calcium, leading to severe reactions [3,4].

●In addition, for most neonates with CSF pleocytosis and negative CSF Gram stain, empiric acyclovir therapy for herpes simplex virus is warranted (after performing appropriate testing), as discussed separately. (See "Neonatal herpes simplex virus infection: Management and prevention", section on 'Initial antiviral therapy'.)

Adding an expanded-spectrum cephalosporin to the empiric regimen provides optimal activity in the CSF against ampicillin-resistant gram-negative enteric organisms and pneumococci. High rates of ampicillin resistance among Escherichia coli isolates and a link between maternal intrapartum ampicillin prophylaxis and E. coli resistance have been reported among preterm neonates managed in the neonatal intensive care unit setting [5,6]. However, this not the case in late preterm or term newborn infants, in whom group B streptococcus (GBS) remains the most likely early-onset pathogen [2,5]. Ampicillin resistance is a concern in community-acquired late-onset infections in term and preterm neonates. In one survey of febrile infants <90 days old presenting to an emergency department, nearly 80 percent of infants with meningitis had ampicillin-resistant pathogens [7]. The authors of the study conclude that the initial regimen should contain both ampicillin and an expanded-spectrum cephalosporin because of the risk of GBS and Listeria monocytogenes infection in this age group, as well as for ampicillin-resistant gram-negative enteric organisms. We agree with this practice.

Antibiotic coverage generally should not be narrowed based on the Gram stain results, because they are subject to observer misinterpretation. Empiric broad-spectrum therapy should be continued pending laboratory identification of the organism and susceptibility results (see 'Definitive therapy' below). However, if the Gram stain suggests a pathogen that is not adequately covered with the empiric regimen, the regimen should be adjusted accordingly. For example, if initial empiric therapy consists of vancomycin plus an extended-spectrum cephalosporin and the Gram stain suggests L. monocytogenes, ampicillin should be added. If meningitis resulting from a multidrug-resistant (MDR) gram-negative organism is strongly suspected (eg, when the CSF Gram stain reveals gram-negative bacilli and positive screen for extended-spectrum beta-lactamase), the empiric regimen should substitute meropenem for the cephalosporin [8].

In the neonatal intensive care unit setting, ongoing use of an expanded-spectrum cephalosporin should be restricted to neonates with suspected bacterial meningitis based on clinical findings and CSF parameters. When use of third- or fourth-generation cephalosporins is routine (eg, when they are used more broadly for all neonates treated for "rule out sepsis"), rapid emergence of resistant strains (especially Enterobacter cloacae, Klebsiella pneumoniae, and Serratia species) can occur [9].

Definitive therapy — Once the causative agent and the in vitro antimicrobial susceptibility results are known, empiric antimicrobial therapy should be altered accordingly. Guidance for treating the most common causes of neonatal meningitis is provided below.

Group B streptococcus — GBS is uniformly susceptible to penicillin and ampicillin. Penicillin G monotherapy is the appropriate definitive therapy once the neonate is improving clinically and repeat lumbar puncture (LP) documents CSF sterilization (see 'Repeat lumbar puncture' below). Duration of treatment is 14 to 21 days (see 'Duration' below). Treatment of neonatal GBS infections, including GBS meningitis, is summarized in the table and discussed in detail separately (table 3). (See "Group B streptococcal infection in neonates and young infants", section on 'Definitive therapy'.)

Escherichia coli and other gram-negative organisms — Ampicillin is the agent of choice for neonatal meningitis resulting from ampicillin-susceptible strains of E. coli.

Ampicillin-resistant E. coli and other gram-negative organisms usually are initially treated with a combination of an extended-spectrum cephalosporin (eg, cefotaxime if available) plus an aminoglycoside, usually gentamicin; the aminoglycoside is discontinued once sterility of the CSF is documented. Appropriate monotherapy is continued to complete a minimum of 21 days.

The prevalence of antimicrobial resistance among gram-negative isolates has increased markedly since the 1990s [10]. Resistance in association with production of AmpC beta-lactamases or extended-spectrum beta-lactamases (ESBLs) occurs primarily, but not exclusively, in E. coli, Klebsiella species, and Enterobacter species. Carbapenemase-producing gram-negative organisms, especially Klebsiella pneumoniae, E. coli, and Enterobacter cloacae, have also emerged but are uncommon in newborn infants [8].

Infections caused by noncarbapenemase-producing MDR enteric organisms are generally treated with meropenem. Meropenem should be administered for the entire course of therapy for neonates with meningitis that is caused by MDR gram-negative organisms. The hospital microbiology laboratory should provide assistance for appropriate testing of carbapenemase-producing gram-negative organisms for MDR patterns.

Other pathogens

●Coagulase-negative staphylococci – Vancomycin is the antimicrobial of choice for proven meningitis caused by coagulase-negative staphylococci. These organisms rarely invade the meninges except as a complication of bacteremia accompanying intraventricular hemorrhage in very low birth weight infants (birth weight <1500 g) or as a result of surgical manipulations or placement of a ventriculoperitoneal shunt. Such infections invariably are of late onset.

Sterilization of the CSF usually is achieved promptly after initiation of vancomycin unless shunt material has not been removed. If the CSF is persistently positive, consideration should be given to adding rifampin (5 mg/kg every 12 hours) to vancomycin for synergy. However, it is uncertain whether combination therapy truly improves clearance of the infection. (See "Infections of cerebrospinal fluid shunts", section on 'Staphylococci'.)

●Listeria – The combination of ampicillin and gentamicin is more effective than ampicillin alone in vitro and in animal models of infection, and it is appropriate for initial therapy (table 4). When the CSF has been sterilized and the infant has improved clinically, a 14- to 21-day course of treatment can be completed with ampicillin monotherapy. (See "Treatment and prevention of Listeria monocytogenes infection", section on 'Antibiotic therapy'.)

●Staphylococcus aureus – The standard therapy for methicillin-susceptible S. aureus (MSSA) meningitis is nafcillin or oxacillin [11,12]. Treatment duration is typically 14 days. The preferred therapy for methicillin-resistant S. aureus (MRSA) meningitis is vancomycin [11-13]. Treatment duration is at least 14 days. Some experts suggest adding rifampin to vancomycin for treatment of MRSA infection, but there are no clinical studies to suggest efficacy. Treatment of MSSA and MRSA infections is discussed in greater detail separately. (See "Staphylococcus aureus in children: Overview of treatment of invasive infections", section on 'Definitive antimicrobial therapy'.)

Duration — The duration of antibiotic therapy depends upon the results of CSF and blood cultures, the clinical course, and whether the neonate was pretreated with antibiotics prior to the LP.

●Positive CSF culture – The suggested duration of antibiotic therapy for different causative organisms is as follows:

•GBS or other gram-positive organisms (eg, L. monocytogenes or Enterococcus) – A 14-day course is sufficient for neonates with an uncomplicated course [14].

•E. coli or other gram-negative enteric pathogens – A 21-day course is the minimum for neonates with an uncomplicated course [8].

•A longer course of therapy is required for neonates with meningitis whose course is complicated. Prolonged treatment, sometimes for as long as eight weeks, may be required for neonates with ventriculitis, abscesses, or multiple areas of infarction or hemorrhage with resulting encephalomalacia. (See "Bacterial meningitis in the neonate: Neurologic complications".)

●Negative CSF culture with positive blood culture and CSF pleocytosis – For neonates with CSF pleocytosis and bacteremia but a negative CSF culture (obtained before antibiotic therapy), we usually continue meningeal doses of antimicrobial therapy for 10 days for gram-positive bacteremia (eg, GBS) and 14 days for gram-negative bacteremia.

●Negative CSF and blood cultures – For neonates in whom cultures were obtained before antibiotic therapy and both blood and CSF cultures are negative after 48 hours, we suggest discontinuing antibiotic therapy. Negative cultures in the setting of CSF pleocytosis should prompt consideration of nonbacterial causes of meningitis (eg, herpes simplex virus [HSV], enterovirus). (See "Bacterial meningitis in the neonate: Clinical features and diagnosis", section on 'Differential diagnosis'.)

●Pretreated or LP delayed – Some neonates may be exposed to antibiotics prior to undergoing LP (eg, because the LP is delayed due to clinical instability, because the mother received intrapartum antibiotic prophylaxis, or because the infant was receiving antibiotics for another reason [eg, prophylaxis for vesicoureteral reflux]). This can result in negative CSF culture. However, in most cases, other CSF parameters, (eg, cell count and protein concentration) will permit accurate diagnosis so long as the LP is not traumatic. (See "Bacterial meningitis in the neonate: Clinical features and diagnosis", section on 'Interpretation of cerebrospinal fluid'.)

Our approach to managing pretreated neonates is as follows:

•For neonates in whom the LP was delayed because of clinical instability, meningeal doses of antimicrobial therapy should be continued until the LP can be safely performed.

•The total duration of therapy in pretreated neonates depends upon the CSF evaluation and blood culture result:

-Neonates with CSF pleocytosis and positive blood culture are treated for 10 days for gram-positive infection (eg, GBS) and 14 days for gram-negative infection.

-For neonates with normal CSF profile and negative blood and CSF cultures, we usually discontinue antibiotic therapy when cultures are sterile after 48 hours.

-For neonates who have a CSF pleocytosis and negative blood and CSF cultures, we individualize the duration of meningitic doses of antimicrobial therapy based on clinical parameters, including whether there is a nonbacterial (eg, HSV, enterovirus) or noninfectious explanation for the pleocytosis (eg, intraventricular hemorrhage). (See "Bacterial meningitis in the neonate: Clinical features and diagnosis", section on 'Differential diagnosis'.)

ADJUNCTIVE THERAPY — Adjunctive immunomodulatory therapy is not a routine part of management of neonatal meningitis. Though there is evidence from experimental models that immune modulation may positively impact the outcome of neonatal meningitis, the modalities that have undergone clinical investigation thus far have either been shown to be ineffective or they have not been adequately studied.

In particular, we suggest not using glucocorticoid therapy to reduce the risk of neurologic sequelae in neonatal meningitis. There are limited data on the use of glucocorticoids in this setting. In a clinical trial involving 52 neonates randomly assigned to dexamethasone plus antibiotic therapy or antibiotic therapy alone, rates of mortality, neurologic disability, and hearing loss at two years were similar in both groups [15]. In an observational study of 263 infants <90 days old with bacterial meningitis, glucocorticoid therapy was not associated with lower mortality; however, only 8 percent of patients in this cohort received glucocorticoids [16]. The role of dexamethasone therapy in older infants and children with bacterial meningitis is controversial and is discussed separately. (See "Bacterial meningitis in children: Dexamethasone and other measures to prevent neurologic complications".)

Other adjunctive therapies for neonatal sepsis are discussed separately. (See "Management and outcome of sepsis in term and late preterm neonates", section on 'Adjunctive therapies' and "Treatment and prevention of bacterial sepsis in preterm infants <34 weeks gestation", section on 'Adjunct therapy to antibiotics'.)

MONITORING RESPONSE TO THERAPY

Ongoing evaluation — The response to therapy and the potential development of complications are monitored with:

●Serial neurologic examinations.

●Assessment of the overall clinical status (eg, vital sign trends, temperature stability, need for hemodynamic or respiratory support).

●Repeat blood cultures – In bacteremic neonates, a repeat blood culture should be performed to document sterility of the blood stream. The follow-up blood culture is usually obtained at the time when the initial blood culture is reported as positive.

●Repeat examination of the cerebrospinal fluid (CSF). (See 'Repeat lumbar puncture' below.)

●Neuroimaging (See 'Neuroimaging' below.)

Most neonates with uncomplicated bacterial meningitis have clinical improvement within 24 to 48 hours of receiving appropriate antibiotic therapy. Clinical deterioration or failure to improve in this timeframe may suggest development of a complication (eg, obstructive ventriculitis, subdural effusion, brain abscess, intraventricular hemorrhage) or inadequate antimicrobial therapy. (See "Bacterial meningitis in the neonate: Neurologic complications".)

Repeat lumbar puncture — We suggest repeating the lumbar puncture (LP) 24 to 48 hours after initiating antimicrobial therapy for neonates with meningitis caused by group B streptococcus (GBS), E. coli, other gram-negatives, and Listeria and for neonates with a complicated course [11,14]. This encompasses the majority of patients with neonatal meningitis.

Reevaluation of the CSF 24 to 48 hours after initiation of antimicrobial therapy is important for several reasons [17,18]:

●In severe cases, gram-negative organisms may persist for several days. Delayed sterilization of the CSF is associated with an increased risk of developing neurologic sequelae (see 'Outcome' below). This clinical scenario is uncommon with the routine use of an extended-spectrum cephalosporin as initial empiric therapy. By contrast, gram-positive bacteria usually clear from the CSF rapidly (within 24 hours) after initiation of appropriate antimicrobial therapy unless there is high bacterial burden in the CSF.

●The persistent identification of organisms on a Gram stain may be an early indication of inadequacy of antimicrobial therapy (eg, the organism is not susceptible to the concentration of antibiotic that is attained in the CSF).

●Persistence of viable organisms more than 48 hours after initiation of antimicrobial therapy is an indication for diagnostic neuroimaging because it can indicate a purulent focus (eg, obstructive ventriculitis) that can require additional intervention or increased duration of antimicrobial therapy. (See 'Neuroimaging' below and "Bacterial meningitis in the neonate: Neurologic complications".)

●Sterilization of the CSF is a criterion for discontinuing combination therapy for some pathogens (eg, GBS, Listeria). (See 'Definitive therapy' above and "Group B streptococcal infection in neonates and young infants", section on 'Definitive therapy'.)

In uncomplicated neonatal meningitis, repeat CSF culture should generally be sterile. A positive culture obtained 48 hours after initiation of therapy raises a concern for obstructive ventriculitis or intraventricular hemorrhage. The additional evaluation and management of such infants should be individualized and undertaken in consultation with specialists in pediatric infectious diseases and pediatric neurosurgery. (See "Bacterial meningitis in the neonate: Neurologic complications", section on 'Ventriculitis'.)

Neuroimaging — We perform neuroimaging (typically with magnetic resonance imaging [MRI]) 48 to 72 hours before the anticipated end of therapy in all neonates with confirmed bacterial meningitis, even those with an apparently uncomplicated course. Neuroimaging may be warranted earlier in the course for neonates with signs suggesting neurologic complications. Neurologic complications should be considered if the neonate fails to improve clinically after 24 to 48 hours of appropriate antibiotic therapy. (See "Bacterial meningitis in the neonate: Neurologic complications".)

MRI is preferred over contrast-enhanced computed tomography (CT) because MRI provides better anatomic detail, optimizes assessment of injury to white matter, and avoids radiation exposure [19]. If there are focal findings that require extension of the course of antimicrobial therapy, treatment can be continued without interruption.

●Cranial ultrasound – Early in the course of meningitis, cranial ultrasound is the most practical neuroimaging technique, as it can be performed at the bedside. It is helpful for assessing ventricular size and the presence of intraventricular hemorrhage. Cranial ultrasound also can demonstrate ventriculitis, echogenic sulci, abnormal parenchymal echogenicities, and extracerebral fluid collections [20,21]. In addition, because serial ultrasound studies can be performed safely at the bedside, it is useful in defining the progression of complications in infants with prolonged seizure activity or focal neurologic deficits [22]. (See "Bacterial meningitis in the neonate: Neurologic complications".)

●MRI – Similar to ultrasonography, early in the treatment course, MRI can demonstrate cerebral edema and ischemic lesions effectively, as well as leptomeningeal enhancement, ventriculitis, and infarction. Pus within the lateral ventricles can be seen as restricted diffusion. Later in the treatment course, contrast-enhanced MRI is useful in detecting cerebral abscesses, persistent cerebritis, areas of infarction or encephalomalacia, subdural empyema, hydrocephalus and degree of cerebral cortical, and white matter injury. (See "Bacterial meningitis in the neonate: Neurologic complications".)

These findings may influence duration of antimicrobial therapy. (See 'Duration' above.)

Contrast-enhanced neuroimaging (MRI or CT) is integral to the care of all neonates with meningitis caused by organisms that have a propensity for formation of intracranial abscesses. These include Citrobacter koseri, Serratia marcescens, Proteus mirabilis, and Cronobacter sakazakii [23-28]. (See "Bacterial meningitis in the neonate: Neurologic complications", section on 'Brain abscess'.)

FOLLOW-UP — Long-term follow-up for survivors of neonatal meningitis includes monitoring of hearing, vision, and developmental status. Hearing should be evaluated by auditory brainstem response within four to six weeks of completing therapy (or when feasible in extremely preterm neonates) [29]. (See "Hearing loss in children: Screening and evaluation".)

Survivors of neonatal meningitis are at risk for developmental delay and may be eligible to receive early intervention services in the United States (eligibility criteria vary by state). Appropriate referrals should be made as indicated. Developmental surveillance should continue throughout childhood. (See "Developmental-behavioral surveillance and screening in primary care", section on 'Approach to surveillance'.)

OUTCOME — Neonatal meningitis is a devastating disease. Advances in infant intensive care have reduced mortality, but morbidity remains high.

●Mortality and disability – In the contemporary era, mortality from neonatal meningitis is approximately 10 percent [1,16,30-34]. However, survivors remain at high risk for neurologic sequelae and lifelong impairment, as discussed below [31,33-35]. Approximately 15 to 20 percent of survivors have moderate to severe disability, and approximately 30 to 35 percent have mild disability. (See "Bacterial meningitis in the neonate: Neurologic complications".)

In a review of 101 term and late preterm neonates (ie, gestational ≥35 weeks) diagnosed with bacterial meningitis between 1979 and 1998, mortality declined from 17 percent in the early era (1979 to 1988) to 9 percent in the later era (1989 to 1998) [36]. Among survivors, 19 percent had moderate or severe disability at one year of age (defined as severe cerebral palsy, moderate to severe developmental delay, blindness, and/or deafness).

●Prognostic factors – Factors predictive of death or serious adverse sequelae from bacterial meningitis include [32,33,36-44]:

•Low birth weight (<2500 g) or preterm birth (<37 weeks gestation)

•History of clinical signs for >24 hours before hospitalization

•Leukopenia (white blood cell <5000/microL) and neutropenia at presentation

•Very high cerebrospinal fluid (CSF) protein (>300 mg/dL) and/or very low CSF glucose (<10 percent of blood glucose value)

•Seizures continuing >72 hours after hospitalization

•Focal neurologic deficits noted during the acute illness

•Requirement for mechanical ventilation or inotropes

•Delayed sterilization of the CSF

Neuroimaging findings of meningeal inflammation are generally not predictive of neurologic outcome, but the presence and size of parenchymal lesions (eg, thrombi, encephalomalacia) do have prognostic significance. In particular, abscess formation is associated with neurologic sequelae [21].

●Outcomes in preterm neonates – The outcome is generally worse in preterm low birth weight infants compared with term infants [33,35]. In one study of 113 infants with bacterial meningitis, long-term motor disability (spasticity or paresis) was noted in 27 percent of preterm infants compared with 10 percent of term infants [33].

Similarly, outcomes for preterm neonates with bacterial meningitis are generally worse compared with preterm neonates without meningitis. In one study of very low birth weight infants (birth weight <1500 g), after controlling for birth weight, intraventricular hemorrhage, chronic lung disease, and social risk factors, survivors of meningitis had a twofold higher risk of major neurologic disability (eg, cerebral palsy or abnormal tone, hydrocephalus, blindness, deafness, and severe developmental delay) at 20 months of age [45]. In a large cohort study of extremely low birth weight (birth weight <1000 g) infants, those with neonatal meningitis were more likely than uninfected infants to have cerebral palsy, neurodevelopmental impairment, and low mental developmental index scores [46]. Long-term neurologic outcome in preterm infants is discussed in greater detail separately. (See "Long-term neurodevelopmental impairment in infants born preterm: Risk assessment, follow-up care, and early intervention", section on 'Predicting outcome'.)

●Outcomes by causative organism – Outcomes vary according to the causative organism:

•Group B streptococcus (GBS) – Reported mortality rates of neonatal GBS meningitis range from 5 to 11 percent [39,46-48]. Long-term neurologic sequelae occur in approximately 20 to 40 percent of survivors [49,50].

In a study of 90 neonates diagnosed with GBS meningitis from 1998 through 2006, 5 percent of patients died acutely and an additional 5 percent died by three years of age [48]. Among the survivors, 56 percent had age-appropriate development, 25 percent had mild-to-moderate impairment, and 19 percent had severe impairment.

Outcomes of GBS infection in neonates are discussed in greater detail separately. (See "Group B streptococcal infection in neonates and young infants", section on 'Outcome'.)

•E. coli – In a report of 325 young infants (71 percent were neonates, 35 percent were preterm) diagnosed with E. coli meningitis from 2001 to 2013, the overall mortality rate was 9 percent [32]. Mortality was approximately threefold higher in preterm infants compared with term infants (17 versus 5 percent, respectively). Short-term morbidities included seizures (13 percent), subdural empyema (6 percent), intraventricular hemorrhage or hydrocephalus (5 percent), and cerebral venous thrombosis or stroke (3 percent). Long-term morbidities were not described. Preterm birth and severe hypoglycorrhachia (ie, CSF glucose <10 percent of blood glucose level) were the strongest predictors of death.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Sepsis in neonates" and "Society guideline links: Bacterial meningitis in infants and children".)

SUMMARY AND RECOMMENDATIONS

●Supportive care – Neonates with bacterial meningitis should receive initial care in an intensive care unit. Supportive care may include (see 'Supportive care' above):

•Management of cardiovascular instability or shock (see "Neonatal shock: Etiology, clinical manifestations, and evaluation")

•Appropriate respiratory support as needed (see "Respiratory support, oxygen delivery, and oxygen monitoring in the newborn" and "Overview of mechanical ventilation in neonates")

•Careful fluid therapy, avoiding both hypo- and hypervolemia (see "Fluid and electrolyte therapy in newborns")

•Prevention and management of hypoglycemia (see "Management and outcome of neonatal hypoglycemia")

•Control of seizures (see "Treatment of neonatal seizures")

•Nutritional support (see "Approach to enteral nutrition in the premature infant" and "Parenteral nutrition in infants and children")

●Empiric antimicrobial therapy – Broad-spectrum antimicrobial therapy should be initiated as soon as possible in neonates with suspected meningitis. The choice of the initial empiric regimen is based on the likely pathogens (table 1) and local susceptibility patterns. (see 'Empiric therapy' above):

•Initial empiric coverage for neonatal sepsis – Often, at the first sign of illness, empiric therapy for neonatal sepsis is initiated as summarized in the table (table 2) and discussed separately. (See "Management and outcome of sepsis in term and late preterm neonates", section on 'Initial empiric therapy'.)

•Adjusting the empiric coverage once meningitis is suspected – For neonates with clinical evidence of meningitis (eg, critical illness, cerebrospinal fluid [CSF] pleocytosis, organism present on CSF Gram stain, or other suggestive CSF parameters), we suggest modifying the empiric regimen by substituting an extended-spectrum cephalosporin (eg, cefotaxime [where available], ceftazidime, or cefepime) for the aminoglycoside (Grade 2C).

•Coverage for herpes simplex virus (HSV) – For most neonates with CSF pleocytosis and negative CSF Gram stain, empiric acyclovir therapy for HSV is warranted (after performing appropriate testing), as discussed separately. (See "Neonatal herpes simplex virus infection: Management and prevention", section on 'Initial antiviral therapy'.)

●Definitive antibiotic therapy – Once the causative agent and its in vitro antimicrobial susceptibility pattern are known, empiric antimicrobial therapy should be altered accordingly (see 'Definitive therapy' above):

•Group B streptococcus (GBS) – Treatment of neonatal GBS infections, including GBS meningitis, is summarized in the table and discussed in detail separately (table 3). (See "Group B streptococcal infection in neonates and young infants", section on 'Definitive therapy'.)

•Escherichia coli and other gram-negatives – Treatment depends on the susceptibility pattern. Ampicillin is used for ampicillin-susceptible strains of E. coli. Ampicillin-resistant organisms usually are initially treated with a combination of an extended-spectrum cephalosporin plus an aminoglycoside (eg, gentamicin); the aminoglycoside is discontinued once sterility of the CSF is documented. Infections caused by multidrug resistant pathogens (eg, AmpC beta-lactamase or extended-spectrum beta-lactamase producing organisms) are treated with meropenem. Treatment for carbapenemase-producing gram-negative organisms should be based upon susceptibility testing by the microbiology laboratory. Treatment duration is for a minimum of 21 days. (See 'Escherichia coli and other gram-negative organisms' above and 'Duration' above.)

•Other – Antibiotic regimens for other causes of neonatal meningitis are summarized above. (See 'Other pathogens' above.)

●Monitoring – The response to therapy and the potential development of complications are monitored clinically, through serial neurologic examinations, repeat lumbar puncture (LP), and neuroimaging (see 'Monitoring response to therapy' above):

•Most neonates should undergo repeat LP 24 to 48 hours after initiation of antibiotic therapy to document sterilization of the CSF. (See 'Repeat lumbar puncture' above.)

•We perform neuroimaging (typically with magnetic resonance imaging [MRI]) 48 to 72 hours before the anticipated discontinuation of antimicrobial therapy in all neonates with bacterial meningitis, even in those with an apparently uncomplicated course. Neuroimaging may be warranted earlier in the course for neonates with signs suggesting neurologic complications. (See 'Neuroimaging' above and "Bacterial meningitis in the neonate: Neurologic complications".)

●Outcome – In the contemporary era, the mortality of neonatal bacterial meningitis is approximately 10 percent. Approximately 15 to 20 percent of survivors have moderate to severe disability, and approximately 30 to 35 percent have mild disability. (See 'Outcome' above.)

●Follow-up – Long-term follow-up for survivors of neonatal meningitis includes monitoring of hearing, visual acuity, and developmental milestones. (See 'Follow-up' above.)