Bacterial meningitis

OVERVIEW: What every practitioner needs to know

Are you sure your patient has bacterial meningitis? What are the typical findings for this disease?

Bacterial meningitis represents an inflammation of the meninges affecting the pia, arachnoid, and subarachnoid space and develops in response to bacteria and bacteria products in the brain. Suspected bacterial meningitis is a medical emergency; thus, immediate steps must be taken to establish the specific diagnosis, and empirical antimicrobial treatment must be started rapidly.

Patients at risk of higher mortality and morbidity include newborns, those living in low-income countries, those infected with gram-negative bacilli and Streptococcus pneumoniae, and those with delayed diagnosis and therapy.

The symptoms and signs of bacterial meningitis depend on the age of the child, the duration of illness and the host response to infection.

The clinical features of bacterial meningitis in infants and young children can be subtle, variable, nonspecific, or even absent. In infants, they might include fever, hypothermia, lethargy, irritability, poor feeding, vomiting, diarrhea, respiratory distress, seizures, or bulging fontanelles.

Continue Reading

In older children, clinical features might include fever, headaches, stiff neck, photophobia, nausea, vomiting, confusion, lethargy, or irritability. Other signs of bacterial meningitis include the Kernig sign, the Brudzinski sign, focal neurologic deficit, seizures, and increased intracranial pressure.

Signs of meningeal irritation are present in about 75% of children with bacterial meningitis at the time of presentation. Absence of meningeal irritation in children with bacterial meningitis is more common in those younger than 12 months.

The constellation of systemic hypertension, bradycardia, and respiratory depression (Cushing triad) is a late sign of increased intracranial pressure.

What other disease/condition shares some of these symptoms?

Viral meningitis and fungal meningitis can have some of the symptoms and signs. These symptoms and signs can also occur in children with intracranial complications of sinusitis and mastoiditis (e.g., epidural and subdural abscess, brain abscess, otic encephalopathy), as well as in children with bacterial sepsis without meningitis, those with pneumonia, particularly in the upper lobe, children with febrile seizure, those who have ingested toxins, and patients with chemical- or drug-induced meningitis, brain tumors, or rheumatologic or metabolic disorders.

What caused this disease to develop at this time?

The age of the patient, vaccination status, and specific risk factors (e.g., splenectomy, skull fracture with cerebrospinal fluid [CSF] leakage, cochlear implant) are important in predicting the likely pathogens.

In neonates, group B Streptococcus, Escherichia coli, and Listeria monocytogenes account for most cases of acute bacterial meningitis, whereas Streptococcus pneumoniae, and Neisseria meningitidis are predominant in children older than 3 months.

The absence of an opsonic or bactericidal antibody is a major risk factor in most cases of meningitis caused by group B streptococci, E. coli, Haemophilus influenza type b (Hib), S. pneumoniae, and
N. meningitidis.

The advancement of vaccine design in enhancing immunogenicity has been shown to be important in preventing meningitis caused by Hib, S. pneumoniae, and N. meningitidis. Since introduction of Hib and pneumococcal polysaccharide-protein conjugate vaccines for infants, N. meningitidis and nonvaccine types of
S. pneumoniae have become the leading causes of bacterial meningitis in children. For example, routine immunization in young infants and children with Hib conjugate vaccines has virtually eradicated meningitis due to Hib in the United States.

Invasive Hib disease occurs in the United States primarily in underimmunized children and in infants too young to have completed the primary immunization series.
Hib, however, remains an important pathogen in resource-limited countries where vaccines are not available routinely.

Introduction of the 7-valent (now 13-valent) pneumococcal conjugate vaccine (PCV) has led to a substantial reduction in the incidence of pneumococcal meningitis due to vaccine serotypes in infants and children younger than 5 years.

Protein-conjugated N. meningitidis vaccines have also been shown to be highly immunogenic in young infants and adolescents and are currently licensed for children older than 2 years in the United States. Recent data, however, demonstrate that invasive N. meningitidis disease is very low (annual incidence of ~0.3/100,000 population in the United States ), although the reasons for this decline remain unclear.

High-risk factors for invasive pneumococcal disease include children with sickle cell disease, congenital or acquired asplenia or splenic dysfunction, HIV infection, cochlear implants, and children younger than 2 years of age who are indigenous Alaskans, and some Native American populations.

Individuals with complement dysfunction are at an increased risk for meningitis caused by
S. pneumoniae and N. meningitidis, and outbreaks of meningococcal meningitis have occurred in freshmen living in dormitories.

What laboratory studies should you request to help confirm the diagnosis? How should you interpret the results?

Bacterial meningitis requires early diagnosis and empirical antimicrobial treatment. The symptoms and signs as well as CSF findings depend on the age and condition of the child, the duration of illness, and the host response to infection.

CSF examination is of paramount importance for the diagnosis of all forms of meningitis. Patients with suspected meningitis should receive a lumbar puncture (LP) after a central nervous system mass lesion has been ruled out on clinical grounds or by computed tomography (CT) of the head, and if there is no cardiopulmonary compromise.

Evidence for mass lesions will include focal neurologic signs and evidence of increased intracranial pressure. CSF pressure should be recorded during the LP.

A CSF profile consistent with bacterial meningitis includes high white blood cell counts (pleocytosis, with predominant polymorphonuclear leukocytes), high protein levels (indicative of the increased blood-brain permeability and/or defective CSF flow), and lower glucose concentrations relative to blood glucose concentrations. CSF cell count and differential and concentrations of protein and glucose are helpful in the differential diagnosis of various forms of meningitis (Table I).

Table I.
Organismn Opening pressure(cm H2O) WBCs(µL in CSF) Glucose (mg/dL) Protein (mg/dL)
GBS, E. coli, S. pneumoniae,N. meningitidis >20 >1000 <10 >100
M. tuberculosis >20 100-500 10-45 >100
Borrelia, Treponema <20 5-500 10-45 50-150

GBS = group B streptococci

A Gram stain of CSF will show whether bacteria are present, and a positive Gram stain is indicative of higher bacterial counts in CSF. Gram staining is positive in about 90% of children with pneumococcal meningitis, about 80% of children with meningococcal meningitis, half of patients with gram-negative bacillary meningitis, and a third of patients with Listeria meningitis.

Cytospin centrifugation increases the chances of detecting organisms in Gram-stained CSF. A low CSF white blood cell count with positive Gram staining is a risk factor for an unfavorable outcome.

CSF culture results can be negative in children who receive antibiotic treatment before CSF examination. For example, complete sterilization of N. meningitidis from CSF occurred within 2 hours of giving a parenteral third-generation cephalosporin, and the beginning of sterilization of S. pneumoniae from CSF took place by 4 hours into treatment.

In such children, increased CSF white blood cell counts and increased CSF protein concentration are usually sufficient to establish the diagnosis of bacterial meningitis. Blood cultures or nonculture diagnostic tests might help in identifying the infecting pathogen.

Nonculture methods

Nonculture tests should be considered for patients who need earlier identification of pathogens or who have previously received antibiotics. Such tests include latex agglutination, polymerase chain reaction (PCR), and immunochromatography.

Latex agglutination uses latex beads absorbed with microbe-specific antibodies. In the presence of homologous antigen there is visible agglutination of the antibody-coated latex beads. Latex agglutination assays have been sensitive for detecting Hib antigen but are less sensitive with N. meningitidis antigen.

In the multicenter pneumococcal meningitis surveillance study, latex agglutination was positive in 49 of 74 (66%) CSF samples that grew S. pneumoniae, and in 4 of 14 (28%) CSF samples that produced culture-negative results.

The use of standard or sequential-multiplex PCR has been shown to be useful in identification of infecting pathogens in patients who have previously received antibiotics or in resource-poor settings.

Multiplex real-time PCR or broad-range PCR aimed at the 16S ribosomal RNA gene of eubacteria is promising for the detection of pathogens from CSF.

The detection rate was substantially higher with PCR than with cultures in patients who had previously received antibiotics. However, the limit of detection differs between assays.

Real-time PCR has been shown to detect as few as two copies of E. coli,
N. meningitidis, and S. pneumoniae, 16 copies of
L. monocytogenes, and 28 copies of group B streptococci, whereas the sensitivity for broad-range 16S ribosomal DNA PCR was about 10-200 organisms/mL of CSF. The time needed for the whole process from DNA extraction to the end of real-time PCR was 1.5 hours, an attractive time frame for its application in clinical practice.

A rapid test for S. pneumoniae was evaluated in 122 patients with pneumonococcal meningitis. Compared with CSF culture (sensitivity of 71%) and latex agglutination (86% sensitivity), immunochromatography was 100% sensitive and specific for the diagnosis of pneumococcal meningitis, suggesting that immunochromatography might be useful in the diagnosis of pneumococcal meningitis.

Would imaging studies be helpful? If so, which ones?

Imaging is usually not needed for the diagnosis or management of acute bacterial meningitis. CT and magnetic resonance imaging (MRI) are the two imaging modalities commonly used in the setting of suspected bacterial meningitis.

CT is widely available and is useful for rapid assessment of hydrocephalus, mass lesions, hemorrhage, or acute brain edema before LP. In contrast, MRI is required to detect more subtle findings. MRI is more sensitive for assessing CSF involvement, leptomeningitis, empyema, ventriculitis, and infarctions, but MRI is not as widely available and can be logistically challenging to obtain in the acutely ill patient.

Noncontrast CT and MRI can be normal in early cases of bacterial meningitis. Administration of contrast may be helpful to detect meningeal enhancement, with MRI being more sensitive than CT. Meningeal enhancement, however, is not specific to the diagnosis of bacterial meningitis and can be seen with other diagnoses, such as leptomeningeal carcinomatosis.

Fluid-attenuated inversion recovery weighted MRI can demonstrate high signal in the subarachnoid space, reflecting high protein content in the CSF, but high signal in the subarachnoid space can also be seen with leptomeningeal carcinomatosis and subarachnoid hemorrhage. The findings of rhombencephalitis suggest Listeria monocytogenes as the likely causative organism.

The most important role of neuroimaging is to identify potential complications of bacterial meningitis, such as infarction, hydrocephalus, ventriculitis, brain empyema, abscess, and venous sinus thrombosis. MRI is the method of choice for the detection of venous sinus thrombosis, with a high signal intensity seen on spin echo sequences in the venous sinuses, reflecting thrombus formation.

Confirming the diagnosis

The ability to distinguish between bacterial and nonbacterial aseptic meningitis in infants and children in the emergency department could contribute to limiting hospital admissions or unnecessary use of antibiotics, but still remains a challenge.

The bacterial meningitis score has been developed for assessing infants and children with meningitis, and outpatient management might be considered for children who have pleocytosis (≥7 white blood cells/µL of CSF) and none of the following five criteria on presentation: history of a seizure with the illness, blood neutrophil count of at least 104/µL, positive CSF Gram staining, CSF protein level of at least 80 mg/dL, or CSF neutrophil count of at leat 103/µL of CSF.

This proposed diagnostic tool did not achieve 100% sensitivity, however. For example, five patients with bacterial meningitis who had pleocytosis were found to have a bacterial meningitis score that indicated low risk, and 5.5% of meningitis cases were without pleocytosis.

Because bacterial meningitis is defined as an inflammation that develops in response to bacteria and bacterial products, patients with CSF culture positivity without pleocytosis or increased CSF protein concentrations are likely to be in the early stages of bacterial meningitis.

If you are able to confirm that the patient has bacterial meningitis, what treatment should be initiated?

In patients considered high risk for herniation on the basis of clinical features (e.g., focal neurologic deficit, seizures, papilledema), antibiotics and interventions to control intracranial pressure, followed by CT without LP, are advised. In the case of obtaining CT before LP, this should not delay the first dose of antibiotics after blood has been obtained for culture.

Antimicrobial therapy

Eradication of the infecting organism from the CSF is entirely dependent on bactericidal antibiotics; therefore, antibiotics should be administered intravenously at the highest clinically validated doses to patients with suspected bacterial meningitis. A delay in antibiotic treatment is associated with adverse outcomes.

In patients with suspected bacterial meningitis for whom immediate LP is delayed because of a pending brain imaging study or the presence of disseminated intravascular coagulation, blood cultures must be obtained and antimicrobial treatment should be initiated immediately.

Selection of empirical antimicrobial regimens is designed to treat the likely pathogens, based on the age of the patient and specific risk factors (Table II), with modifications if CSF Gram staining results are positive. Treatment of neonates with L. monocytogenes meningitis must include a synergistic regimen containing ampicillin and an aminoglycoside (e.g., gentamicin).

Table II.
Age Likely Pathogensnn Antimicrobial Regimen
Less than 1 month GBS, E. coli, L. monocytogenes(neonatal pathogens) Ampicillin (50-100 mg/kg every 6 hours) plus gentamicin (2.5 mg/kg every 8 hours), or cefotaxime (50 mg/kg every 6-8 hours for suspected gram-negative bacilli
1-3 months Neonatal pathogens,S. pneumoniae, N. meningitis Ampicillin (50-100 mg/kg every 6 hours) plus cefotaxime (75 mg/kg every 6-8 hours) or cefriaxone (50 mg/kg every 12 hours), or vancomycin (20 mg/kg every 6 hours) for suspected pneumococcal meningitis
3 months- 21 years S. pneumoniae, N. meningitis Cefotaxime (75 mg/kg every 6-8 hours) or cefriaxone (50 mg/kg every 12 hours) plus vancomycin (20 mg/kg every 6-8 hours), or rifampin (10 mg/kg every 12 hours) if dexamethaone is administered

GBS = group B streptococci

Antibacterial killing activity in CSF also depends on the bacterial burden at the start of treatment. The antimicrobial susceptibility tests (e.g., determinations of minimal inhibitory concentrations and minimal bactericidal concentrations) are conducted in laboratories by using a bacterial inoculum size of 104-105 organisms/mL.

Some patients with bacterial meningitis (e.g., group B streptococci and S. pneumoniae) who have many organisms on CSF Gram staining are likely to yield 107-10
8 organisms/mL, and minimal inhibitory concentrations can be many times higher than would normally be expected. Careful clinical and laboratory monitoring of the response to antimicrobial treatment (e.g., repeat CSF examination) is therefore warranted for patients with bacterial meningitis who have high bacterial burden on the basis of initial CSF Gram stain.

It is also important to point out that antimicrobial susceptibility patterns must be established for all organisms isolated from the CSF. For example, group B Streptococcus is commonly responsible for neonatal bacterial meningitis and has been shown to be uniformly susceptible to β-lactam antibiotics (e.g., penicillin, MIC of 0.1 μg/mL or less), and thus penicillin is at present the drug of choice for invasive group B streptococcal infection, including meningitis. However, clinical isolates of group B streptococci with penicillin minimal inhibitory concentrations of 0.12-1.0 μg/mL have been reported, and those isolates had mutations in the target penicillin-binding proteins similar to the mechanisms involved in penicillin-resistant S. pneumoniae.

Careful monitoring of antimicrobial susceptibility patterns must be carried out for all organisms isolated from the CSF. Similarly, penicillin has been the standard treatment for meningococcal meningitis, but penicillin resistance has evolved, with an implication of treatment failures.

Adjunctive therapy

Dexamethasone given shortly before or when antibiotics are first given has been shown to reduce the rate of hearing loss in children with Hib meningitis, but its beneficial effects on hearing and other neurologic sequelae are not as clear against meningitis caused by other organisms.

The American Academy of Pediatrics Committee on Infectious Diseases suggests that dexamethasone treatment might be considered for infants and children older than 6 weeks with pneumococcal meningitis after considering the potential benefits and possible risks.

The widespread use of dexamethasone in children with bacterial meningitis needs careful monitoring of clinical (e.g., fever curve, resolution of symptoms and signs) and bacteriologic responses to antimicrobial treatment, particularly for patients with meningitis caused by pneumococci that are resistant to third-generation antibiotics, in whom bacteriologic killing in the CSF depends on vancomycin.

Monitoring of the clinical response (e.g., fever curve) can be complicated by the use of dexamethasone. For example, secondary fever (recurrence of fever after at least 24 hours without fever) happens more commonly in patients treated with dexamethasone than in those who are not (52% versus24%).

In addition, concomitant administration of dexamethasone and vancomycin can reduce penetration of vancomycin into the CSF by virture of the antiinflammatory activity of dexamethasone, resulting in treatment failure. However, CSF bactericidal activity has been shown in children who have meningitis due to cephalosporin-resistant pneumococci, and such cases can be treated with dexamethasone as well as vancomycin and ceftriaxone.

Another issue with adjunctive dexamethasone treatment is the possibility of neuronal injury, including hippocampal apoptosis in experimental animals with pneumococcal and E. coli meningitis who received dexamethasone. Long-term follow-up studies are thus needed to address the effect of dexamethasone treatment on any cognitive and neuropsychological outcomes in patients with bacterial meningitis.

What are the adverse effects associated with each treatment option?

See the above section on treatment.

What are the possible outcomes of bacterial meningitis?

The complications of bacterial meningitis include cerebritis, abscess, cranial nerve involvement, hearing loss, hydrocephalus, vasculitis (up to 20% of cases), thrombosis, infarct, ventriculitis, and empyema. A mild, transient hydrocephalus is seen in most patients with bacterial meningitis, and only a minority of patients with hydrocephalus require ventricular shunting.

Cranial nerve dysfunction is the result of direct stretching or pressure resulting from a shift of intracranial pressure. Transient sixth nerve palsy can occur as a sign of increased intracranial pressure. Hearing loss is probably the result of the spread of infection to the inner ear through the cochlear aqueduct or modiolus and can develop at an early stage of meningitis.

The ensuing labyrinthritis is thought to be responsible for the sensorineural hearing loss. Imaging is indicated in patients whose clinical course is not improving as expected or who experience new neurologic symptoms or signs. Neurologic sequelae are common in survivors of meningitis and include hearing loss, cognitive impairment, and developmental delay.

The Metropolitan Atlanta Developmental Disabilities Surveillance Program in 1991 identified bacterial meningitis as the leading postnatal cause of developmental disabilities, including cerebral palsy and mental retardation. Hearing loss develops in 22%-30% of survivors of pneumococcal meningitis compared with 1%-8% after meningococcal meningitis.

What causes this disease and how frequent is it?

The age of the patient is useful in predicting the causative organism (Table I). In the neonatal period, group B streptococci and
E. coli are responsible for most cases of acute bacterial meningitis. Less common causes for neonatal meningitis include L. monocytogenes.
S. pneumoniae and N. meningitidis are the two most common causes of acute bacterial meningitis in immunocompetent infants, children, and adults, accounting for almost 80% of cases.

Individuals with sickle cell disease, congenital or acquired asplenia, splenic dysfunction, HIV infection, or cochlear implants are at an increased risk for meningitis due to
S. pneumoniae. Individuals with complement dysfunction are at an increased risk for meningitis due to S. pneumoniae and N. meningitidis.

Immune activation is initiated by the recognition of different bacterial pathogen-associated molecular patterns by receptors, including Toll-like receptors (TLRs) and Nod-like receptors. The downstream signaling cascades involve the common adaptor molecules such as IRAK-4 and MyD88. Deficiences and polymorphisms in the pattern recognition receptors such as TLRs and adaptor molecules such as MyD88 have been shown to be associated with pneumococcal meningitis.

How do these pathogens/genes/exposures cause the disease?

The pathogens infect the CNS by hematogenous spread, direct extension from paranasal and dental infections, as well as skull base fracture, direct implantation (e.g., after surgery, cochlear implant), or rarely, secondary to infections in the epidural or subdural spaces.

Meningitis-causing bacteria cross the blood-brain barrier transcellularly, paracellularly, or by means of infected phagocytes (so-called Trojan horse mechanism). Transcellular traversal of the blood-brain barrier has been shown for most meningitis-causing pathogens in infants and children, including E. coli and group B streptococci. L. monocytogenes uses transcellular and Trojan horse mechanisms.

Traversal of the blood-brain barrier by meningitis-causing bacteria requires specific microbial factors and their interaction with the host factors as well as specific host signal transduction pathways.

How can bacterial meningitis be prevented?


Routine immunizations in infants and children with Hib, S. pneumoniae, and N. meningitis are the most effective means of preventing bacterial meningitis caused by these organisms. On February 24, 2010, the US Food and Drug Administrationlicensed a 13-valent PCV (PCV13, brand name Prevnar 13) for children 6 weeks through 71 months of age.

In addition to the 7 serotypes included in the 7-valent PCV (PCV7), PCV13 contains 6 additional pneumococcal serotypes. These 13 serotypes currently are responsible for 63% of invasive pneumococcal disease (IPD) cases in children younger than 5 years of age.

PCV13 should replace PCV7, using the same recommended routine and catch-up PCV7 immunization schedules for healthy children 2 months through 59 months of age and for children through 71 months of age who have certain chronic diseases or immunocompromising conditions that increase the risk for IPD.

A single supplemental dose of PCV13 is recommended for all children who already are immunized completely with PCV7 as follows: (1) healthy children 14-59 months of age and (2) children at high risk for PID 14-71 months of age.


The risk of invasive Hib disease is increased in unimmunized household contacts younger than 4 years of age. Rifampin eradicates Hib from the pharynx in approximately 95% of carriers and decreases the risk of secondary invasive illness in exposed household contacts. Nursery and child care center contacts may also be at increased risk of secondary disease. Secondary disease in child care contacts is rare when all contacts are older than 2 years of age.

Close contacts of all individuals with invasive meningococcal disease, whether endemic or in an outbreak situation, are at high risk and should receive chemoprophylaxis (Table III). The attack rate for household contacts is 500 to 800 times the rate for the general population. The decision to give chemoprophylaxis to contacts of individuals with meningococcal disease is based on the risk of contracting invasive disease.

Table III.
Organism Contacts Regimen Dose Duration
Hib For all household contacts with at least 1 contact younger than 4 years of age who is unimmunized or incompletely immunized, or a contact who is immunocompromsiedFor child care centers with 2 or more cases of invasive Hib disease within 60 daysFor index case, if younger than 2 years of age or members of a household with a susceptible contact and treated with a regimen other than cefotaxime or ceftriaxone Rifampin 20 mg/kg once a day 4 days
N. meningitid Household or child care contact during 7 days before onset of illness Direct exposure to index case’s secretions during 7 days before onset of illness Mouth-to-mouth resuscitation, unprotected contact during 7 days before onset of illness Frequent contact in the same dwelling as index case during 7 days before onset of illness Passengers seated directly next to the index case during airline flights more than 8 hours RifampinCeftriaxone Ciprofloxacine <1 month, 5 mg/kg every 12 hours >1 month, 10 mg/kg every 12 hours <15 years, 125 mg intramuscularly>15 years, 250 mg intramuscularly>18 years, 500 mg 2 days Single dose Single dose Single dose

Daily antimicrobial prophylaxis is recommended for children with functional or anatomic asplenia, regardless of their immunization status, for prevention of pneumococcal disease. Oral penicillin V (125 mg twice a day, for children younger than 5 years of age; 250 mg twice a day for children 5 years of age and older) is recommended.

Although overall incidence of invasive pneumococcal infection is decreased after penicillin prophylaxis, cases of penicillin-resistant invasive pneumococcal infections and nasopharyngeal carriage of penicillin-resistant strain in patients with sickle cell disease have increased in recent years. Parents should be informed that penicillin prophylaxis may not be effective in preventing all cases of invasive pneumococcal infections.

The age at which prophylaxis is discontinued often is an empirical decision. Most children with sickle cell disease who have received all recommended pneumococcal vaccines for age, who have received penicillin prophylaxis for prolonged periods, who are receiving medical attention, and who have not had a previous severe pneumococcal infection or a surgical splenectomy may discontinue prophylactic penicillin at 5 years of age. However, they must be counseled to seek medical attention for all febrile events.

The duration of prophylaxis for children with asplenia attributable to other causes is unknown, but prophylaxis may continue throughout childhood.

What is the evidence?

Wald, ER, Kaplan, SL, Mason, EO. “Meningitis Study Group: Dexamethasone therapy for children with bacterial meningitis”. Pediatrics. vol. 95. 1995. pp. 21-8.

Kim, KS. “Neurological diseases: pathogenesis of bacterial meningitis: from bacteremia to neuronal injury”. Nature Rev Neurosci. vol. 4. 2003. pp. 376-85.

Kim, KS. “Treatment strategies for central nervous system infections”. Exp Opin Pharmacother. vol. 10. 2009. pp. 1307-17.

Kim, KS. “Acute bacterial meningitis in infants and children”. Lancet Infect Dis. vol. 10. 2010. pp. 32-42.

Ongoing controversies regarding etiology, diagnosis, treatment

The pathogenesis of neuronal injury associated with bacterial meningitis remains incompletely understood, and there is no adjunctive treatment that shows consistent benefit against meningitis-associated neuronal injury caused by bacteria common in infants and children (group B streptococci, E. coli, N. meningitidis, S. pneumoniae).