You are on page 1of 5

Defining bacterial meningitis and other infections of the central

nervous system
Gary D. Overturf, MD

Objective: To define central nervous system infections of in- association of a compatible clinical syndrome or cerebrospinal
fants and children that occur as co-morbid or predisposing con- fluid changes associated with bacterial meningitis or other cen-
ditions of sepsis. tral nervous system infection, and confirmed as an anatomically
Design: Standard pediatric infectious disease references and defined infection by imaging or surgery, in association with
the pertinent literature in English were reviewed from 1960 to positive blood cultures or bacterial antigen from cerebrospinal
2002 to ascertain the previous methods and definitions utilized in fluid. Possible bacterial meningitis may be defined as a compat-
clinical studies of the epidemiology and treatment of bacterial ible clinical syndrome with predefined cerebrospinal fluid
infections of the central nervous system. An accepted definition of changes in the absence of a confirmatory culture or antigen test
bacterial meningitis defined by the Infectious Disease Society of from any site.
America was reviewed and adapted to the previous clinical def- Conclusions: Bacterial meningitis and other central nervous
initions. The information was formulated into a proposed standard system bacterial infections can be defined as definite, probable,
for definite, probable, and possible bacterial infections of the and possible with a combination of a defining compatible clinical
central nervous system. syndrome and an anatomic definition by surgery or imaging,
Results: The diagnosis of definite bacterial infection of the coupled with isolation of the organism, bacterial antigen, or other
central nervous system, including bacterial meningitis, requires defining molecular component of the organism. (Pediatr Crit Care
the isolation of the pathogen from the cerebrospinal fluid or other Med 2005; 6[Suppl.]:S14 –S18)
significant clinical site such as surgical tissue, an implanted KEY WORDS: central nervous system; cerebrospinal fluid; bacte-
device, or blood. Probable bacterial infection is defined by the rial meningitis; infection

T he pyogenic central nervous shunt infections, intracranial abscesses, America (IDSA) was adopted and com-
system infections of children and spinal infections, these infections are pared with those of the treatment studies
include bacterial meningitis, only infrequently accompanied by posi- and the standard definitions used by the
focal abscesses including in- tive blood cultures at the time of diagno- Centers for Disease Control, Active Bac-
tracranial, extracerebral, and intracere- sis, with the exception of certain patho- terial Core, and National Nosocomial In-
bral abscess, ventricular shunt infections gens (e.g., Staphylococcus aureus). fection Surveillance programs.
or ventriculitis, focal extracranial infec- Viral infection and its diagnosis will be
tions of the central nervous system (e.g., discussed because it represents the infec-
spinal, epidural, subdural or paraspinal tion causing the most frequent diagnostic Definitions of Bacterial
abscesses), and postoperative or trau- confusion. Although it is associated with Meningitis Before 1992
matic wound injury. Of these central ner- cerebrospinal fluid (CSF) pleocytosis and
Before 1992, definitions of bacterial
vous system infections, only bacterial overlaps with some clinical and neuro-
meningitis were nonstandardized and de-
meningitis is regularly associated with logic symptoms, it is defined by the ab-
fined for use in clinical antibiotic trials of
sepsis or bacteremia at the time of diag- sence of a recovery of a bacterial organ-
the treatment of bacterial meningitis,
nosis, with rates of positive blood cul- ism or specific bacterial antigen.
which were conducted after the introduc-
tures ranging from ⬍5% to ⬎25%. De-
tion of effective antibiotics in 1950. The
spite cryptic bacteremia as a cause of
Methods of Review definitions of bacterial meningitis in
these studies are represented by studies
Standard pediatric textbook references conducted between 1977 and 1990 (1– 8).
From the Departments of Pediatrics and Pathology,
University of New Mexico School of Medicine, Albu- and studies of the treatment and epide- Bacterial meningitis often was defined as
querque, NM. miology of bacterial meningitis from the a “compatible clinical syndrome” of men-
This work was supported by the Mannion Family English literature, published from 1960 ingitis plus a positive CSF culture, a pos-
Fund—Center for the Critically Ill Child, Division of to 2002, were examined. Representative itive bacterial antigen test, or positive
Critical Care Medicine at Children’s Hospital Boston,
the PALISI Network, and the ISF. studies that provided a complete defini- blood culture. However, because cultures
Copyright © 2005 by the Society of Critical Care tion used in the studies were included in of blood or CSF were negative in cases of
Medicine and the World Federation of Pediatric Inten- the final definitions. A standardized defi- presumed bacterial meningitis in approx-
sive and Critical Care Societies nition of bacterial meningitis formulated imately 10%–30% of cases, a concept of
DOI: 10.1097/01.PCC.0000161933.42822.86 by the Infectious Disease Society of “purulent unknown,” “pyogenic,” or “pre-

S14 Pediatr Crit Care Med 2005 Vol. 6, No. 3 (Suppl.)


sumed” bacterial meningitis was used in called partially treated bacterial meningi- group B streptococci, Haemophilus influ-
some studies (9, 10). This was often de- tis is remarkably consistent and little enzae, Neisseria meningitidis, and Strep-
fined as meningitis with negative cul- changed from that of patients with posi- tococcus pneumoniae, has used a case
tures of blood and CSF in a patient with a tive CSF cultures who have not been pre- definition of “isolation (e.g., culture pos-
compatible clinical syndrome and CSF treated with antibiotics (Table 1). Only itive) of. . . bacterial pathogens from a
changes consistent with bacterial menin- the frequency of positive cultures and normally sterile site” (14). The Centers
gitis with or without confirmation by a Gram-negative stain is reduced in those for Disease Control and National Nosoco-
nonculture method such as bacterial an- children with bacterial meningitis who mial Infection Surveillance have defined
tigen testing. In a review of studies of have received antibiotics before lumbar meningococcal, haemophilus, and pneu-
bacterial meningitis conducted from puncture (12). On occasion, the diagnosis mococcal infection with surveillance def-
1977 to 1990, the criteria for a diagnosis of bacterial meningitis can be con- initions requiring positive cultures or
of presumed bacterial or purulent un- founded by the accidental traumatic lum- positive antigen tests for meningococcal
known meningitis included a compatible bar puncture, which has been estimated infections in the cases of purpura, with
clinical syndrome consistent with bacte- to occur in 15%–20% of children. Some negative bacterial cultures or haemophi-
rial meningitis plus one or two of the authors have suggested that if determina- lus antigen positivity only with infection
following: an elevated CSF leukocyte tion of the ratio of white blood cells/mm3 due to Haemophilus influenzae in bacte-
count, defined as varying from 5 to ⬎500 to red blood cells/mm3 in CSF exceeds rial meningitis. Because of poor sensitiv-
white blood cells/mm3; a low CSF glucose the ratio in the peripheral blood, the ex- ity and specificity, pneumococcal positive
of either ⱕ40 or 50 mg/dL or a CSF/ cess white blood cells could represent in- antigen tests have not been included as
serum ratio of ⬍0.5; or an elevated CSF flammatory pleocytosis (13). However, criteria in Centers for Disease Control
protein varying from ⱖ50 to 150 mg/dL; prospective studies have often found that and National Nosocomial Infection Sur-
or autopsy evidence for bacterial menin- this determination is inaccurate. There- veillance definitions.
gitis. fore, in those cases in which the CSF In 1992, IDSA formulated definitive
Other positive blood cultures in asso- findings do not support a bacterial infec- guidelines for the eligibility criteria for
ciation with a compatible clinical syn- tion, a positive culture or other reliable clinical trials of antimicrobial agents for
drome and CSF changes have been used diagnostic study for microorganisms has bacterial meningitis (15), thereby provid-
variously as criteria for definite bacterial remained the standard for a diagnosis of ing the first standardized definition of
or probable bacterial meningitis. Adher- definite bacterial meningitis, with proba- bacterial meningitis. Since that time, tri-
ence to strict conventional diagnostic ble or possible categories requiring in- als of antibiotics for the treatment of bac-
CSF changes have often been less rigor- creasingly less stringent supportive crite- terial meningitis have often cited these
ous in patients with positive blood cul- ria. criteria (Table 2) as the standard for the
tures, requiring as few as a total of only 5 diagnosis of bacterial meningitis (16, 17).
leukocytes/mL. The prevalence of nega- Definitions of Bacterial Thus, the currently proposed definitions
tive CSF cultures with positive bacterial for bacterial meningitis for infants, neo-
Meningitis Since 1990
cultures in patients given the diagnosis of nates, and children proposed here are ex-
bacterial meningitis has varied from Since 1990, published studies have trapolated from both the historical defi-
⬍5% to ⬎25%. Many tests for detection usually employed a definition of bacterial nitions and the definitions provided by
of bacterial antigens have been used, meningitis that required the isolation of the 1992 IDSA guidelines authored by
such as latex agglutination, enzyme im- an organism in culture, either directly McCracken et al (Table 3) (15). Therefore,
munoassay or enzyme linked immunoas- from the CSF or the blood. The Centers as much as possible, the definitions of
say, and more recently, polymerization for Disease Control bacterial core surveil- bacterial meningitis provided coincide
chain reaction (PCR) (11). Only latex ag- lance project, which includes surveil- with those of the IDSA and have been
glutination is widely and commercially lance for all the major pathogens of com- reformulated to provide the definite,
available. Unfortunately, these tests often munity-acquired meningitis, including probable, and possible categories.
have been associated with a significant
lack of sensitivity and specificity for
pneumococcal and meningococcal infec- Table 1. Admission cerebrospinal fluid (CSF) findings in children with meningitis caused by Hae-
tions. Other nonspecific tests to differen- mophilus influenzae with or without previous antibiotic treatment before lumbar puncture
tiate bacterial from nonbacterial disease
such as viral infections, including CSF Without Previous With Previous
lactate, C-reactive protein, procalcitonin, Antibiotics Antibiotics
and some cytokines, or specialized stains CSF Findings (n ⫽ 186) (n ⫽ 94) p Value
for bacteria such as ethidium bromide,
WBC/mm3 3,731 2,524 .13
have remained largely experimental and
Neutrophils, % 95 90 .03
never widely incorporated into the diag- Protein, mg/dL 191 115 ⬍.001
nosis of bacterial meningitis. Glucose, mg/dL 29 23 .28
In the absence of a confirmatory cul- CSF/blood glucose ratio 0.25 0.22 .51
ture or other reliable diagnostic test for Positive Gram stain, % 90 82 .05
Positive culture, % 98 94 .03
bacterial organisms in patients pretreated Positive bacterial antigen test 88 88 ⬎.90
for a presumed bacterial infection, the
spinal fluid profile (e.g., cell count, dif- WBC, white blood cells.
ferential, protein, and glucose) in so- Table is modified from Saez-Llorens and McCracken (12).

Pediatr Crit Care Med 2005 Vol. 6, No. 3 (Suppl.) S15


Table 2. Eligibility criteria for clinical trials in bacterial meningitis of bacterial meningitis only when a bac-
terial culture of the CSF or blood is pos-
Patients Without
itive in association with an abnormality
Bacterial
of CSF leukocytes, glucose, or protein
Patients With Bacterial Meningitis Meningitis
consistent with bacterial meningitis. If
Criteria A B C D 1 2 3 there is a negative blood or CSF culture,
the diagnosis is accepted when both the
Blood culture ⫹ ⫺ ⫹ ⫾ ⫺ ⫺ ⫹ CSF antigen test and CSF abnormalities
CSF culture ⫺ ⫹ ⫺ ⫹ ⫺ ⫺ ⫺ are positive and a compatible clinical syn-
CSF WBC or glucose/protein ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹
CSF antigen ⫹ ⫾ ⫹ ⫾ ⫺ ⫺ ⫺
drome. Because this category is less
Previous antibiotics Yes ⫾ ⫹ ⫾ ⫺ ⫺ ⫺ stringent but provides a highly probable
Eligibility Eligiblea Eligible Eligible Eligible NE NE NE pathogenic diagnosis of bacterial infec-
tion, it fulfills the diagnostic category of
CSF, cerebrospinal fluid; WBC, white blood cell; NE, not eligible. probable bacterial meningitis. Because
a
For the purposes of the 1992 criteria, the case was defined as bacterial meningitis, eligible for the IDSA guidelines were to fulfill the
enrollment as a case in phase III antibiotic treatment studies. Reproduced with permission from
requirements for eligibility for entry into
McCracken GH et al (15).
phase III U.S. Food and Drug Administra-
tion studies of bacterial infections, dem-
onstration of the organism by culture or
Table 3. Proposed definitions of bacterial meningitis and other central nervous system infections
antigen detection was required for eligi-
Bacterial meningitis in children and infants ⬎8 wks old bility. Therefore, possible or probable
Definite bacterial meningitis bacterial meningitis as listed in Table 3
Compatible clinical syndrome, plus does not fulfill any of the IDSA criteria
All ages: fever, 94%
because detection of bacteria is not re-
1–5 mos: irritability, 85%
6–11 mos: impaired consciousness, 79% quired, but this definition does match
⬎12 mos: vomiting, 82%; neck rigidity, 78% those of previous studies of the treatment
(note: many other compatible signs and symptoms) plus and epidemiology of bacterial meningitis
Positive culture of cerebrospinal fluid (CSF), or positive CSF Gram stain or bacterial antigen, that utilized definitions of pyogenic or
or
purulent unknown meningitis. Depend-
Probable bacterial meningitis
Compatible clinical syndrome, plus ing on the clinical setting, frequency of
Positive culture of blood, plus previous antibiotics, and quality of mi-
One of the following CSF changes crobiological technical services, the pos-
⬎5 leukocytes sible category has been estimated to rep-
Glucose of ⱕ40 or 0.5 CSF/serum ratio
Protein of ⱖ100 mg/dL resent 5%–20% of cases.
Possible bacterial meningitis Neonatal bacterial infections have re-
Compatible clinical syndrome, plus quired the isolation of the organism be-
One of the following CSF changes cause neither the clinical syndrome nor
⬎100 leukocytes
the CSF changes associated with bacterial
Glucose of ⱕ40 or CSF/serum glucose ratio ⱕ0.5
Protein of ⱖ100 mg/dL plus infection in older subjects are consis-
Negative cultures or antigen for bacteria, viral, fungal, or mycobacteria tently present during bacterial infection
Neonatal meningitis (⬍8 wks of age) of the central nervous system in neonates
Compatible clinical syndrome, plus (Table 2) (18, 19). Although coagulase-
Isolation of likely pathogenic organism from CSF, or positive specific bacterial antigen, or
Abnormal CSF consistent with bacterial infection negative staphylococci are frequent
Shunt infection or device associated ventriculitis causes of sepsis in neonates, they have
Definite shunt infection been infrequently associated with infec-
Compatible clinical signs and symptoms, plus tions of the central nervous system, un-
Isolation of bacterial pathogen from device puncture, lumbar puncture or other significant
less the infant has a device implanted in
site (e.g., overlying shunt wound, cellulites, or shunt tubing)
Probable shunt infection the central nervous system, such as a
Compatible clinical signs and symptoms ventriculostomy or ventricular shunt
CSF consistent with bacterial infection (20). Therefore, the isolation of these or-
Negative blood, CSF and device cultures for bacteria ganisms in infants without such devices
Intracranial or extracranial (spinal) infection
nearly always represents a contaminant
Compatible neurologic signs and symptoms, plus
Acceptable radiologic imaging or surgical/anatomic evidence or positive cultures or histology in the CSF culture and not a true patho-
from site, or gen.
Autopsy confirmation Viral infection is the most common
cause of inflammation of the central ner-
vous system and acute febrile encepha-
lopathy of children and are discussed
The IDSA criteria provided a diagnosis bacterial culture, bacterial antigen tests, briefly because of the diagnostic confu-
of bacterial meningitis in four scenarios, and abnormalities of the CSF consistent sion they engender in the differential of
combining the results of a consistent with bacterial meningitis. The IDSA cri- bacterial infection. Infections of viral or-
clinical syndrome plus either a positive teria assumed a diagnosis (e.g., definite) igin are generally classified as meningitis,

S16 Pediatr Crit Care Med 2005 Vol. 6, No. 3 (Suppl.)


meningoencephalitis, or encephalitis cause encephalitis has even a greater varied from ⬍15% to 94% (31, 32). Be-
based on the relative clinical contribu- range in clinical findings and epidemio- cause shunt infections are essentially for-
tions of meningeal inflammation or in- logic settings, the diagnosis will depend eign-body infections, alternating periods
volvement of the brain parenchyma. Dif- on the clinical syndrome and CSF of quiescent and active symptoms, with
ferentiation from bacterial infection of changes consistent with a viral pathogen- an initial response and subsequent re-
the central nervous system is made on esis, and when available, a specific CSF lapse in response to antibiotic therapy,
the basis of signs and symptoms, includ- culture, viral antibody, or PCR result that are characteristic. Also, the infections are
ing clinical epidemiology and CSF is associated with high sensitivity and usually caused by commensal flora (e.g.,
changes. Clinical symptoms consistent specificity, such as those for herpes and staphylococci), and there is often a diffi-
with meningeal involvement are milder West Nile meningoencephalitis (28, 29). culty in recovery the organisms in fluids
but overlap with those of bacterial infec- In addition, for most cases, neither the or culture of tissues not directly contig-
tion, whereas in meningoencephalitis CSF or cultures of either blood or CSF uous with the shunt. Therefore, only cul-
and encephalitis, the cerebral symptoms will meet the criteria for diagnosis of ture of the pathogen from the shunt res-
and signs predominate with variable as- bacterial infection. Therefore, in those ervoir or from fluids from the distal or
sociated meningeal involvement. Clinical bacterial and viral syndromes with over- proximal shunt tubing is the gold stan-
symptoms may vary from mild to severe lapping CSF findings, the diagnosis of the dard for diagnosis of shunt infection,
nonfocal generalized encephalopathy to likely specific pathogenesis would depend whereas positive cultures from other
that of only focal findings. CSF pleocyto- on either the isolation of bacteria from an sites, such as lumbar CSF or blood, are
sis and changes in CSF chemistry are appropriate CSF or blood culture or the considered evidence for probable shunt
often distinct, generally with a mononu- demonstration of a virus from CSF by infection if combined with clinical and
clear pleocytosis of 100 –300 white blood either polymerase chain reaction, specific CSF changes consistent with bacterial in-
cells/mm3, but can range up to a few antiviral antibodies, or culture of the virus. fection.
thousand, a normal or modestly elevated
protein, and a normal glucose (21). How- CSF Shunt Infections Cerebral Abscess and Other
ever, early in the illness, CSF white cell Focal Central Nervous System
counts may exceed 1000/mm3, with a pre- Infection of the ventricular shunt oc- Infections
dominance of polymorphonuclear cells curs in up to 27% of children with hy-
(22, 23). Therefore, the demonstration of drocephalus, and although shunt infec- Other central nervous system infec-
viral antigens or nucleic acids, or viral tions may have a “typical” clinical tions such as cerebral abscess, subdural
specific antibodies, in CSF is required to syndrome and CSF findings (vomiting, empyema, intracerebral abscess, subdural
definitively establish a diagnosis of viral fever, central nervous system dysfunc- abscess, paraspinal and spinal abscess,
disease of the central nervous system tion, and CSF leukocytosis), these may and other central nervous system ana-
(24 –26). When compared with viral cul- vary by the time of onset after shunt tomic infections are usually suspected on
ture, PCR for enteroviruses in CSF sam- placement, the pathogen causing infec- the basis of localizing, peripheral, or cen-
ples is frequently positive, compared with tions, and site of infection in either the tral neurologic symptoms and confirmed
only about 10% for viral culture, yielding proximal (ventricular) or distal (atrial or by imaging techniques such as magnetic
a sensitivity and specificity of 85.7% and peritoneal) catheter (30, 31). Therefore, resonance imaging or computerized to-
93.9%, respectively (26). Other studies the diagnosis of bacterial shunt infection mography (33, 34). A full review of each
using enteroviral PCR in young infants cannot be established by CSF and clinical of these clinical entities is beyond the
have demonstrated sensitivities ap- findings alone. Positive cultures acquired scope of this work but are available in
proaching 100%, with specificities ex- from the shunt fluid confirms a diagnosis standard pediatric, pediatric infectious
ceeding 90%. Because enteroviruses are of shunt infection, whereas positive cul- disease, neurology, and neurosurgical
the predominant cause of the “aseptic tures of CSF acquired from a lumbar site textbooks. In almost all cases of focal
meningitis” syndrome, this single test is in the presence of typical signs and symp- neurologic abscesses, there is known or
likely to exclude a viral etiology in ⬎90% toms and CSF pleocytosis (with the pres- suspected predisposing cause. In chil-
of children. Coupled with CSF, clinical, ence of a shunt), may be consistent with dren, the most frequent are pyogenic dis-
and bacteriologic findings that exclude presumptive shunt infection. Blood cul- ease of the sinuses and chronic otitis me-
bacterial infection (e.g., culture or bacte- tures are very infrequently positive with dia. Unless associated with a known
rial antigen testing), a diagnosis of a viral shunt infections but may occur in up to hematogenous source for bacteremia
etiology can be established in most chil- 42% of ventriculoatrial and 10% of ven- (e.g., congenital heart disease with endo-
dren. However, the category cited above triculoperitoneal shunt infections (30). carditis, lung, or abdominal abscess or
of possible bacterial meningitis is likely Other sites rarely positive on culture in- previous septic episode), these infections
to significantly overlap in clinical symp- clude urine or overlying wounds of the are infrequently associated with positive
toms and CSF changes with that of viral shunt incision. Clinical symptoms and blood cultures. The diagnosis is con-
meningitis. CSF changes are inconsistent (31, 32). In firmed with the site-specific appropriate
For causes of viral central nervous sys- previous studies, fever has been present imaging technique (e.g., computerized
tem infection, other than enteroviral in- in 64%–92% of children with shunt in- tomography or magnetic resonance im-
fections, many other PCR tests are in fection, whereas other symptoms, such as aging, or both), but confirmation of the
development, and some, such as those for shunt dysfunction, changes in senso- bacterial pathogenesis requires demon-
herpes viruses, are widely available but rium, vomiting, meningismus, irritabil- stration of the organism in culture of
not yet approved by the U.S. Food and ity, cellulitis over the shunt, wound in- spontaneously drained or surgically ac-
Drug Administration (27). However, be- flammation, and abdominal tenderness, quired fluids, culture of tissue, or culture

Pediatr Crit Care Med 2005 Vol. 6, No. 3 (Suppl.) S17


of appropriate autopsy materials. Paraspi- 3. Modai J, Vittecoq D, Decazes JM, et al: Pen- 18. Unhanand M, Mustafa MM, McCracken GH,
nal or spinal abscesses in children are etration of aztreonam I the cerebrospinal et al: Gram-negative enteric bacillary menin-
rare clinical infections and most often fluid of patients with bacterial meningitis. gitis: A twenty-one year experience. J Pediatr
Antimicrob Agents Chemother 1986; 29: 1993; 122:15–17
occur as a result of associated infections
281–283 19. Klinger G, Chin CN, Beyene J, et al: Predict-
of the vertebrae, the intervertebral disk,
4. Peltola H, Anttila M, Renkonen OV: Random- ing the outcome of neonatal bacterial men-
or from trauma or infection of retroper- ised comparison of chloramphenicol, ampi- ingitis. Pediatrics 2000; 106:477– 480
itoneal organs or soft tissues (35). cillin, cefotaxime, and ceftriaxone for child- 20. Gladstone IM, Ehrenkranz RA, Edberg SC, et
hood bacterial meningitis. Finnish Study al: A ten year review of neonatal sepsis and
Discussion Group. Lancet 1989; 1:1281-1287 comparison with the previous fifty-year ex-
5. Lecour H, Seara A, Miranda M, et al: Treat- perience. Pediatr Infect Dis J 1990;
The presence of a clinically consistent ment of 160 cases of acute bacterial menin- 9:819 – 825
syndrome and CSF changes in associa- gitis with cefotaxime. J Antimicrob Che- 21. Sawyer MH: Enteroviral infections: Diagno-
tion with appropriate culture findings mother 1984; 14(Suppl B):195–202 sis and treatment. Pediatr Infect Dis J 1999;
from relevant materials is necessary to 6. Schaad UB, Suter S, Gianella-Borradori A, et al: 18:1033–1040
confirm a definite clinical infection of the A comparison of ceftriaxone and cefuroxime for 22. Rorabaugh ML, Berlin LE, Heldrich F, et al:
central nervous system, including bacte- the treatment of bacterial meningitis in chil- Aseptic meningitis in infants younger than 2
dren. N Engl J Med 1990; 322:141–146 years of age: Acute illness and neurologic
rial meningitis. The expanding array of
7. Overturf GD, Cable DC, Forthal DN, et al: complications. Pediatrics 1993; 92:206 –211
laboratory techniques that can demon- Treatment of bacterial meningitis with cefti-
strate infection of the central nervous 23. Singer JL, Maur PR, Riley JP, et al: Manage-
zoxime. Antimicrob Agents Chemother 1984; ment of central nervous system infections
system with specific bacterial antigens, 25:258 –262 during and epidemic of enteroviral aseptic
antiviral antibodies, viral antigens, and 8. Valmari P, Peltola H, Ruuskanen O, et al: meningitis. J Pediatr 1980; 96:559 –563
PCR for specific bacterial or viral nucleic Childhood bacterial meningitis: Initial symp- 24. Robinson CC, Willis M, Meagher A, et al:
acids have augmented the diagnostic ar- toms and signs related to age and reasons for Impact of polymerase chain reaction results
mamentarium, but these may not be consulting a physician. Eur J Pediatr 1987; on management of pediatric patients with
readily available in all clinical settings. 146:515–518
enteroviral meningitis. Pediatr Infect Dis J
Thus, this necessitates the procurement of 9. Underman AE, Overturf GD, Leedom JM:
2002; 21:283–286
Bacterial meningitis: 1978. Dis Mon 1978;
CSF or other fluids or tissue for bacterial 25. Ahmed A, Brito F, Goto C, et al: Clinical
24:1– 63
culture from presumptively infected sites. 10. Carter PE, Barclay SM, Galloway WH:
utility of the polymerase chain reaction for
In some cases, positive blood cultures may diagnosis of enteroviral meningitis in in-
Changes in bacterial meningitis. Arch Dis
provide a definitive etiologic diagnosis of a fancy. J Pediatr 1997; 131:393–397
Child 1990; 65:495–598
bacterial infection. Bacterial PCR is in the 26. van Vliet KE, Glimaker M, Lebon P, et al:
11. Gray LD, Fedorko DP: Laboratory diagnosis
Multicenter evaluation of Amplicor enterovi-
process of development and, in the future, of bacterial meningitis. Clin Microbiol Rev
rus PCR test with cerebrospinal fluid from
may supplant culture as the diagnostic mo- 1992; 5:130 –145
patients with aseptic meningitis. J Clin Mi-
dality of choice. 12. Saez-Llorens X, McCracken GH: Acute bac-
crobiol 1998; 36:2652–2657
The IDSA definition of bacterial menin- terial meningitis beyond the neonatal period.
27. Read SJ, Jeffery KJM, Bangham CRM: Aseptic
gitis (Table 2) was based on principals that In: Principles and Practices of Pediatric In-
meningitis and encephalitis: The role of PCR
fectious Diseases. Second Edition. Long SS,
were learned in nearly three decades of in the diagnostic laboratory. J Clin Microbiol
Pickering LK, Prober CG (Eds). New York,
clinical studies of the treatment and epide- Churchill Livingstone, 2003, pp 264 –271 1997; 35:691– 696
miology of bacterial meningitis. However, 13. Yogev R: Meningitis. In: Pediatric Infectious 28. Smalling TW, Sefers SE, Li H, et al: Molec-
they were provided to meet the require- Diseases: Principles and Practices. Second ular approaches to detecting herpes simplex
ments of enrollment into phase III U.S. Edition. Jenson HB, Baltimore RS (Eds). virus and enteroviruses in the central ner-
Food and Drug Administration studies and New York, WB Saunders Company, 2002, pp vous system. J Clin Microbiol 2002; 40:
fall somewhat short of the clinical defini- 630 – 650 2317–2322
14. Active Bacterial Core Surveillance. August 29. Marfin AA, Gubler DJ: West Nile encephalitis:
tion of bacterial meningitis that may not
29, 2003, Centers for Disease Control, Na- An emerging disease in the United States.
always be supported by isolation of the bac- Clin Infect Dis 2001; 33:1713–1719
terial agent. The definitions cited in Table 3 tional Center for Infectious Diseases, Divi-
sion of Bacterial and Mycotic Diseases. Avail- 30. Schoenbaum SC, Gardner P, Shillito J: In-
are an attempt to combine and expand the fections of the cerebrospinal fluid shunts:
able at: http://www.cdc.gov/ncidod/dbmd/
1992 IDSA criteria to those that are com- abcs/default.htm. Accessed March 12, 2005 Epidemiology, clinical manifestations and
patible with studies of sepsis. Thereby, cat- 15. McCracken GH, Sande MA, Lentnek A, et al: therapy. J Infect Dis 1975; 131:543–552
egories that include definite, probable, and Evaluation of new anti-infective drugs for the 31. Odio C, McCracken GH, Nelson JD: CSF
possible criteria for definition are provided. treatment of acute bacterial meningitis. Clin shunt infections in pediatrics. Am J Dis Child
Infect Dis 1992; 15(Suppl 1):S182–S188 1984; 138:1103–1108
16. Saez-Llorens X: Quinolone treatment for pe- 32. Ronan A, Hogg GG, Klug GL: Cerebrospinal
REFERENCES
diatric bacterial meningitis: A comparative fluid shunt infections in children. Pediatr
1. Overturf GD, Steinberg EA, Underman AE, et study of trovafloxacin and ceftriaxone with or Infect Dis J 1995; 14:782–786
al: Comparative trial of carbenicillin and am- without vancomycin. Pediatr Infect Dis J 33. Saez-Llorens X: Brain abscess in children.
picillin for purulent meningitis. Antimicrob 2002; 21:14 –22 Semin Pediatr Infect Dis 2003; 14:108 –114
Agents Chemother 1977; 11:420 – 426 17. Odio CM, Puig JR, Feris JM, et al: Prospec- 34. Mathisen GE, Johnson JP: Brain abscess.
2. Baken JS, Bruun JN, Gaustad P, et al: Pene- tive, randomized, investigator blinded study Clin Infect Dis 1997; 25:763–781
tration of amoxicillin and potassium clavu- of the efficacy and safety of meropenem vs 35. Auletta JJ, John CC: Spinal epidural ab-
lanate into the cerebrospinal fluid of patient cefotaxime therapy in bacterial meningitis in scesses in children: A 15 year experience and
with inflamed meninges. Antimicrob Agents children. Pediatr Infect Dis J 1999; 19: review of the literature. Clin Infect Dis 2001;
Chemother 1986; 30:481– 484 219 –222 32:9 –16

S18 Pediatr Crit Care Med 2005 Vol. 6, No. 3 (Suppl.)

You might also like