REVIEW

Niger J Paed 2013; 40 (1): 6 –14

Ogunlesi TA

Diagnosis and treatment of bacterial

meningitis in the newborn

DOI:http://dx.doi.org/10.4314/njp.v40i1.2

Accepted: 29th May 2012

Abstract Background: Bacterial

meningitis in the newborn is glob-

ally renowned for high mortality.

The associated morbidities also

include audiologic, motor, visual

and mental deficits.

Objective: To highlight the peculi-

arities in the current diagnostic and

management strategies in newborn

meningitis.

Methods: Relevant literature on the

subject published only in English

language or translated to English

language was searched manually

and electronically. The Medline,

PUBMED and HINARI were

searched for the period between

riologic culture in the diagnosis of

meningitis can be improved with

serologic method like polymerase

chain reaction. Widespread resis-

tance of pathogens may be threaten-

ing the use of penicillins and gen-

tamicin for empirical treatment of

newborn meningitis. No sufficient

evidence presently supports the

current practices of fluid restriction,

prolonged duration of antibiotic

treatment and non-use of adjuvant

steroid therapies in the newborn.

Conclusion: Efforts to reduce the

incidence of newborn meningitis

cannot be separated from the pre-

vention of newborn sepsis gener-

ally. In addition, more controlled

trials are required in the developing

world with respect to the various

aspects of management of newborn

meningitis, particularly fluid man-

agement and the use of adjuvant

steroids.

(

)

Ogunlesi, TA

Department of Paediatrics,

Olabisi Onabanjo University Teaching

Hospital, Sagamu

P. O. Box 652, Sagamu-121001

Ogun State. Nigeria.

Email: tinuade_ogunlesi@yahoo.co.uk

1

966 and 2012. The following key

words were used during the search:

newborn/neonatal, bacterial/

pyogenic meningitis, central nerv-

ous system infections, antibiotics,

dexamethasone and fluid

restriction.

Results: The pattern of bacterial

aetiology and mortality differ be-

tween the developed and develop-

ing world. The usefulness of bacte-

Keywords: Antibiotics, dexa-

methasone, meningitis, newborn,

lumbar puncture.

Introduction

of neonatal meningitis in the middle belt was reported to

be 1.9/1000 live births f₅our and 6.5/ 1000 live births in

north-eastern Nigeria. Similarly, the incidence rate

reported in the Pana₆ma in the mid-nineties was 3.5

cases/1000 live births.

Meningitis is an extensive inflammatory condition of the

leptomeninges. Although, it is generally described as

uncommon, meningitis occurs commonly in the neonatal

period due to the increased susceptibility of newborn

babies to severe infections.1 Neonatal meningitis is as-

sociated with significant mortality and severe morbid-

ities.

While the incidence has not changed remarkably, there

has been a significant decrease in the mortality associ-

ated with meningitis over the last two decades due to

improved antibiotic therapy and supportive care in most

parts of the developed and industrialized world. On the

contrary, morbidity has not changed significantly over

this same period, even, in the developed world. The per-

sistently high morbidity rate among the survivors of

neonatal meningitis remains a major clinical issue a₂n_,₃d

the need to minimize these morbidities is a challenge.

The overall incidence of meningitis in England and

Wales has not changed remarkably from the known 0.2

–

0.4 cases/1000 live births over the last three decades.

,3

These are similar to data obtained from other parts of

2

the developed world. Although, consistent incidence

rates are not available in most parts of the developing

world, the few available ones are higher compared to the

developed world. For instance, in Nigeria, the incidence

Although, mortality rates in newborn meningitis vary by

7

region, the known rates in the various regions of the

world are not markedly different. For instance, mortality

rates included 0.7–1.9 ⁄ 1000 live births in sub-Saharan

Africa, 0.33–1.5 in the Middle East and North Africa

form the core of clinical diagnosis of meningitis in the

newborn. Due to the low sensitivity and specificity of

the clinical predictors of this severe₁i₂n_,f₁e₃ctious disease,

laboratory diagnosis is indispensable.

7

and 0.4-2.8 in the Americas and Caribbean. Mortality in

the developed world had dropped from close to 50% to

about 10% over the last decade whereas morbidi₈ty

remains high at 15 to 60% among the survivors.

Although, under-reporting is a challenge in the develop-

ing world, mortality associated with neonatal meningitis

also varies between 30% and 60% while morbidity fig-

ures vary greatly. Five to seven a high proportion of the

survivors in neonatal meningitis develop chronic handi-

capping conditions with serious medical and psycho-

social implications such as cerebral palsy, mental retar-

dation₅,_,₉seizure disorder, hemiplegia, deafness and blind-

ness.

Bacteriology of newborn meningitis

Microbiology

In consonance with sepsis, the bacterial aetiology of

neonatal meningitis differs between the developing and

the developed world. In most developed countries, the

leading pathogen in newborn meningitis is Group B

Streptococcus (GBS). Others include Escherichia coli,

Listeria monocytogenes, o₁th_,₈e_,r₁₄ coliforms and lately,

Streptococcus pneumoniae.

Data obtained from

England and Wales from 1985 to 1987 and from 1996 to

1997 showed very little change in the pattern of bacterial

aetiology of neonatal meningitis over this period. GBS

accounted for 39% to 48% of all cases. Others included

Escherichia coli (18% to 26%), Streptococcus pneum₂o_, -₃

niae (6%) and Listeria monocytogenes (5% to 7%).

In a more recent French National survey, the leading

pathogens in newborn meningitis included GBS (59%),

Escherichia coli (28%), Gram-negative bacilli apart

from Escherichia coli (4%), other Streptococcus apart

from GBS (4%), Neisseria menin₁₅gitidis (3%) and

Listeria monocytogenes (1.5%).

Although, bacterial meningitis is an important cause of

newborn death globally, the burden of neonatal sepsis

and meningitis is most pronounced in the resource-poor

parts of the world where the disease constitutes a₁s₀i_,g₁₁nifi-

cant proportion of neonatal admission and deaths.

In spite of the similarity in the pathology and patho-

genesis of meningitis in childhood, there are peculiari-

ties with regard to the diagnosis and management of

newborn bacterial meningitis. This review aims to high-

light the challenges in the diagnosis and management of

bacterial meningitis in the newborn in comparison with

older children. Research issues which may possibly

proffer solutions to the alarming high prevalence of neu-

rologic sequelae and poor quality of life among the sur-

vivors of newborn bacterial meningitis will also be

raised.

The pattern of bacteriological aetiology in neonatal men-

ingitis differs in terms of the spectrum of organisms as

well as the relative prevalence of individual organisms

causing meningitis in most parts of the developing

world. Escherichia coli₁w₆ as earlier reported as the lead₁₇-

ing aetiology in Nigeria and more recently in Kenya.

Although E. coli was also commonly isolated in Nigeria,

Staphylococcu_,₅s aureus predominated in two other Nige-

Diagnosis

Symptoms and signs

4

rian reports. In addition, some reports of GBS pre-

dominance have also been made in parts of th₁e₇_-d₁e₉velop-

Bacterial meningitis can be extremely difficult to diag-

nose in the newborn because the symptoms and signs

are often subtle and non-specific at the early stage of the

disease. The clinical presentation of meningitis is indis-

tinguishable from that of sepsis without meningitis. The

early symptoms of meningitis include pyrexia, poor

feeding, vomiting, lethargy or irritability. These clinical

features may also characterize other newborn disorders

such as hypoxic-ischaemic encephalopathy and meta-

bolic derangements such as hypoglycaemia. Therefore,

distinguishing between early meningitis and other neo-

natal illnesses mentioned above may be challenging and

this is associated with the tendencies to frequently under

ing world like Zimbabwe, Kenya and China.

Inter-

4, 5, 16

estingly, none of the studies from Nigeria

reported

GBS as aetiology of neonatal meningitis. Even most

recent studies of newborn sepsis₂₀in_-₂N₃ igeria did not find

GBS or reported very few cases.

Listeria monocytogenes is also u₇ncommonly encoun-

1

tered in this part of the world. On the other hand,

Gram negative bacilli (with the exception of E. coli)

have a global distribution but appear more common in

the developing world, probably for reasons of poor hy-

giene.

-diagnose meningitis in the newborn.

The emergence of unusual organisms like Hae₅m_,₂o₄ philus

influenza in Nigerian babies with meningitis

poses

With the progression of the disease, bulging fontanelle,

shrill c₁r₂y, apnoeas, seizures, opisthotonus and coma may

occur. Although, these features are more specific fea-

tures of meningitis and thus, facilitate diagnosis on clini-

cal grounds, they occur quite late in the disease. Instruc-

tively, the classic signs of meningeal irritation in the

older children, such as neck stiffness and positive Ker-

nig sign or Brudzinski sign are often absent and unreli-

able among infants. Thus, the latter features should not

new challenges in the treatment of newborn meningitis

in this population. This organism is not usually taken

into consideration when planning empirical antibiotic

treatment for newborn meningitis in Nigeria. This may

constitute technical delay in the commencement of ap-

propriate antibiotic therapy in possible cases of Haemo-

philus influenzae infection among newborns.

The highlighted differences in the aetiology of meningi-

tis between the developed and developing world may be

8

explained in terms of population differences in the rate

and pattern of colonization, genetic differences in im-

mune response and differences in lab₇_,o₂r₅atory techniques

for pathogen isolation and reporting.

Similar caution has been raised with respect to CSF glu-

cose and protein₃₀concentrations among preterm babies

with meningitis. Indeed, there have been reservations

about the use of the CSF glucose and protein parameters

in making the diagnosis of meningitis or assessing the

success of treatment of meningitis because of the₆ high

Lumbar puncture

2

variability in the values of the CSF parameters. In a

Positive bacteriologic culture of the cerebrospinal fluid

large study, babies with meningitis had CSF cell count

of 0 to 15,900/mm , glucose ranging from 0 to 199mg/dl

3

(

CSF) is the gold standard diagnostic procedure for bac-

terial meningitis. In the absence of bacteriologic culture

facilities, Gram stain of CSF may also provide useful

information upon which initial decisions on diagnosis

and management can be based. Blood cult₁u₇r_, e₁₉_,m₂₆ay also

be positive in 40–60% of meningitis cases.

and protein ranging from 41 to 1964mg/dl compared to

0 to 90,000/mm , 0 to 1089mg/dl₂a₆nd 3 to 4122mg/dl

3

respectively among normal babies. Instructively, stud-

ies are yet to show to what extent this variability truly

affects the diagnosis of meningitis in the newborn. In

spite of the challenges posed by the variability high-

lighted above, most practitioners still stick to the tradi-

tional cut-off values₃of the various parameters: white

cell count > 32/mm with more than 60% polymor-

phonuclear cells, proteins > 170mg/dl in preterm babies

(or > 150mg/dl among term babies) and glucose < 50%

of serum glucose, determined just prior to lumbar tap.

The definitive method for diagnosis of meningitis lies in

the chemistry and bacteriologic culture of the CSF since

the clinical features of newborn meningitis are largely

non-specific and blood culture may not always be as

stated above. There have been controversies about the

indications for lumbar puncture (LP) in newborn sepsis.

Current concessions appear to be pivoted on the fact

that, LP is only useful as part of the work-up for late-

onset sepsis or when infants with early-onset sepsis have

clinical features suggestive of meningitis. Available

evidence is not in support of the use of LP in routine

s₂₇epsis work-up for healthy babies with presumed sepsis.

Unlike in the older child, raised intracranial pressure

rarely occurs in newborn meningitis since the open su-

tures and fontanel allows expansion of the cranium

when pressure is building up. Thus, the exclusion of

elevated intracranial pressure prior to LP may not be a

major concern in the newborn. Nevertheless, compro-

mised cardio-respiratory functions are clear contraindi-

cations to performing LP in critically ill infants. How-

ever, a modified left lateral position with the hip flexed

to 90⁰ but without flexion of the neck has been proposed

as useful₂f₈or LP in the presence of respiratory embar-

rassment. In all situations, when meningitis is sus-

pected, inability to perform LP should not delay appro-

priate antibiotic treatment.

The Interpretation of neonatal CSF parameters is also

difficult when the fluid is blood stained, as this obvi-

ously changes both the cellular and biochemical con-

stituents of the fluid. One of the traditional methods of

resolving the debacle is to count both the erythrocytes

and leucocytes in the fluid and use the physiological

ratio of 700 erythrocytes to one leucocyte as the cut-off

point from normal. In a more recent US study, it was

reported amongst others that, in a cohort of newborn

infants, CSF protein increased by 2mg/dl for every 1000

erythrocytes ₃in₁ the CSF even in spite of the presence of

pleocytosis.

This method may not require sophisti-

cated kits or personnel and thus, can be studied further

for its clinical applicability in the developing world.

The CSF gets sterilized rapidly following exposure to

antibiotics. It takes just about two hours for m₂eningo-

3

coccus, and up to six hours for pneumococcus. This is

important in the situations of pre-hospital care antibiotic

use which is common in parts of the world where antibi-

otic use is poorly regulated. The CSF culture findings

may be altered by prior antibiotic use but antibiotics

rarely interfere with CSF protein or glucose. In this

situation, bacterial antigen detection serological methods

would be most useful. Even in the absence of prior ex-

posure to antibiotics, delay in processing CSF for up to

four hours, has been reported to cause progre₃s₃sive de-

cline in CSF glucose and white cell counts. This is

very relevant to the practice in the developing world

where inadequate laboratory services (in terms of per-

sonnel, infrastructures and equipment) may warrant de-

lays in processing CSF samples. Therefore, practitioners

in this part of the world need to take these variables into

consideration when interpreting CSF parameters.

When it is available, the CSF specimen should be exam-

ined macroscopically for turbidity and microscopically

for the presence of bacterial organisms. Instructively,

turbidity of the CSF is determined by the number of pus

cells present in the fluid though the number may vary

widely. Therefore, when clinical suspicion is strong,

microscopic examination should take precedence over

macroscopic examination. Although, Gram stain may

reveal the pathogens in more than three-quarter of cases,

bacteriological culture is the gold standard for diagnos-

ing meningitis in the absence of prior exposure to antibi-

otics. Abnormalities in cell count and biochemical pa-

rameters such as glucose and proteins are also important

for making diagnosis. Due to physiological variations in

the biochemical and cellular parameters of newborn

CSF, caution needs to be exercised in interpreting the

parameters. The white cell counts₂₉in the newborn CSF

has been shown to be age-specific.

To a large extent, the use of molecular and serologic

methods such as the polymerase chain reaction (PCR),

which does not require the presence of live organisms

for diagnosis, have improved the diagnosis of meningitis

on the CSF. Commercial kits for performing the sero-

Thus, CSF cell count needs to be interpreted with cau-

tion in the diagnosis of neonatal bacterial meningitis.

9

logic tests for Streptococcus pneumoniae, Group B

Streptococcus, E. coli and Haemophilus influenzae are

available in most laboratories in the developed world but

not yet in the resource-poor parts of the developing

world. PCR has been shown to detect 100% cases of

bacterial meningitis in a cohor₃t₄ of Greek children while

culture detected only 21.4%. In addition, PCR, may,

in the future be used to prognosticate since a study of

quantitative PCR on blood, showed a correlation be-

tween disease severity and me₃n₅ingococcal bacterial

DNA load among older children. Although, meningo-

coccus is not a common cause of newborn meningitis.

However, it is very likely that a similar principle may

apply to other pathogens with respect to PCR and sever-

ity of illness.

platelets aggregation factor (PAF). These agents in turn,

offset an inflammatory cascade characterized by in-

creased vascular permeability, polymorphonuclear mi-

gration and activities and production of exudates and

cellular debris. Ultimately, these inflammatory events

result in cerebral oedema, elevated intracranial pressure,

reduced₉cerebral perfusion, cerebritis, neuritis and vas-

3

culitis. The end-result of all these pathologic features

include ischaemia, infarction and atrophy of neural tis-

sues. Short and long-term morbidities occur in infants

with bacterial meningitis as a result of the aforemen-

tioned pathologic changes. Some of the acute morbid-

ities in meningitis include cerebral oedema, subdural

effusion, subdural empyema, venous sinus thrombosis,

cranial nerve palsies and hydrocephalus. Long term neu-

rologic deficits in meningitis include hearing loss, corti-

cal blindness, strabismus, speech disorders, behavioural

disorders, motor deficits particula₅_,r₉ly hemiplegia, mental

retardation and seizure disorders.

Neuro-imaging

Neuro-imaging is required for the diagnosis of intra-

cerebral collections, structural focal lesions and ven-

tricular dilatation. These features may affect response to

antibiotic treatment of meningitis. Therefore, poor clini-

cal response despite adequate therapy is an indication

for neuro-imaging in newborn meningitis. The most

basic of such imaging method in use in the developing

world include trans-fontanelle ultrasonography and com-

puterized tomographic (CT) scans of the brain. Unfortu-

nately, the wide variation in the size of normal lateral

ventricles make CT scans less reliable₃₆ in diagnosing

truly dilated ventricles in the newborn. More efficient

imaging facilities like the Magnetic Resonance Imaging

are unfortunately, not available for routine use in most

centres in this part of the world.

Antibiotic treatment

The goal of antibiotic treatment in meningitis is rapid

sterilization of the CSF. This explains why antibiotic

therapy is highly recommended when meningitis is

clinically suspected even when definite investigations

are not feasible or delayed. In standard practice where

microbiological diagnosis could be reliably made, the

choice of antibiotic depends on the organism isolated.

However, in most cases especially in most parts of the

developing world, the initial treatment is usually empiri-

cal pending the availability of sensitivity reports from

the laboratory. This empirical treatment depends on the

known epidemiology of the likely organisms as well as

the local antibiotic resistance patterns. When the blood-

brain barrier is inflamed as it occurs in meningitis, the

permeability is increased and the penetration of most

antibiotics into the CSF is improved. It is important that,

the antibiotic chosen for empirical treatment should have

good penetration into the CSF and achieve adequate

minimum₀bactericidal concentration (MBC) for the or-

Other investigations

Full blood count, C-reactive protein (CRP), clotting

studies, and urea and electrolytes are useful ancillary

tests in sepsis work-up. Leucopaenia and elevated CRP

are known to be more consistent with the d₃i₇agnosis of

severe bacterial infection in the newborn.

Elevated

4

procalcitonin (with sensitivity of 90% and specificity of

5%) was recently shown, in a systematic review, to

ganism. These properties are important for rapid ster-

6

ilization of the CSF which impacts on the survival of the

affected infants.

have good diagnostic accuracy especially₃₈for late-onset

sepsis where meningitis occur commonly.

The traditional treatment of neonatal meningitis involves

ampicillin and gentamicin. Although, the penicillins

generally have₁poor CSF penetration when the meninges

Treatment

4

The three major aspects of treatment of bacterial menin-

gitis include (1) antibiotic therapy (2) fluid restriction

are inflamed, adequate concentrations can be delivered

into the CSF with more frequent and higher doses.

(

3) adjunctive therapy. Central to the understanding of

On the other hand, gentamicin readily penetrates the

inflamed blood-brain-barrier but rarely achieves the

minimum bactericidal concentration for the pathogens.

Increasing the therapeutic dose of gentamicin in order to

achieve higher minimum bactericidal concentration in

the CSF may be harmful since the drug ordinarily has a

narrow therapeutic inde₁x and can readily cause ototoxic-

the relevance of these major aspects is the basic pathol-

ogy and pathogenesis of bacterial meningitis. Following

the penetration of pathogens into the subarachnoid

space, the destruction of the bacterial cell walls in the

meninges may occur spontaneously or following antibi-

otic treatment. The destruction of the bacterial cell walls

results in the release of antigenic components such as

peptidoglycan and teichoic acid into the subarachnoid

space. The released bacterial toxins and cell wall com-

ponents provoke the release of cytokines such as inter-

leukins (IL-1β), tumour necrosis factor-γ (TNF) and

4

ity and nephrotoxicity. This is particularly challenging

in the resource-constrained parts of the world where

facilities for monitoring serum drug levels are not avail-

able. In an attempt to circumvent this challenge, trials of

intraventricular administration of antibiotics in meningi-

1

0

tis were conducted but the finding₄s₂_,₄w₃ ere disappointing

with observed increased mortality.

shown to₁₇_,h₁a₉ve high resistance to ampicillin and gen-

tamicin. These reports have significant implications

for the principle of using ampicillin and gentamicin for

the empirical treatment of newborn meningitis in this

part of the world.

The WHO recommended initial antibiotic combination

of a penicillin (e.g. ampicillin or penicillin G) and an

aminoglycoside (e.g. gentamicin) or a third-generation

cephalosporin (e.g. ceftriaxone or cefotaxime) for the

treatment of meningitis in young₄₅ infants aged <

Duration of treatment and choice of antibiotic

4

6

0days. A recent systematic review which compared

Antibiotic therapy often requires modification once anti-

biotic susceptibility testing becomes available. The anti-

biotic treatment of meningitis in the newborn is tradi-

tionally prolonged because of the challenges of penetra-

tion of CSF and achievement of minimum bactericidal

concentration earlier mentioned. At present, there is no

statistical evidence to specifically guide the duration of

antibiotic treatment in neonatal meningitis. However,

the general principle is that the duration of antibiotic

treatment depends on the organism isolated. For Gram

negative bacilli, parenteral administration of antibiotics

should be continued for a minimum of three we₃e₇ ks or

the use of third generations and the traditional antibiot-

ics (ampicillin, penicillin and chloramphenicol in vari-

ous combinations), found no significant difference in

terms of death, deafness and treatment failures among

subjects with meningitis. However, the review covered

all studies of bacterial meningitis irrespective of age

hence the difficulty in generalizing the results of the

review.

In the developed world, the first line therapy is usually

ampicillin with gentamicin or ampicillin with cefo-

taxime or ceftriaxone although the latter has the ten-

dency₇to cause displace bilirubin from albumin binding

for 14 days after the sterilization of the CSF.

This

prolongation of therapy is important because of delayed

sterilization of the CSF in Gram-negative bacilli menin-

gitis. The latter fact may explain the high mortality and

poorer outcome associated with Gram negative bacilli

meningitis. For GBS, the minimum duration of treat-

ment is 14 days while Staphylococcus aureus meningitis

should be treated for up to three weeks in order to re-

duce the risk of cerebral abscess formation.

3

sites. The third generation cephalosporins are effective

on a wide range of pathogens causing meningitis, in-

cluding the aminoglycoside-resistant strains. Unfortu-

nately, this class of antibiotics has not been shown to be

particularly effective on Listeria monocytogenes, thus,

they are not recommended for monotherapy in places

where Listeriosis is common. Nevertheless, they have

good blood-brain-barrier penetration and achieve ade-

quate minimum bacte₄r₆icidal concentrations for most

organisms in the CSF. Ampicillin is effective on GBS,

the coliforms and Listeria monocytogenes. Although,

Listeria monocytogenes and GBS are reportedly uncom-

mon in most parts of the developing world, most physi-

cians also adopt the combination of ampicillin and cefo-

taxime in the empirical treatment of meningitis in the

newborn. The third-generation cephalosporins

It is recommended that LP should be repeated 24 to 48

hours after the commencement of therapy to ascertain

sterilization of the CSF. However, with cefotaxime and

ceftriaxone, an Australian study of childhood meningitis

demonstrated that the lowest CSF concentrations of both

drugs were several times higher than the minimum bac-

tericidal concentrations for the organisms

(meningo₇coccus, pneumococcus and Haemophilus influ-

4

(

cefotaxime and ceftriaxone) are active against the major

enzae). On the basis of these findings, the authors rec-

pathogens of neonates worldwide, including aminogly-

coside-resistant strains. A recent Nigerian study reported

remarkable sensitivity of pathogens in newborn sept₂i-₃

caemia to ceftriaxone, cefotaxime and ceftazidime.

Thus, these drugs can also be reliably used in the em-

pirical treatment of meningitis.

ommended that repeat LP was not necessary when treat-

ing meningitis due to these organisms with either of

these drugs. Thus, it may be attractive to recommend

either of these drugs for the empirical treatment of new-

born meningitis in places where significant resistance to

the drugs have not been reported and avoid repeat LP.

Antimicrobial resistance

In some unusual cases, inflammatory changes in the

CSF may persist despite adequate antibiotic treatment as

a result of obstructive ventriculitis, subdural empyema,

cerebral abscess or intracranial vessel thrombi. Persis-

tent CSF inflammation is an indication for neuroimaging

and review of antibiotic therapy. From the results of

neuroimaging, other therapeutic measures like surgical

interventions may be required in the care of the meningi-

tic infant. In addition to neuroimaging, antibiotic ther-

apy may need to be prolonged. Subsequent CSF exami-

nation is unnecessary if the CSF has been sterilized by

48 hours of therapy and the clinical course has been

satisfactory.

Antibiotic resistance is a major challenge confronting

2

5

practitioners worldwide. This, to a large extent, makes

global recommendations on the choice of drugs which

are useful for the empirical treatment of severe infec-

tions in the newborn difficult. A recent study of new-

born septicaemia in Sagamu, Nigeria reported overall

resistance₂₃of 70.3% to ampicillin and 43.6% to gen-

tamicin.

Specifically, the resistance of Staphylo-

cocccus aureus to ampicillin and gentamicin were 66.7%

and 37.5% respectively in the report from Sagamu.

This observation is important in view of the predomi-

nance of Staphylococcus aureus as a leading pathogen in

newborn meningitis as reported in another Nigerian

Repeat LP at the end of therapy, hi₄t₈herto recommended,

is currently unpopular in the UK. A study of a small

5

study. In addition, the Gram negative bacilli have been

1

number of babies also showed that a repeat LP was un-

necessary if clinical response to treatment was satisfac-

tory. This suggestion was based on the fact that neither

normal CSF at the end of treatment nor abnorm₉ al CSF

dexamethasone in improving the survival and reducing

neurologic deficits including hearing loss in meningiti₅s₅

among children, particularly in low-income countries.

At present, the best evidence for the benefits of dexa-

methasone is in H influenzae type b meningitis where

better ₅a₆udiologic outcome has been clearly demon-

strated.

4

findings accurately predicted cure or relapse. There-

fore, practitioners need to weight the benefit of a repeat

LP at the end of treatment against the attendant risks.

Trials of shorter duration of treatment are on-going

among older children. A randomized trial among 100

infants with meningitis showed that four days of ceftri-

axone treatment is as effective as seven days with₅₀ no

Studies of the use of dexamethasone in meningitic new-

borns are particularly sparse. Therefore, current treat-

ment guidelines for newborn meningitis exclude adjunc-

tive dexamethasone therapy. Nevertheless, there are

reports of likely usefulness of dexamethasone in the

reduction of overall mortality as well as reduction in

neurologic sequelae among survivors of newborn men-

ingitis. A non-randomized study of Australian infants

between 1953 and 1961 reported mortality rate of 41%

in the steroid-treated group compared with 75% of non-

treated group. Despite better survival, there was no im-

pressive difference in t₇he occurrence of neurologic se-

difference in complications and treatment failures.

A

more recent double-blind randomized trial in Malawi,

showed no significant difference in relapse or treatment

failure rate among children with meni₅n₁gitis treated with

antibiotics for five days or 10 days. This study sug-

gested that discontinuation of antibiotics after the fifth

day of ceftriaxone among children of post-neonatal age

who remained stable was safe. Similar trials in the new-

born period are highly desired.

5

quelae in both groups. Another non-randomized study

of newborns with bacterial meningitis in Nigeria be-

tween 1992 and 1995 revealed lower mortality and

higher frequency of full recove₅ry among babies treated

with adjuvant dexamethasone. These two studies, de-

spite their demonstration of usefulness of dexa-

methasone therapy in the newborn, could not be used as

a universal guide for treatment because they were not

randomized controlled trials. Unfortunately, the most

frequently cited study which questioned the usefu₈lness

Similarly, a randomized controlled trial of seven-day

versus 14-day antibiotic treatment for neonatal sepsis

showed similar treatment failure rates in both groups

prompting ₂a recommendation of shorter duration of

5

treatment. For clinical usefulness, large scale con-

trolled trials are also required. It will be interesting to

know if these findings may be safely extrapolated to

other invasive newborn diseases like meningitis.

5

of dexamethasone was randomized and controlled. The

Use of dexamethasone

latter study of babies with meningitis in Jordan showed

similar case fatality rates (22% Vs 28%) and neurologic

sequelae (30% Vs 39%) among babies treated with or

without dexamethasone.

Bacterial meningitis is characterized by high mortality

and severe neurologic sequelae among most survivors. It

is believed that most of the sequelae are as a result of the

damage to neural tissue during the acute inflammatory

process that characterizes bacterial meningitis. There-

fore, the use of corticosteroids as adjuncts in the treat-

ment of bacterial meningitis is intended to attenuate the

acute inflammatory process, minimize tissue damage

and ultimately improve clinical outcomes both in the

short and long term.

Several other studies among older infants and children

have demonstrated conflicting reports about the e₅₃f_,f₅ic₉ acy

of adjunct dexamethasone therapy in meningitis.

The

findings from these studies could not be scientifically

extrapolated to newborns because subgroup analysis was

not carried out for children aged 6 weeks to 12 weeks.

The extrapolation could have been useful given the un-

derstanding that the immune characteristics and range of

pathogens causing serious infections at that age (6

weeks to 12 weeks) are usually similar to those of the

newborn period. Nevertheless, till date, there is no clear

high-power statistical evidence that the use of corticos-

teroids in neonatal meningitis truly improves the out-

come and prevents neurologic complications of the dis-

ease. Neither is there any clear high-power statistical

evidence that the use of corticosteroids in neonatal men-

ingitis truly lacks benefits in terms of the outcome of the

disease. Systematic reviews of the available controlled

trials are desired.

The timing of administration of corticosteroids in men-

ingitis is also important in the determination of its effi-

cacy. In childhood bacterial meningitis, the administra-

tion of intravenous dexamethasone before or along with

the first dose of antibiotics has been reported to be more

beneficial compared to ad₃ministration after the com-

5

mencement of antibiotics. It is standard practice to

administer dexamethasone in divided doses for four

days. There is no evidence that the drug is more effec-

tive when given for longer or shorter periods but the risk

of adverse effects₅₄appears to increase with the duration

of administration.

Use of intravenous fluids

Although, the use of adjuvant corticosteroids has been

traditionally employed in the treatment of meningitis

among children of post-neonatal age and adults, ran-

domized and non-randomized studies have given con-

flicting reports concerning the effectiveness of adjuvant

Other important supportive care in the management of

meningitis includes anticonvulsant therapy, use of ino-

tropes and fluid management. Unfortunately, there are

no controlled clinical trials with respect to the use of

1

2

these measures among newborns. In general, it is a com-

mon practice to restrict fluids to two thirds or three quar-

ters of the daily maintenance during the management of

childhood meningitis. The basis for this practice is the

need to reduce the likelihood of the syndrome of inap-

propriate secretion of antidiuretic hormone (SIADH).

SIADH is characterized by hyponatraemia, fluid reten-

tion and a tendency to worsen cerebral oedema in men-

ingitis. Therefore, practitioners reduce fluid therapy in

children with meningitis in the hope of preventing

SIADH.

pressure are only used among older children who are

more at risk of significantly raised intracranial pressure

as a result of non-expansible cranium. Osmotic agents

shift fluids from the extravascular to the intravascular

space, resulting in a reduction of intracranial pressure. A

recent study among children aged 2 months to 12 years

suggested that glycerol-induced increment in osmolality

and reduction in CSF volume may be mechan₃ism by

6

which glycerol reduces intra-cranial pressure. Initial

placebo-controlled trials of glycerol with or without

dexamethasone in childhood meningitis have demon-

strated overall better neurologic seque₆₄lae among the

cases but not with respect to deafness. More trials of

other osmotic diuretics such as 20% mannitol, glycerol

and hypertonic saline in the treatment of raised ICP are

still on-going in different parts of the world. It may

seem, by virtue of current knowledge, that meningitic

neonates do not desperately require osmotic therapy.

However, controlled trials are still needed to prove this

assumption.

The true benefits of this practice are yet to be docu-

mented. In the first place, the pathogenesis of SIADH in

meningitis remains unclear and the reported incidence of

SIADH in meningitis varies considerably. These facts

cast a lot of doubt on the usefulness of the practice of

fluid restriction in this instance.

In addition, a significant proportion of children with

meningitis particularly after delay in presentation pre-

sents with dehydration or hy₀povolaemia and are in dire

Prevention

6

need of fluid resuscitation. It will be dangerous to rig-

idly restrict fluid therapy in such situation.

Nevertheless, the clinical dilemma of whether fluids

should be restricted or not continues to generate dispari-

ties in the pattern of clinical practice as well as difficul-

ties in interpreting mortality figures in childhood menin-

gitis.

Neonatal meningitis is characterized by high mortality

and severe morbidities which cause handicaps. A study

of 111 children aged between 9 and 10 years who sur-

vived neonatal meningitis in England and Wales be-

tween 1985 and 1989 showed worse general outcome

compared to non-meningitic matched controls. The sur-

vivors of neonatal meningitis had less mean intelligence

quotient, higher frequency of motor impairment, sei-

Current thinking suggests that the increased extracellular

fluid, the appropriate increased secretion of ADH, and

mild systemic hypertension occurring in situations of

raised intracranial pressure are compensatory mecha-

nisms. These physiologic changes are required to over-

come the raised intracranial pressure and to maintain

adequate cerebral blood flow and perfusion. Therefore,

fluid restriction is likely to reduce the efficacy of the

compensatory mechanisms an₁d thus, increase the likeli-

9

zures, hydrocephalus and hearing loss. The spectrum of

neurologic sequelae reported in the UK study₅was not

different from the pattern reported from Nigeria.

Given the background of generally poor outcome in neo-

natal meningitis, prevention is highly desired. The pre-

ventive measures in newborn meningitis cannot be sepa-

rated from those of newborn sepsis generally. These

may include routine screening of pregnant women for

urinary tract infection, early intervention following

spontaneous rupture of fetal membranes and hygienic

birth generally. For the developed world and other

places where GBS is reportedly predominant, prenatal

screening for GBS, appropriate treatment and₅intrapar-

6

hood of adverse outcome. A systematic reviews of

controlled trials among children showed no significant

difference in number of deaths and acute severe neuro-

logical sequela₆e₂ among maintenance-fluid and restricted

-

fluid groups. Further subgroup analyses in the same

review showed statistically significant difference in fa-

vour of the maintenance-fluid group in terms of spastic-

ity, seizures at 72 hours and 14 days and chronic severe

neurological sequelae at three-months follow up.

The current body of scientific evidence no longer sup-

ports the practice of fluid restriction in the management

of childhood meningitis, as there are no benefits either

immediately or on long term basis. Rather, it is more

attractive to carefully assess infants with meningitis for

possible dehydration, correct dehydration appropriately,

ensure adequate fluid intake but prevent overhydration.

6

tum antibiotic prophylaxis are useful practices. Instruc-

tively, these practices only reduce the burden of early-

onset neonatal disease whereas GBS meningitis com-

monly presents as late-onset sepsis.

Early recognition and prompt commencement of appro-

priate antibiotic therapy are also desired to minimize the

risk of poor outcome in neonatal meningitis. This consti-

tutes major challenge in most parts of the developing

world where most ill babies present late to the hospitals.

The advent of vaccines against Haemophilus influenzae

and Streptococcus pneumoniae infections might have

changed the epidemiology of childhood meningitis in

the developed world but these have not impacted on the

newborn period where the vaccines are not routinely

indicated.

Treatment of raised intracranial pressure

Raised intracranial pressure (ICP) is a well recognized

complication of meningitis among older children but

this is not a cause for concern among newborns with

meningitis. Measures aimed at reducing intracranial

1

3

Conclusion

evidences for guidance with respect to fluid manage-

ment and use of steroids and osmotic agents in neonatal

meningitis.

While efforts are on-going to reduce the incidence of

newborn meningitis in the resource-poor parts of the

toworld, management strategies also need to be im-

proved enhance survival and reduce neurologic seque-

lae among the survivors. To achieve this, more research

is coveted particularly with r₆e₆spect to optimal antibiotic

therapy as earlier suggested. Specifically, large scale,

multi-centred controlled trials are desired to provide

Conflict of interest: None

Funding: None

References

1

1. Chowdhury HR, Thompson S, Ali

21. Mokuolu AO, Jiya N, Adesiyan

OO. Neonatal septicaemia in

Ilorin: bacterial pathogens and

antibiotic sensitivity pattern. Afr J

Med Med Sci 2002;31:127-130.

22. Ojukwu JU, Abonyi LE, Ugwu J,

Orji IK. Neonatal septicaemia in

high risk babies in South-Eastern

Nigeira. J Perinat Med

1

.

World Health Organization. An-

timicrobial and support therapy for

bacterial meningitis in children.

Report of the meeting of 18-20

June 1997, Geneva Switzerland.

WHO/CHD/98.6; WHO/EMC/

BAC/98.2. 1998.

Delouvois J, Blackbourn J, Hurley

R et al. Infantile meningitis in

England and Wales: a two year

study. Arch Dis Child 1991;66:

Mohammed, Alam N, Yunus M,

Streatfield PK. Causes of neonatal

deaths in a rural subdistrict of

Bangladesh: Implications for inter-

vention. Hlth Popul Nutr

2

010;28:375-382.

1

2. Weber MW, Carlin JB, Gatchalian

S et al. Predictors of neonatal sep-

sis in developing countries. Pedi-

atr Infect Dis J 2003;22:711–716.

3. Best J, Hughes S. Evidence behind

the WHO Guidelines: hospital care

for children–what are the useful

clinical features of bacterial men-

ingitis found in infants and chil-

dren? J Trop Pediatr 2008;54:83–

2

3

.

2006;34:166-172.

23. Ogunlesi TA, Ogunfowora OB,

Osinupebi OA, Olanrewaju DM.

Changing trends in newborn sepsis

in Sagamu, Nigeria: Bacterial eti-

ology, risk factors and antibiotic

susceptibility. J Paediatr Child

Health 2011;47:5-11.

24. Adeboye MA, Obasa TO, Fadeyi

A, Adesiyun OO, Mokuolu OA.

Haemophilus meningitis in an

African neonate: time for active

surveillance and institution of ap-

propriate control measure. West

Afr J Med 2010;29:275-277.

25. Osrin D, Vergnano S, Costello A.

Serious bacterial infections in new-

born infants in developing coun-

tries. Current Opinion Infect Dis

2004;17:217–224.

26. Garges HP, Moody A, Cotten M et

al. Neonatal Meningitis: what is

the correlation among cerebrospi-

nal fluid cultures, blood cultures,

and cerebrospinal fluid parame-

ters? Pediatr 2006;117: 1094–

1100.

27. Ajayi OA, Mokuolu OA. Evalua-

tion of neonates with risk for infec-

tion/suspected sepsis: Is routine

lumbar puncture necessary in the

first 72 hours of life? Trop Med Int

Health 1997;2:284–8.

6

03-607.

Holt D, Haket S, Louvouis JD et

al. Neonatal meningitis in England

and wales: ten years on. Arch Dis

Child Fetal and Neonatal Ed

2

001;84: F85–F89.

8

6.

4

5

.

Airede AI. Neonatal bacterial men-

ingitis in the middle belt of Nige-

ria. Dev Med Child Neurol

1

4. Chien HC, Chiu N, Li W, Huang

F. Characteristics of neonatal bac-

terial meningitis in a teaching hos-

pital in Taiwan from 1984-1997. J

Microbiol Immunol Infect

1

993;35:424-430.

Airede KI, Adeyemi O, Ibrahim T.

Neonatal bacterial meningitis and

dexamethasone adjunctive usage in

Nigeria. Niger J Clin Pract

2

000;33:100–104.

1

5. Gaschignard J, Levy C, Romain O

et al. Neonatal bacterial meningi-

tis: 444 Cases in 7 years. Pediatr

Infect Dis J 2000;30:212 – 217.

6. Longe C, Omene J, Okolo A. Neo-

natal meningitis in Nigerian in-

fants. Acta Paediatr Scand

2

008;11:235-245.

6

.

Moreno MT, vargas S, Poveda R,

Saez-liorens X. Neonatal sepsis

and meningitis in a developing

Latin American country. Pediatr

Infect Dis J 1994;13:516-520.

Stoll B. The global impact of neo-

natal infection. Clin Perinatol

1

984;73:477–481.

7

8

.

7. Laving AMR, Musoke RN,

Wasunna AO, Revathi G. Neonatal

bacterial meningitis at the newborn

unit of Kenyatta national hospital.

East Afr Med J 2003;80:456–462.

8. Nathoo K, Pazvakavamba I, Chid-

ede O et al. Neonatal meningitis in

Harare, Zimbabwe: a 2-year re-

view. Ann Trop Paediatr

1

997;24:1-21.

Harvey D, Holt D, Bedford H.

Bacterial meningitis in the new-

born: a prospective study of mor-

tality and morbidity. Semin Peri-

natol 1999;23:218-225.

Stevens JP, Eames M, Kent A,

Halket S, Holt D, Harvey D. Long

term outcome of neonatal meningi-

tis. Arch Dis Child Fetal Neonatal

Ed 2003;88:F179-F184.

1

9

1

.

1

991;11:11–15.

1

9. Chang CJ, Chang WN, Huang LT

et al. Neonatal bacterial meningitis

in southern Taiwan. Pediatr Neu-

rol 2003;29: 288 – 294.

28. Weisman LE, Merenstein GB,

Steenbarger JR. The effect of lum-

bar puncture position in sick neo-

nates. Am J Dis Child

0. Ambe JP, Gasi IS, Mava Y. Re-

view of neonatal infections in Uni-

versity of Maiduguri Teaching

Hospital: common bacterial patho-

gens seen. Niger J Clin Pract

2

0. Adejuyigbe OE, Adeodu OO, Ako

1983;137:1077–9.

-Nai KA, Taiwo O, Owa JA. Septi-

29. Kestenbaum LA, Ebberson J, Zorc

JJ, Hodinka RL, Shah SS. Defining

cerebrospinal fluid white blood

count reference values in neonates and

young infants. Pediatr 2010;125:257-

caemia in high-risk neonates at a

teaching hospital in Ile-Ife, Nige-

ria. East Afr Med J 2001;78:540-

2

007;10:290-293.

5

43.

2

64.

1

4

3

0. Smith PB, Garges HF, Cotton CM,

Walsh TJ, Clark RH, Benjamin

DK Jr. Meningitis in preterm neo-

nates: importance of cerebrospinal

fluid parameters. Am J Perinat

41. Rang H, Dale M, Ritter J, Moore

P. Pharmacology, 5th edn. Chur-

chill Livingstone, London.2003.

42. McCracken GH, Mize SG,

Threlkeld N. Intraventricular gen-

tamicin therapy in gram-negative

bacillary meningitis of infancy.

Report of the Second Neonatal

Meningitis Cooperative Study

group. Lancet 1980;1:787-791.

43. Shah SS, Ohlsson A, Shah VS.

Intraventricular antibiotics for

bacterial meningitis in neonates.

Cochrane Database of Systematic

Reviews 2008 Issue 4, Art No.:

CD004496

44. WHO. Pocket Book of Hospital

Care for Children - Guidelines for

the Management of Common Ill-

nesses with Limited Resources.

World Health Organisation, Ge-

neva. 2005.

45. Prasad K, Kumar A, Singhal T,

gupta PK. Third-generation cepha-

losporins versus conventional anti-

biotics for treating acute bacterial

meningitis. Cochrane Database of

Systematic Reviews 2007 Issue 4,

Art No.: CD001832.

46. Cherubin CE, Eng RHK, Norrby

R, Modai J, Humbert G, Overturf

G. Penetration of newer cepha-

losporins into cerebrospinal fluid.

Rev Infect Dis 1989;11: 526–546.

47. Goldwater PN. Cefotaxime and

ceftriaxone cerebrospinal fluid

levels during treatment of bacterial

meningitis in children. Int J Antim-

icrob Agents 2005;26:408-11.

48. Agarwal R, Emmerson AJ. Should

repeat lumbar punctures be rou-

tinely done in neonates with bacte-

rial meningitis? Results of a survey

into clinical practice. Arch Dis

Child 2001;84:451–2.

49. Schaad UB, Nelson JD,

McCracken GH Jr. Recrudescence

and relapse in bacterial meningitis

of childhood. Pediatr 1981;67:188

–95.

50. Roine I, Ledermann W, Foncea

LM, et al. Randomised trial of four

vs seven days of ceftriaxone treat-

ment for bacterial meningitis in

children with rapid initial recov-

ery. Pediatr Infect Dis J

53. Odio CM, Faingezight I, Paris M

et al. The beneficial effects of early

dexamethasone administration in

infants and children with bacterial

meningitis. New Eng J Med

2

008;25:421-426.

1991;324:1525 – 1531.

1. Hines EM. Nigrovic LE, Neuman

MI, Shah SS. Adjustment of cere-

brospinal fluid protein for red

blood cells in neonates and young

infants. J Hosp Med 2012;7:325-

54. McIntyre PB, Berkey CS, King

SM et al. Dexamethasone as ad-

junctive therapy in bacterial men-

ingitis. A meta-analysis of random-

ized clinical trials since 1988.

JAMA 1997;278:925 – 931.

55. Van de Beek D, de Gans J, McIn-

tyre P, Prasad K. Corticosteroids

for acute bacterial meningitis.

Cochrane Database of Systematic

Reviews 2007 24(1): CD004405.

56. Schaad UB, Lips U, Gnehm HE, et

al. Dexamethasone therapy for

bacterial meningitis in children.

Lancet 1993;342:457–61.

57. Yu JS, Grauaug A. Purulent men-

ingitis in the neonatal period. Arch

Dis Child 1963;38: 391 – 396

58. Daoud AS, Batieha A, Al-Sheyyab

M, Abuekteish F, Obeidat A,

Mahafza T. Lack of effectiveness

of dexamethasone in neonatal bac-

terial meningitis. Eur J Pediatr

1999;158: 230 – 233.

59. Molyneux EM, Walsh AL, Forsyth

H et al. Dexamethasone treatment

in childhood bacterial meningitis

in Malawi: a randomized con-

trolled trial. Lancet 2002;360: 211

– 218.

60. Prince AS, Neu HC. Fluid manage-

ment in Hameophilus influenza

meningitis. Infect 1980;8:5–7.

61. Singhi SC, Singhi PD, Srinivas B,

et al. Fluid restriction does not

improve the outcome of acute

meningitis. Pediatr Infect Dis J

1995;14:495–503.

3

28.

2. Kanegaye JT, Soliemanzadeh P,

Bradley JS. Lumbar puncture in

pediatric bacterial meningitis:

defining the time interval for re-

covery of cerebrospinal fluid

pathogens after parenteral antibi-

otic pretreatment. Pediatr

2

001;108:1169–74.

3

3. Rajesh NT, Dutta S, Prasad R,

Narang A. Effect of delay on neo-

natal cerebrospinal fluid parame-

ters. Arch Dis Child Fetal Neoona-

tal Ed 2010;95:F25-F29.

4. Papavasileiou K, Papavasileiou E,

Tzanakaki G, Voyatzi A, Kre-

mastinou J, Chatzipanagiotou S.

Acute bacterial meningitis cases

diagnosed by culture and PCR in a

children’s hospital throughout a 9-

year period (2000-2008) in Athens,

Greece. Mol Diagn Ther

2

011;15:109-113.

3

5. Hackett SJ, Guiver M, Marsh J, et

al. Meningococcal bacterial DNA

load at presentation correlates with

disease severity. Arch Dis Child

2

002;86:44–6.

3

6. Haslam RH. Role of computed

tomography in the early manage-

ment of bacterial meningitis. J

Pediatr 1991;119:157–9.

7. Stephenson T, Marlow N, Watkin

S, Grant J (eds). Fetal and neonatal

infection. In: Pocket Neonatology.

62. Maconochie IK, Bauner JH, Stew-

art M. Fluid therapy for acute bac-

terial meningitis. Cochrane Data-

base of Systematic Reviews 2008,

Issue 1 Art No.: CD004786

1

st ed. Edinburgh: Churchill Liv-

ingstone, 2008:241-69.

3

8. Vouloumanou EK, Plesssa E, kara-

georgopoulos DE, Mantadakis E,

Falagas ME. Serum procalcitonin

as a diagnostic marker for neonatal

sepsis: a systematic review and

meta-analysis. Intensive Care Med

63. Singhi S, Jarvinen A, Peltola H.

Increase in serum osmolality is

possible mechanism for the benefi-

cial effects of glycerol in child-

hood bacterial meningitis. Pediatr

Infect Dis J 2008; 27: 892-896.

64. Peltola H, Rome I, Fernandez J et

al. Adjuvant glycerol and/or dexa-

methasone to improve the outcome

of childhood bacterial meningitis:

a prospective, randomized, double-

blinded, placebo-controlled trial.

Clin Infect Dis 2007;45:1277-12686.

65. American Academy of Pediatrics.

Revised guidelines for prevention of

early-onset Group B Streptococcal

infection. Pediatr 1997;99:489–96.

2

011;37:747-762.

3

9. Mustafa MM, Ramilo O, Saez

Liorens X, Olsen KD, Magness

RR, McCracken GH Jnr. Cerebro-

spinal fluid prostaglandins, inter-

leukins-1 beta and tumor necrosis

factor in bacterial meningitis:

clinical and laboratory correlations

in placebo treated and dexa-

methasone treated patients. Am J

Dis Child 1990;144: 883 – 887.

0. Tauber MG, Sande MA. General

principles of therapy of pyogenic

meningitis. Infect Dis Clin North

Am 1990;4:661–76.

2000;19:219–22.

51. Molyneux E, Nizami SQ, Saha S

et al. 5 versus 10 days of treatment

with ceftriaxone for bacterial men-

ingitis in children: a double-blind

randomized equivalence study.

Lancet 2011;377:1837-1845.

52. Chowdhary G, Dutta S, Narrang

A. Randomized controlled trial of

7-day vs 14-day antibiotics for

neonatal sepsis. J Trop Pediatr

2006;52:427-432.

4

6

6. Furyk JS, Swann O, Molyneux E.

Systematic review: neonatal meningitis

in the developing world. Trop Med Int

Health 2011;16:672 – 679.