Quantitative measurements of CSF C-Reactive protein for the differentiation

of Bacterial Meningitis from Aseptic Meningitis in children

Dr. Sultana Nadira Rahman

Introduction
Acute bacterial meningitis which is a pediatric emergency with high mortality and morbidity must
be diagnosed and treat promptly. Often diagnosis of bacterial meningitis from viral meningitis is
difficult after some days. Determination of some inflammatory mediators example IL-6 and CRP
were useful in differential diagnosis of bacterial and viral meningitis.1

Meningitis is one of the most potentially serious infections occurring in infants and older children
and is an important cause of morbidity and mortality. Case fatality rates for bacterial meningitis
range from 4.5% in developed countries to1550% in developing countries.2 A further 1520%
of survivors sustain neurological sequelae.3 The mortality from meningitis is close to 100% in
untreated individuals and can still be up to 40% in children who received appropriate antibiotic
therapy in developing countries.2

Rapid and accurate diagnosis coupled with early appropriate therapy is of most importance in
reducing morbidity and mortality of the patients.4 Culture and sensitivity, Gram stain, cytology
and biochemistry of cerebrospinal fluid (CSF) sample are traditionally being done to diagnose and
to differentiate pyogenic from aseptic meningitis. Proper culture is affected by prior antibiotic
therapy, delay in transportation and inoculation. It takes more than 24 hours to isolate the organism.
Gram stain lacks specificity and has interpretative errors.5 Possible causes of false positive result
of Gram stain include contamination of tubes from lumber puncture trays, glass slides and or
Grams reagents. Further more probability of visualization of bacteria on Gram stain is dependent
upon the number of organism present. The overall sensitivity of Gram stain to detect bacterial
meningitis was 67% with a positive predictive value of 60%. Most patients without bacterial
meningitis have negative Gram stain (specificity 99.9%) with a negative predictive value of
99.9%.6

Because of limitation of Gram stain regarding sensitivity and specificity and also culture of the
CSF sample especially in partially treated cases, several rapid diagnostic tests have been developed
to aid in the diagnosis of Acute Bacterial Meningitis (ABM).5

Latex agglutination test (LAT) and other rapid diagnostic test are available but costly and present
only in selective area. Detection of nuclear polymorph leukocytes in the CSF is a fairly reliable
indicator of pyogenic meningitis. Leukocyte count in bacterial meningitis may be elevated to
greater than 1000/mm3 and typically there is neutrophilic predominance (75 95%).7

A CSF leukocyte count < 250/mm3 may be present in as many as 20% of patients with bacterial
meningitis.7 Pleocytosis may be absent in patient with severe overwhelming sepsis and is a poor
prognostic sign. Pleocytosis with a lymphocytic predominance may be present during the early
stage of acute bacterial meningitis; conversely, neutrophilic

pleocytosis may be present in patients during the early stages of acute viral meningitis. Use of
antibiotics makes the gram stain and culture negative and may alter the CSF cytology from
neutrophilic to lymphocytic predominance. Empirical antibiotic therapy is often given. In such
circumstances the detection of C-reactive protein in CSF appears to provide a new dimension to
the diagnosis of meningitis.8

In young children with meningitis, blood or CSF analysis cannot differentiate all cases of aseptic
from bacterial meningitis. Consequently patient with aseptic meningitis generally received
expensive antibiotic for prolong duration causing financial burden to poor parents and lengthening
of hospital stay.9 Serum CRP is an acute phase-reactant that has been utilized clinically to aid in
the diagnosis of neonatal sepsis, urinary tract infection, pneumonia, meningitis.10 Carrol et al
detected CSF C-reactive protein by latex slide agglutination test which was 100% sensitive and
94% specific in differentiating bacterial meningitis from aseptic meningitis. CRP estimation can
help in diagnosing cases of ABM more effectively than culture.11

Celik & et al, in 2007, find no standard method for rule out of bacterial meningitis in patients
whom CSF cytology is according to bacterial meningitis, but CSF culture and gram stain is
negative. Therefore, some factors in CSF need to determine. Aims of this study, is retrospective
comparison of WBC count, CRP, ESR, WBC & neutrophil count of CSF were independent
variable that in regression model, these have 45% positive predictive value in bacterial meningitis
and 93.2% PPV in viral meningitis.12 In another study, Dr Taskin & et al, in 2004, determined
pre-calcitonin and another cytokines level in CSF of children to differentiate bacterial meningitis
from viral meningitis.13 Gendral & et al in 1998, compared serum pre calcitonin and CSF CRP
and IL6, to differentiate these diseases, pre calcitonin level measured in 23 children with bacterial
meningitis and 51 patients with viral meningitis (age of patients, from 2 mo-12yrs.14

Bangladesh is a developing country, with limited resources and skilled manpower particularly in
peripheral set up. An easy and comprehensive test to diagnose ABM would be an alternative tool
to diagnose ABM. Routine use of CSF CRP in diagnosing ABM could be a reliable and easy
method and can be done for rapid diagnosis of meningitis. It is not an alternative of CSF culture,
cytology and biochemistry, but for initial quick assessment it can be considered as first line of
investigation for suspected meningitis to differentiate ABM from aseptic cases in rural or remote
area where investigation facilities are limited. The test does not require much expertise to conduct
and interpret the result.15

So this study was conducted with the objective to measure the specificity, sensitivity, positive and
negative predictive values of CSF-CRP in the diagnosis of bacterial meningitis.

Rationale of the research

The simple Gram stain smear of CSF was the most useful single test for identifying bacterial

meningitis. LAT was more sensitive compared to conventional Gram stain and Culture technique

in identifying the fastidious organisms like H.influenzae, S.pneumoniae and Group B

Streptococcus. However, the combination of Gram stain, Culture and LAT proved to be more

productive than any of the single tests alone. CSF- CRP positive patients demonstrated

significantly higher mortalities and morbidities whereas; CSF- CRP negative patients had much

higher recovery. This thoroughly highlighted the importance of the CSF-CRP level in ABM to be

used as a bad prognostic criterion. The determination of CSF-CRP have significant role in

differentiating bacterial meningitis from aseptic meningitis. Its presence significantly favoured the

diagnosis of acute bacterial meningitis and predicted the possibility of the treatment. The aim of

the study to identify the importance of cerebrospinal fluid C- reactive protein (CSF-CRP) to

establish the diagnosis of ABM, and to measure the specificity, sensitivity, positive and negative

predictive values of CSF-CRP in the diagnosis of Acute Bacterial Meningitis. Easy and early

diagnostic tool is required for rapid detection of acute bacterial meningitis to reduce mortality and

morbidity.
Research question/ hypothesis

Does quantitative measurement of CSF C Reactive protein differentiate bacterial meningitis from

aseptic meningitis in children?

Objective

General

To find out the role of CSF C-Reactive protein for the differentiation of Bacterial

Meningitis from Aseptic Meningitis in Children.

Specific

To measure the specificity, sensitivity, positive and negative predictive values of CSF-CRP

in the diagnosis of Acute Bacterial Meningitis from Aseptic Meningitis. To identify the

importance of cerebrospinal fluid C- reactive protein (CSF-CRP) to establish the diagnosis

of ABM.

Materials and methods:

Method: Cross sectional study
Study period: Six months after approval of protocol
Place of the study: The study will be conducted in the indoor patients of
Department of Pediatric Medicine, Dhaka shishu Hospital (DSH).
 Main outcome variable to the studied: C-reactive protein, Cerebrospinal
fluid, Bacterial Meningitis, Aseptic meningitis.

Confounding variables , if applicable: N/A

Study population: Children admitted between age 0-12 years with fever
and convulsion will be screened in the indoor patients Department of
Pediatric Medicine, DSH.

Sample size and statistical basis of it: Sample size determination depends
on time and resources. As prevalence of meningitis are not known in our country, so
estimated population was calculated by using the following statistical formula:
n=z2p (1-p)/d2
Where n= the desired sample size
Z=the standard normal deviate, usually set at 1.96
P means prevalence = 0.5 (50%), (In unknown prevalence it can be
regarded as 50%)
The degree of accuracy or precision level is d which is considered at
10%.
Using the above formula the expected sample size will be n=96 .

Screening method (if applicable): N/A

Sampling method (s): Purposive sampling from the attending at hospitalized
meningitis patients.
Inclusion and exclusion criteria:
Inclusion criteria:
All children admitted in the age group of 0-12years with fever and
convulsion of short duration.
The patient who has characteristics sign & symptom of meningitis &
had not received any antibiotics prior admission.
Children with bacterial /asceptic meningitis on the basis of CSF
findings.
Both sexes.

Exclusion criteria:
Severely ill patient
Meningitis in patients suffering from other systemic illness(as cardiac,
kidney diseases, severe acute malnutrition etc)
After doing LP if a pt fail to fulfill the criteria of bacterial or asceptic
meningitis according to oparetional definition.
Parents of pt who would not give consent of doing LP to their children.

h) Operational definitions:
Bacterial Meningitis: In this study only those cases will included whos
CSF WBC count between 100-10,000 /cumm or more with PMN
predominance & also increased CSF Protein count (usually 100-500mg/dl)
& markedly reduced glucose (<40mg/dl)

Aseptic Meningitis: Only those patient will be included in this group

whos CSF WBC count rarely >1000/cumm with lymphocytic
predominance, glucose generally normal (may be <40mg/dl) protein count
raised (50-200mg/dl).16

i) Flow chart showing the sequence of task: Appendix-I

j) Procedure of preparing and organizing materials: Sample selection
by purposively l- interview- taking consent - result collection -preparing
for tabulation.
k) Nature of control (if applicable): N/A.
l) Randomization and binding methods, if and as applicable: N/A
m) Equipments to be used: Interview schedule -check list -investigation
form.
N) Procedure of collecting data (Including methods of intervention,
measurement, estimation etc.): The patient who has characteristics sign
& symptom of meningitis will be taken as study population. After proper
clinical examination complete blood count, random blood sugar and other
relevant tests along with CSF cytology, biochemistry and culture-
sensitivity will also be done. Patients will be divided into two groups on
the basis of CSF findings- bacterial meningitis and aseptic meningitis. CSF
CRP will be measured in all cases.CSF CPR will be measured by auto
analyzer at the laboratory of DSH. All patients will be treated empirically
by injectable antibiotics & in some cases drug will be changed according
to culture & sensitivity. According to the report of CSF study we will
decide how long we continue the treatment with injectable antibiotics.
Asceptic meningitis pt will be discharged following standard protocol.
Outcome will be assessed clinically during discharge. The parents of the
patients would be interviewed face to face by the researcher for the purpose
of collection of data. Then the patients would be examined by the
researcher for certain signs and those would be recorded in the check-list.
Few investigations would be done for supporting the diagnoses.
o) Professional assistants for experts, if applicable: N/A
p) Procedure of data analysis interpretation: After collection, data editing
and clearing will be done manually and prepared for data entry and
analysis by using SPSS version 17.
q) Quality assurance strategy: N/A
References:
1. Hamedi A, A Hosain, Nakhaee AA, Evaluation of Il-6 and High
Sensitive C Reactive Protein Value in CSF and Serum Children Suspected
Meningitis Referred to Pediatric Emergency Room. Iranian Red Cres Med
J. 2014:14(12). DOI: 10.5812/ircmj.4726
2. Salih MAM, Khaleefa OH, Bushara M, Taha ZB, Musa ZA, Kamil I,
et al. Long term sequelae of childhood acute bacterial meningitis in a
developing country. Scand J Infect Dis 1991; 23: 17582.
3. Wanyoike MN, Waiyaki PG, McLiegeyo SO, Wafula EM.
Bacteriology and sensitivity pattern of pyogenic meningitis at Kenyatta
National Hospital, Nairobi Kenya. East Afr Med J 1995; 72: 65860.
4. Prober CG, Dyner L. Infections of the central nervous system. In:
Kliegman RM, Stanton B, Behrman RE, Geme SJ, Schor N editors.
Nelsons Textbook of Pediatrics. 19th ed. Philadelphia: WB Saunders Co;
2011; p. 2086 90.
5. Chowdhury MZU, Rahman KM, Miah RA, Sattar H, Hussain T.
Bacterial Meningitis in children. Bangladesh Medical Journal 1992; 21: 3
7.
6. Neuman MI, Tolford S, Harper MB. Test characteristics and
interpretation of cerebrospinal fluid gram stain in children. Pediatr Infect
Dis J 2008 Apr; 27(4): 309-13.
7. Whittle HC, Tugwell P. Rapid bacteriological diagnosis of pyogenic
meningitis by LPAT. The Lancet 1974; 303(7858): 61921.
8. Pemde HK, Harish K. C - Reactive Protein in Childhood Meningitides.
Indian J Pediatr 1996; 63: 737.
9. Anh DD, Riewpaiboon A, Tho LH, Kim SA, Nyambat B, Kilgore P.
Treatment Costs of Pneumonia, Meningitis, Sepsis, and Other Diseases
among Hospitalized Children in Viet Nam. J Health Popul Nutr 2010;
28(5):43642.
10.Clarke D, Cost K. Use of serum C-reactive protein in differentiating
septic from aseptic meningitis in children. J Pediatr 1983; 102: 71820.
11.Singh N, Arora S, Kahlon PS. Cerebrospinal fluid C- reactive protein
in meningitis. Indian Pediatrics 1995 June; 32: 687-88.
12.Celik N, Tanir G, Aydemir C, Tuygun N, Zorlu P. [Differential
diagnosis of bacterial and viral meningitis in childhood acute meningitis:
a statistical model]. Mikrobiyoloji bulteni. [Comparative Study]. 2007;
41(1):63-9.
13.Taskin E, Turgut M, Kilic M, Akbulut H, Aygun AD. Serum
procalcitonin and cerebrospinal fluid cytokines level in children with
meningitis. Mediators of inflammation. 2004; 13(4):269-73.
14.Gendrel D, Raymond J, Assicot M, Avenel S, Lefevre H, Ravilly S, et
al. [Procalcitonin, C-reactive protein and interleukin 6 in bacterial and viral
meningitis in children]. Presse Med. [Comparative Study].
1998;27(23):1135-9.
15.Khanam R, Hanif M, Hoque MM, Tawfique M, Ahmed ASMNU. Role
of CSF C-Reactive Protein for the Differentiation of Bacterial Meningitis
from Aseptic Meningitis in Children, Bangladesh J Child Health 2012;
VOL 36 (3): 126-132
16. Prober CG, Srinivas NS, Mathew R, Cerebrospinal
fluid findings in Central Nervous System Disorder In Central Nervous
System Infections. In: Kleigman RM, Stanton BF. Schor
17. NF, Behrman RE editors. Nelson Textbook of Paediatrics 20th Ed.
Philadelphia: ELSEVIER; 2015; p.2937