ORIGINAL ARTICLE EPIDEMIOLOGY

The use of dried cerebrospinal uid lter paper spots as a substrate for
PCR diagnosis of the aetiology of bacterial meningitis in the Lao PDR

I. Elliott1,2, S. Dittrich1,2, D. Paris2,3, A. Sengduanphachanh1, P. Phoumin1 and P. N. Newton1,2

1) Lao-Oxford-Mahosot Hospital-Wellcome Trust Research Unit (LOMWRU), Microbiology Laboratory, Mahosot Hospital, Vientiane, Lao PDR, 2) Centre for
Tropical Medicine, Nufeld Department of Medicine, Churchill Hospital, University of Oxford, Oxford, UK and 3) Mahidol-Oxford Tropical Medicine Research
Programme, Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand

Abstract

We investigated whether dried cerebrospinal uid (CSF) conserved on lter paper can be used as a substrate for accurate PCR diagnosis of
important causes of bacterial meningitis in the Lao PDR. Using mock CSF, we investigated and optimized lter paper varieties, paper punch
sizes, elution volumes and quantities of DNA template to achieve sensitive and reliable detection of bacterial DNA from lter paper
specimens. FTA Elute Micro CardTM (Whatman, Maidstone, UK) was the most sensitive, consistent and practical variety of lter paper.
Following optimization, the lower limit of detection for Streptococcus pneumoniae from dried mock CSF spots was 14 genomic equivalents
(GE)/lL (interquartile range 5.5 GE/lL) or 230 (IQR 65) colony forming units/mL. A prospective clinical evaluation for S. pneumoniae, S. suis
and Neisseria meningitidis was performed. Culture and PCR performed on fresh liquid CSF from patients admitted with a clinical diagnosis of
meningitis (n = 73) were compared with results derived from dried CSF spots. Four of ve fresh PCR-positive CSF samples also tested PCR
positive from dried CSF spots, with one patient under the limit of detection. In a retrospective study of S. pneumoniae samples (n = 20), the
median (IQR; range) CSF S. pneumoniae bacterial load was 1.1 9 104 GE/lL (1.2 9 105; 1 to 6.1 9 106 DNA GE/lL). Utilizing the
optimized methodology, we estimate an extrapolated sensitivity of 90%, based on the range of CSF genome counts found in Laos. Dried CSF
lter paper spots could potentially help us to better understand the epidemiology of bacterial meningitis in resource-poor settings and guide
empirical treatments and vaccination policies.

Keywords: Bacterial meningitis, cerebrospinal uid, lter paper, Lao PDR, Streptococcus pneumoniae
Original Submission: 5 January 2013; Revised Submission: 24 March 2013; Accepted: 29 April 2013
Editor: M. Drancourt
Article published online: 2 May 2013
Clin Microbiol Infect 2013; 19: E466E472
10.1111/1469-0691.12260

portion of bacterial meningitis cases seen worldwide. As

Corresponding author: I. Elliott, Microbiology Laboratory,
elsewhere in SE Asia, Streptococcus pneumoniae, Streptococcus
Mahosot Hospital, Vientiane, Lao PDR
E-mail: [email protected] suis, Neisseria meningitidis and Haemophilus inuenzae account
for the majority of cases of community-acquired bacterial
meningitis in Laos (LOMWRU unpublished) [25]. An
improved understanding of the epidemiology of these diseases
would support evidence-based national vaccination policies
Introduction and empirical treatment guidelines. Vaccination programmes
against H. inuenzae type b, S. pneumoniae (seven and thirteen
Bacterial meningitis is a life-threatening disease that carries valent) and N. meningitidis (serotypes C, A, Y and W135) have
signicant mortality and considerable risk of severe neurolog- had a profound impact on the incidence of bacterial meningitis
ical complications. In developing countries, such as the Lao cases in many regions [610].
Peoples Democratic Republic (Laos), the combined mortality In Laos, conventional bacteriological and molecular micro-
rates of all causes of bacterial meningitis are likely to be >20% biological facilities for the diagnosis of central nervous system
[1]. Vaccine-preventable diseases account for the major disease are conned to the capital city of Vientiane, and

2013 The Authors. Clinical Microbiology and Infection published by John Wiley & Sons Ltd on behalf of the European Society of Clinical Microbiology and Infectious Disease.
Open access under CC BY license.
CMI Elliott et al. PCR diagnosis of meningitis using dried CSF on lter paper E467

antibiotic treatment is often initiated prior to lumbar puncture CardTM (Cat. no. WB120210; Whatman) (referred to as FTA
[11]. This may reduce the sensitivity of cerebrospinal uid (CSF) hereafter) and the FTA Elute Micro CardTM (referred to as FTA
culture by up to 33% [12]. The polymerase chain reaction (PCR) Elute hereafter) (Cat. No. WB120401; Whatman). Whatman
has been shown to be superior to conventional methods, 903 requires elution of the organism from the paper followed
particularly when antibiotics have been administered [13]. by lysis and DNA extraction. FTA and FTA Elute papers
Reliable procedures are necessary to preserve, transport contain a proprietary mix of reagents that lyse cell walls but
and test CSF samples, frequently requiring a challenging and stabilize and bind nucleic acids. FTA paper requires a washing
costly cold chain to be in place. Effective disease surveillance is protocol, with the paper disc itself being used as a DNA
thus severely hampered in resource-limited settings, such as template in the PCR.
those currently present in rural Laos, where procedures for A mock CSF was prepared by spiking tryptic soy broth
appropriate laboratory diagnosis are suboptimal. Innovative, (TSB) with a few discrete colonies of National Collection of
simple and inexpensive rapid diagnostic tests for detecting Type Culture (NCTC) strains of either S. pneumoniae (NCTC
pathogen antigens in CSF hold promise [1416]. In West 12977) or N. meningitidis (NCTC 10025) grown overnight on
Africa, centralized PCR systems for CSF PCR have been 5% goat blood agar. This was incubated for 4 h in 57% CO2 at
developed [17], but it is likely that prolonged transport of CSF 37C and then diluted with sterile TSB to achieve a standard-
at high temperature will reduce sensitivity. ized initial concentration (optical density) using the Nanodrop
Blood dried onto lter paper to diagnose infectious diseases 2000 (Thermo Scientic, Wilmington, DE, USA). Ten-fold
dates back to as early as 1939 [18] and has proved to be an serial dilutions were prepared and aliquots spotted onto lter
important tool for diagnosis, epidemiology and monitoring in paper (Whatman 903 and FTA, 125 lL/1 inch circle; FTA
settings with limited laboratory infrastructure. However, CSF Elute, 40 lL/11 mm circle) and onto blood agar (100 lL) to
has rarely been collected on lter paper. In the 1970s counter- determine the bacterial load (colony forming units/mL) for
immunouorescence was used to detect capsular polysaccha- each dilution using a micropipette.
ride antigen in dried purulent CSF samples and more recently Optimization studies were performed using S. pneumoniae
an enzyme-linked immunosorbent assay used to diagnose and to demonstrate inter-species reproducibility nal methods
neurocysticercosis [19,20]. Peltola et al. [21] successfully were repeated with N. meningitidis.
identied nucleic acids of S. pneumoniae and H. inuenzae type
b in CSF-impregnated lter paper strips after 8 months DNA extraction
storage at room temperature in sealed plastic bags. Discs (3 or 8 mm) were punched out of the centre of a lter
Until laboratory capacity is developed in rural Laos, lter paper circle using skin biopsy punches (Stiefel, Maidenhead,
paper could serve as a sample preservation method that can UK) and transferred to 1.5-mL microcentrifuge tubes. Three
easily be delivered to a central laboratory for CSF PCR and and 8-mm punched discs would contain c. 3 lL and 20 lL of
therefore allow better understanding of the geographical CSF, respectively. Biopsy punches were sprayed with 70%
epidemiology of bacterial meningitis. We investigated whether ethanol after each punch, dried with tissue paper and then
dried CSF spots would serve as a sensitive and reliable method punched into a stack of sterile lter paper a further ve times
for dening three of the major causative agents of bacterial to eliminate cross-contamination [22,23].
meningitis in Laos: S. pneumoniae, S. suis and N. meningitidis. The FTA Elute manufacturers instructions were followed
We investigated different types of lter paper and optimized with the following modications: 1 mL of sterile water was
the processing of dried CSF spots to obtain the maximum used to wash the disc by pulse-vortexing three times for a total
possible yield of nucleic acid. A prospective pilot study was of 5 s. The DNA was eluted with sterile water after 22.5 min
conducted utilizing these optimized methods to compare PCR at 95C, the mid-point of the manufacturer-recommended
results on DNA extracted from dried CSF spots with routine time period. FTA was processed following the manufacturers
liquid CSF and conventional bacterial culture. instructions, except for the use of a slightly larger disc (3 mm
in diameter rather than the recommended 2 mm). Whatman
Grade 903 discs were processed following the QIAamp DNA
Methods
Mini kit Dried Blood Spot Protocol (Qiagen, Crawley, UK)
with the following modications: elution was carried out in
Filter paper selection 180 lL of buffer ATL with proteinase K incubated at 56C
Three varieties of lter paper were selected for performance overnight.
evaluation: the Whatman Grade 903 (Cat. no. 10535097; Liquid CSF (n = 25) was initially processed using the
Whatman), the Flinders Technology Associates (FTA) Micro QIAamp DNA Mini kit Blood or Body Fluids Protocol

2013 The Authors. Clinical Microbiology and Infection published by John Wiley & Sons Ltd on behalf of the European Society of Clinical Microbiology and Infectious Disease. CMI, 19, E466E472
Open access under CC BY license.
E468 Clinical Microbiology and Infection, Volume 19 Number 10, October 2013 CMI

(Qiagen) after centrifuging 200 lL of CSF at 16 000 g and re- were performed as previously described, with slight modica-
suspending the pellet in buffer ATL. Subsequently (n = 48), tion [3,24,25]. A 25 lL reaction volume contained 1x Amp-
routine extraction was performed on 200 lL unspun CSF with liTaq Gold reaction buffer, 5 mM MgCl, 200 lM of each
the EZ1 Virus Mini Kit v2.0 for automated, simultaneous dNTPs (Applied Biosystems, Paisley, UK) and 1 U of AmpliTaq
purication of viral DNA and RNA as well as bacterial DNA Gold polymerase (Applied Biosystems). Forward and reverse
(Qiagen). primers/probes were included with the following concentra-
tions: S. pneumoniae, 200 nM/100 nM; S. suis, 400 nM/100 nM;
Optimization of detection N. meningitidis, 300 nM/25 nM. The cycling conditions were as
To optimize DNA yield from lter paper, a series of follows: denaturation at 95C for 10 min, followed by 40
experiments were performed in triplicate. Three different cycles of amplication, each consisting of 95C for 15 s and
quantities of DNA eluate were used in the PCR: 3, 5 and 60C for 60 s. Three lL of DNA extracted from liquid CSF
10 lL. A larger punch size of 8 mm (50.2 mm2) was compared samples served as a template for each PCR assay. All runs
with the standard size of 3 mm (7.1 mm2) (c. seven-fold were performed on a Rotor-Gene 6000 (Qiagen) and com-
increase in disc surface area). Finally, three different elution prised at least two no-template controls as well as duplicates
volumes were also investigated: 40, 80 and 145 lL. These of serial dilutions of positive controls, which were quantitated
experiments assumed an even distribution of pathogen DNA using the Quant-iT PicoGreen assay (Invitrogen, Paisley, UK),
across the paper after application. corresponding to c. 103, 102, 101 and 100 genomic equivalents
(GE)/lL.
Prospective evaluation: patients and samples
Patients admitted to Mahosot Hospital (n = 73), Vientiane, Cerebrospinal uid genome counts
between July 2011 and March 2012 with a clinical diagnosis of Previously characterized frozen CSF samples collected from
meningitis, who underwent lumbar puncture, had fresh CSF patients with bacterial meningitis admitted to Mahosot Hos-
spotted onto FTA Elute. Four circles were spotted with CSF pital between 2005 and 2012 were identied retrospectively.
and allowed to dry for 4 h before being stored in an air- These samples served to quantify the DNA loads in S. pneu-
conditioned laboratory at c. 20C in sealed plastic bags moniae PCR-positive CSF samples, using external standards in a
containing silica granules to reduce humidity. PCR on dried quantitative real-time PCR assay.
CSF spots was subsequently compared with PCR performed
on liquid CSF and routine culture. Statistical analysis
Statistical analysis was performed with Stata (v11, College
Culture Station, TX, USA) and the gure created using GraphPad
Patients CSF samples were cultured on both 5% goat blood software (v6, La Jolla, CA, USA).
and chocolate agar (and MacConkey agar in those <1 year old)
and incubated at 37C in 57% CO2. Organism identication Ethics
was performed following standard biochemical methods with Lumbar punctures were performed as part of a routine
additional conrmation provided using the API system (bio- diagnostic service if verbal (20032006) or written (2006
Merieux, Marcy lEtoile, France). 2011) consent was given by patients or their parents/guardian.
Ethical clearance was granted by the Ethical Review Commit-
Molecular identication tee of the Faculty of Medical Sciences, National University of
Three separate real-time quantitative PCRs were performed Laos, and the Oxford University Tropical Ethics Research
for S. pneumoniae, N. meningitidis and S. suis (Table 1). Assays Committee, Oxford, UK.

TABLE 1. Oligonucleotide primers

Sequence (53) Reference
and probes for detection of S. pneu-
moniae, S. suis and N. meningitidis S. pneumoniae Carvalho et al. [24]
ctrA-Forward ACGCAATCTAGCAGATGAAGCA
ctrA-Reverse TCGTGCGTTTTAATTCCAGCT
ctrA Probe ROX-GCCGAAAACGCTTGATACAGGGAG-BHQ2
S. suis Mai et al. [3]
cps2J-Forward GGTTACTTGCTACTTTTGATGGAAATT
cps2J-Reverse CGCACCTCTTTTATCTCTTCCAA
cps2J Probe FAM-TCAAGAATCTGAGCTGCAAAAGTGTCAAATTGA-TAMRA
N. meningitidis Corless et al. [25]
ctrA-Forward GCTGCGGTAGGTGGTTCAA
ctrA-Reverse TTGTCGCGGATTTGCAACTA
ctrA Probe FAM-CATTGCCACGTGTCAGCTGCACAT-BHQ1

2013 The Authors. Clinical Microbiology and Infection published by John Wiley & Sons Ltd on behalf of the European Society of Clinical Microbiology and Infectious Disease. CMI, 19, E466E472
Open access under CC BY license.
CMI Elliott et al. PCR diagnosis of meningitis using dried CSF on lter paper E469

detection c. three-fold, without any negative effect on reaction

Results
efciency.
In addition, increasing the punch size of FTA Elute from 3 mm
Filter paper selection (7.1 mm2) to 8 mm (50.2 mm2) reduced the Ct value by a
Eight 10-fold dilutions of mock CSF were prepared ranging median of 2.7 (IQR 0.55), equivalent to increasing the sensitivity
from 1 9 107 cfu/mL to 1 cfu/mL. When applied to quantita- of detection c. ten-fold (Fig. 1). Of the three elution volumes
tive real-time PCR assay with external controls, the DNA tested, the optimal volume was 40 lL, which led to the highest
templates eluted from FTA Elute consistently led to lower DNA concentration. An additional increase of the elution
cycle thresholds than DNA eluted from Whatman 903 for the volume to 80 lL minimally decreased DNA concentration
same mock CSF dilutions (median reduction 2.2 Ct values, IQR (median Ct value increase 0.22 (IQR 0.25)), whereas increasing
1.68) (Fig. 1). Additionally, triplicate sets of Ct values for each the elution volume to 145 lL resulted in larger increases in Ct
mock CSF dilution were more consistent with FTA Elute value (up to 1.1) and thus loss of sensitivity. As we aimed to
(median Ct 0.42, IQR 0.37) compared with Whatman 903 detect several pathogens from dried CSF spots using single
(median Ct 4.44, IQR 3.53), suggestive of a more efcient and target real-time PCR assays, in order to have sufcient DNA
reliable DNA yield of S. pneumoniae from the FTA Elute template, we selected a nal elution volume of 100 lL, accepting
(Fig. 1). a very small reduction in sensitivity. The quantied limit of
The FTA Micro Card was excluded after the rst phase of detection of the nal conditions, using an 8 mm disc punched out
experiments due to the limitation of paper size, which of FTA Elute paper, with 100 lL elution volume and 10 lL in the
provided insufcient DNA template for performing the PCR, was a median of 14 S. pneumoniae GE/lL spiked mock CSF
necessary number of single PCR assays required as a diagnostic (IQR 5.5), equivalent to a median mock CSF concentration of
routine for bacterial meningitis in a clinical setting. 230 cfu/mL (IQR 65). Studies to optimize the detection of
S. pneumoniae from lter paper were reproducible using
Optimization of detection N. meningitidis (data not shown).
Increasing the quantity of DNA eluted from FTA Elute paper
used in the PCR from 3 lL to 10 lL reduced the Ct values by Prospective evaluation of clinical samples
a median of 0.94 (IQR 0.41), increasing the sensitivity of FTA Elute performed consistently with a high level of
sensitivity throughout the optimization process and was
therefore selected for further evaluation. A total of 73 dried
50 CSF spots were prospectively collected from patients admitted
with a clinical diagnosis of bacterial meningitis over a 9-month
period. Dried CSF spots were stored at c. 20C in sealed
40 plastic bags containing silica granules for up to 8 months
before being processed. Of these, 5/73 (6.8%) tested positive
by quantitative real-time PCR on DNA extracted from liquid
30 CSF (four S. pneumoniae and one S. suis) and 4/73 (5%) tested
Ct value

positive by quantitative real-time PCR on DNA from dried

CSF spots (three S. pneumoniae and one S. suis). The positive
20
samples were identical, resulting in a sensitivity of 80% (95% CI
2997%) for PCR using FTA Elute. One sample was not
903 3mm
detected by FTA Elute due to a low bacterial load of <10 GE/
10 903 8mm
lL. Of the ve positive samples, two were culture positive
FTA 3mm
(both S. pneumoniae). No false-positives occurred, resulting in
FTA 8mm
a specicity of 100% (95% CI 95100%).
0
0

0
00

00

24

Cerebrospinal uid genome counts

00

10
10

Stored frozen ( 80C) DNA extracts of CSF samples from 20

CFU/ml
well-characterized PCR-positive S. pneumoniae patients col-
FIG. 1. Median (and IQR) Ct values at different concentrations of lected over a 7-year period were retrieved and their bacterial
S. pneumoniae (cfu/mL) in mock CSF on FTA Elute Micro Card and load quantitated. Genome equivalent counts in this population
Whatman Grade 903 with two punch sizes. ranged widely from 1 9 100 to 6.1 9 106 DNA GE/lL

2013 The Authors. Clinical Microbiology and Infection published by John Wiley & Sons Ltd on behalf of the European Society of Clinical Microbiology and Infectious Disease. CMI, 19, E466E472
Open access under CC BY license.
E470 Clinical Microbiology and Infection, Volume 19 Number 10, October 2013 CMI

(median 1.1 9 104 GE/lL, IQR 1.2 9 105 GE/lL) (Table 1). can simplify the process of detecting multiple pathogens and
Given the bacterial loads from this small but representative potentially reduce costs. However, careful optimization must
sample of patients with S. pneumoniae meningitis presenting to be performed to ensure that sensitivity and specicity are not
a hospital in Vientiane, we would theoretically expect a reduced [26]. We estimated that c. 230 cfu/mL or 14 GE/lL of
positive result in 90% of the cases, applying the optimized CSF could be reliably detected by PCR from CSF dried on
methods. Very few positive N. menigitidis (n = 4), S. suis lter paper. In a study by La Scolea et al. [27] in the USA, 85%
(n = 5) and H. inuenzae (n = 6) samples were available (data of CSF samples from children with bacterial meningitis had
not shown). counts >103 cfu/mL and 56% had counts >105 cfu/mL. Samples
collected in Laos over a 7-year period showed a wide range in
GE/lL of S. pneumoniae with a median of 1.1 9 104. This is
Discussion
consistent with other studies that report wide ranges and
medians between 4.6 9 104 and 5.7 9 105 DNA copies/lL in
In this study, we aimed to evaluate the use of dried CSF spots Latin America, Australia and Malawi [21,2830] (Table 2).
for the identication of three of the major causative agents of One single study reports the PCR identication of H. inu-
bacterial meningitis in Laos and to identify the role/potential enzae and S. pneumoniae from lter paper with sensitivities of
limitations of using a lter paper-based surveillance approach 70% and 92% and specicities of 99% and 100%, respectively
to investigate the epidemiology of CNS disease aetiologies. [21]. CSF was taken from children in South America,
Our ndings suggest that this can be accomplished using FTA transported to Finland at 20C and subsequently thawed
Elute with the optimized methods, yielding an estimated and spotted onto an unspecied Whatman lter paper. Sixty-
sensitivity of 90% when compared with the reference standard four per cent of CSF samples were taken before antibiotics
real-time diagnostic PCR assay on liquid CSF. Investigation with were administered. The range of DNA copies/lL for S. pneu-
mock CSF suggested that FTA Elute gave the most consistent moniae was 5 9 100 to 9.2 9 106, with a median of 1.6 9 104,
and sensitive results compared with Whatman 903. FTA paper very similar to that seen in Laos (Table 1).
discs used as a DNA template in the PCR proved impractical An additional limitation was that we were only able to
for testing several pathogens from a single dried CSF spot using prospectively test our methods on a limited number of
single assays. We developed our methods using TSB that does samples and pathogens. Other studies have shown that
not contain all the constituents of human CSF. However, the H. inuenzae is readily identied from dried CSF spots [21]
prospective evaluation of clinical samples conrmed our and experiments using mock CSF detected N. meningitidis
ndings during the optimization process. without difculty. Cryptococcus spp. are also important causes
Despite using a larger than manufacturer-recommended of meningitis in Laos and preliminary investigations show that
punch size to increase the lower limit of detection, the volume this would also be detectable by PCR from dried CSF spots
of CSF used from paper is c. 10 times less than that used for (unpublished data). If other central nervous system pathogens
routine PCR on liquid CSF (20 lL vs. 200 lL). Even this may were detectable in CSF using the optimized lter paper
be slightly overestimated due to possible diffusion outside the method, this would further increase the potential of using this
marked circle. As a result, despite the use of highly sensitive methodology as a surveillance tool in areas without laboratory
quantitative real-time PCR assays, those CSF samples with very capacity.
low bacterial loads can fall under the limit of detection. The We did not investigate the practicalities and suitability of
optimization studies demonstrated that, beyond the optimal collecting dried CSF spots in smaller regional hospitals and
cut-off, DNA concentrations in the eluate declined with transporting these for analysis. Although the effects of
increasing elution volumes. A small reduction in DNA humidity and storage conditions on dried CSF spots were
concentration was accepted for the nal methodology to not assessed, DNA was readily detected after 8 months
allow sufcient PCR template to be available. Multiplex PCR storage in an air-conditioned room. A simple standard

TABLE 2. Studies reporting median and ranges of S. pneumoniae copies/lL in CSF from different geographical regions

Country Age Study size Median copies/lL Range Reference

Laos All 20 patients 1.1 9 104 126.1 9 106 This manuscript

Venezuela & Paraguay 2 months14 years 129 CSF samples 1.6 9 105 59.3 9 106 Peltola et al. [21]
Latin Americaa 2 months14 years 121 patients 4.6 9 104 09.3 9 106 Roine et al. [28]
Malawi 2 months16 years 82 patients 5.8 9 104 446.2 9 105 Carrol et al. [30]
Australia All 6 patients 1.1 9 104 7.66 9 105 van Haeften et al. [29]
a
Argentina, Dominican Republic, Brazil, Ecuador, Paraguay and Venezuela.

2013 The Authors. Clinical Microbiology and Infection published by John Wiley & Sons Ltd on behalf of the European Society of Clinical Microbiology and Infectious Disease. CMI, 19, E466E472
Open access under CC BY license.
CMI Elliott et al. PCR diagnosis of meningitis using dried CSF on lter paper E471

operating procedure was provided to laboratory staff for the

References
preparation of dried CSF spots.
FTA Elute paper inactivates pathogens within a short period
of application onto the paper [31], thus increasing the safety of 1. Durand ML, Calderwood SB, Weber DJ et al. Acute bacterial
handling and allowing easy transportation to a central labora- meningitis in adults. A review of 493 episodes. N Engl J Med 1993;
328: 2128.
tory in resource-limited settings. As a simple, inexpensive and 2. Moore CE, Sengduangphachanh A, Thaojaikong T et al. Enhanced
reliable sample specimen preservation tool, the lter paper determination of Streptococcus pneumoniae serotypes associated with
methodology has the potential to contribute to an improved invasive disease in Laos by using a real-time polymerase chain reaction
serotyping assay with cerebrospinal uid. Am J Trop Med Hyg 2010; 83:
understanding of the epidemiology of bacterial meningitis in
451457.
resource-limited and rural regions of the world. This approach 3. Mai NTH, Hoa NT, Nga TVT et al. Streptococcus suis meningitis in Adults
could support and guide vaccination programmes and empirical in Vietnam. Clin Infect Dis 2008; 46: 659667.
4. Taylor WR, Nguyen K, Nguyen D et al. The spectrum of central
treatment protocols to reduce the morbidity and mortality
nervous system infections in an adult referral hospital in Hanoi,
associated with this life-threatening disease. Vietnam. PLoS ONE 2012; 7: e42099.
5. Ho Dang Trung N, Le Thi Phuong T, Wolbers M et al. Aetiologies of
central nervous system infection in Viet Nam: a prospective provincial
Funding hospital-based descriptive surveillance study. PLoS ONE 2012; 7: e37825.
6. Peltola H. Worldwide Haemophilus inuenzae type b disease at the
beginning of the 21st century: global analysis of the disease burden
Wellcome Trust of Great Britain. 25 years after the use of the polysaccharide vaccine and a decade after
the advent of conjugates. Clin Microbiol Rev 2000; 13: 302317.
7. Peltola H, Salo E, Saxen H. Incidence of Haemophilus inuenzae type b
meningitis during 18 years of vaccine use: observational study using
Acknowledgements routine hospital data. BMJ 2005; 330: 1819.
8. Whitney CG, Farley MM, Hadler J et al. Decline in invasive pneumo-
coccal disease after the introduction of protein-polysaccharide conju-
We are very grateful to all the doctors and nursing staff of gate vaccine. N Engl J Med 2003; 348: 17371746.
Mahosot Hospital and participating hospitals, the staff of the 9. Kyaw MH, Lyneld R, Schaffner W et al. Effect of introduction of the
Microbiology Laboratory, especially Rattanaphone Phetsou- pneumococcal conjugate vaccine on drug-resistant Streptococcus pneu-
moniae. N Engl J Med 2006; 354: 14551463.
vanh, David Dance, Anisone Chanthongthip and Sengmany 10. Snape MD, Pollard AJ. Meningococcal polysaccharide-protein conjugate
Symanivong, and to the Minister of Health and the Director of vaccines. Lancet Infect Dis 2005; 5: 2130.
the Curative Department, Ministry of Health, for their support 11. Khennavong M, Davone V, Vongsouvath M et al. Urine antibiotic
activity in patients presenting to hospitals in Laos: implications for
for this study. We also thank Dr Pieter Smit for useful
worsening antibiotic resistance. Am J Trop Med Hyg 2011; 85: 295302.
comments on the manuscript. Part of this work was presented 12. Talan DA, Hoffman JR, Yoshikawa TT, Overturf GD. Role of empiric
as a poster at the 8th International Healthcare Infection parenteral antibiotics prior to lumbar puncture in suspected bacterial
meningitis: state of the art. Rev Infect Dis 1988; 10: 365376.
Society (HIS) and Federation of Infection Societies (FIS)
13. Schuurman T, de Boer RF, Kooistra-Smid AM, van Zwet AA.
meeting on 1921 November 2012 at the BT Convention Prospective study of use of PCR amplication and sequencing of 16S
Centre, Liverpool, UK, as The use of dried cerebrospinal uid ribosomal DNA from cerebrospinal uid for diagnosis of bacterial
lter paper spots to diagnose the major causes of bacterial meningitis in a clinical setting. J Clin Microbiol 2004; 42: 734740.
14. Rose AM, Mueller JE, Gerstl S et al. Meningitis dipstick rapid test:
meningitis in the Lao PDR. evaluating diagnostic performance during an urban Neisseria meningitidis
serogroup A outbreak, Burkina Faso, 2007. PLoS ONE 2010; 5: e11086.
15. Rose AM, Gerstl S, Mahamane AE et al. Field evaluation of two rapid
Author Contributions diagnostic tests for Neisseria meningitidis serogroup A during the 2006
outbreak in Niger. PLoS ONE 2009; 4: e7326.
16. Chanteau S, Dartevelle S, Mahamane AE, Djibo S, Boisier P, Nato F.
IE and PNN conceived the idea. IE, SD, DP and PNN New rapid diagnostic tests for Neisseria meningitidis serogroups A,
W135, C, and Y. PLoS Med 2006; 3: e337.
developed the study methodology. IE performed the labora-
17. Chanteau S, Sidikou F, Djibo S, Moussa A, Mindadou H, Boisier P.
tory work and wrote the manuscript. AS and PP performed Scaling up of PCR-based surveillance of bacterial meningitis in the
the routine diagnostic laboratory work on CSF specimens. All African meningitis belt: indisputable benets of multiplex PCR assay in
the authors contributed to reviewing the nal manuscript. Niger. Trans R Soc Trop Med Hyg 2006; 100: 677680.
18. Zimmermann E. Die Trockblutprobe auf Syphilis; ein Beitrag zu ihrer
Vereinfachung. Munch Med Wochenschr 1939; 2: 17321733.
19. Whittle HC, Greenwood BM, Gabrielse L, Crockford J. The
Transparency Declaration bacteriological diagnosis of pyogenic meningitis from cerebrospinal
uid dried on lter paper. Trans R Soc Trop Med Hyg 1976; 70: 217
218.
No conicts of interest.

2013 The Authors. Clinical Microbiology and Infection published by John Wiley & Sons Ltd on behalf of the European Society of Clinical Microbiology and Infectious Disease. CMI, 19, E466E472
Open access under CC BY license.
E472 Clinical Microbiology and Infection, Volume 19 Number 10, October 2013 CMI

20. Fleury A, Bouteille B, Garcia E et al. Neurocysticercosis: validity of cases of Meningitis and Septicemia using real-time PCR. J Clin Microbiol
ELISA after storage of whole blood and cerebrospinal uid on paper. 2001; 39: 15531558.
Trop Med Int Health 2001; 6: 688693. 26. Gunson RN, Collins TC, Carman WF. Practical experience of high
21. Peltola H, Roine I, Leinonen M et al. Diagnosis of Streptococcus throughput real time PCR in the routine diagnostic virology setting.
pneumoniae and Haemophilus inuenzae type b meningitis by identifying J Clin Virol 2006; 35: 355367.
DNA from cerebrospinal uid-impregnated lter paper strips. Pediatr 27. La Scolea LJ Jr , Dryja D. Quantitation of bacteria in cerebrospinal uid
Infect Dis J 2010; 29: 111114. and blood of children with meningitis and its diagnostic signicance.
22. Driver GA, Patton JC, Moloi J, Stevens WS, Sherman GG. Low risk of J Clin Microbiol 1984; 19: 187190.
contamination with automated and manual excision of dried blood 28. Roine I, Saukkoriipi A, Leinonen M, Peltola H. Microbial genome count
spots for HIV DNA PCR testing in the routine laboratory. J Virol in cerebrospinal uid compared with clinical characteristics in pneu-
Methods 2007; 146: 397400. mococcal and Haemophilus inuenzae type b meningitis in children.
23. Mitchell C, Kraft K, Peterson D, Frenkel L. Cross-contamination during Diagn Microbiol Infect Dis 2009; 63: 1623.
processing of dried blood spots used for rapid diagnosis of HIV-1 29. van Haeften R, Palladino S, Kay I, Keil T, Heath C, Waterer GW. A
infection of infants is rare and avoidable. J Virol Methods 2010; 163: 489491. quantitative LightCycler PCR to detect Streptococcus pneumoniae in
24. Carvalho MdGS, Tondella ML, McCaustland K et al. Evaluation and blood and CSF. Diagn Microbiol Infect Dis 2003; 47: 407414.
improvement of real-time PCR assays targeting lytA, ply, and psaA 30. Carrol ED, Guiver M, Nkhoma S et al. High pneumococcal DNA loads
genes for detection of pneumococcal DNA. J Clin Microbiol 2007; 45: are associated with mortality in Malawian Children with invasive
24602466. pneumococcal Disease. Pediatr Infect Dis J 2007; 26: 416422.
25. Corless CE, Guiver M, Borrow R, Edwards-Jones V, Fox AJ, 31. Abdelwhab EM, Luschow D, Harder TC, Hafez HM. The use of FTA
Kaczmarski EB. Simultaneous detection of Neisseria meningitidis, lter papers for diagnosis of avian inuenza virus. J Virol Methods 2011;
Haemophilus inuenzae, and Streptococcus pneumoniae in suspected 174: 120122.

2013 The Authors. Clinical Microbiology and Infection published by John Wiley & Sons Ltd on behalf of the European Society of Clinical Microbiology and Infectious Disease. CMI, 19, E466E472
Open access under CC BY license.