See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/24418367

Differential diagnosis of acute central nervous system infections in children

using modern microbiological methods

Article  in  Acta Paediatrica · June 2009

DOI: 10.1111/j.1651-2227.2009.01336.x · Source: PubMed

CITATIONS READS

11 558

7 authors, including:

Pasi Huttunen Maija Lappalainen

Helsinki University Central Hospital Helsinki University Central Hospital
44 PUBLICATIONS   431 CITATIONS    127 PUBLICATIONS   3,688 CITATIONS   

SEE PROFILE SEE PROFILE

Eeva Salo Tuula Lönnqvist

Helsinki University Central Hospital Helsinki University Central Hospital
67 PUBLICATIONS   1,379 CITATIONS    104 PUBLICATIONS   1,967 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Central Nervous System CNS View project

Ceftazidime View project

All content following this page was uploaded by Tuula Lönnqvist on 12 December 2017.

The user has requested enhancement of the downloaded file.

 Acta Pædiatrica ISSN 0803–5253

REGULAR ARTICLE

Differential diagnosis of acute central nervous system infections in children

using modern microbiological methods
Pasi Huttunen (pasihut@iki.fi)1, Maija Lappalainen2, Eeva Salo3, Tuula Lönnqvist4, Pia Jokela2, Timo Hyypiä5, Heikki Peltola3
1.Department of Pediatric Hematology, Oncology and Stem Cell Transplantation, the Hospital for Children and Adolescents, Helsinki, Finland
2.Department of Virology, Laboratory Services (HUSLAB), Helsinki University Hospital, Helsinki, Finland
3.Department of Pediatric Infectious Diseases, the Hospital for Children and Adolescents, Helsinki, Finland
4.Department of Pediatric Neurology, the Hospital for Children and Adolescents, Helsinki, Finland
5.Department of Virology, University of Turku, Turku, Finland

Keywords Abstract
Central nervous system infections, Children, Aim: Except bacterial meningitis, the agents causing acute central nervous system (CNS) infections in
Encephalitis, Meningitis, Specific microbiological
children are disclosed in only approximately half of the cases, and even less in encephalitis. We
diagnosis
studied the potential of modern microbiological assays to improve this poor situation.
Correspondence
Methods: In a prospective study during 3 years, all children attending hospital with suspected CNS
Pasi Huttunen, M.D., Ph.D., Hospital for Children
and Adolescents, Helsinki University Hospital, 281, infection were examined using a wide collection of microbiological tests using samples from the
00029 HUS, Finland. cerebrospinal fluid, serum, nasal swabs and stool.
Tel: +358-40-52 88581
Results: Among 213 patients, 66 (31%) cases suggested CNS infection and specific aetiology was
Fax: +358-9-471-74707
Email: pasihut@iki.fi identified in 56 patients. Of these microbiologically confirmed cases, viral meningitis ⁄ encephalitis was
diagnosed in 25 (45%), bacterial meningitis in 21 (38%) and neuroborreliosis in 9 (16%) cases
Received
15 December 2008; revised 29 March 2009; while 1child had fungal infection. In meningitis patients, the causative agent was identified in 85%
accepted 1 April 2009. (35 ⁄ 41) cases and in encephalitis in 75% (12 ⁄ 16). The most common bacteria were Streptococcus
DOI:10.1111/j.1651-2227.2009.01336.x agalactiae, Streptococcous pneumonie and Neisseria meningitidis, while the most frequently
detected viruses were enteroviruses and varicella zoster virus.

Conclusion: In 75% to 85% of paediatric CNS infections, specific microbiological diagnosis was obtained with
modern laboratory techniques. The results pose a basis for prudent approach to these potentially serious
diseases.

INTRODUCTION some molecular methods may be too expensive for routine

Specific aetiological diagnosis of acute central nervous sys-
tem (CNS) infection in a child is a continuous challenge to
paediatrician (1,2). The agents causing bacterial meningitis
in industrialized countries are identified in more than 80% of
cases (3), but in aseptic meningitis, and especially encepha-
litis, the aetiology remains more often unknown than settled
(1). Several reasons explain this shortcoming: clinical sam-
ples are not always optimal for the assays used, some agents
are not detected with the methods available, and lamentably,
Abbreviations
CNS, central nervous system; CSF, cerebrospinal fluid; PCR,
polymerase chain reaction; FNP, facial nerve paresis; HSV, hu-
man herpesvirus; VZV, varicella zoster virus; HHV6, human
herpesvirus 6; M. pneumoniae, Mycoplasma pneumoniae; B.
burgdorferi, Borrelia burgdorferi; Inf, influenzavirus; Pin, parain-
fluenzavirus; RSV, respiratory syncytial virus; Aden, adenovirus;
hEV, human enterovirus; CBV, coxsackie B virus; CMV, cy-
tomegalovirus; EBV, Epstein Barr virus; Toxo, Toxoplasma
gondii; C. pneumoniae, Chlamydia pneumoniae; HIV, human
immunodeficiency virus; S. pneumoniae, Streptococcus pneu-
moniae; S. pyogenes, Streptococcus pyogenes; GBS, Group B
Streptococcus; E. coli, Escherichia coli.
use.
Traditionally, the agents causing bacterial meningitis are
detected in cerebrospinal fluid (CSF) samples by Gram stain-
ing, culture or by utilizing various antigen detection meth-
ods. Virus culture and serology have also been used, but
these approaches are often too slow to benefit the patient,
although important epidemiological information is gathered
by this way. During the past two decades, techniques based
on the polymerase chain reaction (PCR) have revolutionized
the specific diagnostics of CNS infections (4–6). However,
infections such as Lyme neuroborreliosis (7) and tick-borne
encephalitis still require antibody determination from CSF
and ⁄ or serum. Among viral CNS infections, herpes group,
arbo- and enteroviruses pose a worldwide problem, and
should therefore not be neglected (8).
All this means that modern aetiological diagnosis of CNS
infections requires a collection of microbiological tests that
cover at least the majority of pathogenic agents in a par-
ticular geographic region. Therefore for the development of
a relevant test panel, information on the microbes causing
CNS disease in the area is needed. We carried out a prospec-
tive study in Finland to characterize the whole spectrum of
the CNS pathogens in detail by using modern diagnostic
methods.

1300 ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1300–1306
Huttunen et al.
MATERIALS AND METHODS
All immunocompetent patients at the age from 1 month
to 16 years attending the Helsinki University Central Hos-
pital, Hospital for Children and Adolescents, Helsinki,
Finland, with suspected acute CNS infection from January
1, 2004 to December 31, 2006, comprised the study popu-
lation. Approximately 300 000 children less than 16 years
live in the area, and more than 10 000 on-call patients
attend the outpatient ward annually. The child was included
in the study only if a written consent was obtained from the
legal guardian. Five children were excluded from the study
because written consent was not obtained.
Clinical diagnosis
The patients were diagnosed as having meningitis if they
had symptoms and signs of meningeal inflammation without
evidence of brain parenchymal involvement for less than 2
weeks, and the first CSF white blood cell (WBC) count was
‡5 · 106 ⁄ L. This limit for WBC count was chosen because
it has been previously used in both paediatric (9) and adult
(10) studies.
Acute encephalitis ⁄ meningoencephalitis was defined as a
disease with symptoms and signs suggesting cerebral in-
volvement: altered conciousness, epileptic seizures, ataxia
(cerebellitis) or focal neurological signs, and either an ele-
vated CSF WBC count (‡5 · 106 ⁄ L) or electroencephalo-
graphic (EEG) changes, these sometimes with neuroradiological
findings, and no evidence of other aetiology.
Microbiological studies
The set of microbiological tests used is shown in Table 1
panel A. In case of CSF pleocytosis (CSF leucocytes ‡5 ·
106 ⁄ L) or suspicion of encephalitis, tests shown in Table 1
panel B was added to the panel. If deemed necessary by the
attending paediatrician, some further tests (Table 1 panel
C) were still performed. Recent disclosure of human pare-
choviruses (HPEVs) led to its search by PCR from frozen
CSF samples, which were occasionally used also for some
other individual tests (Table 1 panel D).
The samples for CSF bacteria and virus culture, PCR
and the antibody assays were taken on admission, and, if
clinically relevant, a second sample for serology was taken
2 weeks later. Faecal and throat swab specimens were
collected during the first days in hospital. All specimens
were analyzed at the Laboratory Services (HUSLAB) of
the Helsinki University Hospital. Lyme neuroborreliosis was
classified as a group of its own, because the clinical picture
is rather distinctive (11).
Microbiological methods
The PCR for HSVs was performed according to Piiparinen
and Vaheri (12) using the probes previously reported (13).
HHV-6 PCR was done using the primers designed by Gopal
et al. (14) and the HHV-6-probe reported by Pitkäranta
et al. (15). PCR for VZV was done as previously described
(16,17). For enteroviruses and parechoviruses, the primers
and probes derived from the highly conserved 5’noncoding
region of the genome were used (18,19).
ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1300–1306
Differential diagnosis of acute central nervous system infections
Table 1 The microbiological tests used
A. Diagnostical tests carried out for all the studied patients (n ¼ 213)
CSF (cerebrospinal fluid)
• Basic tests (cells, protein, glucose)
• Bacterial staining and culture
• Virus culture
• Herpes virus PCR (HSV1 ⁄ 2, VZV, HSV6)
• Entero- and rhinovirus RT-PCR
• M. pneumoniae PCR
• B. burgdorferi PCR and IgG ⁄ IgM antibodies
• HSV1, HSV2, VZV, HHV6, M. pneumoniae IgG ⁄ IgM antibodies
• IgG antibodies studied by an immunoassay (InfA, InfB, Pin1, RSV,
Aden, hEVs, CBV5)
Serum
• B. burgdorferi IgG ⁄ IgM antibodies
• HSV1, HSV2, VZV, HHV6, M. pneumoniae IgG ⁄ IgM antibodies
• IgG antibodies studied by an immunoassay (InfA, InfB, Pin1, RSV, Aden,
hEVs, CBV5)
Stool
• Virus culture
B. Additional diagnostic tests from the patients with probable CNS
infection (n ¼ 55)
CSF
• Sample to freezer for additional studies
Serum
• Parvovirus, CMV, EBV, Toxo antibodies
• Entero- and rhinovirus RT-PCR
• Antibodies in paired sera samples (n ¼ 15) studied by
complement fixation (CF) technique (Aden, InfA, InfB, Pin, RSV, HSV,
VZV, CMV, hEV, M. pneumoniae, C. pneumoniae, Toxo). Obtained only
if clinically indicated for a check up appointment.
• Sample to freezer for additional studies
NPS (nasopharyngeal secretion)
• Aden, RSV, InfA, InfB, Pin1, Pin2 ja Pin3 antigens
• M. pneumoniae and C. pneumoniae PCR
Stool
• Electron microscopy from the stool specimens
C. Additional serology at the clinician’s discretion
• Tick-borne encephalitis IgG ⁄ IgM antibodies (n ¼ 9)
• Inkoovirus IgG ⁄ IgM antibodies (n ¼ 9)
• HIV antibodies (n ¼ 2)
D. Additional test performed from the frozen CSF samples
• Parechovirus 1–6 PCR (n ¼ 24)
• Additional tests for certain clinical samples according to the clinical
suspicion (n ¼ 5)
Mycoplasma pneumoniae nucleic acid test was based
on a 16S rRNA amplification (20). The genome of Borre-
lia burgdorferi was amplified by two different PCR meth-
ods. One amplified an area of the 16S ribosome coding
gene, and the other the OspA gene, as described by Nocton
et al. (21). Amplicons were sequenced at the Core Facil-
ity Sequencing Unit of Haartman Institute, University of
Helsinki, to verify the identity of the result.
Virus culture of the CSF, faecal and throat swab speci-
mens were performed in cell cultures designed for the de-
tection of HSV, VZV, enteroviruses and common respiratory
viruses. Antibodies were measured using both commer-
cial and inhouse enzyme immunoassays. In detail, HSV1
and HSV2 antibody tests were performed according to the
1301
Differential diagnosis of acute central nervous system infections
manufacturer’s instructions using HerpeSelect 1 ELISA IgG,
HerpeSelect 2 ELISA IgG (Focus Technologies, Cypress,
CA, USA) and EIAgen Herpes Simplex Virus IgM kits
(Adaltis, Casalecchio Di Reno, BO, Italy). For VZV IgG
antibodies an inhouse immunoassay was used (22), whereas
VZV IgM antibodies were detected according to the manu-
facturer’s instructions (Adaltis EIAgen Varicella IgM kit,
Adaltis Italia). HHV6 IgG and IgM antibodies were mea-
sured using commercial kit (ELISA Test, Panbio, Brisbane,
Australia). For measurement of Borrelia antibodies, pre-
viously described method was used (23). M. pneumoniae
immunoassays were performed according to the manufac-
turer’s instructions (M. pneumoniae IgG EIA and M. pneu-
moniae IgM EIA, AniLabsystems Ltd. Oy, Vantaa, Finland).
Antibodies against respiratory viruses were detected as
described earlier (1).
For Toxoplasma, CMV and EBV antibody detection
commercial kits were used; for Toxoplasma: VIDAS
Toxo IgG II, VIDAS Toxo IgM and VIDAS Toxo IgG
Avidity (bioMérieux, Marcy-l’Etoile, France); for CMV:
VIDAS CMV IgG, VIDAS CMV IgM and VIDAS CMV IgG
Avidity (bioMérieux, Marcy-l’Etoile, France), and for EBV:
Dade Behring Enzygnost Anti-EBV ⁄ IgG ja Anti-EBV ⁄ IgM
II (Marburg, Germany). The EBV IgG Avidity test was per-
formed as described earlier (24). For parvovirus IgG and IgG
avidity, an inhouse assay was used (25). Parvovirus IgM test
was performed according to the manufacturer’s instructions
(Parvovirus B19 IgM Enzyme Immunoassay; Biotrin Ltd.,
Dublin, Ireland).
Respiratory viruses were identified with a commercial
antigen detection assay using immunofluorescent antibod-
ies (Respiratory DFA Viral Screening & identification Kit,
Light Diagnostics, Chemicon International, Inc) according
to the manufacturer’s instructions.
The diagnosis was deemed confirmed if a microbe or
microbe-specific nucleic acid was detected from the CSF, or
in the case of neuroborreliosis, if specific IgM (with or with-
out IgG) antibodies were found in the CSF. The diagnosis
was considered probable if specific intrathecal antibody pro-
duction was found, or if simultaneous CSF pleocytosis and
serum IgM antibodies for only one specific microbe were ob-
served. The criterium for specific intrathecal IgG production
was an at least 20-fold CSF-to-serum ratio of a microbe spe-
cific antibody without evidence of other intrathecal antibod-
ies (26,27). If CNS infection caused by EBV or adenovirus
was suspected based on serological findings or intrathecal
antibody production, respectively, the diagnosis was con-
firmed by detecting microbe-specific nucleic acid from the
CSF samples.
RESULTS
Acute CNS infection was suspected in 213 patients ad-
mitted to the hospital. Meningitis, meningoencephalitis or
encephalitis was found in 66 (31%) patients of whom 41
(62%) were diagnosed as meningitis, 16 (24%) as encephali-
tis (meningoencephalitis) and 9 (14%) as neuroborreliosis,
and the patients were treated accordingly. Acute CNS infec-
1302 ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1300–1306
Huttunen et al.
tions comprised 0.2% of all the diagnoses by the paediatric
on-call service in 2004–2006.
Aetiology
An agent was identified in 56 cases (85% of the CNS infec-
tions): 21 (38%) were from bacterial meningitis, 25 (45%)
from viral meningitis ⁄ encephalitis, 9 (16%) from neurobor-
reliosis, while 1 child had a fungal infection. The CSF find-
ings are listed in Table 2. For all meningitides, the aetiology
was revealed in 35 ⁄ 41 (85%) patients, and for encephali-
tides ⁄ meningoencephalitides in 12 ⁄ 16 (75%) of the cases.
Except neuroborreliosis, an agent was identified only by cul-
ture in 21 (45%; mainly bacteria) cases, only by PCR in 19
(40%; all viruses) cases, and by both methods in 7 (15%)
cases. For neuroborreliosis, 8 ⁄ 9 (89%) cases were identified
only by intrathecal antibody production (one case was also
PCR-positive). There were three probable microbiological
diagnoses (Table 2) added to the 56 confirmed cases.
Bacterial meningitis
The three most common bacterial agents, accounting for
90% of meningitis, were Streptococcus agalactiae (n ¼ 9),
Streptococcus pneumoniae (n ¼ 6) and Neisseria meningiti-
dis (n ¼ 4). One case was due to Streptococcus bovis, and
in one case of probable bacterial meningitis the pathogen
remained unidentified.
Aseptic meningitis ⁄ encephalitis
The microbial aetiology of illness is shown agentwise in
Table 2. Three-quarters of all viral CNS infections were
caused by enteroviruses, VZV or HHV6. Enteroviruses
accounted for 44% (n ¼ 13) of the confirmed cases (9 menin-
gitis, and 4 encephalitis ⁄ meningoencephalitis), and a typical
seasonal peak was observed in September and October (n
¼ 8). VZV was detected in seven patients (5 meningitis, 2
encephalis; 24% of the all aseptic meningitis) and interest-
ingly, all the 5 patients manifested with VZV meningitis had
chickenpox or zoster within 2 weeks before the onset of neu-
rological symptoms. HHV6 was identified in three patients;
four cases showed reactivation of this virus in the context of
another infection (coinfection detected from the CSF sam-
ples). In addition, in one confirmed and one probable case
adenovirus was found, one confirmed case was positive for
Candica albicans, one for EBV, and one for M. pneumoniae.
One patient had probably CMV infection, another one Toxo-
plasma gondii. Interestingly, we found no evidence from the
CSF samples for such pathogens as HSV 1 ⁄ 2, rhinoviruses,
influenzavirus A ⁄ B, parainfluenzavirus 1–3 or RSV.
Initial blood and CSF leucocyte count, CRP
The initial blood and CSF leucocyte counts and serum CRP
values are shown in Figure 1. In bacterial meningitis, the med-
ian blood and CSF leucocyte counts were 28.8 (range 1.8–
31.9) · 109 ⁄ L, and 852 (1–11.360) · 106 ⁄ L, while the
median CRP values were 203 (7–345) mg ⁄ L. In aseptic menin-
gitis ⁄ encephalitis, the corresponding values were 6.8 (3.3–
14.8) · 109 ⁄ L, 30 (0–1760) · 106 ⁄ L, and <5 (<5–44) mg ⁄ L.
In neuroborreliosis and other cases (Table 2), median blood
Huttunen et al. Differential diagnosis of acute central nervous system infections

Table 2 The CNS infections and aetiological agents found in the study
Aetiology Age (y) at diagnosis Confirmed n Probable n % of all the CNS infections (n ¼ 66)

A. Meningitis (n ¼ 41) 62
1. Bacteria
S. agalactiae (GBS) 0.1–0.2 9 –
S. pneumoniae 0.4–3.3 6 –
S. bovis 0.1 1 –
N. meningitidis B ⁄ C 0.5–8.2 4; 2B ⁄ 2C –
2. Viruses
Enteroviruses (n) 0.1–14.9 9 –
Echovirus 9 (2)
Echovirus 11 (1)
Echovirus 25 (1)
Coxsackievirus B4 (1)
Coxsackievirus B5 (1)
Untypeable (3)
Varicella zoster virus (VZV) 3.4–7.2 5 –
3. Others
Candida albicans 11.8 1 –
B. Encephalitis (n ¼ 16) 24
1. Bacteria
M. pneumoniae 3.3 1 –
2. Viruses
Enteroviruses 2.3–12.0 4 (untypeable) –
Human herpesvirus 6 (HHV6) 1.0–1.2 3 primary infections* –
Varicella zoster virus (VZV) 4.2–16.2 2 –
Adenovirus 3.2–15.9 1 1
Epstein Barr virus (EBV) 5.0 1 –
Cytomegalovirus (CMV) 0.9 – 1
3. Others
T. gondii 13.5 – 1
C. Neuroborreliosis (n ¼ 9) 14
B. burgdorferi 5.5–12.2 9** –
*
In addition reactivation of HHV6 was detected in four cases: in two patients with neuroborreliosis, one during EBV infection and one associated with enterovirus
meningoencephalitis.
**
Six out of nine neuroborreliosis meningitis cases were also manifested with facial nerve paresis (FNP).

WBC was 12.3 (3.2–16.3) · 109 ⁄ L, CSF WBC 30 (0–521) ·
106 ⁄ L, and serum CRP value <5 (<5–53) mg ⁄ L. Large varia-
tion on the values was characteristic of all groups. How-
ever, high (>20 · 109 ⁄ L) blood or CSF (>2000 · 106 ⁄ L)
leucocyte count, and especially, serum CRP value exceeding
20 mg ⁄ L (28) were valuable hints suggesting bacterial in-
fection. All patients with bacterial meningitis showing CRP
value below 20 mg ⁄ L (n ¼ 3) were under 2 years old.
Differential diagnosis
As shown in Table 3 panel A, no less than 147 patients
showed symptoms and signs suggestive of a CNS infection,
but based on normal CSF (and sometimes EEG) findings
such diagnosis was not finally reached. In bacterial infec-
tions, this occurred most often for by S. pneumoniae (n ¼
9), M. pneumonie (8), E. coli (4), or S. pyogenes (4), while
in viral infections so was the case in influenza (4), and in
manifestations due to RSV (4), rotavirus (4), adenoviruses
(4) and enteroviruses (3).
The three most common differential clinical diagnoses
(Table 3 panel B) were confirmed or suspected septicaemia
(n ¼ 20), febrile convulsion (15) and epilepsy (15). Occa-
sionally, rarities such as opsoclonus ataxia (shown finally
to be caused by neuroblastoma), acute disseminated en-
cephalomyelitis (ADEM), and Kawasaki disease were di-
agnosed.
Facial nerve paresis (FNP)
This manifestation was found in 19 patients of whom
6 (32%) had neuroborreliosis and 3 (16%) VZV infection
within prior to 2–4 weeks, while 2 (11%) patients were
IgM-positive for M. pneumoniae. In the remaining eight
(42%) cases, the likely aetiology remained unidentified. The
FNP cases associated with neuroborreliosis also met the
criteria for borrelia meningitis (meningeal inflammation
together with pleocytosis and specific IgM in the CSF),
whereas other FNP cases were not associated with CNS
infections.

ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1300–1306 1303
Differential diagnosis of acute central nervous system infections Huttunen et al.

A 35
Table 3 Microbial agents found, and the most common clinical manifestations
of the patients with no proven CNS infection
30
n (% of patients
25
Age at with no proven
Microbe ⁄ diagnosis diagnosis (y) CNS infection)
WBC count x 10E9

20
A. Pathogens in differential diagnosis
15 Bacteria
S. pneumoniae 0.9–8.5 9 (6.1)
10
M. pneumoniae 0.6–8.0 8 (5.4)
5 Escherichia coli 0.1–5.2 4 (2.7)
Streptococcus haemolyticus A 3.2–12.8 4 (2.7)
0
S. agalactiae (GBS) 0.1–0.2 2 (1.4)
Bm Vm/enc N-borr/oth Salmonella enteritidis 1.9–8.1 2 (1.4)
Yersinia pseudotuberculosis 5.0 1 (0.7)
B 100000 Viruses
Influenzavirus A ⁄ B 0.1–9.3 4 (2.7)
WBC count in the CSF x 10E6/L

10000 Respiratory syncytial virus 0.1–3.3 4 (2.7)

Rotavirus 0.2–4.6 4 (2.7)
1000 Adenovirus 1.7–7.9 4 (2.7)
Enterovirus 0.1–8.3 3 (2.0)
100 Herpes simplex virus 1 2.3–5.5 2 (1.4)
Norovirus 0.7–7.3 2 (1.4)
10 Parainfluenzavirus 1 ⁄ 2 ⁄ 3 3.8–15.7 2 (1.4)
Varicella zoster virus 1.1–6.3 2 (1.4)
1
Parechovirus 2 0.1 1 (0.7)
Epstein Barr virus 5.2 1 (0.7)
Bm Vm/enc N-borr/oth
Positive findings 59 (40)
C 400 No positive findings 88 (60)
350
B. Most common differential diagnoses
Infections
300
Septicaemia ⁄ suspected septicaemia 0.1–8.5 20 (14)
CRP count (mg/L)

250
Febrile convulsion 1.1–9.3 15 (10)
200 Viral infection NAS 0.1–15.5 14 (9.5)
Gastroenteritis + convulsion 0.5–11.2 13 (8.8)
150
Upper respiratory infection* 0.1–15.5 13 (8.8)
100
Pyelonefritis ⁄ urosepticaemia 0.1–5.2 8 (5.4)
50 Lower respiratory infection 0.1–2.4 5 (3.4)
Other diagnoses
0

Bm Vm/enc N-borr/oth Epilepsy 0.7–14.3 15 (10)

Cephalgia ⁄ migraine 7.3–15.8 14 (9.5)
Figure 1 (A) Blood leucocyte counts on admission to the hospital of the children Facial nerve paresis 2.4–15.9 12 (8.2)
with confirmed or probable CNS infection. Bm ¼ bacterial meningitis, Vm ⁄ enc ¼ Opsoclonus ataxia 0.6–1.2 2 (1.4)
viral meningitis ⁄ encephalitis, N-borr ⁄ oth ¼ neuroborreliosis and others Ataxia ⁄ cerebellitis 2.8–5.2 2 (1.4)
(Table 2); (B) CSF leucocyte count of the same patients; (C) serum C-reactive Intoxication 0.5–0.7 2 (1.4)
protein value of the same patients. ADEM** 8.2 1 (0.7)
Other rare diagnoses*** 11 (7.5)
DISCUSSION *
Includes otitis media; **ADEM ¼ acute disseminated encephalomyelitis;
***
In this prospective study, we searched for the specific aeti- Including periodic fever, Kawasaki disease.
ology of acute CNS infections of childhood using a broad
microbiological test pattern. Opposite to what has been the
approach in most previous reports, we examined all acute and only seven cases remained aetiologically entirely un-
CNS infections, not only one clinical manifestation. Despite known. These results compete well with other studies (1,3),
this, only 66 cases were found during 3 years. We think that one of those being a recent analysis of adult patients in
including more centres, or continuing for a few more years, Finland (10), in which the aetiology of aseptic meningitis
would hardly have added much to the information what we was identified in 66%. We did so in 70% (14 ⁄ 20) of our
already obtained. Evidently, CNS infections were just rare paediatric patients.
in Finland in the early 2000s. Enteroviruses of the Picornaviridae family are globally the
Among the 213 enrolled patients, a pathogenic agent was most common cause of aseptic meningitis in all age groups,
identified in 56 cases (85% of the 66 individuals with CNS and in epidemic and nonepidemic conditions (29–31). Here
infection). In addition, three probable cases were identified, also enteroviruses were common accounting for 22% of all

1304 ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1300–1306
Huttunen et al.
cases, and for 44% of proven meningitis ⁄ encephalitis. As
in other studies (31) coxsackie B viruses and echoviruses
caused the majority of cases based on the typing of isolated
viruses. Parechovirus 2 was detected in a 1-month-old baby
with febrile illness without CNS infection, but never in chil-
dren with CNS infection. Since parechoviruses (32) were
actively searched for with RT-PCR from the CSF (17) (n ¼
24) and virus culture from the stool samples, those seem a
rare cause of meningitis ⁄ meningoencephalitis in Finland.
Acute CNS infection was found in only 0.2% of the on-call
patients admitted to the paediatric outpatient ward. Obvi-
ously, Haemophilus influenzae type b (33) and MMR vac-
cines (34) (both in the paediatric calendar), and to some
extent (used only in private practice), S.pneumoniae con-
jugates and varicella vaccines have done well in Finland.
We cannot explain why no cases of herpesvirus encephalitis
was found, but suspect that improved perinatal care played a
role. Lyme neuroborreliosis was rare perhaps partly because
of active public education on its tick-borne contagiousness.
Various PCR methods are essential in the diagnostics of
acute CNS infections (4–6). Here, PCR was positive in 27 ⁄ 56
(48%) of the samples in which an agent was found. In 19
cases (70% of the PCR-positive samples), only PCR dis-
closed the agent. Even though the exact accuracy is not
known for most of the PCR tests used in this study, we
consider aetiological diagnoses reliable based on the clin-
ical pictures of each individual case. However, in case of
lymphotropic herpesviruses (especially HHV6), a positive
PCR test always has be considered suggestive, not defini-
tive, evidence of virus encephalitis (35). Still other tests
than PCR are needed, too. Antibody determination was es-
pecially valuable in neuroborreliosis (7) (only 1 ⁄ 9 case was
PCR-positive), but paired sera brought little further informa-
tion in any case. In addition to neuroborreliosis, intrathecal
antibody production was also detected in all the 5 VZV
meningitis cases (which were also PCR-positive), in one en-
terovirus meningitis (also PCR-positive) and in one proba-
ble Toxoplasma encephalitis case. In enterovirus infections,
CSF and ⁄ or stool culture helped in typing the virus in half
(6 ⁄ 13, 46%) of the cases. Seven cases remained untyped, but
if knowledge of the serotype is essential, sequence analysis of
the gene coding for VP1 capsid proteins would be valuable
(36). We did not find nasal swabbing useful in the diagnosis
of CNS infections.
Here, we come to the cost issue: not even in affluent
Finland it is possible to finance virology of this kind on
a routine basis. The overall cost of all the microbiological
tests we run here was approximately 2000€ per patient,
corresponding to 2600 USD (400€, 500 USD, for bacteriol-
ogy, and 1600€, 2100 USD, for virology), and in many other
countries, it would have been even more expensive. This is
well above the resources in most centres of the world. One
has to balance between the clinical and academic needs. We
reasoned that this analysis was needed to chart the current
epidemiological situation in Finland, but do not claim this
wide microbiology is essential in all cases. In the future, the
diagnostics of the CNS infections will probably be based on
multiarray tests where several pathogens can be detected at
ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1300–1306
Differential diagnosis of acute central nervous system infections
the same time from a small sample volume. This hopefully
will also reduce the overall costs.
Once an acute CNS infection is possibly bacterial
meningitis, the clinician has to rush; we lost three
lives because of this disease. When dealing with CNS
infections–manifested as 15 entities in this patient material
(Table 2), young residents often deem themselves insuffi-
cient with their skills. Therefore, simple and fast laboratory
methods are of paramount importance. Not underrating the
value of quick spinal tap (essential for culture), the past 25
years (28) have taught us to rely much on serum CRP in
distinguishing between CNS infections due to bacteria ver-
sus viruses (Fig. 1C). This simple test, requiring only a small
whole-blood specimen from a finger tip does not disclose
the agent, but is a very good yardstick (37,38) provided 8–
12 h (39) have elapsed from the onset of illness. In the great
majority of cases, this is the case.
CONCLUSIONS
In up to 85% of paediatric CNS infections, specific micro-
biological diagnosis can be obtained with modern labora-
tory techniques. However, no current laboratory or other
tests are 100% sensitive and specific for all CNS infections.
Therefore, the clinical skills outweigh everything else in im-
portance in their diagnosis and treatment. The skills improve
in time and with the number of patients the doctor meets,
and hence, we strongly advise young colleagues to actively
search for an opportunity to see these patients; there are
plenty of opportunities to do this in developing countries.
Much remains to be done to decrease the grim grip of these
potentially devastating diseases.
ACKNOWLEDGEMENTS
The study was financially supported by an EVO grant from
the Helsinki University Hospital and grants from the Finnish
Medical Foundation, the Academy of Finland and the Sigrid
Juselius Foundation.
References
1. Koskiniemi M, Rantalaiho T, Piiparinen H, von Bonsdorff CH,
Färkkilä M, Järvinen A, et al. Infections of the central nervous
system of suspected viral origin: a collaborative study from
Finland. J Neurovirol 2001; 7: 400–8.
2. Glaser CA, Gilliam S, Schnurr D, Forghani B, Honarmand S,
Khetsuriani N, et al. In search of encephalitis etiologies:
diagnostic challenges in the California Encephalitis Project,
1998–2000. Clin Infect Dis 2003; 36: 731–42.
3. Narkeviciute I, Bernatoniene J, Mikelionyte A, Baliukynaite V,
Eidukevocius R. Aetiological diagnostics of acute bacterial
meningitis in children. Scand J Infect Dis 2006; 38: 782–7.
4. Rådström P, Bäckman A, Qian N, Kragsbjerg P, Påhlson C,
Olcén P. Detection of bacterial DNA in cerebrospinal fluid by
an assay for simultaneous detection of Neisseria meningitidis,
Haemophilus influenzae, and streptococci using a seminested
PCR strategy. J Clin Microbiol 1994; 32: 2738–44.
5. deBiasi RL, Tyler KL. Polymerase chain reaction in the
diagnosis and management of central nervous system
infections. Arch Neurol 1999; 56: 1215–19.
1305
Differential diagnosis of acute central nervous system infections
6. Vuorinen T, Vainionpää R, Hyypiä T. Five years experience of
RT-PCR in daily diagnosis of entero- and rhinovirus
infections. Clin Inf Dis 2003; 37: 452–5.
7. Panelius J, Lahdenne P, Saxén H, Carlsson SA, Heikkilä T,
Peltomaa M, et al. Diagnosis of Lyme neuroborreliosis with
antibodies to recombinant proteins DbpA, BBK32, and OspC,
and VlsE IR6 peptide. J Neurol 2003; 250: 1318–27.
8. Whitley RJ, Gnann JW. Viral encephalitis: familiar infections
and emerging pathogens. Lancet 2002; 359: 507–13.
9. Waisman Y, Lotem Y, Hemmo M, Wietzen T, Garty BZ,
Mimouni M. Management of children with aseptic meningitis
in the emergency department. Pediatr Emerg Care 1999; 15:
314–17.
10. Kupila L, Vuorinen T, Vainionpaa R, Hukkanen V, Marttila R,
Kotilainen P. Etiology of aseptic meningitis and encephalitis in
an adult population. Neurology 2006; 66: 75–80.
11. Oschmann P, Dorndorf W, Hornig C, Schäfer C, Wellensiek
HJ, Pflughaupt KW. Stages and syndromes of neuroborreliosis.
J Neurol 1998; 245: 262–72.
12. Piiparinen H, Vaheri A. Genotyping of herpes simplex viruses
by polymerase chain reaction. Arch Virol 1991; 119:
275–83.
13. Vesanen M, Piiparinen H, Kallio A, Vaheri A. Detection of
herpes simplex virus DNA in cerebrospinal fluid samples using
the polymerase chain reaction and microplate hybridization.
J Virol Meth 1996; 59: 1–11.
14. Gopal MR, Thomso BJ, Fox J, Tedder RS, Honess RW.
Detection by PCR of HHV-6 and EBV DNA in blood and
oropharynx of healthy adults and HIV-seropositives. Lancet
1990; 335: 1598–9.
15. Pitkäranta A, Piiparinen H, Mannonen L, Vesaluoma M,
Vaheri A. Detection of human herpesvirus 6 and
varicella-zoster virus in tear fluid of patients with Bell’s palsy by
PCR. J Clin Microbiol 2000; 38: 2753–5.
16. Echevarria JM, Casas I, Tenorio A, de Ory F, Martinez-Martin
P. Detection of varicella-zoster virus-specific DNA sequences
in cerebrospinal fluid from patients with acute aseptic
meningitis and no cutaneous lesions. J Med Virol 1994; 43:
331–5.
17. Koskiniemi M, Mannonen L, Kallio A, Vaheri A. Luminometric
microplate hybridization for detection of varicella-zoster virus
PCR product from cerebrospinal fluid. J Virol Meth 1997; 63:
71–9.
18. Hyypiä T, Auvinen P, Maaronen, M. Polymerase chain
reaction for human picornaviruses. J Gen Virol 1989; 70:
3261–8.
19. Jokela P, Joki-Korpela P, Maaronen M, Glumoff V, Hyypiä T.
Detection of human picornaviruses by multiplex reverse
transcription-PCR and liquid hybridization. J Clin Microbiol
2005; 43: 1239–45.
20. van Kuppeveld FJ, Johansson KE, Galama JM, Kissing J, Bölske
G, Hjelm E, et al. 16S rRNA based polymerase chain reaction
compared with culture and serological methods for diagnosis of
Mycoplasma pneumoniae infection. Eur J Clin Microbiol Infect
Dis 1994; 13: 401–5.
21. Nocton JJ, Dressler F, Rutledge BJ, Rys PN, Persing DH, Steere
AC. Detection of Borrelia burgdorferi DNA by polymerase
chain reaction in synovial fluid from patients with Lyme arthri-
tis. New Engl J Med 1994; 330: 229–34.
22. Mustonen K, Mustakangas, P, Smeds M, Mannonen L, Uotila
L, Vaheri A, et al. 1998. Antibodies to varicella zoster virus in
1306 ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1300–1306
View publication stats
Huttunen et al.
the cerebrospinal fluid of neonates with seizures. Arch Dis
Child Fetal Neonatal Ed. 1998; 78: F57–61.
23. Jäämaa S, Salonen M, Seppälä I, Piiparinen H, Sarna S,
Koskiniemi M. Varicella zoster and Borrelia burgdorferi are the
main agents associated with facial paresis, especially in chil-
dren. J Clin Virol 2003; 27: 146–51.
24. Korhonen MH, Brunstein J, Haario H, Katnikov A, Rescaldani
R, Hedman K. A new method with general diagnostic utility
for the calculation of immunoglobulin G avidity. Clin Diagn
Lab Immunol 1999; 6: 725–8.
25. Kaikkonen L, Lankinen H, Harjunpää A, Hokynar K,
Söderlund M, Oker-Blom C, et al. Acute phase specific
heptapeptide epitope for parvovirus B19 diagnosis. J Clin
Microbiol 1999; 37: 3952–6.
26. Koskiniemi M, Piiparinen H, Rantalaiho T, Eränkö P, Färkkilä
M, Räihä K, et al. Acute central nervous system complications
in varicella zoster virus infections. J Clin Virol 2002; 25:
293–301.
27. Steiner I, Budka H, Chaudhuri A, Koskiniemi M, Sainio K,
Salonen O, et al. Viral encephalitis: a review of diagnostic
methods and guidelines for management. Eur J Neurol 2005;
12: 331–43.
28. Peltola H. C-reactive protein for rapid monitoring of infections
of the central nervous system. Lancet 1982; 1: 980–3.
29. Andreoletti L, Blassel-Damman N, Dewilde A, Vallee L,
Cremer R, Hober D, et al. Comparison of use of cerebrospinal
fluid, serum, and throat swab specimens in diagnosis of
enteroviral acute neurological infection by a rapid RNA
detection PCR assay. J Clin Microbiol 1998; 36: 589–91.
30. Chambon M, Archimbaud C, Bailly JL, Henquell C, Regagnon
C, Charbonné F, et al. Circulation of enteroviruses and
persistence of meningitis cases in the winter 1999–2000. J Med
Virol 2001; 65: 340–7.
31. Lee BE, Davies HD. Aseptic meningitis. Curr Opin Infect Dis
2007; 20: 272–7.
32. Joki-Korpela P, Hyypiä T. Parechoviruses, a novel group of
human picornaviruses. Ann Med 2001; 33: 466–71.
33. Peltola H, Salo E, Saxén H. Incidence of Haemophilus
influenzae type b meningitis during 18 years of vaccine use:
observational study using routine hospital data. BMJ 2005; 330:
18–19.
34. Peltola H, Heinonen OP, Valle M, Paunio M, Virtanen M, Kar-
anko V, et al. The elimination of indigenous measles, mumps
and rubella from Finland by a 12-year, two-dose vaccination
program. New Engl J Med 1994; 331: 1397–402.
35. Boivin G. Diagnosis of herpesvirus infections of the central
nervous system. Herpes 2004; 11 Suppl 2: 48A–56.
36. Oberste MS, Maher K, Kilpatrick DR, Flemister MR, Brown
BA, Pallansch MA. Typing of human enteroviruses by partial
sequencing of VP1. J Clin Microbiol 1999; 37: 1288–93.
37. Roine I, Banfi A, Bosch P, Ledermann W, Contreras C, Peltola
H. Serum C-reactive protein in childhood meningitis in
countries with limited laboratory resources: a Chilean
experience. Pediatr Infect Dis J 1991; 10: 923–8.
38. Sormunen P, Kallio MJT, Kilpi T, Peltola H. C-reactive protein
is useful in distinguishing Gram stain-negative bacterial
meningitis from viral meningitis in children. J Pediatr 1999;
134: 725–9.
39. Peltola H. C-reactive protein in rapid differentiation of acute
epiglottitis from spasmodic croup and acute laryngotracheitis: a
preliminary report. J Pediatr 1983; 102: 713–15.