Acute_childhood_encephalitis_and_Mycoplasma_pneumoniae.pdf

7/25/2019 Acute_childhood_encephalitis_and_Mycoplasma_pneumoniae.pdf

1/11

1674 CID 2001:32 (15 June) Bitnun et al.

M A J O R A R T I C L E

Acute Childhood EncephalitisandMycoplasma pneumoniae

Ari Bitnun,1 Elizabeth Lee Ford-Jones,1 Martin Petric,3 Daune MacGregor,2 Helen Heurter,1 Susan Nelson,4

Grant Johnson,3 and Susan Richardson3

1Division of Infectious Diseases and 2Division of Neurology, Department of Pediatrics, and 3Division of Microbiology, Department of Pediatric

Laboratory Medicine, The Hospital for Sick Children, and 4Aventis-Pasteur, Ltd., Toronto

In a prospective 5-year study of children with acute encephalitis, evidence of Mycoplasma pneumoniaeinfection

was demonstrated in 50 (31%) of 159 children. In 11 (6.9%) of these patients,M. pneumoniaewas determined

to be the probable cause of encephalitis on the basis of its detection in cerebrospinal fluid (CSF) by polymerasechain reaction (PCR) or by positive results of serologic tests for M. pneumoniaeand detection of the organism

in the throat by PCR. CSF PCR positivity correlated with a shorter prodromal illness ( ) and lack ofPp .015

respiratory symptoms ( ). Long-term neurologic sequelae occurred in 64% of probable cases. ThirtyPp .06

children (18.9%) who were seropositive for M. pneumoniaebut did not have the organism detected by culture

or PCR had convincing evidence implicating other organisms as the cause of encephalitis, suggesting that

current serologic assays for M. pneumoniaeare not sufficiently specific to establish a diagnosis ofM. pneu-

moniaeencephalitis.

Acute childhood encephalitis is a potentially devastating

illness with an incidence of 10.5 cases per 100,000

child-years in developed countries [1]. Among children

!1 year old, the incidence may be as high as 27.7 casesper 100,000 child-years [2]. In the prevaccine era, mea-

sles, mumps, polio, and rubella were commonly im-

plicated as etiologic agents of acute encephalitis [3, 4].

However, in more recent prospective studies, nonpolio

enteroviruses, respiratory viruses, herpesviruses, and

Mycoplasma pneumoniae have accounted for the ma-

jority of etiologically confirmed cases [1, 5, 6].

Most large series published to date have relied ex-

Received 5 May 2000; revised 5 October 2000; electronically published 21 May

2001.

Presented in part: 39th Interscience Conference on Antimicrobial Agents and

Chemotherapy, San Francisco, 2529 September 1999 (abstract 2095).

Grant support: The Meningitis D.R.E.A.M. (Douglas Russell Evely Annual

Memorial) Fund, The Hospital for Sick Children, Toronto.

Reprints or correspondence: Dr. Susan Richardson, Division of Microbiology,

Dept. of Pediatric Laboratory Medicine, The Hospital for Sick Children, 555

University Ave., Toronto, Ontario, M5G 1X8, Canada (susan.richardson@sickkids

.on.ca).

Clinical Infectious Diseases 2001;32:167484

2001 by the Infectious Diseases Society of America. All rights reserved.

1058-4838/2001/3212-0002$03.00

clusively on serologic tests for the diagnosis ofM. pneu-

moniae encephalitis [1, 3, 5, 713]. Unfortunately, the

glycolipid antigen extract ofM. pneumoniaethat is used

in such assays is not specific, and false-positive resultsare a significant concern [1419]. Glycolipids that

cross-react serologically occur commonly in plants [16]

and in human brain tissue [17]. Fourfold or greater

rises in CF titer to M. pneumoniaehave been dem-

onstrated in 40% of adult patients with bacterial men-

ingitis and in 10% of those with other bacteremic

infections [18]. False-positiveM. pneumoniaeIgM anti-

body responses have been detected in patients with

acute meningoencephalitis and acute pancreatitis [19].

Among patients who have acute encephalitis, the de-

tection ofM. pneumoniaein the CSF by means of cul-

ture or PCR provides strong evidence of causality. How-ever, absence of the organism in the CSF does not rule

out M. pneumoniae as a cause, because CNS disease

may be immunologically mediated. In this respect, the

detection ofM. pneumoniaein the respiratory tract of

a patient with acute encephalitis provides supportive

evidence of causality.

Since January 1994, The Hospital for Sick Children,

Toronto, has maintained a prospective encephalitis reg-


2/11

Encephalitis and M. pneumoniae CID 2001:32 (15 June) 1675

istry, and any child who is admitted to the hospital with acute

encephalitis undergoes an extensive neurologic and microbi-

ological investigation [6]. In this report, we summarize data

compiled from January 1994 through December 1998 that per-

tain to the role of M. pneumoniae in acute childhood en-

cephalitis.

PATIENTS AND METHODS

Enrollment criteria, data collection, and investigations.

Enrollment criteria, data collection, and investigations have

been described elsewhere [6]. In brief, patients were identified

by the registry nurse in a daily review of admissions. Patients

were excluded if they were 4 weeks old or 18 years old, if

they were known to have a neurologic disorder, or if they were

known to be immunodeficient. Children with postvaricella cer-

ebellar ataxia were also excluded. In addition to the microbi-

ological investigations detailed below, electroencephalographic

(EEG) analyses, cranial CT scans, MRI scans, and single-photonemission CT were performed, as indicated.

Microbiology. Most of the microbiological methods have

been described elsewhere [6]. Acute-phase and convalescent-

phase serum samples were tested for M. pneumoniaeby use of

the CF test and for M. pneumoniaespecific IgM by use of EIA

(GenBio). Serologic evidence ofM. pneumoniae infection was

determined by seroconversion, which was defined by a 4-fold

increase in titer or a single titer of 1:128, as determined by

CF, or by the presence ofM. pneumoniaespecific IgM. Anti-

body titers for herpes simplex virus (HSV), enteroviruses, ad-

enovirus, and influenza A and B viruses were determined by

CF, and those for arboviruses (Powassan, St. Louis encephalitis,and eastern and western equine encephalitis) were determined

by use of a hemagglutination inhibition assay. Serologic assays

for Epstein-Barr virus (EBV) included the test for heterophile

antibody (Oxoid) and EIAs for early antigen, viral capsid an-

tigen, and Epstein-Barr nuclear antigen (Diasorin). Serologic

testing for human herpesvirus 6 (HHV-6; ABI), varicella-zoster

virus (VZV), and measles (Gull) was done by use of EIA,

whereas serologic testing for parvovirus B19 (Biotrin) was per-

formed by use of indirect immunofluorescence microscopy un-

til 1997 and, thereafter, by use of EIA. Serologic testing for

Bartonella henselae was performed by use of indirect immu-

nofluorescence microscopy (Central Public Health Laboratory,

Ontario Ministry of Health, Toronto).

Detection of M. pneumoniae in throat and CSF specimens

was performed by means of culture before 1996 [6] and, there-

after, by PCR. The PCR assay was performed using the P11

and P13 primers for the P1 adhesin gene, which resulted in

a 209-bp amplicon [2022]. PCR reactions were performed in

a total volume of 50 mL with 2.0 U of AmpliTaq Gold (Perkin

Elmer), 200 mMof each dNTP, 2.0 mMof MgCl2, 25 pmol of

each primer, and 10 mL of sample DNA overlaid with mineral

oil. PCR involved a 2-step reaction that was performed on the

Robocycler 96 (Stratagene). The cycles were as follows: 1 cycle

of 10 min at 95C (for activation of AmpliTaq Gold), 40 cycles

of 1 min at 95C (for denaturation) and 1 min at 72C (for

annealing and extension), and 1 cycle of 3 min at 72 C (for

final extension). The 209-bp product was then detected by gel

electrophoresis by means of ethidium bromide staining. Anextraction control (PBS) was run with each test to detect con-

tamination. In addition, a reagent control and sensitivity con-

trol of 1 color-changing unit (CCU) were run. Each specimen

was also tested to detect inhibition of the PCR reaction (1 CCU

ofM. pneumoniaewas added to a duplicate PCR reaction). The

sensitivity of our PCR assay was determined to be 0.1 CCU by

use of a stock ofM. pneumoniaewith known concentration (as

determined by culture of serial dilutions to determine 1 CCU).

Virus cultures were performed with AGMK and RD cell lines,

for CSF and stool specimens, and with AGMK, MDCK, and

HEp-2 cell lines, for throat and nasopharyngeal specimens. CSF

was analyzed by reverse-transcriptase PCR and PCR for entero-

viruses and all known herpesviruses, respectively [23]. Naso-

pharyngeal specimens were examined for the presence of res-

piratory syncytial virus, parainfluenza viruses (1, 2, and 3),

influenza A and B viruses, and adenovirus by means of direct

immunofluorescence microscopy and virus isolation in cell cul-

ture. Stool specimens were examined by use of electron micro-

scopy for the presence of viral particles, and the specimens

were inoculated into cell culture as described elsewhere [6].

Definitions. Encephalopathy was defined as a depressed or

altered level of consciousness, including lethargy, extreme irri-

tability, or a significant change in personality or behavior thatpersisted for 24 h. Encephalitis was defined by the presence of

encephalopathy plus 2 of the following criteria: fever (tem-

perature, 38.0C), seizure(s), focal neurologic findings, pleo-

cytosis (WBC count, 15 cells/mL), EEG findings compatible with

encephalitis, or abnormal neuroimaging.

All children who fulfilled the study criteria for encephalitis

and in whom evidence ofM. pneumoniae infection was detected

either by means of serologic testing or by culture or PCR of

throat or CSF specimens were included. Depending on the

strength of the evidence that implicatedM. pneumoniae,cases

were classified as probable, possible, or indeterminate

with respect toM. pneumoniaebeing the etiologic agent. Prob-able M. pneumoniaeencephalitis was defined by either of the

following 2 criteria: detection ofM. pneumoniaein CSF spec-

imens by means of PCR, culture, or both, with or without

confirmatory results of serologic tests for M. pneumoniae; or

detection ofM. pneumoniaein throat specimens by means of

PCR, culture, or both, with confirmatory results of serologic

tests for M. pneumoniae. Possible M. pneumoniaeencephalitis

was defined by either serologic evidence of M. pneumoniae


3/11


Table 1. Microbiological data for 11 children with probable Mycoplasma pneumoniaeencephalitis.

Patient

Age,

years

Results of serologic testing

for M. pneumoniae

Other potential etiologic agentsa

Acute-phase

specimens

Convalescent-phase

specimens

Results of PCR

for detection of

M. pneumoniae

IgM EIA CF IgM EIA CF CSF Throat

1 7 1:256 1:1024 None

2 10.5 1:128 1:512 None

3 12 1:256 1:512 Parainfluenza virus 3 NP by IF; HHV-6 IgM

and IgG EIA; Bartonella henselaeserology

(1:128/1:128)

4 6 1:128 1:128 Enterovirus serology (1: 4/1:16)

5 13 NSQ NSQ NSQ None

6 8 1:64 NSQ 1:32 Parainfluenza virus 3 NP by IF; enterovirus

serology (1:16/!1:4)

7 13 NSQ 1:64 HSV-2 CSF PCR; HSV serology (1:16/1:16);

EBV Monospot; EBV EA, IgM VCA, IgG

VCA; EBNA

8 1 !1: 8 !1: 8 MMR vaccine 3 weeks before admission

9 4 NSQ NSQ NSQ Influenza A virus NP by IF; influenza A virus

serology (1:256/1:64)

10 1.5 !1: 8 !1: 8 HHV-6 IgM and IgG EIA

11 1 !1: 8 !1: 8 HSV-2 CSF PCR; HSV serology (!1:8/!1:8)

NOTE. The criteria for classification of cases of M. pneumoniaeencephalitis as probable are described in the Patients and Methods section. EA, early

antigen; EBV, Epstein-Barr virus; EBNA, Epstein-Barr nuclear antigen; HHV-6, human herpesvirus 6; HSV, herpes simplex virus; HSV-2, herpes simplex virus type

2; IF, immunofluorescence testing; NP, nasopharyngeal swab; MMR, measles-mumps-rubella; NSQ, sample insufficient to perform assay; VCA, viral capsid antigen;

, positive finding; , negative finding.a

Values in parentheses are acute-phase serologic titer/convalescent-phase serologic titer.

infection with negative results of culture and PCR of throat

and CSF specimens and the absence of convincing evidence for

other potential etiologic agents, or as positive results of cultureor PCR for M. pneumoniaein throat specimens but negative

results of serologic tests for M. pneumoniae.IndeterminateM.

pneumoniaeencephalitis was defined as serologic evidence of

M. pneumoniaeinfection with negative results of culture and

PCR of throat and CSF specimens and convincing evidence

that implicated at least 1 other pathogen.

Convincing evidence for the presence of other pathogens was

determined by the presence of one of the following: (1) de-

tection of the organism in the CSF by means of culture or PCR;

(2) detection of the organism in samples from a site other than

the CSF by means of culture, PCR, antigen detection tech-

niques, or electron microscopy; (3) serologic evidence of acute

infection that was equally or more convincing than the evidence

forM. pneumoniae; (4) in cases of suspected varicella enceph-

alitis, a clinical diagnosis of varicella made at the time of or

just before the development of encephalitis; or (5) a clinical

presentation consistent with tuberculous meningitis with an

appropriate exposure history and favorable response to anti-

tuberculous therapy.

RESULTS

Evidence forM. pneumoniaeinfection was detected in 50 (31%)of 159 children who were hospitalized with acute encephalitis

from January 1994 through December 1998. Eleven of these

patients (22%) had sufficient evidence ofM. pneumoniaeen-

cephalitis for their cases to be classified as probable; among the

remaining 39 patients, 9 cases (18%) were categorized as pos-

sible, and 30 cases (60%) were classified as indeterminate. Pa-

tients with probable M. pneumoniaeencephalitis included 6

boys and 5 girls, who had a mean age of 7 years (range, 113

years). Patients with possible M. pneumoniae encephalitis in-

cluded 3 boys and 6 girls, who had a mean age of 9.5 years

(range, 117 years).

Microbiological findings. M. pneumoniaewas detected by

PCR in CSF samples from 6 children and in throat specimens

obtained from an additional 5 children who were classified as

having probableM. pneumoniaeencephalitis (table 1). No pa-

tient had M. pneumoniaedetected at both sites. Three of the

6 children in whom M. pneumoniaewas detected in CSF by

means of PCR lacked detectable antibody to this organism in

both acute-phase and convalescent-phase serum samples. All 3


4/11


Table 2. Microbiological data for 9 children with possible Mycoplasma pneumoniaeencephalitis.

Patient

Age,

years

Results of

PCR of throat

specimens for

M. pneumoniae

Results of serologic testing for M. pneumoniae

Other potential etiologic agentsa

Acute-phase

specimens

Convalescent-phase

specimens

IgM EIA CF IgM EIA CF

12 7 1: 8 1:32 None

13 17

1:128 NSQ

1:128 None14 1.2 1:128 1:128 Parainfluenza virus 2 NP by culture;

parainfluenza virus 1 NP by IF

15 17 1:64 NSQ 1:128 None

16 12 1:2048 NSQ 1:4096 Bartonella henselaeserology

(1:128/1:128)

17 7 1:128 NSQ 1:32 None

18 4 1:128 NSQ 1:128 None

19 5 1:128 NSQ 1:128 None

20 15 1: 8 1:16 Influenza A virus NP by IF; influenza

A virus serology (!1:8/1:512)

NOTE. The criteria for classification of cases of M. pneumoniaeencephalitis as possible are described in the Patients and Methods section. IF, immunoflu-

orescence testing; NP, nasopharyngeal swab; NSQ, sample insufficient to perform assay; , positive finding; , negative finding.

a Values in parentheses are acute-phase serologic titer/convalescent-phase serologic titer.

Table 3. Etiologic agents implicated in 30 children withindeterminate Mycoplasma pneumoniaeencephalitis.

Etiologic agent

No. of patients

(n p 30)a

Enteroviruses 9

Herpes simplex virus 1

Varicella-zoster virus 2

Epstein-Barr virus 3

Human herpesvirus 6 7Influenza A virus 7

Adenovirus 4

Bartonella henselae 5

Mycobacterium tuberculosis 2

NOTE. The criteria for classification of cases of M. pneumoniae

encephalitis as indeterminate are described in the Patients and Methods

section.a

More than 1 agent was implicated in 10 patients.

of these children were !2 years old, and only 1 of them had a

documented respiratory prodrome.

Seven of 11 patients who had probable M. pneumoniaeen-

cephalitis had evidence of coinfection with at least 1 other

organism (table 1). In 2 patients, coinfection of the CNS by

M. pneumoniaeand HSV-2 was demonstrated by the detection

of both organisms in the CSF by PCR. Amplicons produced

by HSV-specific primers formed products characteristic for

HSV-2 when digested with restriction endonucleases [23].

When digested with the restriction endonuclease AluI, the 209-bp amplicon produced by the M. pneumoniaespecific primers

yielded 2 base-pair fragments of 134 and 75 bp, which is char-

acteristic ofM. pneumoniae.Cross-reactivity between the 2 or-

ganisms was excluded by demonstrating that the M. pneumo-

niaeDNA was not amplified by the HSV-2 primers and that

the HSV-2 DNA was not amplified by the M. pneumoniae

primers.

The microbiological evidence for the possible cases of M.

pneumoniae encephalitis is summarized in table 2. Viral res-

piratory coinfections were documented in 2 children. For 1

child, influenza A virus antigen was detected by immunoflu-

orescence in a nasopharyngeal specimen, and a 14-fold rise in

titer to the influenza A virus was demonstrated between acute-

phase and convalescent-phase serum samples. The other child

had parainfluenza virus 2 isolated in culture of a specimen

from the nasopharynx. Results of serologic tests forB. henselae

in patient 16 were considered equivocal because of the lack of

change in titer between acute-phase and convalescent-phase

serum samples.

In all 30 of the children who were classified as having in-

determinateM. pneumoniaeencephalitis, the evidence that im-

plicated other agents was equally or more compelling than that

for M. pneumoniae. The possible etiologic agents are summa-

rized in table 3. Nineteen children had particularly strong ev-

idence that implicated agents other than M. pneumoniaeas the

cause of encephalitis. Enteroviruses were the likely etiologic

agent in 6 children. In 1 child, enterovirus was isolated in

culture of the CSF, and in another child it was detected by

means of reverse-transcriptase PCR of the CSF. The remaining4 children had a 4-fold increase or decrease in the CF titer to


5/11


enterovirus. In 1 patient, HSV was detected in the CSF by PCR,

and a 4-fold rise in the CF antibody titer was demonstrated

between acute-phase and convalescent-phase serum samples.

For 3 children, detection of adenovirus in the nasopharynx or

stool was accompanied by positive results of serologic testing

for adenovirus. Two patients exhibited a 4-fold rise in CF an-

tibody titer to influenza A virus. VZV was the likely cause of

encephalitis in 2 children who had active clinical varicella, 1of whom also had VZV detected in the CSF by PCR. EBV was

implicated in 1 patient in whom heterophile antibody as well

as IgM and IgG to the EBV viral capsid antigen were detected

in acute-phase serum samples. Two children had a 4-fold rise

in antibody titer toB. henselae.Tuberculous meningitis was the

likely diagnosis for 2 children. Although neither case was con-

firmed microbiologically, the patients clinical presentation,

CSF findings (lymphocytic pleocytosis and elevated levels of

CSF protein), and response to antituberculous therapy strongly

implicated Mycobacterium tuberculosisas the cause of the en-

cephalitic presentation. The remaining 11 patients had sero-

logic evidence of infection with other agents, including hu-

man herpesvirus 6 ( ) , EBV ( ) , influenza A virusnp 2 np 2

( ), enterovirus ( ), adenovirus ( ), andB. hen-np 1 np 1 np 2

selae( ).np 3

Clinical manifestations. Prodromal respiratory manifes-

tations occurred in 7 of 11 patients with probable M. pneu-

moniaeencephalitis and 6 of 9 patients with possible M. pneu-

moniae encephalitis (table 4). The duration of respiratory

manifestations before admission varied from several days to as

long as 4 weeks. Chest radiographs were performed for 6 pa-

tients with probableM. pneumoniaeencephalitis and 3 patients

with possible M. pneumoniaeencephalitis; focal infiltrates andperibronchial thickening were detected in 4 and 2 patients,

respectively. In patients without preceding respiratory symp-

toms, prodromal symptoms included fever, headache, vomiting,

and diarrhea. One patient with probable M. pneumoniaeen-

cephalitis had no documented prodromal illness, and another

had a mild febrile illness 2 weeks before the development of

encephalitis, subsequent to receiving a routine measles-mumps-

rubella vaccine.

The duration of prodromal illness was significantly shorter

among children with positive M. pneumoniaeCSF PCR than

in children who had negative M. pneumoniaeCSF PCR (Pp

by 2-tailed Fishers exact test). In 5 of 6 children who had.015M. pneumoniaedetected in the CSF but not in the throat, the

prodromal illness was 5 days in duration, whereas all 5 chil-

dren in whom M. pneumoniaewas detected in the throat but

not in the CSF had prodromal symptoms 7 days in duration.

Respiratory symptoms occurred in all children in whom M.

pneumoniaewas detected in the throat but in only 2 of 6 chil-

dren in whom M. pneumoniaewas detected in the CSF (Pp

by 2-tailed Fishers exact test)..06

Neurologic manifestations in patients with probable or pos-

sibleM. pneumoniaeencephalitis included seizures, focal motor

deficits, ataxia, and increased intracranial pressure (table 4). Of

the 13 patients with seizures, 8 had generalized tonic-clonic

seizures and 5 had focal motor seizures. Focal neurologic def-

icits, independent of seizures, included hemiparesis, dysarthria,

and aphasia. Encephalopathic manifestations included a de-

creased level of consciousness, disorientation, confusion, visualhallucinations, and extreme irritability. Acute demyelinating

encephalomyelitis (ADEM) was diagnosed in 2 patients with

probableM. pneumoniaeencephalitis, 1 of whom also had optic

neuritis. For both patients, PCR detected M. pneumoniae in

specimens from the throat but not the CSF.

Specific therapy directed againstM. pneumoniaewas admin-

istered to 4 children in whom M. pneumoniaeencephalitis was

suspected (2 probable cases and 2 possible cases); 3 received

macrolides, and 1 received doxycycline. Ten children were

treated with acyclovir for presumed HSV encephalitis. Corti-

costeroids were administered to 1 patient with ADEM.

CSF findings. CSF specimens were available for all of the

patients with probable or possible M. pneumoniae encephalitis

(table 4). For the 12 patients with pleocytosis, the mean WBC

count was cells/L (range, to cells/L)9 9 94710 6 10 98 10

in children with probable M. pneumoniae encephalitis and

cells/L (range, to cells/L) in children9 9 93510 7 10 11010

with possible M. pneumoniaeencephalitis. A predominance of

lymphocytes (87%100%) was documented in 9 of the 10 pa-

tients for whom a differential cell count was available. Among

patients with probable or possible M. pneumoniae encephalitis,

the mean CSF protein level was 42 mg/dL (range, 10119 mg/

dL) and 37 mg/dL (range, 1365 mg/dL), respectively. The CSFglucose level was 2.86.0 mM.

EEG and neuroimaging findings. EEG and neuroimaging

(cranial CT, MRI, or both) were performed for all 20 patients

with probable and possible M. pneumoniaeencephalitis; EEG

and neuroimaging abnormalities were demonstrated in 85%

and 35% of these patients, respectively (table 4). Neuroimaging

abnormalities included focal hemispheric lesions suggestive of

an ischemic injury in 2 patients, focal temporal enhancing and

edematous lesions in 1 patient in each category, and mild uni-

lateral ventricular dilatation in 1 patient. Findings suggestive

of ADEM included diffuse demyelination in 1 patient and

edema of the basal ganglia, cerebral peduncles, and deep whitematter in another. Results of single-photon emission CT scan-

ning were normal in both patients with ADEM. Auditory and

somatic evoked potentials were also normal in these 2 patients,

but visual evoked responses were abnormal in the patient with

concomitant optic neuritis.

Neurologic outcome. There were no deaths among patients

in any of the 3 categories ofM. pneumoniaeencephalitis. The

mean duration of follow-up for the 8 children with probable


6/11

Table 4. Most common clinical and laboratory manifestations of probable, possible, andindeterminate Mycoplasma pneumoniaeencephalitis ( ).np 50

Parameter

No. (%) of patients with specified

type of M. pneumoniaeencephalitis

PaProbable

(n p 11)

Possible

(n p 9)

Indeterminate

(n p 30)

Prodromal symptoms

None 4 (36) 3 (33) 3 (10) NS

Cough/coryza/sore throat 7 (64) 6 (67) 11 (37) NS

Fever 6 (55) 7 (78) 20 (67) NS

Headache 2 (18) 5 (56) 13 (43) NS

Vomiting 4 (36) 2 (22) 8 (27) NS

Neurologic manifestationsb

Seizures 7 (64) 5 (56) 13 (43) NS

Focal motor deficits 4 (36) 4 (44) 8 (27) NS

Ataxia 4 (36) 1 (11) 1 (3) .014

Increased intracranial pressure 1 (9) 0 0 NSHematologic findings

c

Leukocytosis (115 109 leukocytes/L) 5 (45) 3 (33) 8 (27) NS

Anemia (!110 g/L) 1 (9) 1 (11) 0 NS

Thrombocytopenia (!150 109 platelets/L) 1 (9) 0 1 (3) NS

Elevated ESR (25 mm/h)c

5 (63) 5 (63) 6 (40) NS

CSF findings

Abnormal 6 (55) 7 (78) 27 (90) .022

Pleocytosis (15 WBCs/mL)d

5 (45) 6 (67) 22 (76) NS

Elevated protein level (40 mg/dL)e

4 (36) 4 (44) 12 (43) NS

EEG findingsf

Abnormal 9 (82) 8 (89) 19 (90) NS

Generalized slowing 7 (64) 4 (44) 13 (62) NSEpileptic focus 3 (27) 2 (22) 2 (10) NS

PLED 2 (18) 0 0 NS

FIRDA/OIRDA 2 (18) 3 (33) 6 (29) NS

Neuroimaging findings

Abnormal 5 (45) 2 (22) 12 (40) NS

Focal edema or ischemia 3 (27) 1 (11) 4 (13) NS

Focal enhancing lesions 1 (9) 0 2 (7) NS

Demyelination 1 (9) 0 3 (10) NS

Ventricular dilatation 0 1 (11) 2 (7) NS

NOTE. The criteria for classification of cases ofM. pneumoniae encephalitis as probable, possible, or

indeterminate are described in the Patients and Methods section. EEG, electroencephalography; ESR, eryth-

rocyte sedimentation rate; FIRDA, frontal intermittent rhythmic delta activity; OIRDA, occipital intermittentrhythmic delta activity; PLED, periodic lateralizing epileptiform discharge.

aComparison of probable and indeterminate cases by use of x2 statistic (2-tailed Fishers exact test if cell

no. !5).b

By definition, encephalopathy was present in all patients; for details, see the Patients and Methods

section.c

Available for 8 patients with probable M. pneumoniae encephalitis, 8 patients with possible M. pneu-

moniae encephalitis, and 15 patients with indeterminate M. pneumoniaeencephalitis.d

Available for 29 patients with indeterminate M. pneumoniaeencephalitis.e

Available for 28 patients with indeterminate M. pneumoniaeencephalitis.f

Available for 21 patients with indeterminate M. pneumoniaeencephalitis.


7/11


M. pneumoniae encephalitis in whom neurologic deficits were

present at the time of discharge from the hospital was 1.8 years

(range, 04 years). At least 6 months of follow-up data were

available for 7 of 8 children with probable and 4 of 5 children

with possible M. pneumoniaeencephalitis who had neurologic

deficits at the time of discharge. A seizure disorder that required

prolonged anticonvulsant therapy developed in 4 patients with

probableM. pneumoniaeencephalitis and 1 patient with pos-sibleM. pneumoniaeencephalitis. Mild cognitive impairments,

including difficulty with concentration, memory, reading, and

word finding, occurred in 4 of the 5 children who had seizure

disorders. Other persistent neurologic deficits that occurred in

patients with probable M. pneumoniae encephalitis included

hemiparesis ( ), expressive dysphasia ( ), dysarthrianp 1 np 1

( ), and truncal ataxia ( ). The 2 children withnp 1 np 1

ADEM, 1 of whom was treated with corticosteroids, made full

recoveries over the course of several months. Three of the 4

children who were treated with macrolides or doxycycline ex-

perienced neurologic sequelae. Among children with possible

M. pneumoniaeencephalitis, 1 patient experienced severe de-velopmental delay and concomitant hemiplegia. Other abnor-

mal findings in this group of children included a speech dis-

order (word-finding difficulty) in 2 patients and asymptomatic

ventriculomegaly and mild EEG abnormalities in 1.

DISCUSSION

M. pneumoniaewas strongly implicated as the probable etio-

logic agent in 6.9% of 159 children with acute encephalitis.

This figure is consistent with data from previous studies sug-

gesting that 1%10% of cases of acute childhood encephalitisare caused by this organism [1, 711]. The criteria for diagnosis

of probableM. pneumoniaeencephalitis included the detection

of M. pneumoniae by culture or PCR of specimens from the

throat, the CSF, or both to eliminate the potential impact of

false-positive serologic test results. The detection ofM. pneu-

moniaeby PCR in these children provides conclusive evidence

that all of them were infected, in the respiratory tract or CNS,

at the time of or just before the development of encephalitis.

By using more stringent diagnostic criteria than those of pre-

vious studies, we have strengthened the evidence implicating

M. pneumoniaeas a cause of acute childhood encephalitis.

M. pneumoniaeis typically considered to be a potential cause

of disease in children 5 years old who have respiratory symp-

toms. However, in our cohort, respiratory manifestations were

absent in 4 (36%) of 11 patients with probable M. pneumoniae

encephalitis and 3 (33%) of 9 patients with possibleM. pneu-

moniae encephalitis. Among patients with probable M. pneu-

moniae encephalitis, respiratory symptoms occurred in all 5

children in whom M. pneumoniaewas detected by PCR in the

throat but not in the CSF, but in only 2 of 6 children in whom

M. pneumoniaewas detected in the CSF but not in the throat.

The absence of respiratory manifestations has also been doc-

umented in several previous case reports of patients with en-

cephalitis for whom PCR or culture detected M. pneumoniae

in the CSF [2427]. In a recently reported study of patients

with serologically confirmed systemic infections caused byM.

pneumoniae, M. pneumoniaewas detected in serum by PCR in

only 1 of 25 patients with pneumonia but in 10 of 17 patientswithout pneumonia ( ) [28]. With respect to age, 4 ofP! .001

11 children in our cohort who had probable M. pneumoniae

encephalitis and 2 of 9 children who had possible M. pneu-

moniae encephalitis were !5 years old (tables 1 and 2). Four

of these 6 children were !2 years old. Therefore, M. pneumoniae

infection should be considered in the differential diagnosis of

children with acute encephalitis, regardless of whether they have

signs or symptoms of respiratory illness and regardless of their

age.

The detection ofM. pneumoniaein the brain tissue and CSF

of patients with encephalitis indicates that direct invasion of

the brain parenchyma is responsible for some cases of acute

encephalitis [2938]. We are aware of 2 cases in which M.

pneumoniaewas detected in brain tissue; in one, the organism

was isolated in culture [29], and in the other, it was detected

by nucleic acid hybridization [30].M. pneumoniaehas also been

isolated from the CSF by culture in 7 patients [24, 3235] and

detected by PCR in an additional 26 patients, including those

in our cohort [2527, 31, 3638].

M. pneumoniaehas been implicated as a cause of immune-

mediated neurologic syndromes, including ADEM, transverse-

myelitis, and Guillain-Barre syndrome [12, 36, 39 46]. In our

cohort, ADEM was diagnosed in 2 children in whom M. pneu-moniaewas detected in the throat but not in the CSF by PCR.

The underlying pathogenesis of immune-mediated injury in

patients with M. pneumoniae infection is incompletely under-

stood, but it likely stems from the presence of antigenic sim-

ilarities betweenM. pneumoniaeand human tissues. M. pneu-

moniaeinfection has been associated with the production of a

variety of anti-human antibodies, including IgM-class cold ag-

glutinins, cold-reactive antilymphocyte antibodies, and anti-

bodies to cardiolipin, smooth muscle, mitotic spindle appa-

ratus, centriole, lung, liver, and brain [11, 17, 4749]. CF

antibodies to brain tissue antigens have been demonstrated in

patients with and without CNS manifestations [17, 47]. Therole of such antibodies in human disease has not been deter-

mined, although one report suggests that they are more com-

mon in subjects with CNS disease [50]. The deposition of im-

mune complexes formed from such autoantibodies within small

venules in the CNS has been implicated as a possible mecha-

nism of immune-mediated neurologic injury [43, 51].

In autoimmune-induced encephalitis, one would anticipate

a more prolonged duration of symptoms before the onset of


8/11


encephalitis and the presence ofM. pneumoniaein the respi-

ratory tract but not in the CSF. In our cohort, all 5 children

in whom M. pneumoniaewas detected by PCR in the throat

but not in the CSF had prodromal symptoms of 7 days

duration. In contrast, 5 of the 6 children in whom M. pneu-

moniaewas detected by PCR in the CSF but not in the throat

had prodromal symptoms of 5 days duration. These results

are consistent with those of Narita et al. [28, 36], who dem-onstrated a statistically significant association between the de-

tection ofM. pneumoniaein CSF samples and serum by PCR

and fever of7 days duration before the onset of encephalitis.

This lends support to the contention that autoimmune mech-

anisms are likely responsible for some cases ofM. pneumoniae

encephalitis, particularly those with prodromal symptoms of

157 days duration.

Cerebrovascular thromboembolic phenomena have also been

implicated as a potential cause of neurologic injury in patients

infected with M. pneumoniae [52, 53]. Whether the cerebral

infarcts that occurred in 2 of the children who we studied were

caused by thromboembolic events, an autoimmune vasculitis,

or some other pathologic process is uncertain. Neurotoxin-

induced neurologic disease occurs in mice infected withMyco-

plasma neurolyticumand in turkeys infected with Mycoplasma

gallisepticum,but no such toxin has been demonstrated in hu-

mans infected with M. pneumoniae[52].

No immune response to M. pneumoniaecould be demon-

strated by either EIA or CF assays in the acute-phase or con-

valescent-phase serum samples of 3 of 6 children with probable

M. pneumoniaeencephalitis for whom the results of CSF PCR

were positive (table 1). All 3 previously had been healthy, with

no evidence to suggest an immune deficiency. The absence ofa detectable serologic response could have been due to a lack

of sensitivity of the assays or a failure to develop specific an-

tibody. False-negative results of serologic tests for M. pneu-

moniaehave been described elsewhere [29, 5456]. In a study

of 3546 patients with pneumonia, 10% of culture-proven M.

pneumoniae infections were associated with acute-phase and

convalescent-phase CF titers of 1:16, and only 53% had a

4-fold rise in the CF titer [54]. In studies that used the mi-

croparticle agglutination test and indirect immunofluorescence

serologic assays, 15%25% of children with PCR-proven res-

piratory infections caused byM. pneumoniaehad negative re-

sults of serologic testing for the microorganism [55, 56]. False-negative results of CF for M. pneumoniae have also been

documented in a fatal case of culture-proven infection of the

CNS [29] and in a child who had M. pneumoniaedetected in

the CSF and the throat by PCR [25]. Thus, the absence of

detectable antibodies to M. pneumoniaedoes not eliminate it

as a possible cause of infection.

Strong evidence for coinfections, as determined by the direct

detection of an organism by use of culture, PCR, or antigen

detection techniques was demonstrated in 5 (45%) of the 11

patients who were classified as having probableM. pneumoniae

encephalitis (table 1). Respiratory viral infections can have cyto-

toxic effects on respiratory epithelium, and therefore it is con-

ceivable that such infections could have facilitated bloodstream

invasion byM. pneumoniae. Dual CNS infection with HSV-2

and M. pneumoniae, as was demonstrated in 2 of the children

in our cohort, has not been previously reported. The presenceof detectable antibody to HSV in one of these children at the

onset of encephalitis and the lack of a 4-fold rise in HSV anti-

body titer in both children suggest that reactivation of latent

HSV infection, rather than primary HSV infection, occurred

in these 2 children. The pathogenetic role of other infectious

agents detected by means of serologic tests alone in patients

with probable M. pneumoniaeencephalitis is questionable. M.

pneumoniaehas been shown to induce non-antigen-specific T

celldependent antibody production in human B lymphocytes

in vivo [57, 58], and it is likely that the elevated levels of

antibody to these other infectious agents are the result of non-

specific immune activation byM. pneumoniae.

Twenty-eight (93%) of 30 children who had indeterminate

M. pneumoniaeencephalitis had serologic evidence of infection

with 1 other agent. Thirteen (43%) had evidence of infection

with 2 agents, in addition to M. pneumoniae. A minority of

these cases might represent true dual viral mycoplasmal infec-

tions or mycoplasma-induced nonspecific immune activation.

However, in the majority of these cases, the evidence that im-

plicated other agents as the cause of encephalitis was compel-

ling, suggesting that false-positive results of serologic tests for

M. pneumoniaewere likely. Such false-positive results could be

caused by cross-reactivity with human brain antigens [17, 19,59] or by other inherent limitations in the specificity of sero-

logic tests for M. pneumoniae [1415]. For all these reasons,

we believe that in the absence of corroborating culture or PCR

results, seropositivity is insufficient to establish a definitive di-

agnosis ofM. pneumoniaeencephalitis.

The clinical manifestations, CSF abnormalities, and EEG and

neuroimaging findings associated withM. pneumoniaeenceph-

alitis are indistinguishable from those of viral encephalitis [6].

Long-term neurologic sequelae have been documented in

48%54% of serologically confirmed cases ofM. pneumoniae

encephalitis [7, 12]. In our cohort, 7 (64%) of 11 children with

probable M. pneumoniae encephalitis experienced neurologicsequelae.

Antibiotic therapy has been temporally associated with clin-

ical improvement in some cases ofM. pneumoniaeencephalitis

[12, 25, 31, 32, 34, 35, 37, 40, 42, 43] but not in others [11,

12, 26, 38, 4245, 59]. Complete neurologic recoveries without

antibiotic therapy have also been described [24, 36, 41, 42].

Unfortunately, it is not possible to draw any conclusion re-

garding the efficacy of antibiotic therapy from the results of


9/11


our study or from the data in the literature because of the small

number of children treated, the uncertainty regarding the nat-

ural history of untreated infection, and the failure to use an

antibiotic that penetrates the blood-brain barrier in many of

the reported cases. Nevertheless, until more-definitive data be-

come available, it is reasonable to consider antibiotic therapy

for children who present with a short prodromal illness con-

sistent with direct CNS infection. In theory, an antibiotic suchas chloramphenicol, doxycycline, or a quinolone should be

used, because of the antimycoplasmal activity of such anti-

biotics and their ability to traverse the blood-brain barrier.

The optimal management of this condition in children in

whom immune-mediated neurologic injury is likely remains un-

certain. At present, no data support or refute the potential benefit

of antimicrobial therapy for such patients. Corticosteroids have

been temporally associated with clinical improvement in a small

number of patients with ADEM and serologically confirmed M.

pneumoniae infection [12, 4244, 60, 61]. The efficacy of cor-

ticosteroids in the treatment ofM. pneumoniaeassociated trans-

verse myelitis is less well established [45]. Plasmapheresis ap-

peared to be effective in 1 patient withM. pneumoniaetransverse

myelitis [62] and 1 patient with polyradiculitis [63]. No such

benefit was seen with iv immunoglobulin [40]. Despite the lack

of conclusive evidence, it is reasonable to consider the use of

these immune-modulating therapies in addition to antibiotics in

children withM. pneumoniaeencephalitis who have severe dis-

ease.

Because of the difficulties associated with the serologic di-

agnosis and the prolonged incubation period required for cul-

ture, PCR is increasingly used for the diagnosis ofM. pneu-

moniae infections [2022, 28, 36, 37, 55, 56, 6469]. Inpreliminary experiments that involved 114 clinical respiratory

samples, 50 of which had culture results that were positive for

M. pneumoniae, our PCR assay had a sensitivity, specificity,

positive predictive value, and negative predictive value of 100%,

98.4%, 98%, and 100%, respectively, with culture serving as

the gold standard (unpublished data). The high sensitivity and

specificity of our assay are consistent with published data from

other centers that used the same PCR primers [2122].

Although the possibility of false-positive PCR results cannot

be absolutely excluded, such an event is unlikely, given the

design and execution of the assay. The detection ofM. pneu-

moniae in the CSF of 6 children with encephalitis providesstrong evidence that this organism was responsible for their

illness, regardless of the results of serologic tests. The detection

ofM. pneumoniaein the throat but not in the CSF (in 5 prob-

able cases) provided less-conclusive evidence of causality and

therefore required the presence of supportive positive results

of serologic tests for M. pneumoniae.

In conclusion,M. pneumoniaeis responsible for at least 6.9%

of cases of acute childhood encephalitis and is strongly asso-

ciated with 12.5% of cases. Neurologic sequelae are common,

occurring in 48%64% of cases. Direct invasion of the CNS is

the probable pathogenetic mechanism in children with a brief

(5 days duration) prodromal illness. Immunologic phenom-

ena, thromboembolic phenomena, or both may be the mech-

anisms responsible in children with more prolonged (7 days

duration) prodromal illness. Coinfections with respiratory vi-

ruses and herpesviruses are common. Currently available EIAand CF tests for M. pneumoniaeare not sufficiently specific to

reliably diagnose acute childhood encephalitis caused by M.

pneumoniaein the absence of corroborating evidence, such as

positive culture or PCR.

References

1. Koskiniemi M, Korppi M, Mustonen K, et al. Epidemiology of en-

cephalitis in children: a prospective multicentre study. Eur J Pediatr

1997; 156:5415.

2. Beghi E, Nicolosi A, Kurland LT, Mulder DW, Hauser WA, Shuster L.

Encephalitis and aseptic meningitis, Olmsted county, Minnesota,

19501981: I. Epidemiology. Ann Neurol 1984; 16:28394.

3. Koskiniemi M, Vaheri A. Effect of measles, mumps, rubella vaccination

on pattern of encephalitis in children. Lancet 1989;1:314.

4. Lennette EH, Magoffin RL, Schmidt NJ, Hollister AC. Viral diseases

of the central nervous system; influence of poliomyelitis vaccination

on etiology. JAMA 1959; 171:145664.

5. Koskiniemi M, Rautonen J, Lehtokoski-Lehtiniemi E, Vaheri A. Epi-

demiology of encephalitis in children: a 20-year survey. Ann Neurol

1991; 29:4927.

6. Kolski H, Ford-Jones EL, Richardson S, et al. Etiology of acute childhood

encephalitis at The Hospital for Sick Children, Toronto, 19941995. Clin

Infect Dis 1998; 26:398409.

7. Lehtokoski-Lehtiniemi E, Koskiniemi M.Mycoplasma pneumoniaeen-

cephalitis: a severe entity in children. Pediatr Infect Dis J 1989; 8:6513.

8. Urquhart GED. Mycoplasma pneumoniae infection and neurological

complications. Br Med J 1979; 2:1512.

9. Koskiniemi M, Vaheri A. Acute encephalitis of viral origin. Scand J

Infect Dis1982; 14:1817.

10. Rautonen J, Koskiniemi M, Vaheri A. Prognostic factors in childhood

acute encephalitis. Pediatr Infect Dis J 1991; 10:4416.

11. Lind K, Zoffmann H, Larsen SO, Jessen O. Mycoplasma pneumoniae

infection associated with affection of the central nervous system. Acta

Med Scand 1979; 205:32532.

12. Thomas NH, Collins JE, Robb SA, Robison RO. Mycoplasma pneu-

moniaeand neurological disease. Arch Dis Child 1993; 69:5736.

13. Koskiniemi M. CNS manifestations associated withMycoplasma pneu-

moniaeinfections: summary of cases at the University of Helsinki and

review. Clin Infect Dis 1993; 17(Suppl 1):S527.

14. Jacobs E. Serological diagnosis ofMycoplasma pneumoniaeinfections:

a critical review of current procedures. Clin Infect Dis 1993; 17(Suppl

1):S7982.

15. Cimolai N.Mycoplasma pneumoniae. Arch Dis Child 1994;71:281.

16. Kenny GE, Newton RM. Close serological relationship between gly-

colipids of Mycoplasma pneumoniaeand glycolipids of spinach. Ann

N Y Acad Sci 1973; 225:5461.

17. Biberfeld G. Antibodies to brain and other tissues in cases ofMyco-

plasma pneumoniae infection. Clin Exp Immunol 1971; 8:31933.

18. Kleemola M, Kayhty H. Increase in titers of antibodies toMycoplasma

pneumoniaein patients with purulent meningitis. J Infect Dis 1982;

146:2848.

19. Leinikki PO, Panzar P, Tykka H. Immunoglobulin M antibody response


10/11


against Mycoplasma pneumoniaelipid antigen in patients with acute

pancreatitis. J Clin Microbiol 1978; 8:1138.

20. Ursi JP, Ursi D, Ieven M, Pattyn SR. Utility of an internal control for

the polymerase chain reaction: application to detection ofMycoplasma

pneumoniae in clinical specimens. APMIS 1992; 100:6359.

21. Ieven M, Ursi D, Van Bever H, Quint W, Niesters HGM, Goossens H.

Detection ofMycoplasma pneumoniaeby two polymerase chain reac-

tions and role of M. pneumoniae in acute respiratory tract infections

in pediatric patients. J Infect Dis1996; 173:144552.

22. Dorigo-Zetsma JW, Zaat SAJ, Wertheim-van Dillen PME, et al. Com-parison of PCR, culture, and serologic tests for diagnosis ofMycoplasma

pneumoniaerespiratory tract infection in children. J Clin Microbiol

1999; 37:147.

23. Johnson G, Nelson S, Petric M, Tellier R. Comprehensive PCR-based

assay for detection and species identification of human herpesviruses.

J Clin Microbiol2000; 38:32749.

24. Nagayama Y, Sakurai N, Tamai K, Niwa A, Yamamoto K. Isolation of

Mycoplasma pneumoniaefrom pleural fluid and/or cerebrospinal fluid:

report of four cases. Scand J Infect Dis 1987; 19:5214.

25. Abele-Horn M, Franck W, Busch U, Nitschko H, Roos R, Heesemann

J. Transverse myelitis associated with Mycoplasma pneumoniaeinfec-

tion. Clin Infect Dis 1998; 26:90912.

26. Ieven M, Demey H, Ursi D, Van Goethem G, Cras P, Goosens H. Fatal

encephalitis caused byMycoplasma pneumoniaediagnosed by the poly-

merase chain reaction. Clin Infect Dis 1998; 27:15523.

27. Inoue S, Ohfu M, Morooka T, Tsuru N, Mitsudome A, Narita M. A

case of encephalitis due toMycoplasma pneumoniae:detection of spe-

cific DNA from cerebrospinal fluid and elevation of interleukin-6 [in

Japanese]. Kansenshogaku Zasshi 1998; 72:7715.

28. Narita M, Matsuzono Y, Itakura O, Togashi T, Kikuta H. Survey of

mycoplasmal bacteremia detected in children by polymerase chain re-

action. Clin Infect Dis 1996; 23:5225.

29. Koletsky RJ, Weinstein AJ. FulminantMycoplasma pneumoniaeinfec-

tion: report of a fatal case and a review of the literature. Am Rev Respir

Dis 1980; 122:4916.

30. Launes J, Paetau A, Linnavuori K, Iivanaineu M. Direct invasion of

the brain parenchyma byMycoplasma pneumoniae.Acta Neurol Scand

1997; 95:374.

31. Dionisio D, Valassina M, Mata S, et al. Encephalitis caused directly by

Mycoplasma pneumoniae. Scand J Infect Dis 1999; 31:5069.

32. Abramovitz P, Schvartzman P, Harel D, Lis I, Naot Y. Direct invasion

of the central nervous system byMycoplasma pneumoniae:a report of

two cases. J Infect Dis 1987; 155:4827.

33. Fleischhauer P, Huben U, Mertens H, Sethi KK, Thurmann D. Na-

chweis vonMycoplasma pneumoniaeim liquor bei akuter polyneuritis.

Dtsch Med Wochenschr 1972; 97:67882.

34. Suzuki K, Matsubara S, Uchikata M, et al. A case of Mycoplasma

pneumoniae pneumonia associated with meningoencephalitis [ab-

stract]. Yale J Biol Med 1983; 56:8734.

35. Kasahara I, Otsuba Y, Yanase T, Oshima H, Ichimaru H, Nakamura

M. Isolation and characterization of Mycoplasma pneumoniae from

cerebrospinal fluid of a patient with pneumonia and meningoenceph-

alitis. J Infect Dis 1985; 152:8235.

36. Narita M, Itakura O, Matsuzono Y, Togashi T. Analysis of mycoplasmal

central nervous system involvement by polymerase chain reaction. Pe-

diatr Infect Dis J 1995; 14:2367.

37. Narita M, Matsuzono Y, Togashi T, Kajii N. DNA diagnosis of central

nervous system infection by Mycoplasma pneumoniae. Pediatrics

1992; 90:2503.

38. Tjhie JHT, Van de Putte EM, Haasnoot K, Van den Brule AJC, Van-

denbroucke-Grauls CMJE. Fatal encephalitis caused by Mycoplasma

pneumoniaein a 9-year-old girl. Scand J Infect Dis 1997; 29:4245.

39. Pellegrini M, OBrien TJ, Hoy J, Sedal L. Mycoplasma pneumoniae

infection associated with an acute brainstem syndrome. Acta Neurol

Scand1996; 93:2036.

40. Meseguer MA, Aparicio M, Calvo A, de Blas G, Lorenzo G. Mycoplasma

pneumoniaeantigen detection in Guillain-Barre syndrome. Eur J Pe-

diatr1998; 157:1034.

41. Kornips HM, Verhagen WIM, Prick MJJ. Acute disseminated enceph-

alomyelitis probably related to a Mycoplasma pneumoniae infection.

Clin Neurol Neurosurg1993; 95:5963.

42. Nishimura M, Saida T, Kuroki S, et al. Post-infectious encephalitiswith

anti-galactocerebroside antibody subsequent to Mycoplasma pneumo-

niae infection. J Neurol Sci 1996; 140:915.

43. Behan PO, Feldman RG, Segerra JM, Draper IT. Neurological aspects

of mycoplasma infection. Acta Neurol Scand 1986; 74:31422.44. Sheth RD, Goulden KJ. Mycoplasma pneumoniaeassociated enceph-

alitis: extension of the clinical spectrum [abstract]. Ann Neurol

1992; 32:4489.

45. Mills RW, Schoolfield L. Acute transverse myelitis associated with

Mycoplasma pneumoniae infection: a case report and review of the

literature. Pediatr Infect Dis J 1992; 11:22831.

46. Heller L, Keren O, Mendelson L, Davidoff G. Transverse myelitis as-

sociated with Mycoplasma pneumoniae: case report. Paraplegia 1990;

28:5225.

47. Biberfeld G. Infection sequelae and autoimmune reactions in Myco-

plasma pneumoniae infection. In: Razin S, Barile MF, eds. The myco-

plasmas. Vol 4. New York: Academic Press, 1985:293311.

48. Lind K, Hier-Madsen M, Wiik A. Autoantibodies to the mitotic spin-

dle apparatus in Mycoplasma pneumoniae disease. Infect Immun

1988; 56:7145.

49. Cimolai N, Mah D, Roland E. Anticentriolar autoantibodies in children

with central nervous system manifestations ofMycoplasma pneumoniae

infection. J Neurol Neurosurg Psychiatry1994; 57:6389.

50. Lind K. Manifestations and complications ofMycoplasma pneumoniae

disease: a review. Yale J Biol Med 1983; 56:4618.

51. Biberfeld G, Norberg R. Circulating immune complexes in Mycoplasma

pneumoniae infection. J Immunol 1974; 112:4135.

52. Clyde WA. Mycoplasmal diseases. In: Scheld WM, Whitley RJ, Durack

DT, eds. Infections of the central nervous system. 2d ed. Philadelphia:

Lippincott-Raven,1997; 60312.

53. Fu M, Wong KS, Lam WWM, Wong GWK. Middle cerebral artery

occlusion after recent Mycoplasma pneumoniaeinfection. J Neurol Sci

1998; 157:1135.

54. Kenny GE, Kaiser GG, Cooney MK, Foy HM. Diagnosis ofMyco-

plasma pneumoniaepneumonia: sensitivities and specificities of se-

rology with lipid antigen and isolation of the organism on soy pep-

tone medium for identification of infections. J Clin Microbiol

1990; 28:208793.

55. Abele-Horn M, Busch U, Nitschko H, et al. Molecular approaches to

diagnosis of pulmonary disease due toMycoplasma pneumoniae.J Clin

Microbiol 1998; 36:54851.

56. Skakni L, Sardet A, Just J, et al. Detection ofMycoplasma pneumoniae

in clinical samples from pediatric patients by polymerase chain reac-

tion. J Clin Microbiol 1992; 30:263843.

57. Biberfeld G. Activation of human lymphocyte subpopulations byMyco-

plasma pneumoniae.Scand J Immunol 1977; 6:114550.

58. Beberfeld G, Arneborn P, Forsgren M, Von Stedingk LV, Blomqvist S.

Non-specific polyclonal antibody response induced by Mycoplasma

pneumoniae.Yale J Biol Med 1983; 56:63942.

59. Ponka A. Central nervous system manifestation associated with sero-

logically verifiedMycoplasma pneumoniaeinfection. Scand J Infect Dis

1980; 12:17584.

60. Lum G, Kulhanjian J. Severe encephalitis in a three-year-old girl. Pe-

diatr Infect Dis J 1996; 15:181, 1846.

61. Beskind DL, Keim SM. Choreoathetotic movement disorder in a boy

withMycoplasma pneumoniaeencephalitis. Ann Emerg Med 1994;23:

13758.

62. Cotter FE, Bainbridge D, Newland AC. Neurological deficit associated

with Mycoplasma pneumoniae reversed by plasma exchange. Br Med

J (Clin Res Ed) 1983; 286:22.

63. Ponka A, von Bonsdorff M, Farkkila M. Polyradiculitis associated with


11/11


Mycoplasma pneumoniaereversed by plasma exchange. Br Med J (Clin

Res Ed) 1983; 286:4756.

64. Jensen JS, Sndergard-Andersen J, Uldum SA, Lind K. Detection of

Mycoplasma pneumoniae in simulated clinical samples by polymerase

chain reaction. APMIS 1989; 97:10468.

65. Jensen JS, Uldum SA, Lind K. Clinical evaluation of a polymerase chain

reaction assay for detection ofMycoplasma pneumoniae. 9th Interna-

tional Congress of the International Organization for Mycoplasmology,

27 August 1992, Ames, Iowa. Int Organ Mycoplasmol Lett 1992;

2(PSI/65):198.

66. Richardson S, Correa A, Lim C, Matlow A, Quinn P, Penn L. Detection

ofMycoplasma pneumoniaeand Mycoplasma genitaliumin respiratory

specimens using PCR. Clin Invest Med 1994; 4(Suppl 17):B80.

67. Waris ME, Toikka P, Saarinen T, et al. Diagnosis ofMycoplasma pneu-

moniaepneumonia in children. J Clin Microbiol1998; 36:31559.

68. Bernet C, Garret M, De Barbeyrac B, Bebear C, Bonnet J. Detection

ofMycoplasma pneumoniaeby using the polymerase chain reaction. J

Clin Microbiol1989; 27:24926.

69. Leng Z, Kenny GE, Roberts MC. Evaluation of the detection limits of

PCR for identification ofMycoplasma pneumoniaein clinical samples.

Mol Cell Probes 1994; 8:12530.

Acute_childhood_encephalitis_and_Mycoplasma_pneumoniae.pdf

Documents

Transcript of Acute_childhood_encephalitis_and_Mycoplasma_pneumoniae.pdf