A

ar

H

a

©

K

R

cd

a

e

d©

M

0

Disponible en ligne sur

ScienceDirectwww.sciencedirect.com

Médecine et maladies infectieuses 44 (2014) 429–432

Short communication

Impact of serodiagnosis on the management of Lyme borreliosis at AngersUniversity Hospital

Impact du sérodiagnostic sur la prise en charge de la borréliose de Lyme au CHU d’Angers

E. Twizeyimana a,b,∗,c, E. Pichard b, F. Lunel-Fabiani a, S. Fanello b, S.J. De Martino c

a Laboratoire de virologie, CHU d’Angers, 4, rue Larrey, 49933 Angers cedex 9 Franceb Service des maladies infectieuses et tropicales, CHU d’Angers, 4, rue Larrey, 49933 Angers cedex 9, France

c CNR des Borrelia, hôpitaux universitaires de Strasbourg, 1, rue Koeberlé, 67000 Strasbourg, France

Received 14 October 2013; received in revised form 22 May 2014; accepted 19 July 2014Available online 22 August 2014

bstract

Introduction. – Lyme borreliosis (LB) is an emerging arthropod-borne disease the diagnosis of which is made on clinical and biological data. Wessessed the Angers University Hospital physicians’ management of LB, in case of positive serology, and estimated their compliance to Europeanecommendations (EUCALB).

Methods. – We retrospectively included 75 cases with positive ELISA serologies confirmed by Western-Blot, performed at the Angers Universityospital between 2008 and 2012.Results and discussion. – There were 4 cases of early localized phase, 26 of early-disseminated phase (including 17 cases of neuroborreliosis),

nd one case of late phase. The curative management complied with EUCALB guidelines in 28 cases out of 31.Conclusion. – Serology remains a reference diagnostic tool for LB, as long as the practitioner is aware of the main clinical and biological criteria.

2014 Elsevier Masson SAS. All rights reserved.

eywords: Lyme borreliosis; Serodiagnosis; EUCALB

ésumé

Introduction. – La borréliose de Lyme (BL) est une maladie vectorielle émergente dont le diagnostic est posé devant un faisceau d’argumentslinico-biologiques. Nous avons évalué les conduites pratiques des médecins du CHU d’Angers en cas de sérologie positive dans la prise en chargee la BL et apprécié leur conformité au vu des recommandations européennes (EUCALB).

Méthodes. – Nous avons recensé rétrospectivement 75 patients ayant présenté des sérologies Elisa positives confirmées par Western-Blot réaliséesu CHU d’Angers entre 2008 et 2012 inclus.

Résultats et discussion. – Il y avait quatre cas de phase précoce localisée, 26 cas de phase précoce disséminée, dont 17 cas de neuroborréliose,t un cas de phase tardive. Le traitement curatif était conforme aux recommandations dans 28 cas sur 31.

Conclusion. – La sérologie constitue un bon outil diagnostique de la BL à condition de disposer d’une meilleure connaissance des critères deéfinition clinico-biologique.

2014 Elsevier Masson SAS. Tous droits réservés.

ots clés : Borréliose de Lyme ; Sérodiagnostic ; EUCALB

∗ Corresponding author.E-mail address: eterne.twizeyimana@chru-strasbourg.fr (E. Twizeyimana).

1

B

http://dx.doi.org/10.1016/j.medmal.2014.07.011399-077X/© 2014 Elsevier Masson SAS. All rights reserved.

. Introduction

Lyme borreliosis (LB) is an infection due to bacteria of theorrelia genus transmitted by the bite of a tick of the Ixodes

4 malad

gddintalcwiaLtamc

2

2t

U2rdpmsI

2

IoamfcdtGncoa

2

tdaw

niont

3

2rteEfffE1(

lc4itTtirfsotaifIt

a

fiwow7wLda

o

30 E. Twizeyimana et al. / Médecine et

enus [1]. The clinical diagnosis of the disease lacks specificityuring the disseminated stage. During that stage, the biologicaliagnosis is crucial and relies on an indirect serologic diagnosisn two steps: a first screening serological test, using an immu-oenzymatic technique (ELISA), and a second confirmationest using an immunolabelling or Western-Blot technique (WB),llowing defining the specificity of identified antibody [2]. Ana-yzing all the clinical and biological data allows the physician toonfirm the infection or not. The diagnosis relies on strict criteriaell defined on the national level by the consensus conference

ssued by the French Infectious Diseases Society (SPILF), andt the European level by the European Concerted Action onyme Borreliosis (EUCALB) [3–5]. Ignoring the recommenda-

ions may lead to misdiagnosis and inadequate treatment. Wessessed the Angers University Hospital physician’s manage-ent of LB, in case of positive serology, and estimated their

ompliance to EUCALB.

. Materials and methods

.1. Inclusion criteria for cases with positive serologicalests

We collected the LB serologies performed by the Angersniversity Hospital microbiology laboratory between 2008 and012, recorded with GLIMSTM (MIPS) software. We then sepa-ated the negative ELISA results from the significant (positive oroubtful) confirmed by WB. The patient’s clinical data for eachositive serology was documented by consulting the electronicedical record on the CROSSWAYTM (CLM) network and clas-

ified according to the EUCALB diagnostic criteria [5].IgM andgG kit (DIASORIN, Italy)

.2. Kits used for the serology

The ELISA Liaison Borrelia IgM and IgG kit (DIASORIN,taly) was used for the screening test. The test’s principle reliesn the semi-quantitative chemiluminescent immunoassay ofnti-Borrelia burgdorferi sansu lato (Bbsl) IgM and IgG. Theanufacturers defined the significant thresholds of positivity

or IgG and IgM according to European recommendations. Theonfirmation technique, performed only in case of positive oroubtful screening serological test for IgG and/or IgM, relied onhe Euroline–WB Anti-Borrelia + VlsETM kit (EUROIMMUN,ermany) [6]. The interpretation of WB is based on a combi-ation algorithm between the major antigenic bands (OspC inase of IgM and VlsE in case of IgG) and minor antigenic bandsf total Borrelia afzelii extract. The bands were scanned andnalyzed by an automatic scanning system Euroline ScanTM.

.3. Classification of cases in clinical stages

The seropositive clinical cases were classified according to

he EUCALB recommendations [5]: early localized phase, earlyisseminated phase, and late disseminated phase. The compli-nce of treatments undertaken to the same recommendationsas also assessed [3–5]. We created two other groups that could

pfnc

ies infectieuses 44 (2014) 429–432

ot be classified according to EUCALB criteria: aspecific clin-cal presentations and asymptomatic seropositivity. The mainbjective was to determine the compliance of the clinical diag-osis and of the treatment to EUCALB recommendations forhe management of LB.

. Results and discussion

We collected 2524 requests for LB serologies from 2008 to012 included; corresponding to 2369 patients after eliminatingedundant requests (6%). Repeating LB serologies is only jus-ified to screen for seroconversion and not to assess therapeuticffectiveness. There were 253 (10%) significant serologies withLISA for IgG, including 187 positives (74%) and 66 doubt-

ul (26%). There were almost twice less significant serologiesor IgM, 134 (5%), including 94 positives (70%) and 40 doubt-ul (30%). The WB, when performed (only for 80% significantLISA for IgG), confirmed the results of the ELISA for IgG in03 cases out of 204 (50%) and for IgM in 24 cases out of 13418%).

This retrospective study focusing on all requests for sero-ogical tests during 5 years revealed the weak rate ofonfirmed significant (positive or doubtful) serological results:% (103/2524) for IgG and almost 1% (24/2524) for IgM. Fig. 1llustrates the number of negative and significant ELISA andhose confirmed by WB according to various prescribing units.he rate of significant tests ranged at an average of 10% in

he units with the greatest number of requests except for thenfectious and tropical diseases unit where it reached 25%. Theetrospective analysis of all the data proves an overprescriptionor LB serologies that inevitably exposes to risks of misdiagno-is and overtreatment. This is why the systematic prescriptionf LB serology is contra-indicated and that the indications forhis serology are based on well defined epidemiological, clinical,nd biological data [3–5]. The analysis of results allowed us todentify 10 cases for which ELISA serologies were positive bothor IgG and IgM, and despite recommendations, 2 WB (IgG andgM) had been performed. It is recommended to perform onlyhe confirmation test for IgG which is more specific in this case.

Finally, the study of 2524 serologies prescribed between 2008nd 2012 included led us to include only 75 cases.

The diagnosis made for 75 patients (Table 1) were classi-ed by phase. Four cases of single erythema migrans (EM)ere observed. Nevertheless, after carefully reading the medicalbservations, the semiological description of lesions remainedeakly informative. An adequate treatment was administered in5% of the cases. The early disseminated phase presentationsere multiple EM (MEM = 2), early neuroborreliosis (NB = 17),yme arthritis (LA = 4), and cardioborreliosis (CB = 3). The lateisseminated phase presentation was an acrodermatitis chronicatrophica (ACA = 1).

The high rate of NB (54.8%) reflects the clinical epidemiol-gy of LB and corresponds to the most frequent disseminated

hase of Bbsl infections [7,8]. These early NB (most often,acial palsy and meningoradiculitis) were all typical but wereot classified as proven diagnosis because of a missing spe-ific intrathecal synthesis index even though CSF samples

E. Twizeyimana et al. / Médecine et maladies infectieuses 44 (2014) 429–432 431

Fig. 1. Number of serological tests (screening tests ELISA confirmed by WB) prescribed from 2008 to 2012 included (mean and standard deviation).Nombre de tests sérologiques de dépistage Elisa prescrits et confirmés en WB de 2008 à 2012 (moyenne et écarts-type).

Table 1Description of clinical cases with positive serology according to the clinical definition and therapeutic EUCALB guidelines.Description des tableaux cliniques positifs en sérologie selon les critères diagnostiques de définition clinique et thérapeutique de l’EUCALB.

Clinical cases Number of patientfiles(n = 75)

Significant ELISAIgG

Confirmation byWestern-Blot IgG

Significant ELISAIgM

Confirmation byWestern-Blot IgM

Therapeuticcompliance withEUCALBa

EM 4 4 4 2 2 3Multiple EM 2 2 2 2 2 2Early

neuroborreliosis17 17 15 10 5 15

Lyme arthritis 4 4 4 2 1 4Cardioborreliosis 3 3 2 2 1 3Chronic

atrophyingacrodermatitis

1 1 1 0 0 1

Total number ofborreliosiscases

31 31 28 18 11 28

Asymptomaticseropositivity

35 32 29 8 4 27

Aspecific clinical 9 8 8 3 3 0

wco

mrtpbcTEf

a

c

Lod

4

adi

D

cases

a European concerted action on Lyme borreliosis: http://www.eucalb.com.

ere available. The serology was positive for IgG in 88%ases. The antibiotic treatment was inadequate in 2 cases outf 17.

There were 4 cases of seropositive LA; only 1 case was a kneeonoarthritis. All the cases of arthritis were associated to high

ates of specific IgG and were all treated adequately. No inves-igation on synovial fluid or synovial biopsy (specific PCR) waserformed. The cases of cardioborreliosis were atrioventricularlock, myocarditis and atrial fibrillation. Sixty-six percent ofases had specific IgG. All the patients were treated adequately.he ACA corresponded to clinical, biological, and therapeuticUCALB definitions. Furthermore, a specific PCR was positive

or a skin biopsy and confirmed the diagnosis.We also noted 35 cases of asymptomatic seropositivity (46%)

nd 9 cases with aspecific symptoms (12%).The management was compliant with EUCALB therapeutic

riteria for 55 out of 75 patients (73%): 28/31 cases of proven c

B were treated adequately (90%). Twenty-seven over 44 casesut of EUCALB definitions were not treated (61%) and were byefinition, compliant.

. Conclusion

Serology is a good diagnostic tool for LB as long as there is better knowledge of clinical and biological criteria of LB. Theiagnosis of NB remained uncertain because of missing specificntrathecal synthesis index.

isclosure of interest

The authors declare that they have no conflicts of interestoncerning this article.

4 malad

A

adMrbihFc

R

[

[

[

[

[[

[

32 E. Twizeyimana et al. / Médecine et

cknowledgements

The first author, Dr. E. Twizeyimana, collected the clinicalnd biological data and drafted the article: this study was con-ucted for a Medical Biology postgraduate diploma. Dr. S.J. Deartino controlled the original draft and especially the bacte-

iological technique. Pr. E. Pichard prepared the clinical andiological topic and suggested it to the first author who wasn a clinical training session in his unit. Pr. S. Fanello offeredis technical assistance for bibliographic research. Pr. F. Lunel-abiani proofread the article and also suggested technicalorrections.

eferences

1] Stanek G, Wormser GP, Gray J, Strle F. Lyme borreliosis. Lancet2012;379(9814):461–73.

[

ies infectieuses 44 (2014) 429–432

2] De Martino S-J. Place des méthodes biologiques dans le diagnostic desdifférentes manifestations de la borréliose de Lyme. Quelles sont les tech-niques ? Quelles sont celles disponibles actuellement ? Med Mal Infect2007;37(7–8):496–506.

3] 16e Conférence en thérapeutique anti-infectieuse. Borréliose de Lyme :démarches diagnostiques, thérapeutiques et préventives. Texte long. MedMal Infect 2007;37(Suppl. 3):S153–74.

4] Stanek G, Fingerle V, Hunfeld K-P, Jaulhac B, Kaiser R, Krause A, et al.Lyme borreliosis: Clinical case definitions for diagnosis and management inEurope. Clin Microbiol Infect 2011;17(1):69–79.

5] http://www.eucalb.com6] Busson L, Reynders M, Van den Wijngaert S, Dahma H, Decolvenaer

M, Vasseur L, et al. Evaluation of commercial screening tests and blotassays for the diagnosis of Lyme borreliosis. Diagn Microbiol Infect Dis2012;73(3):246–51.

7] Blanc F, Jaulhac B, Fleury M, de Seze J, de Martino SJ, Remy V, et al.

Relevance of the antibody index to diagnose Lyme neuroborreliosis amongseropositive patients. Neurology 2007;69(10):953–8.

8] Blanc F. Epidemiology of Lyme borreliosis and neuroborreliosis in France.Rev Neurol (Paris) 2009;165(8–9):694–701.