Erythema migrans multiloculare 
ERYTHEMA MIGRANS MULTILOCULARE 
IN STYRIA (AUSTRIA) 
R. R. Mtillegger, G. Brunner-Kohler, N. Zochling, H. Reiter, S. Hodl, 
H. P. Soyer and H. Kerl 
ABSTRACT 
To characterize clinical, serologic, and histopathologic features of patients with erythema migrans multiloculare 
from Styria (Austria), an European area well known to be endemic for Lyme Borreliosis, 270 consecutive 
patients (m:f = 128:142, mean age = 54 years) with erythema migrans were studied .at the Department of 
Dermatology in Graz, Austria between March 1993 and October 1995. 17/270 patients (m:f = 9:8, mean age 
= 37 years) (6.3%) presented with a multifocal form of the disease. Several findings in these 17 patients were 
remarkable. Anamnestic data/clinical findings: History of tick or insect bite(s): 10/17 patients (4 singular, 6 
multiple); mean total number of lesions/patient = 4 (ranging 2-18); latency of 1 - 14 days (mean 7 days) in 
9/17 patients (53%) between primary erythema migrans and secondary lesions; in these 9 patients the 
secondary lesions showed a different morphologic aspect in comparison to the primary erythema migrans; 
extracutaneous signs and symptoms: 7 /17 patients ( 41 % ). Serologic findings: Borrelia burgdorferi serum ELISA 
lgG-antibodies: 9/17 patients (53% ), lgM: 9/17 patients (53% ). Histopathologic findings: Complete lack of 
epidermal changes and milder inflammatory dermal infiltrate in all secondary lesions (4 biopsy specimens) in 
comparison to primary erythemata migrantia (5 biopsy specimens). 
Our clinical and serologic findings are largely in accordance with the scarce literature on this condition from 
Europe where erythema migrans multiloculare is much rarer than in the USA The pathogenesis of erythema 
migrans multiloculare which is currently defined to be a disseminated form of Lyme Borreliosis due to 
haematogenous spread of Borrelia burgdorferi will be discussed in consideration of our observations. 
KEY WORDS 
Lyme borreliosis, Borrelia burgdorferi, e,ythema migrans multiloculare, early disseminated disease 
INTRODUCTION 
Erythema migrans (EM) usually occurs as a single 
lesion. However, multiple (secondary) erythemas in 
one patient (erythema migrans multiloculare (EMM)) 
are quite frequent in the USA (25 - 50% of all EM 
acta dermatovenerologica A.P.A. Vol 5, 96, No 3-4 
patients) (1), whereas EMM is rarely seen in Europe 
( 4 - 10% of all EM patients) (2). In the USA up 
to 100 erythematous lesions may appear in · one 
patient with EMM (1). On the contrary, EMM 
patients in Europe normally present with only few 
113 
Erythema migrans multiloculare 
erythemas (up to 10) (3). Differences between the 
primary EM and the secondary lesions concerning 
morphology, predilection sites, dynamics, and local 
symptoms can be observed ( 4,5). Extracutaneous 
signs and symptoms are described to be found more 
often in EMM than in patients with solitary EM 
( 4,5). Elevated serum antibody titres against Borrelia 
burgdorferi (Bb) are more common in EMM than in 
singular EM (6). Bb could be cultivated from single/ 
primary lesions as well as from secondary lesions 
(1). EMM is currently classified as manifestation of 
the secondary stage ( early disseminated stage) of 
Lyme Borreliosis (LB) (7). 
The present study has been undertaken to assess 
the characteristics of EMM patients in Styria the 
southernmost province of Austria which is known to 
be highly endemic for LB. 
PATIENTS AND METHODS 
270 consecutive patients (m:f = 128:142, mean 
age = 54 years) with EM were seen at the 
Department of Dermatology in Graz, Austria between 
March 1993 and October 1995. The diagnosis was 
made on clinical and partly histopathologic (167/270 
patients) grounds. All patients resided in Styria 
(Austria) a LB endemic area. 17/270 patients (m:f 
= 9:8, mean age = 37 years) (6.3%) presented with 
multiple EM lesions. 10/17 patients could recall an 
arthropod bite 18 days before the occurrence of 
EM on the average (ranging 4 - 73 days). 4/10 
patients had singular, 6/10 multiple arthropod bites. 
Analysis for lgG and lgM antibodies to Bb was 
performed in 252/270 patients including all patients 
with multiple lesions. Purified native flagella of Bb, 
strain DK-1, isolated from a human EM lesion 
(DAKOPATTS ELISA Kit, DAKO Diagnostica, 
Glostrup, Denmark) was used as a test antigen. 
In 167/270 patients punch biopsies (4 mm) were 
taken for routine histopathologic examination. After 
formalin-fixation and paraffin-embedding, specimens 
were haematoxylin-eosin stained. 7/167 patients had 
EMM. In these 7 patients punch biopsies were 
taken either from the primary lesion (3/7), or 
secondary lesion (2/7) or primary and secondary 
lesion (2/7). 
RESULTS 
The mean disease duration before hospital visit 
was 18 days in solitary EM (ranging 4 - 55 days) 
114 
compared to 15 days in EMM (ranging 2 - 28 
days). The mean total number of erythemas in one 
patient with EMM was 4 (ranging 2 - 18). In 9/17 
EMM patients (53%) a primary erythema occurred 
on the average of 7 days before secondary lesions 
(ranging 1 - 14 days). In all these 9 patients 
primary EM showed the same morphology as a 
typical solitary EM, whereas the secondary erythemas 
were characterized by a smaller size and milder 
inflammatory reaction. Ali secondary erythemas showed 
a quite similar morphology and had no predilection 
sites. They lacked the typical peripheral expansion 
of solitary lesions. In the other 8/17 EMM patients 
(47%) erythemas developed more or less 
simultaneously and there were no differences in the 
clinical appearance between the lesions. 108/253 EM 
patients (43%) suffered from extracutaneous signs 
and symptoms of LB in comparison to 7/17 EMM 
patients ( 41 % ). A positive Bb serum ELISA lgG 
antibody titre was found in 71/235 EM patients 
(31 % ) and in 9/17 EMM patients (53% ), respectively. 
91/235 EM patients (39%) had a positive Bb serum 
ELISA lgM antibody result, whereas 9/17 EMM 
patients (53%) were lgM reactive. The histopathologic 
examination of the 4 biopsy specimens from secondary 
lesions in EMM revealed a mild to moderate (in 
comparison to solitary or primary erythemas) 
perivascular infiltrate with very sparse interstitial 
extension throughout thi dermis. The infiltrate was 
mainly composed of lymphocytes intermingled with 
some plasma cells. Epidermal changes were not 
observed in secondary lesions, whereas slight atrophy 
of the epidermis and few spongiotic foci could be 
found in primary EM. 
DISCUSSION 
Within a period of 31 months only 17 from a 
total of 270 patients with EM (6.3%) seen at the 
Department of Dermatology in Graz, Austria presented 
with a multifocal form of the disease. A similar low 
incidence of EMM as in Styria is also reported 
from other European areas (2), whereas EMM is 
quite frequent in the USA (1). The low average 
number of 4 erythemas in one EMM patient in 
our series is also consistent with former publications 
from Europe (3). In contrast to these observations, 
a high number of lesions usually occurs in EMM 
patients in North America (1). A latent period 
between primary and secondary erythemas with 
differences concerning their morphology were "present 
in 9 of our 17 patients (53% ). On the other hand 
we were not able to find · an interval or clinical 
acta dermatovenerologica A.P A. Vol 5, 96, No 3-4 
Erythema migrans multiloculare 
variations between erythemas in the other 8 patients 
( 47% ). Berger noted morphological differences of 
the lesions in EMM patients „in some instances" 
( 4) and Melski described this diversity in a very 
distinct manner (5). Weber and coworkers discovered 
secondary EM lesions to „resemble" the primary 
EM (8). However, no clear and comprehensive 
data on the tirne course of multiple erythemas exist 
in the literature. Additional signs and symptoms 
were not found more often in our EMM patients 
(41 %) than in patients with singular EM (43%). 
This contradicts the more frequent and severe 
constitutional symptoms in EMM patients known 
from the European as well as from the American 
literature (4,5,7). The percentage of increased serum 
ELISA IgG and IgM antibodies to Bb in our EMM 
series was far above the sensitivity for patients with 
solitary EM as in former reports (6). All 4 secondary 
EM lesions from which a punch biopsy was obtained 
in our EMM patients clearly differed clinically 
from the primary EM. Thus, the milder dermal 
infiltrate found in these lesions on histopathologic 
examination reflects the distinct clinical aspect of 
the erythemas. 
EMM is defined at present as secondary stage 
manifesta ti on of LB ( early disseminated disease) 
due to a haematogenous spread of Bb after primary 
infection (7). On the other hand it is conceivable 
that EMM is caused by multiple bites of infected 
vectors in some cases. The current definition is 
supported by some characteristics of EMM patients 
in comparison to patients with solitary EM in the 
literature: (i) Delayed occurrence of secondary 
erythemas with distinct morphological features ( 4,5,8). 
(ii) A more frequent appearance of extracutaneous 
signs and symptoms ( 4,5). (iii) Elevated Bb specific 
antibody titres are more common (6). (iv) A higher 
rate of positive Bb serum PCR reactions (9). Features 
(i) and (iii) were also seen in our EMM patients, 
but not feature (ii). However, our data are not 
sufficient enough to ·ctecide which pathogenetic 
mechanism underlies EMM. 
In case the postulate of a haematogenous spread 
of Bb in EMM is correct, one crucial question 
seems to have been addressed insufficiently so far: 
What are the prerequisites for a single patient to 
acquire either solitary EM or EMM? On the one 
hand, infection with a specific Bb subtype is 
conceivable. Associations between the existence of 
distinct Bb subtypes, different clinical manifestations 
of LB and various geographic regions were shown 
(10). As Bb sensu stricto is mainly found in the 
USA, whereas very rarely in Europe (11), and the 
incidence of EMM is much higher in the USA (1), 
one might suspect a relation between EMM and Bb 
sensu stricto. In lesional skin from 19 of our solitary 
EM patients in 1993, only B. afzelii (74%) and B. 
garinii (26%) could be detected ( unpublished), although 
all three Bb subtypes were isolated from the Styrian 
tick population (Pierer K., et al., unpublished). 
However, neither results from the literature nor 
own investigations are available so far to answer 
whether Bb sensu stricto might be the exclusive 
pathogen in EMM. On the other hand, a certain 
precondition of the patient (e.g. immunological status) 
could be responsible for the- development of EMM. 
For example, a German patient receiving radiation 
therapy because of cancer developed around 70 
lesions (12). 
More studies on the subject of EMM are nece-
ssary to better characterize this peculiar variant 
of EM. If the assumption will be further corro-
borated that EMM is a disseminated form of LB 
it appears to be a good model to study several 
pathogenetic aspects of LB (relation between Bb 
and the patient). 
REFERENCES 
l. Steere AC, Bartenhagen NH, Craft JE et al. The 
early clinical manifestations of Lyme disease. Ann 
Intern Med 1983; 99: 76-82 
2. Weber K, Neubert U. Clinical features of early 
erythema migrans disease and related disorders. Zbl 
Bakt Hyg A 1986; 263: 209-28 
3. Asbrink E, Olsson I. Clinical manifestations of 
acta dermatovenerologica A.P A. Vol 5, 96, No 3-4 
erythema chronicum migrans Afzelius in 161 patients. 
Acta Derm Venereol (Stockh) 1985; 65: 43-52 
4. Berger BW. Erythema chronicum migrans of 
Lyme disease. Arch Dermatol 1984; 120: 1017-21 
5. Melski JW. Lyme Borreliosis. Fitzpatrick's J Clin 
Derm 1994; May-June: 14-25 
6. Melski JW, Reed KD, Mitchell PD et al. Primary 
115 
Erythema migrans multiloculare 
and secondary erythema migrans in central Wisconsin. 
Arch Dermatol 1993; 129: 709-16 
7. Asbrink E, Hovmark A. Early and late cutaneous 
manifestations in Ixodes-borne borreliosis ( erythema 
migrans borreliosis, Lyme borreliosis). Ann NY Acad 
Sci 1988; 539: 4-15 
8. Weber K, Neubert U, Buechner SA. Erythema 
migrans and early signs and symptoms. In: Aspects 
of Lyme Borreliosis. Weber K, Burgdorfer W (eds.) , 
Springer 1993; pp 105-121 
9. Goodman JL, Bradley JF, Ross AE et al. 
Bloodstream invasion in early Lyme disease: results 
from a prospective, controlled, blinded study using 
the polymerase chain reaction. Am J Med 1995; 
99: 6-12 
10. Marconi R, Garon CF. Development of 
polymerase chain reaction primer sets far diagnosis 
of Lyme disease and far species-specific identification 
of Lyme disease isolates by 16S rRNA signature 
nucleotide analysis. J Clin Microbiol 1992; 30: 
2830-34 
11. Baranton G, Postic D, Saint Girons I, et al. 
Delineation of Borrelia burgdorferi sensu stricto, 
Borrelia garinii sp. nov., and group VS461, associated 
with Lyme Borreliosis. Int J Syst Bacteriol 1992; 42: 
378-83 
12. Vaillant-Rieder L, Hey D. Atypischer Verlauf 
einer Erythema-migrans Krankheit. Dtsch Med 
Wochschr 1987; 112: 1680-81 
AUTHORS' ADDRESSES 
116 
Robert R. Mullegger, MD, Department of Dermatology, Karl-Franzens-University Graz 
Auenbruggerplatz 8, A-8036 Graz, Austria 
Gudrun Brunner-Kohler, MD, same address 
Natalie Zochling, MD, same address 
Harald Reiter, MD, same address 
Stefan Hodl, MD, professor of dermatology, same address 
H. Peter Soyer, MD, professor of dermatology, same address 
Helmut Kerl, MD, professor of dermatology, chairman, same address 
acta dennatovenerologica A .P.A. Vol 5, 96, No 3-4