close

Вход

Забыли?

вход по аккаунту

?

Lyme arthritis in european children and adolescents.

код для вставкиСкачать
ARTHRITIS & RHEUMATISM Volume 38
Number 3, March 1995, pp 361-368
0 1995, American College of Rheumatology
361
LYME ARTHRITIS IN EUROPEAN CHILDREN AND ADOLESCENTS
HANS-IKO HUPPERTZ, HELGE KARCH, HANS-JOACHIM SUSCHKE, EVA DORING,
GERD GANSER, ANGELIKA THON, WASSILIOS BENTAS,
and the PEDIATRIC RHEUMATOLOGY COLLABORATIVE GROUP
Objective. To evaluate and describe Lyme arthritis in European children and adolescents.
Methods. This was a prospective multicenter
study. The diagnosis of Lyme arthritis required the
exclusion of other diseases and positive findings on
serology for IgG antibodies to Borrelia burgdogen’.
Enzyme-linked immunosorbent assay, immunoblotting,
and polymerase chain reaction techniques to identify
infection by B burgdorferi were used.
Results. Among 62 children and adolescents with
Lyme arthritis, only 1 had a preceding erythema migrans. Arthritis was episodic in 62% and was chronic at
onset in 18%. The most common manifestation was
monarthritis of the knee. Joint involvement in patients
with oligoarthritiswas predominantly unilateral or symmetric. Arthralgia was very rare. Treatment with 1 or 2
courses of different antibiotics resulted in disappearance
of the arthritis in 77% of the patients.
Conclusion. The clinical presentation of Lyme
arthritis in children is different from that in adults. The
calculated incidence of Lyme arthritis in persons under
the age of 17 years (4/100,000) exceeds previous estimations.
Supported by Hoffmann-LaRoche, Grenzach-Wyhlen, Germany, and Stifterverband fur die deutsche Wissenschaft, Essen,
Germany
Hans-Iko Huppertz, MD: Children’s Hospital, University
of Wurzburg, Wiirzburg, Germany; Helge Karch, PhD: Institute of
Hygiene, University of Wurzburg, Wurzburg, Germany; HansJoachim Suschke, MD: Children’s Policlinic, University of Munich,
Munich, Germany; Eva Doring, MD: Children’s Hospital BerlinBuch, Berlin, Germany; Gerd Ganser, MD: Pediatric Department of
the Northwest German Center of Rheumatology, Sendenhorst,
Germany; Angelika Thon, MD: Children’s Hospital, Medical School
of Hannover, Hannover, Germany; Wassilios Bentas, cand.med.:
Children’s Hospital, University of Wurzburg; and members of the
Pediatric Rheumatology Collaborative Group (see text).
Address reprint requests to PD Dr. Hans-Iko Huppertz,
Children’s Hospital, University of Wurzburg, Josef-SchneiderStrasse 2, D-97080 Wurzburg, Germany.
Submitted for publication February 28, 1994; accepted in
revised form October 5, 1994.
Arthritis is a well-known manifestation of Lyme
borreliosis in North America and has been described
in detail in adults and in children (1-3). In contrast, in
European children, erythema migrans and lymphocytic meningitis frequently presenting as facial palsy
are major clinical manifestations of Lyme borreliosis
(43). Only a few European children with Lyme arthritis have been described (6-8).
The diagnosis of Lyme arthritis may be difficult.
The clinical manifestations are sometimes indistinguishable from other forms of oligoarthritis, and laboratory tests for antibodies against Borrelia burgdorferi, the causative agent, ;are sometimes unreliable
(4,9). Procedures used to isolate B burgdorferi from
body fluids are low yield, and it has been proposed that
polymerase chain reaction (PCR) techniques be used
to detect borrelial sequences in synovial fluid (10-12).
In the present study, we used enzyme-linked
immunosorbent assay (ELISA), in addition to clinical
criteria, immunoblot, and nested PCR techniques to
identify infection by B burgdorferi in children suspected of having Lyme arthritis. We describe the
presentation and course of Lyme arthritis in 62 European children who were entered into a prospective
multicenter study initiated iin 1991.
PATIENTS AND METHODS
Beginning July 1, 1991, one of us (H-IH) initiated a
multicenter study of Lyme arthritis in children and adolescents within the Pediatric R.heumatology Collaborative
Group, a body of approximately 100 German-speaking pediatric rheumatologists. *
* Members of the Pediatric Rheumatology Collaborative
Group (Arbeitsgemeinschaft Padiatrische Rheumatologie), their cities, and the number of patients entered into the present study are:
Eva Doring, MD, Monika Schontub’e, MD, Christiane Petrowskaya,
MD, and Barbara Liedtke, MD (B’erlin; 6 patients); Ulrike Seidel,
MD (Chemnitz; 1 patient); Tilman H:acker, MD, Barbara Tautz, MD,
and Sabine Thonig, MD (Dresden; 4 patients); Gerd Horneff, MD
and Volker Wahn, MD (DusseldonF; 1 patient); Ulrich Vetter, MD
HUPPERTZ ET AL
362
Patients with a diagnosis of Lyme arthritis were
eligible for study if they were 5 1 6 years old at the onset of
arthritis. Lyme arthritis was considered to be present if other
causes of arthritis (including juvenile rheumatoid arthritis,
juvenile spondylarthropathy, or infection-associated arthritis) were excluded by clinicall and/or serologic examination,
if serum IgG antibodies to B burgdorferi were detected by
ELISA, as described elsewhere (7), and if the specificity of
these antibodies could be confirmed by immunoblot (see
below).
The erythrocyte sedimentation rate (ESR), concentrations of C-reactive protein (CRP) and IgG, IgA, and IgM
(by nephelometry), HLA-B27, antinuclear antibodies
(ANA), and anti-streptolysinl 0 (ASO) were determined by
routine laboratory methods.
In addition to ELISA., serum was also analyzed by
immunoblot for IgG and IgM antibodies to B burgdorferi (in
the Wiirzburg laboratory) (7). Immunoblot strips were
screened for the presence of low molecular weight bands
(<30 kd), for bands at 31,34,39, and 41 kd, for bands in the
higher molecular weight range (45-85 kd), and for the 95-kd
band. All serum samples were also tested with recombinant
antigens of B burgdorferi (13) (obtained from Biologische
Arbeitsgemeinschaft, Lich, Germany). When available, synovial fluid was probed by nested PCR forfra gene-derived
borrelial sequences (10).
Patients were reexamined several times, including 3
months and 1 year after diagnosis, if appropriate. Arthritis
was defined as “episodic” if it was self-limiting, but recurred
after an interval of time, as “acute” if there was a single
episode of arthritis, and as “chronic” if it persisted for 3
months or longer.
The physicians participating in the study were free to
choose any antibiotic regimen; however, the recommended
treatment was intravenous ceftriaxone, 50 mgkg of body
weight, for 2 weeks. In the event of treatment failure after
several weeks to a few months, a 4-week course of oral
roxithromycin (150 mg twice a day) and cotrimoxazole (4 mg
of trimethoprim/kg of body weight) was recommended. In
patients older than 9 years, doxycycline, 200 mg/day for 4
weeks, was recommended if initial and secondary treatment
had not resulted in disappearance of arthritis.
The study was approved by the ethics committee of
(Frankfurt; 1 patient); Lothar !khuchmann, MD, Margit Speckmaier, MD, Andreas Reiff, MD, and Johannes Forster, MD
(Freiburg; 2 patients); Hartmut Michels, MD and Hans Truckenbrodt, MD (Garmisch-Partenkirchen; 1 patient); Angelika Thon,
MD, Frank Dressler, MD, and Horst von der Hardt, MD (Hannover; 4 patients); Alexander Scliulze-Berge, MD and Uwe Knoop,
MD (Koln; 1 patient); Wolfgang I’ritsch, MD and Georg Tretter, MD
(Kotzting; 3 patients); Jean-Pierre Deslarzes, MD (LeChable, Switzerland; 1 patient); Hans-Joachim Suschke, MD (Munich; 7 patients); Michael Frosch, MD (Minster; 2 patients); Jiirgen Quietzsch, MD (Plauen; 1 patient); Gottschalk Riegel, MD and Herwig
Rumpel, MD (Regensburg; 2 patients); Gerd Ganser, MD and
Willibald Breit, MD (Sendenhorst; 5 patients); Carsten Wurst, MD
and Uwe Marr, MD (Suhl; 1 patient); Ursula Scherer, MD (Tiibingen; 1 patient); Hans-Iko Huppertz, MD, Frank Gohlke, MD,
Jiirgen Pannenbecker, MD, Rfjdiger Krauspe, MD, Herrmann
Girschick, MD, Gerhard Hofmann, MD, and Wassilios Bentas,
cand.med. (Wiirzburg; 18 patients).
the Medical Faculty of the University of Wiirzburg. Informed consent was obtained from the patients’ parents and
from the adolescent patients themselves.
RESULTS
Of the 62 children and adolescents who met the
criteria for Lyme arthritis, 34 were boys and 28 were
girls. Their median age at the onset of arthritis was 11
years (range 3-16 years). Only 4 children were
younger than 6 years.
Two patients with positive results on immunoblot analyses for IgG antibodies to B burgdorferi were
not included among the 62 patients with Lyme arthritis
because they had clinical features of juvenile spondylarthropathy (including dactylitis), oligoarthritis, sacroiliitis, and a family history of psoriasis. The titers of
antibody to B burgdorferi declined after antibiotic
treatment of these 2 patients, but there was no improvement in the clinical presentation during the next
2 years. We concluded that the infection with B
burgdorferi was coincidental to the spondylarthropathy in these patients.
Incidence. An annual incidence of 4/100,000 was
deduced from the assumption that all cases of Lyme
arthritis in the endemic catchment area of about
200,000 persons under the age of 17 years presented to
the center in Wurzburg, Lower Frankonia.
Clinical presentation. Fifteen patients (24%) remembered one or more tick bites, but only 1 patient
had a preceding erythema migrans. The onset of
arthritis did not show a seasonal variation (Figure 1).
All patients experienced swelling and/or limitation of joint motion, but no pain. In 6 patients,
disappearance of arthritis after antibiotic therapy was
followed by the occurrence of arthralgia in the same
joints.
Among 60 patients, the course of arthritis was
I
7
vl
6
z5
k 2
5
4
3
2
5 1
1
+
c 6
a
E L
’c
0 3
n
c
J
F
M A M J J A S O
month o f onset of a r t h r l t l s
N
D
Figure 1. Month of onset of arthritis in 58 European children and
adolescents with Lyme arthritis.
363
PEDIATRIC LYME ARTHRITIS IN EUROPE
episodic in 37 (62%), acute in 12 (20%), and chronic
from the onset in 11 (18%). In 8 patients, the arthritis
became chronic after an initial episodic period lasting
1 month to 2 years; this increases the total percentage
of patients with chronic arthritis to 32% (19 of the 60).
In 11 of these 19 patients, arthritis persisted for 12
months or more.
Fifty-seven of 61 patients (93%) had limitation
of joint motion, and 61 of 62 patients had a joint
effusion. The only patient without a demonstrable joint
effusion had limited range of motion of the hip joint.
The numbers of affected joints are shown in Table 1.
The knee was by far the most frequently involved
joint. Both sides of the body were involved with equal
frequency.
In only 4 patients was the knee joint not involved: 1 had arthritis of the elbow, 2 had hip arthritis
(both possibly quite early in the course of borreliosis),
and 1 had episodic polyarthritis of the ankle, elbow,
wrist, and small finger joints, all on the right side.
Sixty-eight percent of the patients (42 of 62) had
monarthritis, affecting the knee in 41 patients and the
hip in 1 patient. Twenty-six percent (16 of 62) had
oligoarthritis and 6% (4 of 62) had polyarthritis. Three
of the 4 patients with polyarthritis had involvement of
5 or 6 large joints in a symmetric pattern. The pattern
of arthritis in patients with oligoarthritis involving only
the large joints was unique, being predominantly unilateral or symmetric (Figure 2).
The period from the onset of arthritis until the
establishment of the correct diagnosis “Lyme arthritis” was 5 days to 5 years (median 5 months).
Findings of laboratory investigations. Laboratory evidence of inflammation was found in 79% of
patients (49 of 62) at initial presentation. The ESR was
Table 1. Joints affected at the onset and during the course of Lyme
arthritis in 62 European children and adolescents*
Involvement during the
course of arthritis
Involved
joint
No. (%) of patients
with joint involved
at onset
* Monarthritis
No. (%) of
patients
of the knee occurred in 41 patients.
No. of patients
with bilateral
involvement
c
0
($9 0
c
c
0
0
Figure 2. Pattern of joint involvement in 16 children and adolescents with oligoarticular (2-4joints) Lyme arthritis. The distribution
of affected joints was symmetric or predominantly unilateral. Occasionally, patients with predominantly unilateral involvement had
an affected joint on the contralateral side; in those patients, the joint
affected on the contralateral side was always the same as the
affected joint on the ipsilateral side. A, Predominantly right-sided
involvement (6 children). B, Predominantly left-sided involvement
(5 children). C, Symmetric involvement (5 children). In addition, 4
patients had polyarthritis ( 2 5 joints), with either symmetric involvement of the large joints (3 patients) or right-sided involvement of the
large joints and the fingerjoints (I patient) (see text), and 42 patients
had monarticular arthritis.
elevated (>14 mm/hour) in 152%of the patients (37 of
60), with a median value of 35 md h o u r. Measurement
of serum CRP concentrations was not informative
(data not shown). Immunoglobulin levels were slightly
increased above the age-specific upper limit for IgG in
30% (15 of 50), for IgA in 36% (18 of 50), and for IgM
in 39% (20 of 51) of the patients. Ten of the 23 patients
who had a normal ESR had elevated concentrations of
IgG (4 patients), IgA (7 patilents), or IgM (7 patients).
ANA were found in low titers in 7 of 54 patients
(13%), 6 of whom were older than 8 years. The seventh
patient was a 3-year-old girl who had episodic oligoarthritis which disappeared after antibiotic therapy.
None of these children had antibodies to doublestranded DNA. Early-onset pauciarticular juvenile
rheumatoid arthritis and systemic lupus erythematosus were excluded. Rheumatoid factor was not found
in any of the 62 patients.
HUPPERTZ ET AL
364
Four of the 41 patients tested (10%) were HLAB27 positive, but none of them had a family history of
spondylarthropathies. Three of them had episodic
arthritis involving the knee and/or the elbow joints. All
4 patients presented witlh additional manifestations
compatible with Lyme arthritis: 2 developed noninflammatory pain for several months in the joints previously affected with Lyrne arthritis (after antibiotic
treatment and subsequent disappearance of arthritis),
1 experienced a Herxheimer reaction during antibiotic
treatment, and 1 developed keratitis after antibiotic
treatment and disappearance of his arthritis. Spondylarthropathies were excluded in all 4 patients.
Radiographic findings. Radiographs of the most
seriously affected joint were obtained at the request of
the attending physician at the time of the first presentation. The findings were either unremarkable or there
was effusion and soft tissue edema. Gadoliniumenhanced magnetic resonance imaging of the knee was
performed in 2 patients, and showed effusion and
intense synovial enhancement. After 5 years of mild
episodic arthritis, no pannus and no anatomic changes
were found in 1 of these 2: patients.
In the second patient, after 3 years of episodic
arthritis and 2 years of chronic arthritis, a subchondral
defect, but no pannus, was found. Radiographs, which
had been unremarkable :! years before, showed an
uneven femoral surface and 2 small subchondral cysts
at the same location.
Results of synovial h i d analysis. Synovial fluid
analysis, performed in 16 patients, showed a median of
12,000 cells/pl (range 90&53,000). Eight of the 12
patients in whom these cells were further characterized had a preponderance of polymorphonuclear cells.
One patient in whom 2 consecutive synovial fluid
specimens were analyzed showed 53,000 ceWpl(7 1%
polymorphonuclear cells) during the initial evaluation
and 9,600 cellslpl (7% polymorphonuclear cells) after
several months of chronic arthritis.
PCR for borrelial sequences was positive in
only 1 of the 14 synovial fluid specimens tested. The
patient was a 9-year-old g,irl who was experiencing a
second episode of monarthritis of the knee after an
interval of 8 weeks without arthritis. The joint showed
a massive effusion and marked heat. Her ESR was 115
mm/hour, CRP was 110 ,gm/liter, and all Ig classes
were elevated. Analysis of synovial fluid revealed
47,000 cells/pl (92% polyimorphonuclear cells). Routine cultures remained sterile, but PCR for borrelial
sequences yielded positive results. Two months after
treatment with ceftriaxone, the arthritis had disap-
Table 2. Results of immunoblot analysis for IgG antibodies to
Borreliu burgdorferi in 60 European children and adolescents with
Lyme arthritis*
Antigen tested
Low molecular weight
proteins (<30 kd)t
31-kd protein
34-kd protein
39/41-kd protein
High molecular weight
proteins (45-85 kd)+
95/100-kd protein§
Recombinant protein 41/i
Recombinant OspC
Recombinant OspAT
Recombinant protein 41
Recombinant protein loo§
No. of
patients
positive
% of
patients
positive
59
98
47
49
60
60
78
82
100
100
32
58
36
7
60
59
53
97
60
12
100
98
* Two children with transient arthritis were initially seronegative for
antibodies to B burgdorferi and are not included. One of them had
presented with hip arthritis 6 weeks after a tick bite, but seroconverted 3 months later (bands at 19,23,39, and 41 kd and several high
molecular weight bands); the other had facial palsy and lymphocytic
meningitis, and remained without antibiotic treatment for 6 weeks,
when hip arthritis developed. At that time, she had seroconverted
(bands at 19, 39, and 41 kd and several high molecular weight
bands).
t All except 1 serum had at least 1 band (median 4 bands, range 0-8).
Those with more than 6 bands also had bands at <10 kd.
Patients had a median of 5 bands (range 2-8).
8 These proteins associated with late Lyme borreliosis (see ref. 13)
were not always distinguishable in our gel system, but were easily
identified with the recombinant antigen immunoblot. All except 1
serum were positive.
Ti The low frequency of antibodies to recombinant outer surface
protein A (OspA), derived from the tick isolate PKO, has been
found by others investigating European Lyme borreliosis (see ref.
13). We do not know whether the bands frequently seen at 31 and 34
kd in our immunoblot system (using a local cerebrospinal fluid
isolate W1) represent non-cross-reacting variants of these proteins.
+
peared and there was no recurrence during the ensuing
24 months.
Serologic findings. All except 2 patients (97%)
had IgG antibodies to B burgdorferi by ELISA, often
at very high optical densities. The remaining 2 patients
had hip arthritis, perhaps during early Lyme borreliosis, and seroconverted during the ensuing weeks (Table 2). The other 60 patients had 6 or more bands at the
first evaluation of immunoblots for IgG antibodies to
B burgdorferi (Table 2). The specificity of these antibodies was confirmed by testing with recombinant
antigens (Table 2).
Presence of other infectious agents. Serologic
screening for other infectious agents revealed elevated
antibody titers in 18 patients: AS0 in 9 patients,
365
PEDIATRIC LYME ARTHRITIS IN EUROPE
Table 3. Antibiotic treatment of 62 European children and adolescents with Lyme arthritis
Antibiotic
Ceftriaxone
Roxithrom ycin/cotrimoxazole
Doxycycline
Oral penicillin
Penicillin G
Erythromycin
Amoxicillin
Other cephalosporins
Clindam ycin
Imipenem
No. of
courses of
treatment*
53
14
11
8
7
4
4
3
2
1
No. of patients
in whom
antibiotic
appeared to be
effectivet
36
8
3
1
1
0
2
0
1
0
* Except for 2 patients who received ceftriaxone twice, all received
the antibiotic listed only once.
t Effective was defined as disappearance of arthritis. Ten patients
still have arthritis (1 of whom did not receive antibiotics). Of the 52
patients free of arthritis 69% had taken ceftriaxone, and 15% had
taken roxithromycinlcotrimoxazole.
Yersinia (IgA antibodies by immunoblot) in 5 patients,
Salmonella in 2 patients, and Chlamydia and Mycoplasma in 1 patient each. Throat swabs and stool
cultures were negative in all patients with elevated
titers. None of the patients had clinical features of
acute rheumatic fever or enteropathogenic reactive
arthritis. The clinical and serologic data for these 18
patients did not differ from those for the other patients
with Lyme arthritis.
Sixteen of these 18 patients responded to antibiotic therapy. The patient with Chlamydia antibodies
and 1 of the patients with Salmonella antibodies had
ongoing arthritis despite antibiotic treatment, but subsequently lost these antibodies in the continued presence of borrelial antibodies. For these reasons, none
of these 18 patients were excluded from analysis.
However, simultaneous infection with B burgdorferi
and another arthritogenic agent cannot be excluded in
these patients.
Antibiotic therapy. Antibiotic therapy was
started immediately in 59 patients. In 3 cases, the
patients’ parents initially refused antibiotic treatment,
but 2 of them accepted treatment 3 months and 6
months later. Details of antibiotic therapy are given in
Table 3.
Forty-five percent of the patients (28 of 62)
received 2-6 antibiotics consecutively. Arthritis disappeared after 1 or 2 courses of antibiotics in 77% of the
patients (47 of 61). In 4 of 6 patients who took 3
courses of antibiotics and in 2 of 3 patients who took 4
courses, this treatment resulted in disappearance of
arthritis. Two patients who took 5 and 6 courses of
antibiotics have ongoing artlhritis.
Probable Herxheimer reactions to antibiotic
treatment were noted in 9 patients, with fever, arthralgia, myalgia, nonpruritic erythematous eruptions, acrocyanosis, and elevated arterial blood pressure.
These reactions started sometimes with the first encounter of the drug, during or immediately after the
infusion of ceftriaxone, and lasted for several hours.
This pattern was difficult to dlistinguish from an allergic
reaction and resulted in interruption of therapy in 4
patients. However, in those patients in whom treatment was continued, the exlent of the reaction diminished or disappeared during the next few antibiotic
infusions. During therapy with ceftriaxone, 1 girl
showed true drug hypersensitivity with a pruritic rash.
Other treatments. Five patients, 3 of whom
have ongoing chronic arthritis, received intraarticular
steroids before antibiotic therapy. Ten patients, 4 of
whom have recalcitrant arthritis, received intraarticular steroids after unsuccessful antibiotic treatment.
Six patients had a synovectomy prior to antibiotic therapy; all had ongoing arthritis after synovectorny. Five patients underwent synovectomy after
antibiotic treatment, and in 1 of them, the arthritis
disappeared. In 2 of these 5 patients, arthritis disappeared after further antibiotic therapy; 2 patients have
ongoing arthritis.
Outcome. At the last followup, arthritis had
resolved in 49 patients and had not recurred a median
of 4 months later (range 14!4 months). In a further 3
patients, the exact time of diisappearance of the arthritis was not certain. Ten children and adolescents have
ongoing arthritis a median of 10 months (range 4-24
months) after the first antibiotic treatment (Table 4). In
3 of the nonresponders, the followup period has been
short, and the arthritis might ultimately resolve. There
was no obvious difference in the history and presentation of the nonresponders compared with the responders to antibiotic therapy.
DISCUSlSION
Although some aspects of Lyme arthritis in
European children and adolescents are similar to what
has been found in North American children (14) or in
European adults with Lyrne arthritis (15), several
distinctions should be emphasized. Intermittent migrating pain has been frequlently described in Ameri-
366
HUPPERTZ ET AL
Table 4. Clinical features of 10 European children and adolescents
with refractory Lyme arthritis
Disease duration
(months)
No. of
Since
failed
first
antibiotic
antibiotic treatments
Patient
Before
first
antibiotic
1
17
24
4
2
3
16
3
3
2
13
3
6
10
9
12
12
10*
9
4t
3
3
4t
4t
Other therapies
~
2
I
9
10
~
Synovectomy
(before antibiotic)
In traarticular
steroids (before
antibiotic)
Intraarticular
steroids (before
antibiotic)
Synovectomy
Synovectomy
Intraarticular
steroids (before
antibiotic)
2
2
* No. of months since establishment of the diagnosis; patient’s
mother has refused antibiotic treatment.
t Because of the relatively short followup, it is possible that the
arthritis will disappear later.
can and European adults with Lyme arthritis (1,16),
but was not found in the children with Lyme arthritis
in this study. Six of our patients developed arthralgia
only after the disappearance of arthritis. The lower
frequency of arthralgia in children reported in this
study has not previously been noted, although the
available data in American children permit the conclusion that arthralgia might be rare in American children
with Lyme arthritis (2,3,17).
It has been reported that arthralgia in European
adults most frequently o1;curs in the shoulder joint
(15). In one study of American children, the shoulder
joint was frequently affected with arthritis (3), but
another study showed this occurred very rarely (17),
as was true in the patients of the present study. While
the elbow joint was the second most frequently affected joint among patients in this study, and the third
most frequent in Americ,an children (3), the occurrence of arthritis in this joint is rarely found in European adults with Lyme arthritis (15).
Similar to findings of other studies, monarthritis
of the knee and oligoarticular joint involvement that
included the knee were frequently found in this series
of patients with Lyme arthritis. The predominantly
unilateral or symmetric paittern of oligoarthritis found
in this study has not been reported before, however
(15,18). We conclude that the clinical features of
infection with B burgdorferi are unique in children and
can be distinguished from the disease in adults. The
differences in the strains of Borrelia burgdorferi sensu
lato circulating in the USA and in Europe (19) may
account for the varying clinical presentation of American and European children who have Lyme arthritis;
however, an attempt to explain the differences of the
disease in adults and children requires an age-specific
host factor.
In European adult patients who have Lyme
arthritis, dactylitis, heel pain, and possibly sacroiliitis
have been described (15), and it has been claimed that
B burgdorferi may be another infectious agent responsible for inducing reactive arthritis (20). In the present
study, we excluded 2 patients with juvenile spondylarthropathies and IgG antibodies to B burgdorferi in
whom the clinical presentation and further course
differed from those of the other patients. With extended experience, the role of B burgdorferi in these
patients might be elucidated.
Juvenile spondylarthropathy should be considered in the differential diagnosis of Lyme arthritis. Age
at onset in our patients with Lyme arthritis was similar
to that in American children (3) and resembled the age
at onset of the juvenile spondylarthropathies. The
presence of HLA-B27 in our patients did not exceed
the frequency in the general central European population. The hip joint, which is often involved in juvenile
spondylarthropathy , was less frequently affected in
the children in our study as well as in American
children described elsewhere (3); enthesopathy was
not observed. The episodic course of Lyme arthritis
and the pattern of joint involvement with the predominance of monarthritis of the knee or predominantly
unilateral or symmetric oligoarthritis was different
from the asymmetric oligoarthritis seen in patients
with juvenile spondylarthropathies. While Lyme arthritis and juvenile spondylarthropathies show a distinct clinical presentation, it may be difficult to achieve
a correct diagnosis in the individual patient.
Erythema migrans, the characteristic skin rash
of Lyme borreliosis, was found in only 1 patient in this
study, and was not a reliable diagnostic sign. Although
investigators in the US initially described a high percentage of patients with Lyme arthritis who had a
preceding erythema migrans (2,3,14), currently, patients with erythema migrans are treated early and
successfully with antibiotics and probably do not
367
PEDIATRIC LYME ARTHRITIS IN EUROPE
progress to develop arthritis (Steere AC: personal
communication).
The observed radiographic changes, including
subchondral cysts and cartilage erosion in 1 patient,
are consistent with previous reports (21) and show that
chronic Lyme arthritis may lead to joint damage
similar to that found in patients with juvenile rheumatoid arthritis.
In patients with acute monarthritis, synovial
fluid analysis is mandatory to exclude septic arthritis.
PCR for borrelial sequences in synovial fluid has been
advocated to improve diagnosis; however, its sensitivity in this study was very low and inferior to results
obtained in the US (11,12). This difference might be
due to the low volume of synovial fluid used (500 pl),
to the use of primers derived from chromosomalflu
sequences instead of plasmid-coded osp sequences, or
to the use of only 1 primer pair instead of up to 4 pairs.
Serology is still the laboratory tool most frequently used to confirm a clinical diagnosis of Lyme
arthritis. All patients with Lyme arthritis in this study
had highly positive results by immunoblot analysis,
with at least 6 specific bands including the 100-kd
protein. This finding is consistent with results from the
US (22). Numerous children who presented to us
because of suspected Lyme arthritis and results on
commercial antibody tests that were positive for an
infection with B burgdorferi were shown not to have
Lyme arthritis. As determined in our laboratory, the
ELISA results were negative or showed only low
titers, and the findings of immunoblot analyses were
negative. Followup revealed other diagnoses. In our
experience, therefore, positive results on ELISA
should always be confirmed by positive results on
immunoblot analysis when support for a diagnosis of
Lyme arthritis is being sought.
It was not our intention that this study would
investigate different treatment modalities; however,
several valuable features emerged. Ceftriaxone was a
safe and effective agent. Since it is infused only once a
day, it can be administered on an ambulatory basis,
and the problems of possibly declining compliance
over several weeks of oral treatment, as recommended
for doxycycline and amoxicillin (23), do not occur.
Although response to antibiotic therapy may take
several weeks, the possibility of further antibiotic
treatment should be anticipated. Intraarticular steroids
and synovectomy are not effective before antibiotic
treatment and may even blunt the response to antibiotics.
We propose that the incidence of Lyme arthritis
in Europe has been underestimated. The incidence of
pediatric neuroborreliosis in a region of Lower Saxony, Germany, has been reported to be 5.8/100,000 (9,
and the incidence of pediatric Lyme arthritis deduced
from this study is 4/100,000~
In conclusion, this study has shown that the
clinical presentation of Lyme arthritis in children is
distinct from the adult disease by the low frequency of
pain and the predominantly unilateral or symmetric
pattern of oligoarticular disease. Lyme arthritis is not
rare in Europe. All childiren and adolescents with
arthritis of unknown origin should be evaluated for the
presence of Lyme arthritis, both clinically and serologically, including immunoblot analysis of IgG antibodies to Borrelia burgdorj’ki.
ACKNOWLEDGMENTS
The authors are very grateful to Olga Bohler for
excellent technical assistance, and to Prof. H. Bartels
(Wurzburg), Dr. F. Dressler (IHannover), Dr. M . A. Noble,
and especially to Dr. R. E. Petty (both of Vancouver, British
Columbia) for reading the manuscript.
REFERIENCES
1. Steere AC: Lyme disease. N Engl J Med 321:58&596, 1989
2. Eichenfield AH, Goldsmith DP, Benach JL, Ross AH, Loeb
FX, Doughty RA, Athreya BH: Childhood Lyme arthritis:
experience in an epidemic area. J Pediatr 109:753-758, 1986
3. Williams CL, Strobino B, L,ee A, Curran AS, Benach JL,
Inamdar S, Christofaro R: Lyme disease in childhood: clinical
and epidemiologic features of ninety cases. Pediatr Infect Dis J
9:lO-14, 1990
4. Huppertz HI: Childhood Lyme borreliosis in Europe. Eur J
Pediatr 149:814-821, 1990
5. Christen HJ, Hanefeld F, Eiffert H, Thomssen R: Epidemiology
and clinical manifestations of Lyme borreliosis in childhood.
Acta Paediatr 82 (suppl 386):1-76, 1993
6. Schaad UB, Fliieler U, Schaulb H, Suter H, Vischer D, Caflisch
U, Tschumi A, Wick H , Vest M, Durrer D: Durch Ixodes
ricinus-Spirochaten (Borrelia burgdorferi) verursachte Krankheitsbilder (Lyme-Krankheit) bei padiatrischen Patienten in der
Schweiz. Schweiz Med Wochenschr 118:422426,1986
7. Huppertz HI, Karch H: Die Lyme-Arthritis im Kindesalter:
Monarthritis des Kniegelenkes, klinisch nicht unterscheidbar
von Monarthritiden unbekannter Ursache. Monatsschr Kinderheilkd 139:759-764, 1991
8. Hammers-Berggren S, Andersson U, Stiernstedt G: Borrelia
arthritis in Swedish children: clinical manifestations in 10 children. Acta Paediatr 81:921-924, 1992
9. Steere AC, Taylor E, McHugh GL, Logigian EL: The overdiagnosis of Lyme disease. JAMA 269:1812-1816, 1993
10. Karch H,Huppertz HI: Repeated detection of Borrelia burgdorferi DNA in synovial fluid of a child with Lyme arthritis.
Rheumatol Int 12:227-230, 1993
11. Liebling MR, Nishio MJ, Rodlriguez A, Sigal LH, Jin T, Louie
JS: The polymerase chain reaction for the detection of Borrelia
burgdorferi in human body fluids. Arthritis Rheum 36:665475,
1993
368
12. Nocton JJ, Dressler F, Rutledge BJ, Rys PN, Persing DH,
Steere AC: Detection of Balrrelia burgdorferi DNA by polymerase chain reaction in syncivial fluid from patients with Lyme
arthritis. N Engl J Med 330:229-234, 1994
13. Wilske B, Fingerle V , Herzer P, Hofmann A, Lehnert G, Peters
H, Pfister HW, Preac-Mursic V, Soutschek E, Weber K:
Recombinant immunoblot in the serodiagnosis of Lyme borreliosis. Med Microbiol Immunol 182:255-270, 1993
14. Szer IS, Taylor E, Steere AC: The long-term course of Lyme
arthritis in children. N Engl .I Med 325:159-163, 1991
15. Herzer P: Joint manifestations, Aspects of Lyme Borreliosis.
Edited by K Weber, W Burgdorfer. Berlin, Springer-Verlag,
1993
16. Weber K , Neubert U: Clinical features of early erythema
migrans disease and related disorders. Zentralbl Bakteriol Hyg
[A] 263:209-228, 1986
17. Culp RW, Eichenfield AH, Davidson RS, Drummond DS,
Christofersen MR, Goldsmith DP: Lyme arthritis in children. J
Bone Joint Surg [Am] 69:9&99, 1987
HUPPERTZ ET AL
18. Valesova M, Trnavsky K: Joint manifestations of Lyme borreliosis in Czechoslovakian patients. Z Rheumatol 49: 192-196,
1990
19. Baranton G, Postic D, Saint Girons I, Boerlin P, Piffaretti JC,
Assous M, Grimont P: Delineation of Borrelia burgdorferi sensu
stricto, Borrelia garinii sp. nov., and group VS461 associated
with Lyme borreliosis. Int J Syst Bacteriol 42:378-383, 1992
20. Weyand CM, Goronzy JJ: Immune responses to Borrelia burgdorferi in patients with reactive arthritis. Arthritis Rheum
32: 1057-1064, 1989
21. Lawson JP, Steere AC: Lyme arthritis: radiologic findings.
Radiology 154:37-43, 1985
22. Dressler F, Whalen JA, Reinhardt BN, Steere AC: Western
blotting in the serodiagnosis of Lyme disease. J Infect Dis
167:392400, 1993
23. Steere AC, Levin RE, Molloy PJ, Kalish RA, Abraham JH, Lin
NY, Schmid CH: Treatment of Lyme arthritis. Arthritis Rheum
372378488, 1994
Документ
Категория
Без категории
Просмотров
0
Размер файла
756 Кб
Теги
lyme, adolescenta, arthritis, children, european
1/--страниц
Пожаловаться на содержимое документа