Original Article
The Long-Term Course of Lyme Arthritis in Children
N Engl J Med 1991; 325:159-163July 18, 1991
- Abstract
- Abstract
Background and Methods.
The natural history of Lyme disease is not completely known. We studied the long-term course of Lyme arthritis in 46 children in whom the onset of the disease occurred between 1976 and 1979 and who received no antibiotic therapy for at least the first four years of the illness.
Results.
Of the 46 children (age range, 2 to 15 years), 33 (72 percent) initially had erythema migrans, 7 (15 percent) had influenza-like symptoms, and 6 (13 percent) had migratory joint pain. These manifestations were followed by brief attacks of arthritis, particularly affecting the knee. The percentage of children with recurrent episodes of arthritis declined each year. By year 4, only 10 children still had a mean of two episodes of arthritis per year; the duration of arthritis was generally longer in older children (P<0.05). During the sixth year of illness, two children (4 percent) had keratitis, and more than 10 years after the onset of disease, a subtle encephalopathy developed in two other children. Of the 39 children whom we were able to contact in 1988–1989, 12 (31 percent) still had occasional brief episodes of joint pain and 1 (3 percent) had marked fatigue. All 46 children had positive IgG antibody responses to Borrelia burgdorferi throughout the illness and on long-term follow-up. As compared with those who became asymptomatic, the children with recurrent symptoms more often had IgM responses to the spirochete and had significantly higher IgG titers (P<0.05).
Conclusions.
The course of initially untreated Lyme disease in children may include acute infection followed by attacks of arthritis and then by keratitis, subtle joint pain, or chronic encephalopathy. (N Engl J Med 1991; 325:159–63.)
Media in This Article
Figure 1
Percentage of Children with Recurrent Episodes of Lyme Arthritis and the Mean Number of Episodes per Child per Year.Figure 2
Geometric Mean Titers of IgM and IgG Antibody to B. burgdorferi in Children with Recurrent Lyme Arthritis during Each Year of Active Illness and on Long-Term Follow-up, 10 to 13 Years after the Onset of Disease.Article Activity
- Article
LYME disease, a tick-borne spirochetal infection, J typically begins with a characteristic skin lesion, erythema migrans, and is often followed by systemic manifestations involving the heart, nervous system, or joints.1 The illness was recognized as a separate entity in 1975 because of the geographic clustering of children in Lyme, Connecticut, who were thought to have juvenile rheumatoid arthritis.2 Joint involvement in these children was characterized by brief episodes of pauciarticular swelling and pain primarily affecting large joints, most frequently the knees, over a period of several years. During the late 1970s, before the efficacy of antibiotic therapy in Lyme disease was known, both children and adults with erythema migrans or Lyme arthritis were entered in prospective studies to determine the natural history of the illness. We have previously reported the clinical evolution of Lyme arthritis in this cohort of patients.3
In the first half of the 1980s, antibiotics were found to be generally effective for the treatment of both early and late manifestations of Lyme disease.4 5 6 7 During this period, the few patients enrolled in the prospective studies in the 1970s who still had symptoms or in whom symptoms developed during the 1980s were treated with antibiotic therapy. In the majority of patients, however, the illness had apparently resolved, and antibiotics were not prescribed. It has recently become apparent, however, that late neuroborreliosis may begin long after the initial disease, following years of latent infection.8 In the current study, our goals were to describe the clinical course of Lyme arthritis in the children who did not receive antibiotic therapy during the initial illness and to determine whether any of these children still had symptoms 10 to 13 years later that might be attributed to this disorder.
Methods
Prospective Study of Lyme Arthritis
From 1976 to 1979, 58 children (ages, 15 years or less) with Lyme arthritis were entered in a prospective study at the Yale University School of Medicine to determine the natural history of the illness. Lyme arthritis was defined as brief, recurrent attacks of pauciarticular joint swelling, not due to another cause, in a patient from an area in which the disease was endemic.3 (Subsequently, it was shown that all the children had positive IgG antibody responses to Borrelia burgdorferi.) Of the 58 children, 8 did not continue treatment at Yale or were lost to follow-up and 4 received antibiotics for other reasons. The remaining 46 children were followed in the clinic for one to eight years, throughout the period of active arthritis. During the 1970s, the 46 children were treated with nonsteroidal anti-inflammatory drugs, and 4 received intraarticular steroid injections; none received antibiotic therapy for at least the first four years of illness. From 1980 to 1983, the four children who still had arthritis were treated with parenteral penicillin.6 After the children were discharged from the clinic, their parents were asked to call if their children had recurrent episodes of arthritis or other unexplained symptoms. Thus, the children who had subsequent eye involvement or neurologic symptoms were reevaluated and treated by one of us. Yearly telephone contact was maintained with all patients for at least three years after the last episode of arthritis.
Long-Term Follow-up
In 1988–1989, we were able to contact 39 of the 46 children (85 percent) for long-term follow-up (range, 10 to 13 years). They were asked if they had had arthritis, neurologic symptoms, other illnesses, or antibiotic therapy since their last contact with us. Of the 17 patients who had had symptoms after the period of arthritis, 10 were evaluated by us at New England Medical Center in Boston. All patients, regardless of symptoms, were asked to send a serum sample so that their antibody response to B. burgdorferi could be determined.
Serologic Testing
The serial serum specimens, obtained at about yearly intervals during the prospective study, and the long-term follow-up samples were tested for antibody to B. burgdorferi with an enzyme-linked immunosorbent assay, as previously described.9 Positive responses were defined as a titer of more than 1:100 for IgM and more than 1:400 for IgG; these titers were 3 SD above the mean values of eight normal control serum samples included on the same plate. All samples from a given patient were tested on a single plate to assess the change in titer. In patients with late neurologic involvement, the antibody response was determined in serum and cerebrospinal fluid by capture enzyme immunoassay, as previously described.10
Statistical Analysis
Correlation between matched pairs was determined by linear regression analysis, and differences among means were determined by one-way analysis of variance. All P values were two-tailed.
Results
Early Infection
Of the 46 children, 33 (72 percent) had erythema migrans at the beginning of the illness, often accompanied by intermittent headache and stiff neck, fever, lethargy, myalgia, and anorexia (Table 1Table 1
Clinical Course of Lyme Arthritis in 46 Children.*). Twenty-four of the children (52 percent) remembered a tick bite at the site of the skin lesion. In seven children (15 percent), the initial manifestations of the disease consisted of influenza-like symptoms without erythema migrans. Within the first weeks of illness, all but 1 of the 46 children also had intermittent joint pains without objective signs of arthritis, and in 6 (13 percent), arthralgias were the presenting symptom of the illness. The arthralgias, which usually affected several large joints, lasted from a few hours to several days and were interspersed with pain-free days or weeks. During this period, 10 children (22 percent) had aseptic meningitis or facial palsy and 4 (9 percent) had atrioventricular nodal block that usually lasted for several weeks.Arthritis
A median of 3.4 months (range, 2 weeks to 2 years) after the onset of disease, the 46 children began to have brief episodes of frank arthritis (Table 1). Characteristically, one or two large joints suddenly became swollen and warm, but they were often only mildly painful and not erythematous. The joint most frequently affected was the knee (98 percent), which became extremely swollen, followed by the elbow, ankle, hip, and wrist. Neither synovial thickening nor irreducible contractures were noted. Superficial bursae and tendons were not affected. During one or more of the episodes of arthritis, 78 percent of the children had diffuse or regional lymphadenopathy and 40 percent had symptoms reminiscent of the early illness, sometimes including recurrent erythema migrans. The pattern of joint involvement was similar regardless of the initial manifestations of the illness.
During the first year of arthritis, the 46 children had a mean of 5 episodes of joint swelling (range, 1 to 15) (Fig. 1Figure 1
Percentage of Children with Recurrent Episodes of Lyme Arthritis and the Mean Number of Episodes per Child per Year.). Most flares resolved within three weeks, and the majority lasted less than seven days. During the next several years, the number of children who still had bouts of arthritis decreased each year. Although the frequency of attacks diminished after the first year of illness, the duration of each episode of arthritis during the second and third years often increased, so that episodes lasted weeks or months. By year 4, only 10 children had a mean of two episodes of arthritis per year. There was a direct correlation between the age at the onset of infection and the total duration of arthritis (r = 0.3, P<0.05). Among the youngest children (age, 2 to 4 years), the median total duration of arthritis was only 4 weeks, whereas in teenagers (age, 13 to 15 years), it was 22 weeks. In two boys, who were 8 and 10 years old at the onset of disease, joint inflammation lasted more than one year. Both had the HLA-DR4 allele; neither had the HLA-B27 allele. By the sixth year, five children (11 percent) still had recurrent arthritis, but only one child continued to have short attacks lasting several days during the seventh and eighth years (Fig. 1).During the first year of arthritis, all 46 children had elevated IgG and often IgM antibody responses to B. burgdorferi (Fig. 2Figure 2
Geometric Mean Titers of IgM and IgG Antibody to B. burgdorferi in Children with Recurrent Lyme Arthritis during Each Year of Active Illness and on Long-Term Follow-up, 10 to 13 Years after the Onset of Disease.). By the second and third years, the IgM titers were generally lower, whereas the IgG titers were higher. Thereafter, the IgG responses tended to decline, but remained positive throughout the illness.Eye Involvement
Keratitis developed six years after the onset of disease in two patients, in one instance while arthritis was still present and in the other three years after it had resolved (Table 1). One child had acute keratitis, which resolved spontaneously within several weeks. The other patient was found to have corneal opacities during a routine eye examination. The lesions did not change during a one-month course of oral doxycycline, but resolved after treatment with topical steroids. During the next three years, the patient had several episodes of photophobia with conjunctival hyperemia, which resolved quickly with topical-steroid therapy.
Late Neurologic Involvement
Two children had a subtle encephalopathy with memory impairment, headache, and fatigue 11 and 12 years after the onset of disease, and 7 and 9 years after their arthritis had resolved (Table 1). Evaluation showed evidence of intrathecal production of antibody to B. burgdorferi in both and increased protein levels in cerebrospinal fluid in one (Table 2Table 2
Clinical and Laboratory Findings in Four Patients with Neurologic Syndromes after Lyme Arthritis.). Both children were treated successfully with intravenous ceftriaxone. Neither child had had neurologic abnormalities early in the course of the illness.Neurologic syndromes developed in two additional children after the resolution of their arthritis. One had a complex seizure disorder, and the other had apparent demyelinating disease (Table 2). Although both had remarkably high serum titers of IgG antibody to B. burgdorferi, selective concentration of specific antibody in cerebrospinal fluid could not be demonstrated. The child with apparent demyelinating disease, a 12-year-old girl, had a brief episode of transverse myelitis followed during the subsequent seven-year period by one bout of transient vestibular neuronitis and two attacks of optic neuritis. Analysis of cerebrospinal fluid during each episode showed a mild pleocytosis and oligoclonal bands, but not myelin basic protein. Visual or auditory evoked potentials and the CT scan of the head were normal. Neither child seemed to benefit from repeated courses of intravenous antibiotic therapy. Since we could not determine the cause of the neurologic syndromes in these two children with certainty, we excluded them from the analysis of the group with late neuroborreliosis.
Late Arthralgias or Fatigue
Many of the patients had several brief attacks of joint pain after the episodes of arthritis had resolved. However, of the 39 children whom we were able to contact in 1988–1989, 12 (31 percent) still had occasional episodes of joint pain and 1 (3 percent) had marked fatigue 10 to 13 years after the onset of disease. In a pattern similar to that observed early in the course of infection, the patients had brief periods of joint pain, usually in one joint at a time, especially the knee, lasting hours or days, followed by pain-free periods that were sometimes as long as several years. In most instances, the patients could not identify events that triggered the pain, such as exercise. Except for two girls, one of whom had findings characteristic of the patellofemoral syndrome and the other of whom had hypermobile joints, the patients had normal joint examinations. Five of the 12 patients received courses of oral antibiotics for their joint symptoms, but they did not have a clear response to therapy. One patient had only fatigue and a high IgG antibody titer to B. burgdorferi (1:12,800), but he declined further evaluation.
Correlation of Late Symptoms and Serologic Findings
Serum samples were obtained from 16 of the 17 patients who had keratitis, encephalopathy, arthralgias, or fatigue after the period of arthritis and from 20 of the 22 patients who had no symptoms after the arthritis had resolved. All 36 children tested were seropositive (Fig. 3Figure 3
Highest Titers of IgM and IgG Antibody to B. burgdorferi during Active Lyme Arthritis and the Titer on Long-Term Follow-up, a Mean of 10 Years after the Resolution of Arthritis.). Detectable IgM responses to B. burgdorferi were found on long-term follow-up in 6 of the 13 patients with arthralgias or fatigue (46 percent), in 2 of the 4 patients with keratitis or encephalopathy (50 percent), but in only 3 of the 20 asymptomatic children (15 percent) (P<0.05). As compared with the asymptomatic group, the patients with keratitis or encephalopathy tended to have higher IgG titers, and those with arthralgias or fatigue had significantly higher IgG titers (P<0.05). When the highest IgG titer measured during active arthritis was compared with the response obtained on long-term follow-up, the mean titer had declined twofold in the patients with arthralgias or fatigue, fourfold in those with keratitis or encephalopathy, and eightfold in asymptomatic children (P<0.05). Ten of the 17 symptomatic patients (59 percent) received antibiotic therapy after the first four years of illness, as did 10 of the 20 asymptomatic patients (50 percent), but the titers were similar in treated patients and untreated patients.Discussion
In this study, which was begun before the use of antibiotic therapy for Lyme disease, we attempted to trace the natural history of the disorder in 46 children. The study is biased toward patients with more severe disease because joint involvement, a manifestation of systemic illness, was the criterion for study entry. On the other hand, the use of antibiotic therapy later in the illness in one third of the patients may have reduced the frequency of late manifestations.
As has been reported in earlier studies of children and adults,2 , 3 , 11 12 13 the illness in the children in this study often began with erythema migrans and systemic symptoms, followed within a few weeks by early neurologic or cardiac abnormalities or by migratory joint pain. Within months, the children began to have brief, recurrent episodes of arthritis characterized by huge knee effusions. The arthritis usually had a self-limited course over a period of several years, but its
duration was typically shorter in younger children than in adolescents or adults. Several other forms of arthritis, such as natural and vaccine-induced rubella or parvovirus-associated arthritis, are also generally milder in young children than in adults.14 15 16
No child in this study fulfilled the diagnostic criteria for juvenile rheumatoid arthritis, which require the presence of arthritis for six or more consecutive weeks at onset. Instead, joint involvement in our patients usually began with migratory arthralgias followed by brief attacks of frank arthritis. Although antinuclear antibodies may occur in both illnesses, none of the children in this study had iritis, a typical feature of pauciarticular juvenile rheumatoid arthritis. Unlike patients with spondyloarthritis, the adolescent boys in this study did not have low back pain, enthesopathy, a family history of the illness, or a positive test for HLA-B27. Instead, they had the HLA-DR4 allele, which has been associated with chronic Lyme arthritis.17
This study extends the length of observation beyond the period of arthritis. As in past descriptions,18 , 19 we found that keratitis, characterized by multiple focal opacities of the corneal stroma, developed in two children (4 percent) during the sixth year of illness. Treponema pallidum is also known to produce bilateral, diffuse stromal keratitis. Episodic recurrences of syphilis are thought to be due to the persistence of antigen, either in living, sequestered spirochetes or in fragments of the organism; this may also be true in Lyme disease. These lesions appear to respond to topical corticosteroids rather than to antibiotics, as was the case in our patient with recurrent keratitis.
Eleven and 12 years after the onset of disease, two children in this study (4 percent) were found to have a subtle encephalopathy, with memory impairment, headache, and fatigue accompanied by evidence of intrathecal production of antibody to the spirochete, a neurologic picture that is typical of Lyme encephalopathy.8 , 20 Neurologic syndromes developed in two other study patients years after their arthritis had resolved: in one instance, a seizure disorder, and in the other, apparent demyelinating disease. Although both children had high serum titers of antibody to B. burgdorferi, intrathecal production of antibody to the spirochete could not be demonstrated. However, arguing against the diagnosis of multiple sclerosis in the latter case were the onset of symptoms before puberty, the normal evoked-potential responses, and the absence of periventricular white-matter lesions throughout the seven-year period of neurologic symptoms. Although we excluded these patients from the neuroborreliosis group, it remains possible that their neurologic abnormalities were due to infection with B. burgdorferi.
One third of the study patients had sporadic, brief episodes of joint pain as long as 10 years after the period of active arthritis. At long-term follow-up, these patients had IgM responses to the spirochete more often than did the asymptomatic patients, and they also had higher IgG titers. Except in two instances, no other potential cause for joint pain could be ascertained from physical examination. Thus, in some cases, these subtle and subjective joint symptoms may have been caused by residual spirochetal infection of synovial tissue.
In conclusion, the long-term course of Lyme disease in children may include acute infection followed initially by attacks of arthritis and then over a period of several years by keratitis, subtle joint pain, or chronic encephalopathy. A positive antibody response to B. burgdorferi does not, by itself, indicate active infection, since most patients remain seropositive after treatment. However, the combination of subtle musculoskeletal or neurologic symptoms in a patient with a high IgG antibody titer to the spirochete, particularly if accompanied by an IgM response, is of concern. The appropriate treatment for these patients is not yet clear.
Supported in part by a grant (AR-20358) from the National Institutes of Health.
We are indebted to Dr. Cody Meissner and Dr. Jane Schaller for helpful discussions, to Ms. Jennifer Whalen and Mr. Kirk Lum for excellent laboratory assistance, and to Ms. Nina Solomita for assistance in the preparation of the manuscript.
Source Information
From the Departments of Pediatrics and Medicine, Floating Hospital for Infants and Children, New England Medical Center, Tufts University School of Medicine, Boston. Address reprint requests to Dr. Szer at the Department of Rheumatology, Children's Hospital of Los Angeles, 4650 Sunset Blvd., Los Angeles, CA 90054–0700.
- References
References
1
Steere
Full Text | Web of Science | Medline2
Steere
CrossRef | Web of Science | Medline3
Steere
Web of Science | Medline4
Steere
Web of Science | Medline5
Steere
Web of Science | Medline6
Steere
Web of Science | Medline7
Steere
Full Text | Web of Science | Medline8
Logigian
Full Text | Web of Science | Medline9
Craft
CrossRef | Web of Science | Medline10
Steere
CrossRef | Web of Science | Medline11
Williams
CrossRef | Web of Science | Medline12
Culp
Web of Science | Medline13
Eichenfield
CrossRef | Web of Science | Medline14
Yanez
Web of Science15
Swartz
Web of Science | Medline16
Anderson
CrossRef | Web of Science | Medline17
Steere
Full Text | Web of Science | Medline18
Baum
Web of Science | Medline19
Kornmehl
Web of Science | Medline20
Halperin
Web of Science | Medline
- Citing Articles (37)
Citing Articles
1
Laure Seugé, Michel Fischbach, Vincent Laugel, Dan Lipsker. (2011) Borréliose de Lyme chez l’enfant : étude rétrospective de 16 cas. La Presse Médicale 40:9, e359-e364
CrossRef2
Eugene D. Shapiro, Sunil K. Sood. 2011. Prognosis of Persons with Lyme Borreliosis. , 213-223.
CrossRef3
Frank Dressler, Patricia I. Irigoyen, Norman T. Ilowite, Hans-Iko Huppertz. 2011. Lyme Arthritis. , 115-134.
CrossRef4
Hans-Iko Huppertz, Frank Dressler. 2011. LYME DISEASE. , 579-590.
CrossRef5
Sonny Dhanani, Peter N. Cox. 2011. Infectious Syndromes in the Pediatric Intensive Care Unit. , 1336-1348.
CrossRef6
S. C. Li. (2010) The Prognosis of Children Treated for Lyme Arthritis. AAP Grand Rounds 24:2, 18-18
CrossRef7
H. Hofmann, C. Schnopp. (2009) Kinderdermatologie. Der Hautarzt 60:3, 183-193
CrossRef8
Henry M. Feder. (2008) Lyme Disease in Children. Infectious Disease Clinics of North America 22:2, 315-326
CrossRef9
F. Dressler, H.-I. Huppertz. (2008) Lyme-Arthritis bei Kindern und Jugendlichen. Zeitschrift für Rheumatologie 67:2, 121-127
CrossRef10
D. Lipsker, B. Jaulhac. (2008) Manifestaciones cut??neas de las borreliosis. EMC - Dermatolog??a 42:1, 1-11
CrossRef11
Jennifer E. Weiss, Norman T. Ilowite. (2007) Juvenile Idiopathic Arthritis. Rheumatic Disease Clinics of North America 33:3, 441-470
CrossRef12
J. Pourel, I. Chary-Valckenaere. (2007) Borreliosis de Lyme. EMC - Aparato Locomotor 40:2, 1-14
CrossRef13
Kristin M Corapi, Marc I White, Charlotte B Phillips, Lawren H Daltroy, Nancy A Shadick, Matthew H Liang. (2007) Strategies for primary and secondary prevention of Lyme disease. Nature Clinical Practice Rheumatology 3:1, 20-25
CrossRef14
Sunil K. Sood. (2006) What we have learned about Lyme borreliosis from studies in children. Wiener klinische Wochenschrift 118:21-22, 638-642
CrossRef15
Henry M. Feder, Micha Abeles, Megan Bernstein, Diane Whitaker-Worth, Jane M. Grant-Kels. (2006) Diagnosis, treatment, and prognosis of erythema migrans and Lyme arthritis. Clinics in Dermatology 24:6, 509-520
CrossRef16
2005. Hypertension (High Blood Pressure). .
CrossRef17
2005. Diabetes Mellitus. .
CrossRef18
Isabelle Renaud, Claude Cachin, Jean-Charles Gerster. (2004) Good outcomes of Lyme arthritis in 24 patients in an endemic area of Switzerland. Joint Bone Spine 71:1, 39-43
CrossRef19
Eugene D. Shapiro. (2002) Long-Term Outcomes of Persons with Lyme Disease. Vector-Borne and Zoonotic Diseases 2:4, 279-281
CrossRef20
F. Peters, E.E. Petersen, C.J. Kirkpatrick. (2002) Isolated erythema (cellullitis) of the breast. The Breast 11:6, 484-488
CrossRef21
C Phillips. (2001) Lyme disease and preventive behaviors in residents of Nantucket Island, Massachusetts. American Journal of Preventive Medicine 20:3, 219-224
CrossRef22
T. J. Wang, M. H. Liang, O. Sangha, C. B. Phillips, R. A. Lew, E. A. Wright, V. Berardi, A. H. Fossel, N. A. Shadick. (2000) Coexposure to Borrelia burgdorferi and Babesia microti Does Not Worsen the Long-Term Outcome of Lyme Disease. Clinical Infectious Diseases 31:5, 1149-1154
CrossRef23
Bo Qiu, Michael Brunner, Guobao Zhang, Leonard Sigal, Stanley Stein. (2000) Selection of continuous epitope sequences and their incorporation into poly(ethylene glycol)-peptide conjugates for use in serodiagnostic immunoassays: Application to Lyme disease. Biopolymers 55:4, 319-333
CrossRef24
SUNIL K. SOOD. (1999) Lyme disease. The Pediatric Infectious Disease Journal 18:10, 913-925
CrossRef25
Johan Berglund. (1999) Natural history and long-term consequences of Lyme disease in children. Current Opinion in Infectious Diseases 12:3, 265-269
CrossRef26
Avinash K. Shetty, Abraham Gedalia. (1998) SEPTIC ARTHRITIS IN CHILDREN. Rheumatic Disease Clinics of North America 24:2, 287-304
CrossRef27
Leonard H. Sigal. (1998) MUSCULOSKELETAL MANIFESTATIONS OF LYME ARTHRITIS. Rheumatic Disease Clinics of North America 24:2, 323-351
CrossRef28
BRADLEY J. BLOOM, PATRICIA M. WYCKOFF, H. CODY MEISSNER, ALLEN C. STEERE. (1998) Neurocognitive abnormalities in children after classic manifestations of Lyme disease. The Pediatric Infectious Disease Journal 17:3, 189-196
CrossRef29
Allen C. Steere. (1997) DIAGNOSIS AND TREATMENT OF LYME ARTHRITIS. Medical Clinics of North America 81:1, 179-194
CrossRef30
Ariadna Sadziene, A. G. Barbour. (1996) Experimental immunization against Lyme borreliosis with recombinant Osp proteins: An overview. Infection 24:2, 195-202
CrossRef31
Hans-Iko Huppertz, Helge Karch, Hans-Joachim Suschke, Eva DÖRing, Gerd Ganser, Angelika Thon, Wassilios Bentas, . (1995) Lyme arthritis in european children and adolescents. Arthritis & Rheumatism 38:3, 361-368
CrossRef32
Nocton, James J.Dressler, FrankRutledge, Barbara J.Rys, Paul N.Persing, David H.Steere, Allen C.. (1994) Detection of Borrelia burgdorferi DNA by Polymerase Chain Reaction in Synovial Fluid from Patients with Lyme Arthritis. New England Journal of Medicine 330:4, 229-234
Full Text33
Jane D. Cooper, Robert T. Schoen, Stephen E. Malawista. (1993) Treatment of asymptomatic, retrospectively diagnosed, Lyme disease: comment on the report by Christian. Arthritis & Rheumatism 36:11, 1637-1638
CrossRef34
J. Gutiérrez, M. Palermo, M. C. Maroto, M. Abellan. (1993) Atypical bilateral symmetric erosive chronic polyarthritis in the course of lyme disease. European Journal of Clinical Microbiology & Infectious Diseases 12:10, 787-789
CrossRef35
Peter Herzer. (1993) Rheumatic manifestations in Lyme borreliosis. Clinics in Dermatology 11:3, 401-406
CrossRef36
Jane G. Schaller. (1993) Therapy for childhood rheumatic diseases. have we been doing enough?. Arthritis & Rheumatism 36:1, 65-70
CrossRef37
(1991) Long-Term Course of Lyme Arthritis. New England Journal of Medicine 325:26, 1886-1886
Full Text
- Letters
