Lyme Arthritis: A Comparison of Presentation, Synovial Fluid Analysis, and Treatment Course in Children and Adults

Transcription

1 Arthritis Care & Research Vol. 65, No. 12, December 2013, pp DOI /acr , American College of Rheumatology ORIGINAL ARTICLE Lyme Arthritis: A Comparison of Presentation, Synovial Fluid Analysis, and Treatment Course in Children and Adults BRIAN E. DAIKH, 1 FRED E. EMERSON, 2 ROBERT P. SMITH, 2 F. LEE LUCAS, 2 AND CAROL A. MCCARTHY 2 Objective. This case series examines differences in the presentation, management, and outcome of Lyme arthritis between the pediatric and adult population. Methods. We reviewed charts of pediatric and adult patients evaluated for Lyme arthritis by rheumatologists and pediatric infectious disease specialists in Portland, Maine between January 2002 and July Patients included for analysis had documented joint swelling and positive Lyme serology. Data on clinical presentation, synovial fluid and peripheral blood results, treatment, and clinical course were analyzed. Results. Twenty-nine adults and 52 children met case criteria for Lyme arthritis. Children were more likely than adults to present acutely (P < ) and also had higher mean peripheral blood (P 0.05) and synovial fluid white blood cell counts (P < ). Lyme arthritis was more frequently suspected in children at presentation (P 0.04). There was no difference between children and adults with respect to suspicion for septic arthritis, hospitalization, or surgical intervention. Adults received more antibiotic courses (P 0.007) and were more likely to have intravenous antibiotics in subsequent treatment courses (P 0.006). Children were more likely to have normal function within 4 weeks of initiating antibiotic treatment (P < ). Conclusion. Children with Lyme arthritis were more likely to present acutely with higher synovial white cell counts than adults. We did not, however, observe a significant difference in hospitalization or surgical management. Children had more prompt resolution of their joint swelling and received less treatment overall. INTRODUCTION Lyme disease, caused by the spirochete Borrelia burgdorferi and transmitted by the Ixodes scapularis tick, is endemic in many regions of the northeast and midwestern US. An inflammatory mono- or oligoarticular arthritis with a predilection for large, weight-bearing joints, especially the knee, is a common manifestation of the late disseminated stage of the disease (1). Several published studies have reported the clinical presentation, course, and outcome of Lyme arthritis in children (2 9). Lyme arthritis may be initially misdiagnosed as septic arthritis, juvenile idiopathic arthritis, or transient 1 Brian E. Daikh, MD: Rheumatology Associates, Portland, Maine; 2 Fred E. Emerson, MD, Robert P. Smith, MD, F. Lee Lucas, PhD, Carol A. McCarthy, MD: Maine Medical Center, Portland. Dr. Smith has received human immunodeficiency virus diagnostics research support from Alere. Address correspondence to Brian E. Daikh, MD, Rheumatology Associates, 51 Sewall Street, Portland, ME daikhb@maine.rr.com. Submitted for publication December 12, 2012; accepted in revised form July 3, synovitis (2,3,6,9 11). A large percentage of children with Lyme arthritis present with acute, monarticular arthritis sometimes associated with fever, peripheral blood leukocytosis, elevated acute-phase reactants, and significantly elevated synovial white cell count, leading to the erroneous diagnosis of septic arthritis. Some of these children receive unnecessary anesthesia and surgical drainage. The long-term outcome of Lyme arthritis in children is excellent and the majority of children have no long-term sequelae (3,4,8). Although differences in presentation and course of Lyme arthritis have been noted between young children and adolescents (7), no series has compared pediatric and adult Lyme arthritis in North America. It is not known whether differences exist between the pediatric and adult population with respect to clinical presentation, clinical course, and outcome. We have observed children ultimately diagnosed with Lyme arthritis initially undergoing surgical incision and drainage for presumed septic arthritis. We hypothesized that differences in clinical presentation might lead to differences in management and outcome between children and adults with Lyme arthritis. The goals of this study were to address the following 3 questions. Does the presentation of Lyme arthritis differ in 1986

2 Analysis and Treatment of Lyme Arthritis in Children and Adults 1987 Significance & Innovations To our knowledge, this is the first reported series comparing clinical and laboratory features and outcomes of Lyme arthritis between children and adults. Children with Lyme arthritis were found to present more acutely than adults. Synovial white cell counts were found to be significantly higher in children than adults, often approaching counts observed with septic arthritis. Despite the acute, intense inflammatory response observed in children with Lyme arthritis compared to adults, children recovered faster and received less antibiotic therapy. the pediatric and adult populations? Are there differences in synovial fluid white blood cell counts between these populations? Are there differences in treatment course and outcome between children and adults? PATIENTS AND METHODS Study population and data collection. We reviewed charts of adult and pediatric patients with Lyme arthritis seen between January 2002 and June 2008 by Rheumatology Associates in Portland, Maine and/or the Pediatric Infectious Disease Division at Maine Pediatric Specialty Group in Portland. Patients were identified by billing codes for Lyme disease and Lyme arthritis. Inclusion criteria included positive enzyme-linked immunosorbent assay (ELISA) and positive IgG by Western blot to B burgdorferi as defined by the Centers for Disease Control and Prevention criteria (12) and documentation of joint effusion in the medical record. Children evaluated in the pediatric infectious disease clinic for other possible forms of Lyme disease were excluded. Adults were defined as being age 18 years. The study was exempted from review by the Maine Medical Center Internal Review Board. Charts were reviewed by 3 authors (BED, FEE, and CAM). Data were extracted using a standardized tool. Demographic data included age, sex, and residence. Historical data included location and duration of current arthritis, history of other arthritis, date of onset, history of tick bite, prior erythema migrans (EM) type rash, acuity of presentation, and suspicion for Lyme arthritis at the time of initial presentation. Current arthritis was defined as the episode of arthritis that directly led to diagnosis of Lyme disease. The acuity of presentation was categorized as 2 weeks of symptoms, 2 6 weeks of symptoms, and 6 weeks of symptoms. Physical examination data included fever defined as 38 C, joint effusion, joint erythema, and ability to bear weight. Laboratory data included serum white blood cell (WBC) count with differential, erythrocyte sedimentation rate, C-reactive protein level, Lyme ELISA and Western blot, synovial fluid cell count with differential, gram stain, and bacterial culture. Management data included suspicion for septic arthritis, hospitalization, and surgical drainage. The duration and route of antibiotic therapy was recorded. Followup data included presence of normal examination and function at 4 weeks. Statistical analysis. Data were analyzed using the SAS statistical package. The unpaired t-test for unequal variance was used to compare the mean for numerical variables. Pearson s chi-square test or, in cases where the sample size rendered this test inaccurate, Fischer s exact test were used to compare categorical variables. RESULTS The charts of 47 adults and 55 children with suspected Lyme arthritis were reviewed. Twenty-nine adult and 52 pediatric patients met inclusion criteria. Eighteen adult patients were excluded: 16 without documented effusion and 2 without documented positive Lyme serology. Three pediatric patients were excluded because of absence of a documented effusion. All patients were evaluated by either rheumatology or pediatric infectious disease specialists. All 29 adults were seen by rheumatology services and 6 (21%) were also seen by adult infectious disease services. Fifteen (52%) adults were also evaluated by orthopedic services. Twenty-eight (54%) children were evaluated by orthopedic services, but all of these were also evaluated by either rheumatology or pediatric infectious disease services. Twenty-seven (52%) children were evaluated by rheumatology, 32 (62%) were evaluated by pediatric infectious disease, and 7 (13%) were evaluated by both services. Demographic data are displayed in Table 1. Patients ranged in age from years with a mean of 24 years. The mean pediatric age was 9.4 years and the mean adult age was 48.6 years. Sixty-five percent of the patients were male and all but one were Maine residents. The majority resided in Cumberland and York counties, the 2 most southern counties in Maine. The time of diagnosis was distributed throughout the calendar year with 38 of 81 (47%) occurring from September through December. The knee was the most commonly affected joint in both children (73%) and adults (72%). Other affected joints included the hip, elbow, and ankle. Two of 52 (4%) children, but no adults, had multiple joints affected at the time of diagnosis. A significant difference existed in the acuity of presentation in children and adults (Table 2). Eighty-eight percent of pediatric patients had 2 weeks of symptoms prior to evaluation, compared to 52% of adults. Table 1. Demographics of study population* Children Adults Cases, no Age, mean (range) years 9.4 (2.5 18) 48.6 (19 80) Male 34 (65) 19 (66) Maine resident 52 (100) 28 (97) Cumberland County resident 23 (44) 17 (59) York County resident 15 (29) 9 (31) * Values are the number (percentage) unless indicated otherwise.

3 1988 Daikh et al Table 2. History and presentation of patients with Lyme arthritis* Children Adults P History of other arthritis 4/52 (8) 11/28 (39) History of tick bite 18/49 (37) 6/25 (24) 0.40 History of erythema migrans type rash 8/45 (18) 7/25 (28) 0.49 Presentation weeks 46/52 (88) 15/29 (52) 2 6 weeks 3/52 (6) 5/29 (17) 6 weeks 3/52 (6) 9/29 (31) Temperature 38 C at onset of current arthritis 7/44 (16) 1/8 (13) 0.99 Joint erythema 4/38 (11) 3/13 (23) 0.35 Weight bearing 41/48 (85) 25/27 (93) 0.47 Diagnosis suspected at presentation 33/50 (66) 11/28 (39) 0.04 * Values are the number/total number (percentage) unless indicated otherwise. Thirty-one percent of adults had symptoms for 6 weeks. Adults were more likely to have a history of other types of arthritis (P 0.002). There were no differences between pediatric and adult cases with respect to recollection of a tick bite, history of EM-type rash, or fever at presentation. There were also no differences in joint swelling and erythema at presentation. The majority of both pediatric and adult patients were weight-bearing at presentation. The diagnosis of Lyme arthritis was initially suspected in children more often than adults (P 0.04). The results of laboratory evaluations are shown in Table 3. The inflammatory markers were similar in both age groups. The peripheral blood WBC count was higher in children (P 0.05). Adults were more likely to have fluid obtained from their affected joint (P 0.002). The synovial fluid WBC counts were significantly higher in children. The mean cell count for children was 49,924 cells/mm 3 (range 2, ,740), while the mean adult cell count was 12,396 cells/mm 3 (range 1,700 30,000). There was growth of bacteria in 1 adult culture. The organism was coagulase negative Staphylococcus. This organism grew in the broth culture only and was believed to be a contaminant. There were no statistically significant differences between adults and children in the suspicion of septic arthritis, hospitalization, or intraoperative joint drainage. Adults were more likely to have multiple joint aspirations (P 0.02). Ten of 52 children were initially started on intravenous (IV) antibiotics compared with 2 of 29 adults (P 0.20). These 10 children were then transitioned to oral antibiotics after a mean of 3.5 days. Forty children (77%) received a single 1-month course of antibiotics. The oral antibiotics used were amoxicillin (48%), doxycycline (48%), cefuroxime (2%), and amoxicillin (2%) and azithromycin (2%). Nine children (17%) received 2 courses of oral antibiotics. Three children (6%) received a course of ceftriaxone after 2 oral courses of antibiotics. Thirteen adults (45%) received 1 course of doxycycline and 1 adult was treated with a 1-month course of ceftriaxone. Seven (24%) of the adults had 2 courses of antibiotics. Four received doxycycline and 3 received ceftriaxone for the second course of therapy. Five adults (17%) had 3 courses of antibiotics. Of these 5 adults, 4 received ceftriaxone and 1 received doxycycline for the third course of therapy. Three adults (10%) received more than 3 months of antibiotic therapy. Fifty-two percent of adults (15 of 29) had more than 1 course of antibiotics compared with 23% (12 of 52) of children (P 0.02). Two courses of antibiotics were prescribed for 7 of 29 adults (24%) and 9 of 52 (17%) children (P 0.56). There were 8 adults (28%) and 3 children (6%) that received 2 courses of antibiotics (P 0.014). A course of ceftriaxone was used in 10 of 29 (34%) adults and 3 of 52 (6%) children (P ). Table 3. Laboratory evaluation of patients with Lyme arthritis* Children Adults P WBC 10 3 /mm, mean SD ESR, mean SD mm/hour CRP, mean SD mg/dl Lyme serologies, no./total no. (%) ELISA positive 52/52 (100) 29/29 (100) 0.99 Western blot IgM positive 16/45 (36) 11/24 (46) 0.57 Western blot IgG positive 52/52 (100) 29/29 (100) 0.99 Synovial fluid obtained 27/52 (52) 26/29 (90) Synovial fluid WBC count, mean SD cells/mm 3 49,924 37,691 12,396 8, * WBC white blood cell; ESR erythrocyte sedimentation rate; CRP C-reactive protein; ELISA enzyme-linked immunosorbent assay.

4 Analysis and Treatment of Lyme Arthritis in Children and Adults 1989 Children were more likely to have a normal examination (P 0.007) and to have resumed normal function of their affected joint (P ) within 4 weeks of starting antibiotic therapy. Preexisting arthritis of another type did not predict the presence of persistent swelling in the adult population. DISCUSSION Lyme arthritis is a common cause of monarticular and oligoarticular arthritis in the northeastern region of the US. In North America, Lyme disease is caused by one genospecies, B burgdorferi, sensu stricto, while in Europe, Lyme disease has been attributed to at least 5 major genospecies (13). This strain variation may account for some of the reported differences in clinical presentation of Lyme disease between the two continents. We have therefore limited our review and comparison to other reports of Lyme arthritis within the US. In retrospective analyses of children in Connecticut and Massachusetts who have undergone joint aspiration for arthritis, 31% were noted to have Lyme arthritis (11,14). In both series, it was more common than septic arthritis. While others have noted age-related differences in clinical aspects of Lyme arthritis in children and adults (1,7), our study is the first to compare presentation and management between children and adults. Most patients did not recall a history of an EM-type rash, which would typically lead to treatment for early Lyme disease. As in prior studies, the knee was the most commonly involved joint for both age groups, occurring in approximately 70% of the patients and most patients were able to walk (2,14). However, children were more likely than adults to have an acute presentation and had higher peripheral blood WBC counts. Children have higher peripheral WBC counts than adults on average, so the observed difference may not be clinically significant. Adults were more likely to undergo joint aspiration. Referral patterns, differential diagnostic considerations, and ease of aspiration may explain this difference. In children who underwent aspiration of the affected joint, the mean synovial WBC count was significantly higher than that found in adults (mean 49,924 cells/mm 3 versus 12,396 cells/mm 3 ). Children were evaluated earlier in the course of their arthritis, which may have had an influence on the magnitude of the synovial white cell count, but the maximum cell count in the adults was 30,000 cells/mm 3, considerably lower than the average cell count in children. Synovial white cell counts 50,000 cells/mm 3 in children were not uncommon. Others have noted high synovial fluid white cell counts in pediatric Lyme arthritis (2 4,9,10,14). In a recent review by Milewski et al (14), 60 of 123 children with Lyme arthritis had synovial cell counts 50,000 cells/mm 3 and 13% had cell counts 100,000 cells/mm 3. Often these patients have been initially diagnosed and treated for septic arthritis. In the current series, despite the more acute presentation in children than adults, there was no statistical difference in hospitalization, surgical drainage, or initial use of IV antibiotics. It is important for physicians who care for both children and adults to realize that there may be differences in presentation between the 2 groups. In endemic areas, Lyme disease should be considered in all patients with monarticular and oligoarticular arthritis. Given the excellent outcome of Lyme arthritis with antibiotic therapy, an elevated synovial cell count should not be the indication for surgical drainage if the diagnosis of Lyme arthritis is suspected and other factors are atypical for septic arthritis. Some patients who are ambulating may be closely monitored while awaiting results of Lyme serology. Consideration of Lyme arthritis may prevent unnecessary anesthesia and surgery as well as extended courses of IV antibiotics. Cases involving the hip joint may be more difficult, especially if the synovial white cell count is in the range of 50,000 cells/mm 3 and results of the Lyme tests are not yet available. It is therefore important to have access to accurate as well as timely diagnostic tests to assist with diagnosis and management. Further studies would be helpful to guide optimal treatment of Lyme arthritis. One month of oral antibiotics is currently recommended as initial therapy (15). Additional antibiotic therapy has been suggested for patients with persistent joint swelling. The Infectious Disease Society of America treatment guidelines (15) recommend that clinicians should consider waiting several months before initiating re-treatment with antimicrobial agents because of the anticipated slow resolution of inflammation after treatment. In our series, more than one-half of adults, compared with 23% of children, were treated with multiple courses of antibiotics. It is possible that common practice in some endemic areas is to re-treat if there is not prompt resolution of joint swelling, although there is no clinical evidence of benefit of re-treatment. Additional courses of antibiotics have been recommended previously if there is a positive polymerase chain reaction (PCR) result for B burgdorferi DNA in synovial fluid after therapy (16). However, a recent report demonstrated that such positive synovial PCR samples did not predict active spirochetal infection (17). A possible explanation for this has been proposed by Wormser et al (18). They propose the amber hypothesis, which suggests nonviable B burgdorferi spirochetes are trapped in a matrix in the joint, but outside of the synovial compartment, and that an immune response to the nonviable cells leads to the inflammatory arthritis. This hypothesis may help to explain why patients with Lyme arthritis may have a relapsing and remitting pattern of arthritis, sometimes occurring after antibiotic therapy in the absence of evidence for active infection. Approximately 10% of patients with Lyme arthritis develop a persistent joint effusion after appropriate antibiotic therapy (16). Such cases, which may be associated with the presence of several HLA DR4 alleles (16,19,20), may respond to disease-modifying antirheumatic drug therapy (19) or intraarticular corticosteroid injection. This case series examines differences between children and adults with Lyme arthritis. We found that children with Lyme arthritis had more acute presentations with higher synovial white cell counts than adults, yet children were more likely than adults to have normal joint examination and function within one month of oral antibiotic therapy. There may be differences in immune responses

5 1990 Daikh et al between children and adults, which may impact clinical outcome over time. Further studies may elucidate these differences and allow more optimal diagnosis and management of Lyme arthritis. ACKNOWLEDGMENTS The authors wish to acknowledge the contributions of Sarah B. Trinward, BA, in manuscript preparation and Stephen R. Hayes, MD, in manuscript review. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Daikh had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Daikh, Lucas, McCarthy. Acquisition of data. Daikh, Emerson, McCarthy. Analysis and interpretation of data. Diakh, Emerson, Smith, Lucas, McCarthy. REFERENCES 1. Steere AC, Schoen RT, Taylor E. The clinical evolution of Lyme arthritis. Ann Intern Med 1987;107: Culp RW, Eichenfield AH, Davidson RS, Drummond DS, Christofersen MR, Goldsmith DP. Lyme arthritis in children: an orthopaedic perspective. J Bone Joint Surg Am 1987;69: Eichenfield AH, Goldsmith DP, Benach JL, Ross AH, Loeb FX, Doughty RA, et al. Childhood Lyme arthritis: experience in an endemic area. J Pediatr 1986;109: Gerber MA, Zemel LS, Shapiro ED. Lyme arthritis in children: clinical epidemiology and long-term outcomes. Pediatrics 1998;102: Huppertz HI, Karch H, Suschke HJ, Doring E, Ganser G, Thon A, et al, for the Pediatric Rheumatology Collaborative Group. Lyme arthritis in European children and adolescents. Arthritis Rheum 1995;38: Rose CD, Fawcett PT, Eppes SC, Klein JD, Gibney K, Doughty RA. Pediatric Lyme arthritis: clinical spectrum and outcome. J Pediatr Orthop 1994;14: Szer IS, Taylor E, Steere AC. The long-term course of Lyme arthritis in children. N Engl J Med 1991;325: Tory HO, Zurakowski D, Sundel RP. Outcomes of children treated for Lyme arthritis: results of a large pediatric cohort. J Rheumatol 2010;37: Willis AA, Widmann RF, Flynn JM, Green DW, Onel KB. Lyme arthritis presenting as acute septic arthritis in children. J Pediatr Orthop 2003;23: Bachman DT, Srivastava G. Emergency department presentations of Lyme disease in children. Pediatr Emerg Care 1998; 14: Thompson A, Mannix R, Bachur R. Acute pediatric monoarticular arthritis: distinguishing Lyme arthritis from other etiologies. Pediatrics 2009;123: Recommendations for test performance and interpretation from the Second National Conference on Serologic Diagnosis of Lyme Disease. MMWR Morb Mortal Wkly Rep 1995;44: Stanek G, Wormser GP, Gray J, Strle F. Lyme borreliosis. Lancet 2012;379: Milewski MD, Cruz AI Jr, Miller CP, Peterson AT, Smith BG. Lyme arthritis in children presenting with joint effusions. J Bone Joint Surg Am 2011;93: Wormser GP, Dattwyler RJ, Shapiro ED, Halperin JJ, Steere AC, Klempner MS, et al. The clinical assessment, treatment, and prevention of Lyme disease, human granulocytic anaplasmosis, and babesiosis: clinical practice guidelines by the Infectious Diseases Society of America. Clin Infect Dis 2006;43: Steere AC, Angelis SM. Therapy for Lyme arthritis: strategies for the treatment of antibiotic-refractory arthritis. Arthritis Rheum 2006;54: Li X, McHugh GA, Damle N, Sikand VK, Glickstein L, Steere AC. Burden and viability of Borrelia burgdorferi in skin and joints of patients with erythema migrans or Lyme arthritis. Arthritis Rheum 2011;63: Wormser GP, Nadelman RB, Schwartz I. The amber theory of Lyme arthritis: initial description and clinical implications. Clin Rheumatol 2012;31: Puius YA, Kalish RA. Lyme arthritis: pathogenesis, clinical presentation, and management. Infect Dis Clin North Am 2008;22: Steere AC, Gross D, Meyer AL, Huber BT. Autoimmune mechanisms in antibiotic treatment-resistant Lyme arthritis. J Autoimmun 2001;16:263 8.