Lyme arthritis is an increasingly recognized clinical entity that often prompts orthopaedic evaluation in pediatric patients. While Lyme arthritis is most common in the knee, the clinical presentation of Lyme arthritis of the hip can be similar to both acute bacterial septic arthritis and transient synovitis. Accurately distinguishing these clinical entities is important since the definitive treatment of each is distinct. Because there is limited literature on monoarticular Lyme arthritis of the hip, the purpose of this study was to perform a systematic review and meta-analysis of clinical and laboratory parameters associated with Lyme arthritis (LA) of the hip and compare them to septic arthritis (SA) and transient synovitis (TS).
A systematic review of the literature was performed using the following search terms, including the variants and plural counterparts “hip” and “Lyme arthritis.” A final database of individual patients was assembled from the published literature and direct author correspondence, when available. A previously published cohort of patients with hip transient synovitis or septic arthritis was used for comparative analysis. A comparative statistical analysis was performed to the assembled database to assess differences in laboratory and clinical variables between the three diagnoses.
Data on 88 patients diagnosed with Lyme arthritis of the hip was collected and consolidated from the 12 articles meeting inclusion criteria. The average age of patients presenting with Lyme arthritis was 7.5 years (± 3.5 years), the mean erythrocyte sedimentation rate (ESR), and the C-reactive protein (CRP) was 41 mm/hr and 3.9 mg/L, respectively. Peripheral white blood cell (WBC) count averaged 10.6 x 109cells/L with the synovial WBC count averaging 55,888 cells/mm3. Compared to a previous cohort of patients with confirmed transient synovitis or septic arthritis, the 95% confidence interval for ESR was 21 - 33 mm/hr in those diagnosed with toxic synovitis (TS), 37 - 46 mm/hr for Lyme arthritis (LA), and 44 - 64 mm/hr for septic arthritis (SA). Synovial WBC counts (cells/mm3) 95% confidence intervals (CI) were 5,644 - 15,388 cells/mm3 for TS, 47,533 - 64,242 cells/mm3 for LA, and 105,432 - 260,214 cells/mm3 for SA. There was a statistically significant difference in the incidence of fever > 38.5oC (P < 0.001) and refusal to bear weight (P < 0.01) between SA, LA, and TS.
Monoarticular Lyme arthritis can be a cause of hip pain in certain geographic areas and has clinical and diagnostic overlap with transient synovitis and acute bacterial septic arthritis. This study consolidates the available literature and represents the largest series of patients diagnosed with Lyme arthritis of the hip to date. We propose a diagnostic algorithm that serially incorporates ESR, followed by a synovial neutrophil count, when evaluating pediatric patients with an irritable hip in Lyme endemic areas.
Since its first description by Steere et al. in 1977 , Lyme disease has become the most common tick-borne illness in the United States and Europe [2-3]. Lyme disease is caused by the spirochete Borrelia burgdorferi and is transmitted by the Ixodes tick, which is endemic to certain areas of the United States, including the Northeast, parts of the upper Midwest, and the Pacific Northwest . Reported cases of Lyme disease have a bimodal age distribution with average annual rates peaking in those between five to nine years old and 55-59 years old . While Lyme disease can affect multiple organ systems, acute Lyme arthritis (LA) is an increasingly recognized clinical entity that prompts urgent or emergent evaluation in pediatric patients [4-9].
The most common joint affected by LA is the knee, and there have been multiple studies evaluating the diagnosis and treatment of this clinical entity in children [7, 10-14]. LA of the hip has also been described, albeit at a much lower population incidence compared to the knee [15-21]. Evaluation of a child with an acutely irritable hip continues to pose clinical and diagnostic challenges, particularly in geographic regions in which Lyme disease is endemic [17, 22-23]. The clinical presentation of LA of the hip can often be similar to both acute bacterial septic arthritis (SA) as well as transient synovitis (TS). In addition, there is no readily available, validated, and rapid point-of-care testing for Lyme disease. Therefore, accurately distinguishing these clinical entities is important, particularly because the definitive treatment of each is distinct.
Because of the limited available literature on isolated LA of the hip, the primary purpose of this study was to perform a systematic review to consolidate all available studies examining pediatric patients with isolated LA of the hip. We also performed a meta-analysis of the clinical and laboratory parameters associated with LA of the hip compared to SA and TS. The goal of the meta-analysis was to assemble the data available in the literature and combine it to perform a comparative analysis in order to establish clinically useful diagnostic parameters distinguishing LA from bacterial SA or TS of the hip.
Materials & Methods
We queried EMBASE, COCHRANE, and MEDLINE computerized literature databases from the earliest date available in the databases to August 7, 2017 using the following search terms (including variants and plural counterparts): “hip” and “Lyme arthritis.” A final database of individual patients was assembled from the published literature and direct author correspondence, when available. Reference lists from all articles were scrutinized to identify any additional articles of interest. Two authors (JBA, AIC) performed the initial search, while three authors (JBA, AIC, and JMR) independently reviewed the references of identified papers and selected those that fit the following criteria: (1) written in the English language, (2) level I, II, III, IV, and V study classification as detailed by “The Journal of Bone and Joint Surgery” criteria, (3) patient age ≤ 20 years old, and (4) individual patient level information available in the article or by direct correspondence with the original study author(s).
A previously published cohort of patients with hip transient synovitis or septic arthritis, examined at our home institution , was used for comparative analysis. In this cohort, SA was defined as either (1) a child with a positive synovial fluid culture for a bacterial pathogen, (2) synovial fluid with a white blood count (WBC) count > 50,000 cells/µL and positive blood culture results for a bacterial pathogen, or (3) frank purulence found at the time of surgery with subsequent negative cultures. The diagnosis of TS was one of exclusion, based on negative microbiology cultures and negative Lyme tests.
For comparative analysis, univariate P values were calculated by independent sample t-tests when variables were continuous, while Pearson’s Chi-squared and Fisher’s exact tests were used to assess differences in dichotomous, categorical outcome variables. Statistical significance was set at P < 0.05. Statistical analysis was performed using the Statistical Package for Social Sciences (SPSS), version 20.0 (IBM SPSS Statistics, Armonk, NY).
There was a 2:1 predilection of males to females. Over 76% (67/88) of the patients were afebrile, defined as a temperature < 38.5oC. Two patients had a positive culture recorded (Table 2).
One patient was positive for coagulase-negative S. aureus and the other was positive for a non-aureus Staphylococcus species. Table 3 summarizes the laboratory data for patients diagnosed with LA assembled from the available literature.
The average age of patients presenting with Lyme arthritis was 7.5 years (standard deviation (SD): 3.5 years; range: 1.8 - 20 years), the mean and standard deviation for erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) was 41 ± 20 mm/hr and 3.9 ± 5.9 mg/L, respectively. Peripheral white blood cell (WBC) count averaged 10.6 x 109cells/L (95% CI, 10.0 - 11.3) with the synovial WBC count averaging 55,888 cells/mm3 (95% CI, 47,533 - 64,242). The mean synovial polymorphonuclear cell percentage was 89% (95% CI, 86 - 92) in those for whom data were available.
The 95% CI for ESR was 21 - 33 mm/hr in those diagnosed with toxic synovitis (TS), 37-46 mm/hr for LA, and 44 - 64 mm/hr for SA (Figure 2).
Synovial WBC counts (cells/mm3) exhibited a similar trend. The 95% CIs were 5,644 - 15,388 cells/mm3 for TS, 47,533 - 64,242 cells/mm3 for LA, and 105,432 - 260,214 cells/mm3 for SA (Figure 3).
Lyme arthritis of the hip is a relatively rare entity and accounts for 5 - 18% of cases of acute, inflammatory, non-traumatic hip pain in children [15, 17]. Differentiating between SA, LA, and TS of the hip can be challenging, even for the most discerning clinician. While Lyme arthritis of the knee has been well studied, monoarticular LA of the hip is less common; yet, it remains in the differential diagnosis when evaluating an irritable hip in a pediatric patient presenting in endemic areas. Because of this, the purpose of the current investigation was to synthesize the available literature of monoarticular LA in order to improve the clinician's ability to arrive at an accurate and timely diagnosis in patients presenting with an acutely painful hip.
This systematic review identified 88 patients diagnosed with monoarticular LA of the hip. We found that less than a quarter (22.9%) of patients with LA were febrile at presentation compared to more than half (53.8%) of those with SA. Additionally, we found a significant difference in the refusal to bear weight among patients with either SA or TS compared to those diagnosed with LA. Of patients in whom the ambulatory status was recorded, one-third (33.3%) of patients with LA refused to bear weight, while 62.2% and 61.1% of patients with SA and TS, respectively, refused to do so. This supports the authors’ clinical experience that LA of the hip seems to be associated with a less acutely irritable clinical presentation compared to either SA or TS.
In addition to history and physical examination, laboratory evaluation is often performed in patients who present with acute hip pain suspicious for an infectious or inflammatory etiology. Initial laboratory measurements typically include CBC, ESR, and CRP. In our analysis, patients with LA presented with a mean peripheral WBC count of 10.6 x 109 cells/L (95% CI, 10.0 - 11.3) that was not significantly different than the mean serum WBC count in patients diagnosed with SA (12.4 x 109 cells/L; 95% CI, 10.8 - 14.0) or TS (mean 11.7 x 109 cells/L; 95% CI, 10.1 - 13.3). Therefore, our data suggests that peripheral WBC is not a reliable laboratory marker for distinguishing LA from TS or SA.
With regards to ESR, we found that the LA group had ESR (mean: 41 mm/hr; 95% CI, 37 - 46) values that were not significantly different compared to those with SA (mean: 54 mm/hr; 95% CI, 44 - 64); however, compared to the those with TS (mean: 27 mm/hr; 95% CI, 21 - 32), patients with LA had significantly elevated ESR values. Moreover, we found that the range of ESR 95% confidence intervals ascended in an increasing order of diagnostic severity from TS --> LA --> SA (Figure 2). Our results show that while serum WBC count does not help distinguish LA from SA or TS, the ESR may help differentiate patients who would warrant further laboratory testing with either serum Lyme titers or intra-articular hip aspiration.
Based on our data, we propose a diagnostic algorithm in which an ESR level ≥ 40 mm/hr warrants a hip aspiration to rule out SA in the absence of other symptoms or clinical parameters suspicious for the latter (Figure 4).
In other words, in a patient with an equivocal history, physical exam, and laboratory values, an ESR ≥ 40 should prompt the clinician to obtain hip synovial fluid to evaluate for SA. An ESR level < 40 mm/hr could be worked up further with Lyme serology as LA or TS become more likely. Of course, if significant hip irritability or other clinical signs of SA exist, then treatment for SA should be initiated. The authors stress that this proposed algorithm should be used merely as a guide to assist with clinical decision making and it is at the treating physician’s discretion as to the most appropriate diagnostic and treatment course. Figures 2-3 describe potential guideline numbers for ESR and peripheral WBC count when assessing a patient with an irritable hip. The figures illustrate the concept of these three diagnoses existing on a continuum of inflammation, with TS being the least inflammatory and SA the most inflammatory. This clinical picture is consistent with basic science literature supporting the idea that intra-articular neutrophils associated with SA are inherently different from those associated with other inflammatory etiologies [27-28], which seems to be consistent with the results of the present study.
After the aforementioned ESR level, the algorithm presented herein utilized a synovial neutrophil count threshold to assist the clinician in the decision to perform surgical drainage of the hip. As seen with ESR, the 95% confidence intervals for synovial WBC counts also increased with increasing disease severity. Moreover, there was no observed overlap in the 95% confidence intervals between LA and SA (Figure 3). Thus, we propose utilizing the upper end of the 95% confidence interval (65,000 cells/mm3) as a potential cutoff, with values above this number strongly suggestive of bacterial SA. Utilization of gram stain results should also be taken into account; however, the negative predictive value of gram stain results in the evaluation of acute SA limits its utility [29-30].
There are several limitations to this study. This was a systematic review of previously published articles which were retrospective investigations in and of themselves. Therefore, our study was subject to the limitations inherent in retrospective studies. When selecting studies to include in our meta-analysis, we were stringent in our criteria of having patient-level data available from prior studies in order to help determine (as best as possible from the available literature) clinically relevant parameters that may help clinicians distinguish the three different disease processes investigated. When not available from the published manuscript, we made every effort to contact the original study authors to gain access to the raw data from the primary investigation.
Because of the relatively low reported incidence of isolated LA of the hip, our methods allowed us to pool data from a variety of studies and make clinically meaningful comparisons between LA, SA, and TS. This study must be taken in context, however, as one must be cautious when pooling laboratory and clinical data from heterogeneous sources. Different laboratories may have different methods and different investigators may have defined clinical parameters, such as weight-bearing, differently. In addition to heterogeneous sources, there may have also been some bias introduced in this systematic review since three out of the 12 reviewed articles were reported from the same institution [16-17, 21]. This may have inflated the true incidence of LA in the studied sample. We also used a historical cohort for our comparative analyses, and these patients may be inherently different from those pooled together from the literature. We chose to do this primarily because the studies included in our systematic review lacked adequate comparison groups and these historical controls were evaluated at our home institution, which should help limit any variability in data collection. Finally, we must emphasize that the proposed diagnostic algorithm, as described, has not been prospectively validated and is based primarily on the authors’ experience and the findings of the current systematic review. Ideally, multicenter prospective studies or retrospective reviews of prospectively collected data, such as registries, would help counter the limitations of our investigation.
Monoarticular LA can be a cause of hip pain in certain geographic areas and has a clinical and diagnostic overlap with TS and acute bacterial SA. This study consolidates the available literature and represents the largest series of patients diagnosed with LA of the hip to date. We propose a diagnostic algorithm that serially incorporates ESR followed by synovial neutrophil count when evaluating pediatric patients with an irritable hip in Lyme endemic areas.
- Steere AC, Malawista SE, Snydman DR, et al.: Lyme arthritis: an epidemic of oligoarticular arthritis in children and adults in three connecticut communities. Arthritis Rheum. 1977, 20:7–17.
- Mead PS: Epidemiology of Lyme disease. Infect Dis Clin North Am. 2015, 29:187–210. 10.1016/j.idc.2015.02.010
- Rauter C, Hartung T: Prevalence of Borrelia burgdorferi sensu lato genospecies in Ixodes ricinus ticks in Europe: a metaanalysis. Appl Environ Microbiol. 2005, 71:7203–16. 10.1128/AEM.71.11.7203-7216.2005
- Bacon RM, Kugeler KJ, Mead PS; Centers for Disease Control and Prevention (CDC): Surveillance for Lyme disease--United States, 1992-2006. MMWR Surveill Summ. 2008, 57:1–9.
- Aiyer A, Hennrikus W, Walrath J, et al.: Lyme arthritis of the pediatric lower extremity in the setting of polyarticular disease. J Child Orthop. 2014, 8:359–65. 10.1007/s11832-014-0602-3
- Bachman DT, Srivastava G: Emergency department presentations of Lyme disease in children. Pediatr Emerg Care. 1998, 14:356–61.
- Milewski MD, Cruz AI Jr, Miller CP, et al.: Lyme arthritis in children presenting with joint effusions. J Bone Joint Surg Am. 2011, 93:252–60.
- Smith BG, Cruz AI Jr, Milewski MD, Shapiro ED: Lyme disease and the orthopaedic implications of lyme arthritis. J Am Acad Orthop Surg. 2011, 19:91–100.
- Sood SK: Lyme disease in children. Infect Dis Clin North Am. 2015, 29:281–94. 10.1016/j.idc.2015.02.011
- Baldwin KD, Brusalis CM, Nduaguba AM, Sankar WN: Predictive factors for differentiating between septic arthritis and Lyme disease of the knee in children. J Bone Joint Surg Am. 2016, 98:721–28. 10.2106/JBJS.14.01331
- Deanehan JK, Kimia AA, Tan Tanny SP, et al.: Distinguishing Lyme from septic knee monoarthritis in Lyme disease-endemic areas. Pediatrics. 2013, 131:695–701. 10.1542/peds.2012-2531
- Deanehan JK, Nigrovic PA, Milewski MD, et al.: Synovial fluid findings in children with knee monoarthritis in lyme disease endemic areas. Pediatr Emerg Care. 2014, 30:16–19. 10.1097/PEC.0000000000000028
- Thompson A, Mannix R, Bachur R: Acute pediatric monoarticular arthritis: distinguishing lyme arthritis from other etiologies. Pediatrics. 2009, 123:959–65. 10.1542/peds.2008-1511
- Willis AA, Widmann RF, Flynn JM, et al.: Lyme arthritis presenting as acute septic arthritis in children. J Pediatr Orthop. 2003, 23:114–18. 10.1097/00004694-200301000-00023
- Cruz AI Jr, Aversano FJ, Seeley MA, et al.: Pediatric Lyme arthritis of the hip: The great imitator?. J Pediatr Orthop. 2017, 37:355–61. 10.1097/BPO.0000000000000664
- Glotzbecker MP, Kocher MS, Sundel RP, et al.: Primary lyme arthritis of the pediatric hip. J Pediatr Orthop. 2011, 31:787–90. 10.1097/BPO.0b013e31822f1abc
- Bachur RG, Adams CM, Monuteaux MC: Evaluating the child with acute hip pain ("irritable hip") in a Lyme endemic region. J Pediatr. 2015, 166:407-11.e1. 10.1016/j.jpeds.2014.09.040
- Eichenfield AH, Goldsmith DP, Benach JL, et al.: Childhood Lyme arthritis: experience in an endemic area. J Pediatr. 1986, 109:753–58. 10.1016/S0022-3476(86)80688-6
- Miller A, Stanton RP, Eppes SC: Acute arthritis of the hip in a child infected with the Lyme spirochete. Clin Orthop Relat Res. 1993, 286:212–14.
- Saulsbury FT: Lyme arthritis in 20 children residing in a non-endemic area. Clin Pediatr (Phila). 2005, 44:419–21. 10.1177/000992280504400506
- Heyworth BE, Shore BJ, Donohue KS, et al.: Management of pediatric patients with synovial fluid white blood-cell counts of 25,000 to 75,000 cells/mm³ after aspiration of the hip. J Bone Joint Surg Am. 2015, 97:389–95. 10.2106/JBJS.N.00443
- Caird MS, Flynn JM, Leung YL, et al.: Factors distinguishing septic arthritis from transient synovitis of the hip in children. A prospective study. J Bone Joint Surg Am. 2006, 88:1251–57. 10.2106/JBJS.E.00216
- Kocher MS, Zurakowski D, Kasser JR: Differentiating between septic arthritis and transient synovitis of the hip in children: an evidence-based clinical prediction algorithm. J Bone Joint Surg Am. 1999, 81:1662–70.
- Amini B, Geller MD, Mathew M, Gerard P: MRI features of Lyme arthritis of the hips. Pediatr Radiol. 2007, 37:1163–65. 10.1007/s00247-007-0578-y
- Saulsbury FT: Lyme arthritis presenting as transient synovitis of the hip. Clin Pediatr (Phila). 2008, 47:833–35. 10.1177/0009922808318337
- Moak JH, Vaughan AJ, Silverberg BA: Ultrasound-guided hip arthrocentesis in a child with hip pain and fever. West J Emerg Med. 2012, 13:316–19. 10.5811/westjem.2011.10.6797
- Deirmengian C, Lonner JH, Booth RE Jr: The Mark Coventry Award: white blood cell gene expression: a new approach toward the study and diagnosis of infection. Clin Orthop Relat Res. 2005, 440:38–44. 10.1097/01.blo.0000185756.17401.32
- Ramilo O, Allman W, Chung W, et al.: Gene expression patterns in blood leukocytes discriminate patients with acute infections. Blood. 2007, 109:2066–77. 10.1182/blood-2006-02-002477
- Faraj AA, Omonbude OD, Godwin P: Gram staining in the diagnosis of acute septic arthritis. Acta Orthop Belg. 2002, 68:388–91.
- Stirling P, Faroug R, Amanat S, et al.: False-negative rate of gram-stain microscopy for diagnosis of septic arthritis: suggestions for improvement. Int J Microbiol. 2014, 2014:830857. 10.1155/2014/830857
Distinguishing Pediatric Lyme Arthritis of the Hip from Transient Synovitis and Acute Bacterial Septic Arthritis: A Systematic Review and Meta-analysis
Ethics Statement and Conflict of Interest Disclosures
Human subjects: Consent was obtained by all participants in this study. The Children's Hospital of Philadelphia Institutional Review Board (IRB) issued approval N/A. This study was exempt from formal IRB review. Systematic review and meta-analysis of previously published data. . Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.
Cite this article as:
Cruz, Jr. A I, Anari J B, Ramirez J M, et al. (January 25, 2018) Distinguishing Pediatric Lyme Arthritis of the Hip from Transient Synovitis and Acute Bacterial Septic Arthritis: A Systematic Review and Meta-analysis. Cureus 10(1): e2112. doi:10.7759/cureus.2112
Received by Cureus: December 29, 2017
Peer review began: January 08, 2018
Peer review concluded: January 17, 2018
Published: January 25, 2018
© Copyright 2018
Cruz, Jr. et al. This is an open access article distributed under the terms of the Creative Commons Attribution License CC-BY 3.0., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.