The acuity of the patient’s presentation, including significant swelling of the affected joint and height of the inflammatory markers, were suggestive of a septic arthritis, of which Staphylococcus aureus is the major pathogen in children.1 However, Gram stain and culture from joint fluid, along with his initial blood culture, were all negative. On the third day of hospitalization, his Lyme disease screening enzyme immunoassay was positive with an index value of 4.26 (positive at >1.2); Lyme Western blot IgM was positive for 2 of 3 bands, and IgG was strongly positive with 10/10 bands. Based on this result and the clinical picture, the patient was diagnosed with Lyme arthritis.
Lyme disease is caused by the spirochete Borrelia burgdorferi and transmitted by Ixodes deer ticks. Localized disease is characterized by erythema migrans and flu-like symptoms, typically within 2 weeks of infection.1 Early disseminated disease, occurring weeks to months after infection, can include secondary erythema migrans, seventh cranial nerve palsy and less frequently, carditis, aseptic meningitis and radiculoneuritis.1 An intermittent, episodic oligoarthritis of large joints is the most common late manifestation of untreated B. burdorferi, though it may also present as an acute monoarthritis.1,2 The number of counties in the United States with a high incidence of Lyme disease continues to grow.3
The initial clinical features of this case were not consistent with a typical septic arthritis. At presentation, the patient was afebrile, constitutionally well and able to bear weight on the affected knee; he also had substantial symptomatic improvement with joint aspiration alone. In a recent pediatric study, patients with septic arthritis were more likely to have a fever history (79% vs. 41%), not tolerate any weight bearing (46% vs. 25%) and have pain with short-arc joint movement (74% vs. 6%) compared with those with Lyme arthritis.4
The patient’s significantly elevated inflammatory markers but low initial synovial fluid leukocytes, however, created an uncertain picture, which could be consistent with early septic arthritis. Average C-reactive protein levels in septic arthritis are >10 mg/dL, while levels in Lyme arthritis are <4 mg/dL.4,5 While synovial leukocyte counts in septic arthritis are frequently >100,000 cells/mL, they may be as low as 10,000 cells/mL; the range for Lyme arthritis is similarly wide with 53,600 ± 38,700 cells/mL reported in a series of 140 patients.4 The presentations of both septic and Lyme arthritis can vary considerably and have overlapping laboratory results, limiting clinicians’ ability to discriminate.
Specific radiologic features, however, may be a useful method to differentiate Lyme disease from septic arthritis. In a blinded study comparing knee MRI images of children with Lyme arthritis to those with septic arthritis, subcutaneous edema was present in all 7 patients with septic arthritis and only 1 of 11 with Lyme disease.6 Myositis and local adenopathy were present in all 11 Lyme patients, but only in 2 of 7 with septic arthritis. There was no difference in joint effusion size, synovial thickness or marrow edema.6 Based on these findings, the combination of adenopathy, myositis and absence of subcutaneous edema on MRI strongly suggests the diagnosis of Lyme arthritis. In our case, MRI demonstrated inflammation of the vastus lateralis and medialis muscles (Figure 1) and an enlarged superior popliteal lymph node (Figure 2). Subcutaneous edema was strikingly absent.
Regarding additional diagnoses beyond staphylococcal and other typical bacterial arthritides, Bartonella henselae infection (cat-scratch disease) and juvenile idiopathic arthritis were also considered. Musculoskeletal involvement in B. henselae infection has generally been underappreciated and occurs in 10% of cases.7 Nevertheless, arthropathy is primarily seen in adults, with osteomyelitis more common in children.7 Juvenile idiopathic arthritis may first present as a monoarthritis, but usually along with spiking fevers and other systemic features. Also, MRI findings of lymphadenopathy and myositis would be unexpected for juvenile idiopathic arthritis, whereas enthesitis would be common.8
Upon furthering questioning, the patient did report that he had something that looked like an insect bite near his knee recently; nevertheless, he could not recall the exact timing or location of this, and he reaffirmed that there was never any associated erythematous rash. Interestingly, his father also had abrupt, severe knee (and other joint) swelling and had been diagnosed with Lyme arthritis about 2 months earlier. Notably, neither Central nor Western New York are considered high-incidence areas for Lyme disease, but this patient was one in a cluster of acute Lyme arthritis cases encountered at Women and Children’s Hospital of Buffalo in the late summer and early fall of 2016.
Based upon the positive Lyme Western blot, which was available 2 days after arthroscopy, our patient was started on doxycycline. At 2 weeks into therapy, though he could walk freely and had no complaints, his knee flexion was still limited. By 28 days into therapy, the limitation in knee flexion had resolved, but he still had swelling of the superior knee joint and musculature/tissue above the knee. This, along with residual elevation of his CRP and erythrocyte sedimentation rate (3.9 mg/dL and 20 mm/h, respectively), led to continuation of the antibiotic for a total of 56 days, reflecting Infectious Diseases Society of America treatment guidelines.9 After that time, his examination still showed some residual swelling in the previously mentioned areas, but he was fully active without compromise, running around and playing football; he has had no relapse through 8 months of further monitoring.
This case highlights that Lyme disease may mimic acute septic arthritis, but patients are less often febrile or systemically ill and can have distinguishing MRI findings. Early recognition of Lyme arthritis may spare invasive procedures and expedite optimal management.
1. Gutierrez KLong SInfectious and Inflammatory Arthritis. In: Principles and Practice of Pediatric Infectious Diseases. 2012: Chapter 79.London: Elsevier Inc;
2. Hu LLyme arthritis. Infect Dis Clin North Am. 2005;19:947–961.
3. Kugeler KJ, Farley GM, Forrester JD, et alGeographic distribution and expansion of human Lyme disease, United States. Emerg Infect Dis. 2015;21:1455–1457.
4. Baldwin KD, Brusalis CM, Nduaguba AM, et alPredictive factors for differentiating between septic arthritis
and Lyme disease of the knee in children. J Bone Joint Surg Am. 2016;98:721–728.
5. Thompson A, Mannix R, Bachur RAcute pediatric monoarticular arthritis: distinguishing lyme arthritis
from other etiologies. Pediatrics. 2009;123:959–965.
6. Ecklund K, Vargas S, Zurakowski D, et alMRI features of Lyme arthritis
in children. AJR Am J Roentgenol. 2005;184:1904–1909.
7. Maman E, Bickels J, Ephros M, et alMusculoskeletal manifestations of cat scratch disease. Clin Infect Dis. 2007;45:1535–1540.
8. Sheybani EF, Khanna G, White AJ, et alImaging of juvenile idiopathic arthritis: a multimodality approach. Radiographics. 2013;33:1253–1273.
9. Worsmer GP, Ddattwyler RJ, Shapiro E, et alThe 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;43:1090–1134.