Out of the Woods and into the Lyme-light: Lyme Arthritis

Lyme disease is transmitted via a bite from the Ixodes or black-legged tick (with an attachment period of 36 to 48 hours) and infection by the spirochete Borrelia burgdorferi or Borrelia mayonii. With only 25% of patients recognizing a tick bite, it is not surprising that many individuals go on to develop symptoms of the disease.[2] The clinical presentation is similar in adults and children, and the illness is classically divided into three stages: early localized disease, early disseminated disease, and late presentations of the disease.

Table 1: Stages of Lyme Disease Stages[12]

Within 3 to 30 days, the early manifestations begin, with the most common being the classic bull’s eye rash (erythema migrans). Among 275,589 cases (208,834 confirmed and 66,755 probable) from 2008 to 2015, the CDC reported that 70% of the cohort manifested the classic rash of Lyme disease.[3] 

For those without erythema migrans (EM) or misdiagnosed EM, other common early disseminated symptoms mimic viral infections, with children reporting fatigue, headache, arthralgias/myalgias, fevers, gastrointestinal symptoms, and sore throat.[4]

Lyme arthritis refers to the symptomatic aftermath of spirochetal invasion into the joints, triggering an accumulation of neutrophils, immune complexes, complement factors, and cytokines within the synovial fluid as part of the ensuing inflammatory response.[5]

Lyme arthritis typically presents as swelling in the knees, but it might involve the shoulders, ankles, elbows, wrists, or temporomandibular joint. These patients generally have pain out of proportion to the level of swelling. The joint will appear very swollen and have large effusions, but patients will complain of only a minimal amount of pain with passive movement of the joint. The majority of patients tend to develop this late complication approximately six months after the initial disease onset, a pattern consistent with our observed patient; for some, the swelling and pain might fluctuate over the course of years.[6-8]

Lyme arthritis exhibits a higher incidence among pediatric cases, aligning its occurrence closely with that of EM and reaching its peak between the ages of 10 and 14, as displayed in the CDC table below.[3]

Notably, the historical presence of a tick bite does not consistently serve as a reliable indicator, as three of 17 patients recalled a rash, and only four remembered a tick bite.[9]

Lyme arthritis might be easy to miss if you are not given the right information. In our case, the parents were not initially forthcoming with the recent history of Lyme disease until specifically questioned. Septic arthritis of the knee is a “cannot-miss diagnosis” and should be at the top of your differential.

Other pathologies to always consider with isolated joint pain/swelling are traumatic such as dislocations, fractures, meniscus and ligament tears, tendon ruptures, and tendonitis. Also consider overuse injuries such as Osgood-Schlatter disease, patellofemoral syndrome, and pes anserine bursitis. 

Infectious considerations to remember are overlying soft tissue infections, osteomyelitis, acute rheumatic fever, post-streptococcal arthritis, reactive arthritis from recent viral or bacterial illnesses, or even septic arthritis of the hip with referred pain to the knee. Lastly, always consider malignancy. 

Given the overlapping clinical presentations, it becomes imperative to distinguish Lyme arthritis from septic arthritis. Septic arthritis is considered an orthopedic emergency, with the classic presentation being a febrile patient with a warm, red, painful, swollen joint with limitations in both active and passive range of movement. In Lyme arthritis, a good physical examination may reveal minimal pain with movement, as observed in our patient. Additionally, the presence of polyarthritis may provide valuable diagnostic clues favoring Lyme disease over septic arthritis.[6]

The diagnosis of Lyme arthritis is made with blood testing, and there are different combinations of serological testing - ultimately, you’ll use test for IgG and IgM for Lyme disease. You should see positive IgG-positive serologies in the setting of a late manifestation, but you may not see a positive IgM. It is also important that you get a history of prior Lyme exposure, as IgM and IgG may continue to be positive for years.[10]

The decision to perform arthrocentesis to rule out septic arthritis should be made on a case-by-case basis. If the patient has the pathognomonic EM rash followed by the development of arthritis, then it is reasonable to defer arthrocentesis and await serologies while empirically treating Lyme disease, especially in the absence of systemic signs of infection or significant pain on passive range of motion.

One decision tool has been proposed to help determine the need for arthrocentesis in Lyme-endemic regions.

Lyons et al. conducted a study at eight pediatric EDs located in Lyme endemic areas. They included patients 1 to 21 years old who presented with monoarthritis and also had blood work ordered during their ED evaluation. The primary outcome was the diagnosis of a musculoskeletal infection. The authors found that among their cohort of 735 children, none of the children with septic arthritis had both a negative procalcitonin and negative CRP. For those children with a negative procalcitonin but a positive CRP, a positive/equivocal first-tier Lyme C6 enzyme immunoassay test ruled out septic arthritis.[11] The graph below describes lab cutoffs and their ability to differentiate musculoskeletal infections caused by a bacteria defined as polymyositis, septic arthritis, or osteomyelitis (MSKI = musculoskeletal infection).

This study had a large sample size and prospective design, however, there were several limitations. It only included those who presented and had these specific biomarkers obtained, which led to the loss of ~12.5% of their initial sample size; this can introduce selection bias and impact the generalizability of their findings. Whether a patient had monoarthritis was also left up to the discretion of the treating physician, which can lead to issues with interrater reliability. Furthermore, knee monoarthritis made up the majority (68%) of the joints affected, so their rule may be less applicable to other joints. The patient selection was also specific to a lyme endemic region, so the findings are not generalizable to regions with a lower prevalence of Lyme disease. Ultimately, the serology-based algorithm is not yet ready for “prime time” to dictate management, and currently, I would not rely on it to rule out septic arthritis.

The treatment for Lyme arthritis in children is straightforward and outlined below.[5]

For patients < 8 years old, the general practice is to avoid use of doxycycline, and if the patients have persistent symptoms, either a second course of the same antibiotic or observation alone is reasonable.

While we didn’t have lab work, we were in a Lyme endemic area and had a clear history of a tick bite and erythema migrans, an X-ray that didn’t show a traumatic reason for the swelling, positive serologies for Lyme, and the development of arthritis in the most commonly Lyme-affected joint in children. The patient was treated with amoxicillin for 28 days for Lyme monoarthritis.