Serious musculoskeletal infections in children receiving anti-tumor necrosis factor-α therapy: a case series
- First Online:
- Cite this article as:
- Morishita, K., Petty, R., Cairns, R. et al. Clin Rheumatol (2010) 29: 677. doi:10.1007/s10067-010-1410-x
- 84 Views
Anti-tumor necrosis factor-α (anti-TNF-α) agents are widely used to treat children with juvenile idiopathic arthritis (JIA) whose disease is resistant to conventional therapy. Although generally well tolerated, use of these agents has been associated with an increased risk of infection. In particular, in patients treated with anti-TNF-α agents, there is an increased susceptibility to infection by intracellular organisms such as tuberculosis, and common infections may present atypically or be more severe. We report four cases of serious musculoskeletal infections among 31 children with JIA being treated with anti-TNF-α agents, two of which were secondary to group A β-hemolytic Streptococcus.
KeywordsAnti-tumor necrosis factor-αBiologicsInfectionsJuvenile idiopathic arthritis
Tumor necrosis factor-α (TNF-α) is a pro-inflammatory cytokine that has an important role in the pathogenesis of juvenile idiopathic arthritis (JIA) [1, 2]. TNF-α also plays a critical role in host defense against infections, particularly intracellular granulomatous infections such as tuberculosis [3–5]. Anti-TNF-α agents are being increasingly used in the management of patients with JIA who are resistant to conventional therapy [6–8]. Although these agents are generally well tolerated, and often very effective, they have been associated with an increased risk of infection [5, 9–13]. In addition to intracellular infections, an altered immune response may predispose patients to uncommon presentations and more severe infections caused by common organisms . The following four cases highlight this potential risk.
Clinical features and medications at the time of infection
Subtype of JIA
Polyarthritis, RF –
Polyarthritis, RF –
Disease duration (years)
Current active arthritis
20 [0.3 mg/kg/week]
15 [1 mg/kg/week]
25 (subcutaneous) [0.4 mg/kg/week]
15 [0.2 mg/kg/week]
6 mg/kg every 8 weeks
6 mg/kg every 6 weeks
Duration on (months)
Laboratory investigations at the time of infection
WBC (× 109/L)
Platelets (× 109/L)
A 5-year-old girl with JIA, RF negative polyarthritis, was being treated with etanercept and methotrexate with good effect (Table 1). Low-dose prednisone had been discontinued 2 weeks prior to admission. She had never received any intraarticular corticosteroid injections. She had been on azithromycin for 2 days for otitis media. She presented with fever, lethargy, and acute pain and swelling of the left elbow. She appeared unwell and was found to have a swollen, warm left elbow and a large right knee effusion. Results of laboratory studies are summarized in Table 2. She was intravenously treated with ceftriaxone, vancomycin, and stress doses of hydrocortisone. Her blood culture, throat swab, aspirate from left elbow and right knee, and swab from the ear all grew GABHS. She improved on IV antibiotics and was switched to penicillin taken orally upon discharge. She has subsequently been treated with infliximab without complication.
A 15-year-old girl with JIA, systemic arthritis, being treated with etanercept, methotrexate, naproxen, prednisone, and amitriptyline, presented with a painful enlarging lump on her left inner thigh. There was no history of recent fevers or symptoms suggesting other infection. Initially, she was orally treated with cephalexin, which was shortly thereafter switched to cloxacillin. Two days later worsening symptoms required an incision and drainage of the abscess. No organism was cultured from the abscess. The etanercept was subsequently discontinued. Results of laboratory studies are summarized in Table 2.
Three of the four patients had formal immunologic assessments to exclude the possibility of an underlying immune deficiency. Case 4 was evaluated clinically by an immunologist but did not have laboratory investigations done. Lymphocyte phenotyping demonstrated normal B, T, and natural killer cell numbers. Adaptive immune function was intact with all patients having normal lymphocyte proliferation, normal serum immunoglobulins, and the ability to make functional antibodies to vaccine antigens. Complement assays including CH50, C3, and C4 were performed on Cases 1 and 3 and were normal. Case 1 also had a normal oxidative burst neutrophil dihydrorhodamine (DHR-123) test to rule out chronic granulomatous disease. None of the four patients had a prior history of serious or unusual infections.
Anti-TNF-α therapy has become an important therapeutic option for treating patients with JIA who are not responsive to conventional disease-modifying anti-rheumatic drugs. One of the major concerns regarding the use of this therapy has been the increased risk of infection. Randomized trials in patients with rheumatoid arthritis have shown inconsistent results [15–17], but were often underpowered to detect specific risks. Postmarketing surveillance and observational studies have suggested an increased risk of serious infections [5, 10–12]; however, such studies usually lack an adequate control group to infer causal relationships . A systematic review and meta-analysis of randomized clinical trials was performed by Bongartz et al. to address the limitation of small sample size and found an increased risk of serious infection in anti-TNF-α-treated patients compared to placebo (odds ratio 2.0, 95% CI, 1.3–3.1) . Another meta-analysis performed by Leombruno et al. did not identify an increased risk for patients receiving recommended doses of anti-TNF-α therapy .
There are little published data on infectious complications of anti-TNF-α therapy in pediatric patients with rheumatic diseases. Elwood and colleagues reported a case of multifocal septic arthritis and osteomyelitis caused by GABHS in a 12-year-old girl with chronic arthritis receiving etanercept . Other reported cases include: a deep abscess caused by methicillin-resistant Staphylococcus aureus in an 11-year-old girl with polyarticular JIA receiving etanercept ; pyomyositis caused by S. aureus in an 18-year-old male with Behcet’s disease receiving infliximab ; histoplasmosis in a 17-year-old female with JIA receiving infliximab ; and two cases of extrapulmonary tuberculosis in a 5- and 7-year-old both with systemic onset JIA (one on etanercept and the other on infliximab) [22, 23]. In a recent multicenter series of the use of biologic treatment in children with primary systemic vasculitis, six out of 25 patients experienced significant infections while on biologic therapy . In the major pediatric efficacy trials of anti-TNF-α agents, viral infections, particularly mild upper respiratory tract infections, were the most frequently reported infectious adverse event [6–8, 25]. Bacterial sepsis and serious musculoskeletal infections have only been reported rarely [25, 26]. Even in the absence of anti-TNF-α blockade, serious musculoskeletal infections in children with chronic inflammatory arthritis are uncommon .
The most disconcerting feature of all the patients in our series was their apparently “well” appearance disproportionate to the severity of the infection. For example, case 1 presented with a very brief history of fever, malaise, and poorly localized leg pain. She had no history of pharyngitis. Despite prompt initiation of intravenous antibiotics she developed an extensive subperiosteal abscess and osteomyelitis requiring surgical debridement 48 h after admission. The severity and extent of infection were surprising given the brevity of symptoms, negative bone scan, and prompt initiation of therapy.
The cause of multifocal septic arthritis in case 2 was also GABHS. There was history of a preceding otitis media 2 days earlier but no pharyngitis despite a positive throat culture taken during the hospital admission. Multifocal septic arthritis caused by GABHS is uncommon, with only a few reported cases in the pediatric literature [14, 28–30]. The case reported by Elwood et al. was very similar to this case and provides another example of a severe and complicated course of a common infection.
In cases 3 and 4, no causative agents were identified, although case 3 had received a course of amoxicillin for a presumed upper respiratory tract infection a few weeks earlier. In case 3, the finding of frank pus from the initial ultrasound guided aspirate was consistent with infection despite a lack of other infectious signs such as fever or overlying warmth or redness of the area. Her inflammatory markers were elevated at presentation in the absence of any signs or symptoms to suggest a flare of her underlying systemic JIA.
C-reactive protein (CRP) values were obtained in three of the four patients and were markedly elevated in all three cases at the time of hospital admission. Compared to the patients' baseline levels and levels during active disease, these levels were significantly higher in all instances. In our series, the CRP values increased our index of suspicion of a serious underlying infection, particularly in cases 1 and 3, where the patients did not appear especially unwell.
The occurrence of four serious musculoskeletal infections in 31 children with chronic arthritis being treated with anti-TNF-α agents in a single center is a cause for concern. In two cases, the causative agent was GABHS. None of the patients had evidence of an underlying immunodeficiency. Physicians caring for children receiving anti-TNF-α therapy should maintain a high index of suspicion for serious infection and be aware that presenting signs and symptoms may be atypical in these patients. In addition, the diagnosis of a musculoskeletal infection in a patient with chronic arthritis may be delayed because the initial symptoms might be attributed to the underlying active disease.