To the Editor:

Our growing understanding of the role of cytokines, especially tumor necrosis factor α (TNFα), in the intraarticular inflammation of rheumatoid arthritis (RA) and psoriatic arthritis (PsA) has allowed for directed therapy against these cytokines with biologic agents such as infliximab and etanercept. Treatment of persistent monarticular and oligoarticular arthritis, resistant to disease-modifying antirheumatic drugs (DMARDs) and glucocorticoids, with intraarticular anti-TNF therapy appears to be a promising and attractive option. The efficacy and safety of intraarticular administration of infliximab and etanercept in patients with RA have recently been reported (1–4). We report the case of a 35-year-old white man with persistent monarticular PsA in whom a significant postinjection effusion developed after the second of two 25-mg doses of intraarticular etanercept was administered.

The patient was diagnosed as having PsA in 1994, when he presented with right great toe dactylitis. Subsequently, he reported occasional flares in both knees that were relieved by arthrocentesis and glucocorticoid injection. The patient refused use of any DMARDs for treatment of the arthritis. Beginning in July 2001, a persistent left knee effusion developed. Aspiration of the joint fluid showed a white blood cell (WBC) count of 15,000 without any evidence of crystals, and negative acid-fast bacilli, bacterial, and fungal cultures. The effusion persisted despite 3 arthrocenteses and 3 intraarticular injections of methylprednisolone (80 mg each). Magnetic resonance imaging performed in July 2002 confirmed a large effusion as well as synovial proliferation, without evidence of avascular necrosis. On physical examination, a large, tense left knee effusion was present, causing decreased extension to −10°. The knee was not erythematous, and there was minimal increase in warmth. The Westergren erythrocyte sedimentation rate (ESR) was 24 mm/hour.

High-resolution ultrasonography of the suprapatellar bursa revealed a large effusion and synovial proliferation measuring 1.4 cm in height. Fronds of synovium as well as rice bodies were detected throughout the suprapatellar bursa (compared with a normal right suprapatellar bursa). Arthrocentesis of the left knee yielded straw-colored fluid with a WBC count of 20,000 and no evidence of crystals or infection. At this time, 25 mg of intraarticular etanercept was administered under ultrasound guidance. The patient's symptoms decreased dramatically within hours and were minimal within days after the injection.

Physical examination 1 week later revealed decreased left knee effusion and normal range of motion. The ESR was 11 mm/hour. Ultrasonography of the suprapatellar bursa revealed a moderate effusion with decreased synovial thickness of 0.75 cm. There was a 1-cm × 1-cm area of mild dermal erythema at the injection site, similar to a typical injection site reaction.

Two weeks after intraarticular etanercept was administered, the patient reported the return of pain. Physical examination revealed a moderate left knee effusion with full range of motion. The ESR was 8 mm/hour. A second ultrasound-guided intraarticular injection of etanercept, 25 mg, was given. Several hours after the injection, the patient noted an increase in left knee pain and decreased range of motion associated with a tense effusion. Increased warmth was noted on physical examination. Synovial fluid (60 cc) was aspirated, which had a WBC count of 870 (33% neutrophils, 11% lymphocytes, 51% monocytes, 5% eosinophils). The patient received two 500-mg doses of ciprofloxacin over a period of 24 hours. Bacterial, acid-fast bacilli, and fungal stains (and subsequently cultures) were negative. Intraarticular methylprednisolone (80 mg) was administered.

One week later the patient reported decreased pain, and synovial thickness in the suprapatellar bursa was 0.60 cm on high-resolution ultrasound. Three weeks after the second intraarticular injection of etanercept, the patient had full range of motion of the left knee, no articular warmth, and moderate swelling. Ultrasonography displayed enlarged fronds of synovium throughout the suprapatellar bursa.

Prior to our experience, the only reported adverse event associated with intraarticular injection of etanercept was minor swelling on the dorsum of the hand, as observed by Bliddal et al (2, 3) in a patient with RA who received an injection in the wrist.

Unlike the patients described by other investigators, our patient had an adverse event after receiving a second intraarticular injection of etanercept (25 mg). We postulate that the development of effusion and pain in our patient 4 hours after receiving the medication represents an intraarticular reaction similar to the subcutaneous injection site reactions reported in 42% of patients receiving subcutaneous etanercept in controlled trials (5, 6). As suggested by Zeltser et al (7), injection site reactions may represent an example of a T lymphocyte–mediated delayed-type hypersensitivity reaction, with eventual induction of tolerance. Clinically, the reaction is characterized by erythema, edema, pruritus, or pain and occurs within the first several hours after the injection is administered. Histologically, the lesion is characterized by edema and vasodilation in the reticular dermis accompanied by perivascular cuffing with mononuclear cells (eosinophils, neutrophils, and macrophages are also present).

Another possible explanation for the patient's effusion and pain is the formation of antibody to etanercept resulting in immune complex formation and a subsequent arthus-like reaction. This mechanism could also account for the paucity of WBCs (n = 870) in the patient's synovial fluid.

In conclusion, there has been great interest in the use of intraarticular anti-TNF therapy for treatment of resistant monarticular and even oligoarticular arthritis. We used previous observations as a basis for treating our patient who had a persistent large knee effusion attributable to PsA. This letter is intended to alert physicians that an acute intraarticular injection site reaction may be a potential side effect of intraarticular etanercept.

  • 1
    Dreher R, Flaig W, Leitzke D. Treatment of rheumatoid arthritis by intra-articular injections with TNF alpha blockers. Proceedings of the 2002 Annual European Congress of Rheumatology; 2002 Jun 15–20, Stockholm, Sweden.
  • 2
    Bliddal H, Qvistgaard E, Terslev L, Savnik A, Holm CC, Kristoffersen H, et al. Injection of etanercept into arthritic joints. II: dose-response and efficacy. Proceedings of the 2002 Annual European Congress of Rheumatology; 2002 Jun 15–20, Stockholm, Sweden.
  • 3
    Bliddal H, Terslev L, Qvistgaard E, Kristoffersen H, Torp-Pedersen S, Danneskiold-Samsoe B. Injection of etanercept into arthritic joints. I: safety. Proceedings of the 2002 Annual European Congress of Rheumatology; 2002 Jun 15–20, Stockholm, Sweden.
  • 4
    Osborn TG. Intraarticular etanercept versus saline in rheumatoid arthritis: a single injection double blind placebo controlled study. Arthritis Rheum 2002; 46 Suppl 9: S518.
  • 5
    Weinblatt ME, Kremer JM, Bankhurst AD, Bulpitt KJ, Fleischmann RM, Fox RI, et al. A trial of etanercept, a recombinant tumor necrosis factor receptor: Fc fusion protein, in patients with rheumatoid arthritis receiving methotrexate. N Engl J Med 1999; 340: 2539.
  • 6
    Moreland LW, Schiff MH, Baumgartner SW, Tindall EA, Fleischmann RM, Bulpitt KJ, et al. Etanercept therapy in rheumatoid arthritis: a randomized, controlled trial. Ann Intern Med 1999; 130: 47886.
  • 7
    Zeltser R, Valle L, Tanck C, Holyst MM, Ritchlin C, Gaspari AA. Clinical, histological, and immunophenotypic characteristics of injection site reactions associated with etanercept: a recombinant tumor necrosis factor α receptor: Fc fusion protein. Arch Dermatol 2001; 137: 8939.

Erin L. Arnold MD*, Dinesh Khanna MD*, Harold Paulus MD*, Mark P. Goodman MD†, * University of California at Los Angeles, † Beverly Hills, CA.