The common causes of pain and effusion after total knee arthroplasty are infection, instability, loosening, and patellofemoral problems. Fehring et al. reported that the primary cause of failure in a series of 440 revision knee arthroplasties was infection and 63% of the revisions failed within five years after the operation1. In a review of 212 revision total knee arthroplasties performed at one center, Sharkey et al. reported that the three most common reasons for failure were polyethylene wear, aseptic loosening, and instability2.
A carefully recorded history and physical examination can often identify the cause of pain and effusion, especially in patients with instability. Routine radiographs can be used to evaluate the knee alignment and fixation status of the components. Laboratory studies, such as the erythrocyte sedimentation rate and C-reactive protein level, may be helpful in screening for infection3,4. Aspiration at the site of the prosthetic knee joint has been considered the single most helpful test in determining the presence of an early or late metastatic infection4,5. There have also been reports of acute crystalline-related synovitis at the sites of knee arthroplasties6,7.
We report an unusual cause of late pain and effusion due to hypertrophic pulmonary osteoarthropathy in a patient who had had a bilateral total knee arthroplasty. The patient was informed that data concerning the case would be submitted for publication and consented.
A seventy-seven-year-old woman presented in early February 2007 with a two-month history of pain and swelling in both knees. She had undergone a bilateral staged total knee arthroplasty with cement for the treatment of osteoarthritis in January and November 2001, without complications. Routine follow-up examinations and radiographs in December 2005 showed unremarkable findings. The onset of the pain and swelling was insidious, and there was no antecedent trauma. The patient was able to walk outdoors with a cane and did not report any fever or chills.
Nine months before the examination for the pain and swelling in the knees, hoarseness had developed and a chest radiograph made at another institution showed a left hilar mass. Mediastinoscopy and lymph node biopsy at that institution showed small-cell carcinoma. The patient had a vocal cord injection for the hoarseness, and a work-up for metastases revealed negative findings. After extensive discussion of treatment options with her oncologist, the patient decided against radiation or chemotherapy and wanted only supportive care.
On examination for the pain and swelling in the knees, the patient was afebrile and walked without a limp. The knees were both warm, without erythema, and both had effusions. There was tenderness to deep palpation about the distal parts of both femora. The range of motion was from full extension to 120° of flexion bilaterally. There was no instability or crepitus. Inspection of the hands showed clubbing of multiple digits. Radiographs showed periosteal new bone formation at the medial and lateral parts of the distal femoral metaphysis bilaterally (Fig. 1). There were no radiolucent lines and no evidence of osteolysis. The hemoglobin level was 122 g/L (normal, 120 to 160 g/L), the white blood-cell count was 12.3 × 109/L (normal, 4.5 to 11.0 × 109/L), the erythrocyte sedimentation rate was 82 mm/hr (normal, 0 to 30 mm/hr), and the C-reactive protein level was 6.3 mg/dL (normal, 0 to 100 mg/dL). As infection was suspected, the sites of the total knee arthroplasties were sequentially aspirated. Each knee had 20 mL of clear yellow fluid with a total nucleated cell count of 124/mm3 and 1% neutrophils, 76% lymphocytes, and 23% other cells. No crystals were seen. There was no growth after seven days of aerobic or anaerobic culture of the aspirate from either knee.
With the exclusion of infection, loosening, instability, and crystalline synovitis, we made the presumptive diagnosis of pulmonary hypertrophic osteoarthropathy of the distal parts of both femora. Indomethacin (25 mg three times per day) was prescribed, and there was complete relief of the pain and effusions within one week. The patient was evaluated six weeks later and, with the continued absence of symptoms, it was decided that the indomethacin therapy would be continued indefinitely.