Acute crystal arthritis mimicking infection after total knee arthroplasty

Total knee arthroplasty is the most effective treatment to relieve the pain associated with end stage arthritis of the knee.¹ A reported 0.5-2% of people will develop sepsis of the implant, necessitating lavage and debridement of the affected knee with immediate or delayed removal of the implant.²,³ The key to successful management of implant sepsis is early and accurate diagnosis, which allows prompt treatment. Every patient with pain and swelling at the site of a total knee arthroplasty must, therefore, be assessed for infection.¹ We report a case of a patient initially diagnosed as having implant sepsis who in fact had pseudogout of the knee. This shows the diagnostic difficulties and issues surrounding management that may arise when a patient presents with crystal arthritis at the site of previous joint replacement.

Case report

A previously well 72 year old man presented with a three day history of progressive swelling of the left knee with associated pain, erythema of the skin, tachycardia, and a fever of 38.8°C. Two years before, the patient had had primary total knee arthroplasty surgery of the affected joint for a diagnosis of osteoarthritis. Examination found a reduced range of movement at the knee joint, restricted by pain and swelling. All other joints were normal. Radiographs of the knee were unremarkable and showed no evidence of the implant loosening. Erythrocyte sedimentation rate was raised at 120 mm in the first hour and C reactive protein was raised at 333 mg/l. Full blood count and differential found a total white cell count of 18×10⁹/l, and a neutrophil count of 12.0×10⁹/l. Serum electrolytes, hepatic enzymes, bone markers, and urate were all within normal concentrations. Arthrocentesis was done and 60 ml of turbid synovial fluid was aspirated, which was sent for microbiological and biochemical analysis. Initial Gram stain analysis of the synovial fluid was negative. Microscopy showed a synovial fluid white cell count of 60 000 cells × 10⁶/l, with a neutrophil population in excess of 90%. In addition to this, numerous rhomboid shaped crystals were noted, which were positively bifringent under polarised light. The patient had no previous history of crystal arthropathy and in view of the clinical history was admitted with a presumptive diagnosis of implant sepsis.

Later that evening, the patient was taken to theatre for open lavage and debridement. The knee was explored through an anterior approach, and the components seemed well fixed. No frank pus was seen within the joint, and the knee was irrigated thoroughly. Culture of all preoperative joint aspirates, blood cultures, and all intraoperative specimens remained negative for bacterial and fungal growth. Microscopic examination of a specimen of synovial tissue showed a large number of positively bifringent crystals consistent with calcium pyrophosphate dihydrate. After surgery, antibiotics were discontinued and the patient started taking 50 mg diclofenac sodium three times a day. After seven days, the patient had made a full recovery and was discharged to be reviewed at the outpatient clinic.

Discussion

The case shows the diagnostic difficulties that may arise when a person presents with acute crystal arthritis at the site of a previous joint arthroplasty. Both septic and crystal arthritis may present similarly, with joint pain, effusion, erythema, fever, leucocytosis, and raised serum inflammatory markers. In both conditions, analysis of synovial fluid obtained by arthrocentesis typically shows a raised white cell count with a differential neutrophil count typically in excess of 90%.⁴ The diagnosis of crystal arthritis is normally based on the identification of monosodium urate or calcium pyrophosphate dihydrate crystals in synovial fluid examined under polarising light microscopy. Unlike gout, no known serum biochemical markers for pseudogout exist.⁵ Most cases of pseudogout are idiopathic, although it has been associated with ageing, trauma, hyperparathyroidism, Gitelman's syndrome, and haemochromatosis.⁴ The most common sites of deposition of calcium pyrophosphate dihydrate crystals are the knee, wrist, and shoulder.⁶ Non-steroidal anti-inflammatory drugs are the mainstay of treatment, and the prognosis for resolution of acute attack is excellent.⁴

Acute crystal arthritis (acute gout or pseudogout) of a prosthetic joint is a relatively rare occurrence, and, as such, only a few cases have been reported.^7-11 Although it is of prime importance to exclude implant sepsis, diagnosis of acute crystal arthritis must be excluded as this condition can be treated by simple pharmacological means. Preceding episodes of gout or pseudogout or predisposing medical conditions should raise the possibility of crystal induced arthritis. Physical examination should include a detailed survey to exclude other involved joints. Immediate joint aspiration is indicated in all cases of suspected infected joint arthroplasty and should be among the first diagnostic tests done.¹²,¹³,¹⁴ The aspirate should routinely be sent for Gram stain analysis; bacterial and fungal cultures; white cell count and differential; and biochemical and crystal analyses.¹⁵ Synovial fluid crystal analysis, Gram stain, and appropriate cultures are the only reliable means of distinguishing between the acute crystal and septic arthritis. Failure or delay in recognising acute gout or pseudogout can lead to inappropriate treatments, such as the unnecessary use of antibiotics, surgical arthrotomy, or removal of the implant. Acute crystal arthritis should be excluded by polarising light microscopy of synovial fluid aspirate before removing implant components.