28 March 2022

Investigating a painful swollen knee

Pain Rheumatology

Even though the patient has a history of rheumatoid arthritis, differentials for a monoarthritis need to be considered.


Welcome to the latest edition of Rheumatology Reasoning. My aim is to present a common case and work through the reasoning of why I order each test, what questions are most useful to help diagnose this condition and a simple approach to management.

The approach is practical and based on experience and evidence (where it exists). I hope you will find it useful in your day-to-day practice.

A 62-year-old active male presents with a painful, swollen right knee. It has gradually worsened over a couple of weeks and now the pain sometimes wakes him at night. He is able to walk and can continue working as a courier. He is systemically well and has no previous episodes of gout. He has a six-year history of seronegative rheumatoid arthritis that has been well controlled on methotrexate 25mg weekly. His initial rheumatoid arthritis presentation was polyarticular, especially affecting shoulders and knees. At the time of diagnosis, he could not lift his arms in the air and had very large bilateral knee effusions and was struggling to walk. CRP at the time of rheumatoid arthritis diagnosis in 2015 was 95g/L (N<5) and ESR was 82mm/hr (N<20). He also has gradually worsening osteoarthritis of both knees, especially the right, where there is a valgus deformity.

Even though he has a history of rheumatoid arthritis, a broad consideration of differentials for a monoarthritis needs to be considered. Differentials for a monoarthritis include inflammatory arthritis (rheumatoid arthritis in this case), crystal arthritis (gout or pseudogout), septic arthritis, osteoarthritis or hemarthrosis.

Let’s go through each one. Rheumatoid arthritis would be high on my list given his known history; however, his rheumatoid arthritis has been extremely well controlled for a long time and therefore this is less likely. Crystal arthritis usually presents with sudden onset pain and swelling, with patients frequently waking in the morning with severe pain. He has a gradual increase in symptoms over a couple of weeks, making crystal arthritis thus less likely. Bacterial septic arthritis should always be considered; however, his pain is not severe, it was of gradual onset, he is still able to move the knee and he is systemically well.

An infective cause with more indolent bacteria should be considered, and it is even possible that immunosuppression has led to an unusual fungal infection. Methotrexate monotherapy would be considered as relatively low-level immunosuppression by most rheumatologists. Hemarthrosis usually occurs in the setting of trauma or anticoagulation, neither of which was present. He does take aspirin 100mg daily as primary cardiovascular prevention, but it would be unusual for aspirin to cause a large hemarthrosis with no other contributing cause. Lastly, advanced osteoarthritis can cause prominent knee effusions and the gradual onset of pain could be consistent with this.

So, what is the best way to sort this out? The best test for a monoarthritis is a joint aspirate as it will exclude a number of the differentials.

Frequently I see blood tests and imaging (X-ray and ultrasound) ordered as the initial tests. While they are less invasive and may provide some clues, it rarely pinpoints the precise cause of monoarthritis in the patient sitting in front of you. For instance, a high serum uric acid level does not diagnose gout. Uric acid is falsely low in half of cases during an acute attack, and a high serum uric acid does not mean the current knee effusion is due to gout. An ultrasound will show a joint effusion (which you can discern from clinical examination) but not much else. X-ray or MRI is most useful if there has been trauma, where a fracture or ligamentous injury is possible. The presence of osteoarthritis or chondrocalcinosis (faint speckling in the joint line where CPPD crystals deposit in the menisci) won’t help you determine the cause of the effusion as they may be pre-existing. Finally, a raised CRP or ESR can occur with rheumatoid arthritis, crystal arthritis or septic arthritis.

If septic arthritis is considered a strong possibility, the patient should not be given empirical antibiotics. The patient needs a joint aspirate prior to antibiotics and should be sent to the emergency department if you are unable to promptly perform a joint aspirate. Antibiotics prior to a joint aspirate can neutralise the synovial fluid culture and makes it very difficult to exclude sepsis.

I performed a joint aspirate and obtained 40mL of straw-coloured joint fluid. My clinical assumption prior to the joint aspirate was that rheumatoid arthritis or osteoarthritis is the likely cause. As such, I injected the knee with corticosteroids (40mg methylprednisolone) at the time of the aspirate. I would advise against a steroid injection if there is any concern of septic arthritis. It is imperative the fluid is sent for laboratory analysis. The joint fluid showed white cell count 389 and extracellular calcium pyrophosphate crystals. Cultures are negative.

As a reminder, a white cell count greater than 2000 is classified as inflammatory. So, he has a non-inflammatory joint aspirate with CPPD crystals.

I feel this is most likely due to his advanced medial compartment knee osteoarthritis. Pseudogout (acute inflammatory arthritis secondary to CPPD crystals) usually presents with sudden-onset joint swelling (knee and wrist are the most common) with an inflammatory joint aspirate. Some argue that intracellular CPPD is helpful in diagnosing acute pseudogout; however, it is not a reliable enough feature to use this alone. CPPD crystals are a common “innocent bystander” in a joint aspirate, in that they are seen in the synovial fluid but are not the cause of the clinical problem. This is more likely to be the case in the elderly, in the presence of osteoarthritis and in certain conditions such as hyperparathyroidism or chronic kidney disease. The presence of chondrocalcinosis on X-ray (the radiological form of CPPD) would warrant you to consider CPPD disease as a cause, but it is not diagnostic alone.

Knee X-ray in this patient showed grade 4 medial compartment osteoarthritis and the presence of chondrocalcinosis. His knee pain and effusion settled after the steroid injection. I diagnosed a “flare” of osteoarthritis and did not think the CPPD crystals were the cause of the effusion. He is a stoic man and will be delaying a knee replacement for a long as possible.