Workup for Knee Pain Before Cortisone Injection

Essential Initial Imaging

The American College of Radiology recommends plain radiographs as the first-line imaging modality for all patients with chronic knee pain being considered for corticosteroid injection, including standing anteroposterior, lateral, and sunrise views to evaluate for osteoarthritis severity, joint space narrowing, osteophyte formation, loose bodies, and prior osseous injury 1, 2
Plain radiographs should be used to assess for osteoarthritis, loose bodies, or structural abnormalities, and to evaluate for Kellgren-Lawrence grading, joint space narrowing, and osteophyte formation 1, 2, 3

The American College of Radiology recommends joint aspiration if an effusion is present on examination to analyze synovial fluid for cell count and differential, crystal analysis, and Gram stain and culture if infection is suspected 1, 2, 3
Injecting corticosteroids into an infected joint can lead to catastrophic outcomes, making joint aspiration a critical step before proceeding with corticosteroid injection 1

The American College of Radiology recommends MRI in specific circumstances, such as when radiographs show findings suggesting meniscal pathology, ligamentous injury, or osteochondral defects that might change management, or when clinical examination suggests internal derangement that may require surgical intervention rather than injection 1, 2, 3
Ultrasound can be used to detect synovial pathology and effusions, guide aspiration if needed, and demonstrate meniscal extrusion or chondrocalcinosis 1, 2

In patients over 70 years old, bilateral structural abnormalities on radiographs or MRI are common even with unilateral symptoms, making it difficult to discriminate painful from non-painful knees based on imaging alone, and the presence of synovitis, effusion, or Kellgren-Lawrence score ≥2 are the best discriminators between painful and asymptomatic knees in this age group 1, 2

The American Geriatrics Society recommends documenting medication history, particularly anticoagulation status, current NSAID use and response, prior corticosteroid injections, and all over-the-counter medications before proceeding with injection 4, 5, 6
Patients with diabetes must be counseled that corticosteroids can cause transient hyperglycemia for 1-3 days post-injection, and corticosteroid injection should be avoided within 3 months of planned knee replacement surgery due to theoretical infection risk 3, 5

The American Geriatrics Society recommends documenting trial of acetaminophen, topical NSAIDs or counterirritants, and physical therapy or structured exercise program before offering corticosteroid injection 4, 5, 6
Corticosteroid injection is particularly indicated when evidence of acute inflammation with joint effusion is present, systemic NSAIDs are contraindicated, or pain is refractory to non-invasive pharmacologic and non-pharmacologic therapies 3, 4, 6

The American College of Radiology recommends not injecting without ruling out infection if effusion is present, not ordering MRI routinely unless specific clinical findings suggest internal derangement that would change management, and not injecting within 3 months of planned arthroplasty due to infection risk 1, 2, 5
The American College of Radiology also recommends not repeating injections more frequently than every 3-4 months, as benefits typically last only 1-12 weeks, and being aware of dose-dependent risk of progression to total knee arthroplasty 3, 7, 8

The Mayo Clinic recommends counseling patients that pain relief from corticosteroid injection is short-term only, typically lasting 1-12 weeks, with no long-term improvement at 2-year follow-up, and that patients should avoid overuse of the injected joint for 24 hours following injection 5, 7, 8
The presence of effusion predicts better response to injection, and methylprednisolone improves pain and function at 4 and 24 weeks, while triamcinolone shows benefit at 6 weeks but not 12 weeks 5, 7, 8