Which diagnostic tests should be performed for suspected septic arthritis of the knee?

Diagnostic Testing for Suspected Septic Arthritis of the Knee

Begin with plain radiographs of the knee, followed immediately by arthrocentesis with synovial fluid analysis including cell count with differential, Gram stain, and culture—these are the essential diagnostic tests for suspected septic arthritis. 1

Initial Imaging: Plain Radiographs

• Obtain knee radiographs first as the initial imaging study for all suspected cases of septic arthritis 1
• Radiographs may show soft tissue swelling, joint effusion, or exclude alternative diagnoses (fractures, tumors) 1
• Early septic arthritis (<14 days) often shows normal radiographs or only mild soft tissue swelling, so normal films do not rule out infection 1
• Radiographs provide anatomic overview and help interpret subsequent advanced imaging if needed 1

Essential Laboratory Testing: Arthrocentesis

Synovial fluid analysis is the cornerstone of diagnosis and should be performed urgently in all suspected cases 1, 2

Synovial Fluid White Blood Cell Count

• Most powerful diagnostic test available for septic arthritis 2, 3, 4
• WBC >50,000/μL significantly increases probability (likelihood ratio 7.7) 2
• WBC >100,000/μL is highly suggestive (likelihood ratio 28.0) 2
• WBC >64,000-65,000/μL provides optimal sensitivity (40%) and specificity (90%) 5, 6
• WBC <25,000/μL substantially decreases probability (likelihood ratio 0.32) 2

Synovial Fluid Differential

• Polymorphonuclear cells ≥90% increases likelihood (likelihood ratio 3.4) 2
• PMN <90% lowers probability (likelihood ratio 0.34) 2

Synovial Fluid Gram Stain

• Positive Gram stain is highly specific when bacteria are visualized (odds ratio 21.5) 5
• Most accurate test for ruling in septic arthritis alongside synovial WBC count 3

Synovial Fluid Culture

• Gold standard for definitive diagnosis and guides antibiotic selection 1, 2
• Allows identification of causative organism and antibiotic sensitivities 1
• Positive in approximately 75% of cases when obtained before antibiotics 4

Crystal Analysis

• Presence of crystals substantially reduces probability of septic arthritis (odds ratio 0.1) 5
• Essential to exclude gout or pseudogout as alternative diagnoses 7

Serum Laboratory Tests

Serum inflammatory markers have limited diagnostic utility but should be obtained:

• Serum WBC count is not discriminatory—64% of septic arthritis patients have normal serum WBC 6
• ESR and CRP are not useful acutely for diagnosis 4
• These markers are more useful for monitoring treatment response than initial diagnosis 4, 8

Clinical Assessment Limitations

History and physical examination alone cannot reliably diagnose or exclude septic arthritis 2, 3, 4:

• Fever is present in only 57% of cases 2
• 77% of patients are afebrile at presentation 6
• Joint pain (sensitivity 85%) and swelling (sensitivity 78%) are common but nonspecific 2
• Risk factors that increase probability: recent joint surgery (likelihood ratio 6.9), skin infection over prosthetic joint (likelihood ratio 15.0), history of prior septic arthritis (odds ratio 5.0), clinical effusion (odds ratio 4.8), symptom duration >2 days (odds ratio 6.9) 5, 4

Advanced Imaging (When Initial Tests Are Equivocal)

Image-Guided Aspiration

• Use ultrasound or CT guidance when arthrocentesis is technically difficult or to confirm proper needle placement 1
• Imaging guidance reduces vascular complications and prevents contamination of adjacent tissues 1

MRI

• Not recommended for initial evaluation but may be useful when diagnosis remains uncertain after arthrocentesis 1
• Excellent for evaluating soft tissue extent, abscess formation, and adjacent osteomyelitis 1
• High sensitivity for detecting marrow signal changes and soft tissue complications 1

CT with IV Contrast

• Not recommended for initial evaluation 1
• May help differentiate cellulitis, myositis, and abscess when diagnosis is unclear 1
• Most sensitive for detecting soft tissue gas 1

Tests to Avoid

• 3-phase bone scan has insufficient evidence and poor specificity for septic arthritis 1
• Synovial lactate is inaccurate and should not be relied upon 3
• PCR testing is inaccurate for diagnosis in the emergency setting 3

Critical Pitfalls

• Do not delay arthrocentesis based on normal serum inflammatory markers 4, 6
• Do not rely on fever or elevated serum WBC to rule in or rule out septic arthritis 3, 6
• Do not withhold aspiration due to anticoagulation—the diagnostic benefit outweighs bleeding risk 1
• Obtain cultures before antibiotics whenever possible to maximize yield 4
• Consider multivariable prediction models (Septic Arthritis Risk Calculator) that incorporate multiple clinical and laboratory variables for more accurate risk stratification 5, 9