
Arthrocentesis
Definition: Bacterial infection of a joint space and its synovial fluid
Causes
- Hematogenous spread of bacteria to the joint capsule
- Most common mechanism of infection
- 60-90% of cases are Gram positive (S. aureus = 50-70%)
- 5-25% of cases are Gram negative
- Direct inoculation (e.g. recent arthrocentesis, prior joint surgery)
Epidemiology
- Bimodal distribution of cases, peaking for young children and adults >55 years old
- Nongonococcal:
- Intravenous drug use (IVDU)
- Prosthetic joint
- Diabetes
- Immunocompromised states
- Elderly
- Indwelling IV catheters
- Gonococcal
- Sexually active population
- 4:1 female-to-male predominance
- Consider Gram negative, fungal, or mycobacterial arthritis in immunocompromised patients and elderly
Complications
- Cartilage destruction requiring joint replacement
- Osteomyelitis
- Sepsis
- Immunocompromised patients may decompensate rapidly
Differential Diagnosis for monoarticular arthritis
- Gout / pseudogout
- Trauma
- Cellulitis of overlying joint
- Bursitis / tendonitis
- Reactive arthritis
- Lyme arthritis
- Autoimmune disease (SLE, RA, seronegative spondyloarthropathy, etc.)
- Hemarthrosis secondary to coagulopathy
- Osteoarthritis
History
- Risk group from Epidemiology section above
- Large joints more commonly affected (knee most common)
- >80% monoarticular
- Fever (44-97% sensitive) (Carpenter 2011)
- Chronic joint disease (e.g. gout) does not rule out concomitant septic arthritis
- Sexual history should be obtained
- Gonococcal – usually oligoarticular arthritis (generally with rash, migratory arthralgias, tenosynovitis)
Physical Exam
- Joint pain and effusion
- Erythema and warmth
- Tenderness on palpation
- Limited in both active and passive range of movement
- Immunocompromised patients
- Often do not present typically
- May be well-appearing or afebrile
- More likely to have a polyarticular involvement
Diagnostic Testing
- Laboratory Analysis
- Serum lab tests cannot be used to rule in or out the diagnosis
- Suspicion for Gonococcal disease: obtain genital/oral/rectal Gonorrhea and Chlamydia specimens (PCR, Culture), screen for concomitant STIs and HIV
Ranges of Serum Lab Value Probabilities (Margaretten 2007, Horowitz 2011)
+ LR | – LR | Sensitivity (%) | Specificity (%) | |
WBC >10,000 cells/mm3 | 1.4 (1.1-1.8) | 0.28 | 90 | 36 |
ESR >30 mm/h | 1.3 (1.1-1.8) | 0.17 | 95 | 29 |
CRP >100 mg/L | 1.6 (1.1-2.5) | 0.44 (0.24-0.82) | 77 | 53 |
Blood culture
(non-gonococcal) |
50% positive * | |||
Blood culture
(gonococcal) |
Seldom positive * |
- Imaging
- The role of X-ray is limited. It can assess for:
- Joint/bony destruction
- Trauma or foreign body
- Not sensitive for osteomyelitis or other complications
- Findings suggestive of other arthropathies (e.g. osteophytes) do not rule out septic arthritis
- The role of X-ray is limited. It can assess for:
-
Knee Arthrocentesis (medpagetoday.com)
Arthrocentesis
- Relative contraindications:
- Overlying cellulitis: if septic joint is suspected, arthrocentesis should still be performed through cellulitic-appearing skin
- Coagulopathy: if septic joint is suspected, arthrocentesis should still be performed (Thumboo 1998)
- Prosthetic joint: if no orthopedic surgeon is available and septic joint is suspected, arthrocentesis should still be performed by the ED provider
- Do not need to wait for serum lab study results prior to arthrocentesis
- Most important synovial fluid studies: Cell count, Gram stain, aerobic and anaerobic cultures
- Relative contraindications:
Synovial Fluid Interpretation
Normal | Inflammatory | Crystal-Induced | Septic | |
Color | Clear/yellow | Yellow/white | White/turbid | Cloudy/opaque |
Viscosity | Thick, sticky | N/A | N/A | Thin, water-like |
Synovial WBC | <25,000/mm3 | <100,000/mm3 | >50,000/mm3 | >50,000/mm3 |
Synovial PMN | <30% | >50% | <90% | >90% |
Other | Crystals seen | Culture positive |
Pooled Joint Aspirate WBC Probabilities (Carpenter 2011, Margaretten 2007)
+ LR | – LR | Sensitivity (%) | Specificity (%) | Study | |
WBC >25,000 cells/mm3 | 3.2 (2.3-4.4) | 0.35 (0.23 – 0.50) | 73 (64-81) | 77 (73-81) | Carpenter 2011 |
2.9 (2.5-3.4) | 0.32 | 77 | 73 | Margaretten 2007 | |
WBC >50,000 cells/mm3 | 4.7 (2.5-8.5) | 0.52 (0.38-.72) | 56 (49-63) | 90 (88-92) | Carpenter 2011 |
7.7 (5.7-11.0) | 0.41 | 62 | 92 | Margaretten 2007 | |
WBC >100,000 cells/mm3 | 13.2 (3.6-51.1) | 0.83 (0.80 – 0.89) | 19 (14-20) | 99 (96-100) | Carpenter 2011 |
28.0 (12.0-66.0) | 0.72 | 29 | 99 | Margaretten 2007 | |
>90% PMN | 2.7 (2.1-3.5) | 0.49 | 60 (51-68) | 78 (75-80) | Carpenter 2011 |
3.4 (2.8-4.2) | 0.34 | 73 | 79 | Margaretten 2007 | |
Lactate >10 mmol/L | Infinity | 0.14 | 86 | 100 | Carpenter 2011 |
WBC > 1,100 cells/m3 (prosthetic joint) | 7.6 | 0.10 | 91 | 88 | Ghanem 2008 |
WBC PMN > 64% (prosthetic joint) | 17.9 | 0.53 | 95 | 95 | Ghanem 2008 |
- Prosthetic joints have lower diagnostic threshold (Ghanem 2008, Tande 2014)
- WBC > 1,100 cells/mm3 (+LR 7.6) OR
- >64% PMNs (+LR 17.9)
- Gram stain is only positive in 50-60% of infected joints
- Synovial culture is positive in >90% of nongonoccal septic arthritis but only positive in 25-70% of gonococcal septic arthritis (Horowitz 2011)
Treatment Thresholds
- There is no absolute serum or synovial laboratory value to rule in or out septic arthritis
- Decision to initiate treatment is based on overall clinical suspicion derived from
- Patient risk factors
- Clinical findings
- Synovial fluid results
- Low threshold for workup and treatment: (Carpenter 2011)
- If clinical suspicion is 5%, further testing should be pursued
- If probability of septic arthritis is above 39%, treatment should be initiated
- Dry or non-diagnostic tap
- If suspicion for septic joint, start antibiotics and admit
- Presence of crystals in synovial fluid does not rule out septic arthritis
- Suspect septic joint in gout patient with: (Abelson 2010)
- Fever
- Involvement of an new/atypical joint
- New/worsening synovitis despite adherence to outpatient treatment
- Suspect septic joint in HIV+ patient with: (Takhar 2010)
- IVDU – likely MRSA
- Risky sexual behaviors – likely gonococcal
- CD4 <200 – possibly fungal or mycobacterial
- Abnormally low synovial cell counts are possible
ED Management
- Provide analgesia
- Investigate for primary source if hematogenous spread suspected
- Antibiotic management
- Based on suspicion for particular organism
- Gram positive:
- Vancomycin 15 mg/kg IV q12h OR
- Linezolid 600mg q12h
- Gram negative/Gonococcal:
- Ceftriaxone 2g IV q24h OR Cefotaxime 2g IV q8h OR Ceftazidime 1g IV q8h PLUS
- Gentamycin 5mg/kg IV q8h for Pseudomonas coverage (e.g. IVDU)
- Orthopedic surgery consultation for possible operative debridement or hardware removal
- Gonococcal infection often managed non-operatively
Take Home Points
- Septic arthritis is a critical diagnosis that should be considered in all patients with mono- or oligoarticular arthritis
- Be aware that immunocompromised patients often present atypically with septic arthritis. Have a low threshold to obtain a joint aspirate
- Serum laboratory testing and imaging cannot be used to rule in or out the diagnosis. Do not skip the arthrocentesis because these tests are unremarkable
- The decision to start treatment in the ED for septic arthritis is based on overall clinical suspicion derived from patient’s risk factors, clinical picture, and synovial fluid cell count
- If suspicion is still high after equivocal or dry tap, admit the patient and initiate empiric IV antibiotics while the synovial culture results
References
- Liang SY et al. Orthopedic infections and other complications. Orthopedic Emergencies. Cambridge University Press 2013. 178-180.
- Matteson EL, Osmon DR. Infections of Burae, Joints, and Bones. Goldman’s Cecil Medicine, 24th Edition. Elsevier Saunders 2012. 1731-1736.
- Genes N, Chisolm-Straker M. Monoarticular Arthritis Update: Current Evidence for Diagnosis and Treatment in the Emergency Department. Emergency Medicine Practice. EBMedicine.net. 2012; 14(5): 1-20.
- Carpenter CR et al. Evidence based diagnostics: adult septic arthritis. Acad Emerg Med. 2011; 18(8): 781-796. PMID 21843213.
- Margaretten ME et al. Does this patient have septic arthritis? JAMA. 2007; 297:1478-88. PMID 17405973.
- Horowitz DL et al. Approach to Septic Arthritis. Am Fam Physician. 2011; 84(6): 653-660. PMID 21916390.
- Visser S. Septic until proven otherwise. Can Fam Physician. 2009; 55(4): 374–375. PMCID PMC2669005.