Septic joint: Reminders, Updates, and Pitfalls
- May 9th, 2015
- Brit Long
Authors: Matthew A. Heimann, MD (EM Resident Physician, University of Alabama at Birmingham Hospital) and Kevin Barlotta, MD (Attending Physician and Associate Professor of EM, University of Alabama at Birmingham Hospital) // Editors: Alex Koyfman, MD (@EMHighAK, EM Attending Physician, UTSW / Parkland Memorial Hospital) and Brit Long, MD (EM Resident Physician at SAUSHEC; USAF)
The presentation of an acutely painful, warm, and/or swollen joint requires emergent evaluation. In the absence of trauma, the differential is broad ranging from benign to potentially life-threatening. Bacterial arthritis, often used synonymously with septic arthritis, portends the most destructive course and should therefore bear weight on a clinician’s diagnostic considerations.
Septic arthritis typically involves one joint, but can be polyarticular in up to 20% of cases. A predilection for large joints exists with as many as 60% of cases involving the knee or the hip (1). The most common source of infection is hematogenous spread of bacteria from bacteremia. The migration of bacteria into a joint is facilitated by the lack of a basement membrane (or barrier) in the synovial tissues. Other causes of septic arthritis include direct inoculation from either penetrating trauma or iatrogenic introduction of bacteria during procedural intervention and contiguous spread of a local infection including osteomyelitis, septic bursitis, and/or abscess. Differentiating the source of inflammation or pain as articular (within the joint) or periarticular (outside the joint) is an important step in preventing iatrogenic inoculation. Predisposing risk factors for developing septic arthritis underscore these three routes of entry and can be found in TABLE 1.
TABLE 1: Risk Factors for Septic Arthritis (2,3):
- Bacteremia or systemic infection
- Intravenous (IV) drug use
- Overlying skin infection
- Diabetes Mellitus
- Rheumatoid arthritis
- HIV or other immunocompromised state
- Recent joint surgery or procedure
- Prosthetic joint
- Age > 80 years
Although history alone cannot reliably identify or exclude a septic joint, all patients with a joint-related complaint should receive an age-appropriate, targeted history aimed at identifying risk factors associated with septic arthritis (TABLE 1). Patients typically present with a constellation of symptoms including pain, tenderness, swelling, warmth, redness, fever, painful or limited range-of-motion, and immobility. Pain is the most common complaint. A recent review found that in patients diagnosed with septic arthritis, 80% endorsed pain in the joint. In contrast, only 50% of patients were found to be febrile, making the presence of fever a poor indicator of septic arthritis (2).
It is important to establish both a time of onset and progression of symptoms. Although patients are often unreliable in this regard, sudden onset of pain and rapid progression can indicate infection or crystal-induced pathology, while a more protracted course may indicate a less destructive process. Exacerbations of pre-existing disease (e.g. osteoarthritis or gout arthropathy) with similar pain and successful treatment regimens can aid in decision-making and support non-infectious etiologies.
A joint should be examined for obvious deformity, injury, warmth, tenderness, effusion, stability, laxity, range of motion, and pain during both active and passive movement. Careful examination should attempt to determine whether the inflammation is located in the joint capsule (intra-articular) or in the surrounding soft tissue structures (peri-articular) including the bursa, tendons, and skin. This can be difficult to differentiate on exam as joint effusions may not be readily identified on exam, and patients with bursitis may present with inflammation that mimics septic arthritis. Classically, the presence of generalized tenderness with painful limitation of both active and passive range of motion often indicates true joint involvement (3). In contrast, focal tenderness and pain limited to specific movements on active range of motion testing is more typical of periarticular inflammation.
Extra-articular physical exam findings may provide clues to underlying pathology and are important to note on evaluation. For example, the presence of a pustular rash and urethritis may indicate gonococcal arthritis, while patients with longstanding gout may have tophi (deposits of uric acid) on exam. In addition, it is important to note that joint pain may be referred from internal structures (e.g. shoulder pain in the presence of diaphragmatic irritation).
Five major imaging modalities can assist in the work-up of septic arthritis. Plain radiographs are often more helpful in evaluating chronic conditions than acute, atraumatic monoarthritis. They may identify the presence of soft tissue swelling or a joint effusion (e.g. displaced fat pad) as well as bony lesions including osteomyelitis and malignancy. In the later stages of joint disease, they can demonstrate degenerative changes associated with joint inflammation and identify changes specific to certain disease entities (chondrocalcinosis in calcium pyrophosphate dihydrate (CPPD) crystal deposition disease). Ultrasound has a high sensitivity in screening for joint effusions and can aid in safe joint aspiration. Computed tomography (CT) has the power to evaluate joints that are less available for physical exam (e.g. sacroiliac joint, sternoclavicular joint, and the hip). CT has superior sensitivity for effusions, edema, and the presence of abscess formation compared to plain radiographs but can be unreliable early in the disease course. Magnetic resonance imaging (MRI) provides greater resolution for evaluating soft tissues than the other imaging methods and is dependable early in the disease course. Finally, radionuclide scanning offers another modality for pinpointing areas of increased inflammation (4). All of these radiologic studies lack specificity and fail to differentiate septic arthritis from other inflammatory counterparts.
Traditionally clinicians have attempted to utilize the serum markers of white blood cell count (WBC), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) levels to aid in differentiating infectious from non-infectious etiologies of joint pain. However, many studies have questioned whether these markers are sufficiently sensitive to “rule out” septic arthritis and report poor specificity for joint inflammation. Most currently available studies have been performed in post-operative orthopedic patients and result from case-series reporting. As a result, generalizability to the undifferentiated patient population of an emergency department has been questioned and most recommendations caution against using these markers exclusively to exclude the diagnosis of septic arthritis. Hariharan and Kabrhel demonstrated greater than 90% sensitivity for septic arthritis when low thresholds of ESR and CRP levels were used in an emergency department setting, but concluded that further testing to establish clinical significance was required (5). In addition, patients with confirmed septic arthritis have been found to have normal ESR and CRP levels (6).
In the pediatric population, where the hip is a common site for infection, many studies have attempted to establish criteria to distinguish transient synovitis from septic arthritis. The Kocher criteria incorporates both ESR and serum WBC levels and predicts that septic arthritis can be excluded in the absence of four criteria (non-weight bearing, ESR > 40, serum WBC >12K, and fever) (7). In addition, Paakkonen et al. reported sensitivities of 94 and 95% for ESR and CRP levels respectively and a combined sensitivity of 98% in a pediatric population. Repeat markers were obtained over the following three days and no cases of septic arthritis were identified when ESR and CRP levels remained normal (8). Of note, these studies were performed in immunocompetent patients from developed countries and underscore the importance of incorporating risk factors for septic arthritis in decision-making.
As the most common mechanism for bacterial infection of a joint is hematogenous spread from underlying bacteremia, blood cultures may aid in identification of the causative organism and have been shown to be positive in 25 – 50% of patients with septic arthritis (1,6). Serum uric acid levels are not helpful in diagnosing acute exacerbations of gout and have been shown to normalize during acute attacks (9).
The clinical presentation of septic arthritis is often difficult to distinguish from other causes of acute arthritis and may require synovial fluid analysis when infection is suspected. Arthrocentesis performed by an experienced clinician under sterile preparation can yield valuable information with limited risk of injury. Synovial fluid should be inspected for color, clarity, and viscosity and sent for laboratory analysis including WBC count with differential, crystalline analysis, and Gram stain & culture. Synovial fluid culture is considered the standard to establish the diagnosis of septic arthritis. Synovial fluid can be categorized as normal, non-inflammatory, or inflammatory depending on bedside inspection and laboratory analysis.
While normal synovial fluid is typically clear and transparent with WBC counts < 200 cells/mm3 and < 25% PMNs, non-inflammatory synovial fluid is typically straw colored and translucent with 200 – 2,000 WBC/mm3 and < 25% PMNs. Gram stain is negative in both normal and non-inflammatory synovial fluid.
Inflammatory synovial fluid encompasses both crystalline and non-crystalline joint disease as well as infectious pathology. Unfortunately there is significant overlap in WBC counts and predominance of PMNs across all causes of inflammatory synovial fluid, making interpretation of the synovial fluid complex. While markedly elevated WBC counts can be indicative of infection and likelihood ratios (LR) for septic arthritis increase as synovial fluid WBC counts rise, lower WBC counts cannot exclude the presence of septic arthritis. Furthermore, low WBC counts can be found in early presentations of septic arthritis while markedly elevated WBC counts can be found in noninfectious causes of joint inflammation (RA, gout, and pseudogout). Although helpful, cell counts should not be used in isolation to rule in or rule out septic arthritis in a given clinical setting. Instead, decision-making should be guided by a combination of historical and physical exam findings coupled with objective data obtained during a patient’s evaluation.
Traditionally, synovial fluid with a WBC count > 50,000/mm3 with a polymorphonuclear (PMN) cell count > 90% have been associated with infectious arthritis. However, in culture-proven septic arthritis, the WBC count only reached this level in 50 – 75% of cases (10). A synovial fluid WBC count >100,000/mm3 has been shown to be more specific than the traditional 50,000/mm3 and carries a LR of 28 (9). Synovial lactate, glucose, and protein measurements have not been shown to be helpful in the evaluation of septic arthritis, though studies are currently underway evaluating synovial lactate.
A positive Gram stain can be diagnostic; however, a negative result for bacteria does not necessarily rule out bacterial arthritis (sensitivity only 50%). Blood cultures in patients with septic arthritis can be positive in up to 50% of patients, underscoring the most common source of infection: hematogenous spread. Consideration should be given to specific culture media when gonococcal, mycobacterial, or fungal infections are suspected.
The presence of crystals on synovial fluid evaluation can be helpful in identifying underlying disease (e.g. gout, pseudogout); however, the presence of crystals should not exclude the diagnosis of septic arthritis. The two disease states can occur simultaneously (11).
Contraindication and procedural considerations: It has been long espoused that arthrocentesis should not be performed through cellulitis. Dooley performed a broad literature search that could not find any case reports of adverse events from aspirating through cellulitis. The only definitive contraindications include frank abscess or purulence (12). Furthermore, it is safe to perform synovial fluid sampling in patients who are therapeutically anticoagulated (13). Prosthetic joint arthrocentesis should be performed in consultation with the operating surgeon. It is not uncommon for a clinician to encounter a dry tap or failure to aspirate synovial fluid. Several techniques can be employed to augment success in instances of dry taps. Switching to a larger gauge needle with a smaller syringe (preferably 5-10 mL) coupled with compression of the contralateral joint gutter (an attempt to shift any fluid toward the inserted needle) and gentle rotation of the needle bevel may aid in success (14).
Bacterial arthritis can be subdivided broadly into gonococcal or nongonococcal arthritis. Gonococcal arthritis, caused by Neisseria gonorrhoeae, is the most common cause of septic arthritis in the sexually active patient population. Clinicians should be inclined to obtain separate throat, urethral, cervical, and/or anal cultures as indicated to aid in source identification. Risk factors for gonococcal arthritis include pregnancy, menstruation, and patients with complement deficiencies. Gonococcal arthritis is classically described as a migratory polyarthritis, may involve several joints (wrist, knee and ankles), and may include a distinctive rash. There is a 4:1 female to male predominance and a lower incidence of long-term joint pathology.
Nongonococcal bacterial arthritis has a bimodal distribution affecting young children and adults over the age of 55 years. The most common cause of nongonococcal septic arthritis is a Gram-positive organism (75-90%), with other less common sources of infection including Gram-negative bacilli (10-20%), anaerobes, mycobacterium, and fungal infections. Ten percent of cases are polymicrobial (4). Staphylococcus aureus (S. aureus) is the most common cause of septic arthritis, and recent reports have identified methicillin-resistant S aureus (MRSA) contributing to an escalating number of cases of community-associated septic arthritis. Adding further complexity to the interpretation of synovial fluid analysis, patients with MRSA-associated septic arthritis were shown to have lower synovial fluid WBC counts, with an average of close to 15,000 cells/μL (range 3,400 to 34,075 cells/μL) (15). MRSA remains a leading cause of prosthetic joint infections.
Early identification and prompt treatment of bacterial arthritis is essential to minimize morbidity and mortality. Treatment algorithms include empiric antibiotic coverage (typically based on Gram stain results and suspected pathogens based on the patient’s clinical presentation). Hospital admission is indicated for continued IV antibiotics and drainage of the infected joint using one or a combination of needle aspiration and arthroscopic or open surgical drainage techniques.
The following represents a simplified treatment outline (16).
|Host / Gram stain results||Antibiotic Choice|
|Immunocompetent host with Gram positive organisms||15-20 mg/kg IV Vancomycin load|
|Immunocompetent host with Gram negative organisms or no visualized organisms||2 gm IV Ceftriaxone or 500 mg IV Imipenem; Add 15-20 mg/kg IV Vancomycin at discretion of provider|
|At risk hosts (major co-morbidities, immunocompromised or iv drug users)||15-20 mg/kg IV Vancomycin load + 2 gm IV Ceftriaxone or 400 mg IV Ciprofloxacin or 500 mg IV Imipenem|
Intolerance to vancomycin has prompted investigators to assess the efficacy of additional antibiotic agents with strong Gram-positive coverage. Although data is limited, several agents under investigation including, linezolid, daptomycin, quinupristin-dalfopristin and ceftaroline have been used with some success (17).
Septic arthritis presents clinicians with a significant diagnostic challenge. Early identification and prompt treatment of bacterial arthritis is essential to minimize morbidity and mortality. Despite a long history of reports identifying classic symptoms at presentation, characteristic historical elements, physical exam findings on evaluation, and traditional parameters on serum and synovial studies during laboratory testing, septic arthritis remains a diagnostic conundrum.
- Failing to consider septic arthritis in a oligoarticular/polyarticular presentation
- Relying solely on history, physical examination, ESR, and CRP to rule out septic arthritis
- Not recognizing non-bacterial etiologies of septic arthritis
- Allowing an uncomplicated cellulitis or therapeutic INR to thwart joint aspiration
- Not considering MRSA in initial antibiotic choice
- Not considering the overlap in clinical presentation and both serum and synovial studies in differentiating septic and noninfectious arthritis
- MRSA presenting with low synovial fluid cell counts
- Not performing arthrocentesis of a prosthetic joint in consultation with the treating surgeon
References / Further Reading:
- Mathews CJ, Coakley G. Septic arthritis: current diagnostic and therapeutic algorithm. Curr Opin Rheumatol. 2008;20(4):457-462.
- Margaretten ME, Kohlwes J, Moore D, Bent S. Does this adult patient have septic arthritis? JAMA. 2007;297(13):1478-1488.
- Baker DG, Schumacher HR. Acute monoarthritis. N Engl J Med. 1993;329(14):1013-1020.
- Shirtliff ME, Mader JT. Acute septic arthritis. Clin Microbiol Rev. 2002;15(4):527-544.
- Hariharan P, Kabrhel C. Sensitivity of erythrocyte sedimentation rate and C-reactive protein for the exclusion of septic arthritis in emergency department patients. J Emerg Med. 2011;40(4):428-431.
- Weston VC, Jones AC, Bradbury N, Fawthrop F, Doherty M. Clinical features and outcome of septic arthritis in a single UK Health District 1982-1991. Ann Rheum Dis. 1999;58(4):214-219.
- McGillicuddy DC, Shah KH, Friedberg RP, Nathanson LA, Edlow JA. How sensitive is the synovial fluid white blood cell count in diagnosing septic arthritis? Am J Emerg Med. 2007;25(7):749-752.
- Pääkkönen M, Kallio MJ, Kallio PE, Peltola H. Sensitivity of erythrocyte sedimentation rate and C-reactive protein in childhood bone and joint infections. Clin Orthop Relat Res. 2010;468(3):861-866.
- Genes N, Adams B. Arthritis. In: Marx J, ed. Rosen’s emergency medicine : concepts and clinical practice. 8th ed. Philadelphia, PA: Saunders; 2014:1501-1517.
- Brannan SR, Jerrard DA. Synovial fluid analysis. J Emerg Med. 2006;30(3):331-339.
- Yu KH, Luo SF, Liou LB, et al. Concomitant septic and gouty arthritis–an analysis of 30 cases. Rheumatology (Oxford). 2003;42(9):1062-1066.
- Dooley DP. Aspiration of the possibly septic joint through potential cellulitis: just do it! J Emerg Med. 2002;23(2):210.
- Ahmed I, Gertner E. Safety of arthrocentesis and joint injection in patients receiving anticoagulation at therapeutic levels. Am J Med. 2012;125(3):265-269.
- Roberts WN. Primer: pitfalls of aspiration and injection. Nat Clin Pract Rheumatol. 2007;3(8):464-472.
- Frazee BW, Fee C, Lambert L. How common is MRSA in adult septic arthritis? Ann Emerg Med. 2009;54(5):695-700.
- Burton J. Acute Disorders of the Joints and Bursae. In: Tintinalli J, ed. Tintinalli’s Emergency Medicine. 7th ed. New York, NY: McGraw-Hill; 2011:1926-1933.
- Sharff KA, Richards EP, Townes JM. Clinical management of septic arthritis. Curr Rheumatol Rep. 2013;15(6):332.