Septic Arthritis Workup

Updated: Dec 09, 2022
  • Author: John L Brusch, MD, FACP; Chief Editor: Michael Stuart Bronze, MD  more...
  • Print

Approach Considerations

An acutely inflamed joint calls for a rapid workup of the synovial fluid (See Joint Fluid Analysis below). This should include examining the synovial fluid for uric acid crystals with polarizing microscopy and for organisms via Gram stain. Serum levels of uric acid may be normal during an acute gouty attack. If crystals are present and the Gram stain findings are negative, treatment for crystal-associated arthritis should be initiated. An exception to this would be the presence of significant risk factors for infection such as areas of infection elsewhere that could lead to bacteremia (pneumonia or pyelonephritis). Therapeutic decisions should not be delayed until results of the synovial fluid culture are available. The Musculoskeletal Infection Society (MSIS) has updated its criteria for diagnosing joint infection on the basis of culture results, synovial fluid studies, and inflammatory markers. [31]

If crystals are not detected, the patient should presumptively be treated for infection even in the face of a negative Gram stain. Its sensitivity in those who have not received antibiotics may be as low as 65%. Always send the fluid for culture. A joint damaged by gout or pseudogout has increased susceptibility to becoming infected. Culture of synovial tissue may be necessary to detect mycobacterial or fungal involvement.

If the patient's condition does not improve significantly after 5 days, the joint should be reexamined. Most septic joints have a white blood cell (WBC) count that exceeds 50,000/μL with a marked shift. Various sterile inflammatory processes may exhibit the same cellular profile.

Other considerations

The fluid of an infected bursa can mimic that of a bacterial joint infection. [20]

Measurements of C-reactive protein (CRP) and serum procalcitonin are useful in establishing acute inflammation of a joint as well as when following the patient's response to treatment (See below). [32]

Appropriate serologic tests for the diagnosis of various vasculitides or rheumatologic disorders often are required. [33]


Joint Fluid Analysis and Culture

Always perform joint aspiration under the most sterile conditions possible to prevent the introduction of infection. [34]

Normal joint fluid is clear and colorless and produces a stringlike structure when ejected from a syringe, indicating normal viscosity. Infected joint fluid typically is yellow-green due to elevated levels of nucleated cells, and the cell count usually is markedly elevated, demonstrating a predominance of polymorphonuclear leukocytes. An evaluation of the synovial fluid (ie, via leukocyte count, appearance on Gram stain, polarizing microscopy examination, culture) is the most rewarding approach in assessing a potentially infected joint. Additional stains and/or cultures should be obtained depending on the differential diagnosis considered. [27, 34, 35, 36] Alterations in the glucose and protein concentrations of the synovial fluid are nonspecific and so are generally not helpful in establishing a specific diagnosis.

Culture results in patients with non-gonococcal septic arthritis almost always are positive, unless the patient has received antibiotics before the joint aspiration. Cultures of the joint fluid in gonococcal infections yield positive results in only about 25% of cases. If this diagnosis is suspected, the organism can be cultured from other sites, such as the cervix, urethra, or throat.  

Lyme disease

Findings from examination of the synovial fluid in Lyme arthritis are similar to those found in infection caused by any other type of bacterium. Positive serology results (ie, antibody measurements, Western blot, polymerase chain reaction [PCR] for Lyme disease) do not establish the diagnosis of Lyme arthritis. A positive result on any of these tests simply indicates that the patient has encountered B burgdorferi; a positive result does not necessarily establish a connection between the patient's musculoskeletal symptoms and Lyme disease. [18]

Silver stains can be used to detect organisms in 5% of cases of Lyme arthritis.

Prosthetic joint infection

Evaluation of a possibly infected prosthetic joint is similar to that of a native joint. [2, 29, 31]  The presence of leukocytes in the aspirated fluid is variable. Because many of the potential pathogens are also classic contaminating organisms (eg, coagulase-negative S aureus [CoNS], Propionibacterium species, Corynebacterium species), repeat aspirates often are required to confirm the diagnosis. The use of multiple types of media with prolonged incubation times may increase both the sensitivity and specificity of the culture in prosthetic joint infection. The sensitivity of periprosthetic-tissue culture ranges from 65% to 94%. [28]  Material from fistulous tracts is associated with a high rate of contamination and probably is best avoided.

See Other Laboratory Tests, below.

Reactive and tuberculous arthritides

The synovial fluid of reactive arthritis demonstrates few signs of inflammation. PCR may reveal the DNA of the purported causative organism.

The synovial fluid of a joint infected with Mycobacterium tuberculosis shows marked leukocytosis. Although findings on acid-fast stains usually are negative, culture results are positive in 80% of cases. Culture results of synovial biopsies are positive in 94% of specimens.


Blood and Other Cultures

Obtain at least 3 sets of blood cultures to document a bacteremic origin of the septic joint. The bloodstream infection may be continuous or intermittent in nature.

In the setting of possible gonococcal infection, obtaining cultures from the patient's rectum, cervix, urethra, and pharynx and from any skin lesions is most helpful. Immediate plating of the joint fluid directly onto appropriate media and/or rapid delivery of the specimen to the laboratory for appropriate plating and culturing are of benefit in improving the relatively low yield. [37]



OTHER Laboratory Studies

Markers of inflammatory processes are becoming more appreciated in making the diagnosis of infection as well as monitoring its response to various therapeutic interventions. Procalcitonin and C-reactive protein are the most widely clinically available; other examples include the JAK-STAT pathway. Procalcitonin usually is secreted solely by the thyroid. During times of severe inflammation, such as sepsis, it may be produced by a wide variety of organs. Unlike the other inflammatory markers, its secretion is not triggered by viral infections. [38]

Polymerase chain reaction (PCR) holds promise for detection of bacterial DNA in joint fluid and synovial tissue. [32] PCR has led to diagnosis of infective arthritis due to Yersinia species, B burgdorferi, Chlamydia species, N gonorrhoeae, and Ureaplasma species. However, caveats concerning this approach are raised, because it cannot be used to distinguish between live and dead organisms, and it is susceptible to contamination. PCR techniques hold promise in detecting pathogens in patients who have recently received antibiotics. Unfortunately, many patients receive empirically administered antibiotics prior to sampling the synovial fluid. [38]


Radiologic Studies

At times, imaging studies may be required to determine the significance of a given culture.

Radiography and ultrasonography

Plain radiography is of limited value in evaluating a joint for infection. [23]  Periarticular soft-tissue swelling is the most common finding. This imaging modality is most useful in ruling out underlying osteomyelitis or periarticular osteomyelitis caused by the joint infection itself.

In addition, plain radiography can reveal the linear deposition of calcium pyrophosphate. The radiographic findings of reactive arthritis usually are limited to those of soft-tissue swelling or periarticular osteoporosis. 

A 30-year-old man who was taking steroids presente A 30-year-old man who was taking steroids presented with a joint effusion and knee pain. Anteroposterior view of the knee demonstrates patchy demineralization of the tibia and femur and joint-space narrowing. This was caused by tuberculoid infection of the joint.
Septic arthritis. Anteroposterior view of the shou Septic arthritis. Anteroposterior view of the shoulder demonstrates subchondral erosions and sclerosis in the humeral head. These are relatively late findings of septic arthritis. Periosteal reaction due to coincident osteomyelitis is present adjacent to the surgical neck of the humerus.
During the progression of infectious arthritis of During the progression of infectious arthritis of the hip, this image was obtained early in the disease and shows only concentric joint-space loss.

Ultrasonography may be used to diagnose effusions in chronically distorted joints (secondary to trauma or rheumatoid arthritis). It is an underutilized technique. [39]  

CT scanning, MRI, and radionuclide scanning

Computed tomography (CT) scanning and magnetic resonance imaging (MRI) are more sensitive for distinguishing osteomyelitis, periarticular abscesses, and joint effusions. The information gained usually does not justify the increased cost; however, these tests are most helpful in patients with sacroiliac or sternoclavicular joint infection to rule out extension into the mediastinum or pelvis. MRI is preferred because of its greater ability to image soft tissue.

Radionuclide scans (ie, technetium-99m [99m Tc], gallium-67 [67 Ga], indium-111 [111 In] leukocyte scans) are used to nonspecifically localize areas of inflammation. They cannot be used to distinguish infectious from sterile processes. Radionuclide scans may be of use in diagnosing septic arthritis in relatively sequestered areas, such as the hip and sacroiliac joints.

Imaging in prosthetic joint infection

The sensitivity of all the available types of imaging studies is inadequate in documenting PJI. 

In PJI, plain radiography can reveal new subperiosteal bone growth and transcortical sinus tracts. [2, 40] These findings are quite specific for infection. Arthrography can demonstrate loosening of the prosthesis and abscesses. Nuclear scans of all types are of limited diagnostic use in patients with prosthetic joint infection, and MRIs are limited by the type of implanted material (this diagnostic modality can safely image only titanium or tantalum devices). Fludeoxyglucose-positron emission tomography (FDG-PET) scans may hold some promise in diagnosing lower-extremity prosthetic joint infections. However, this approach cannot differentiate aseptic loosening from infection. [35, 40]  CT scans are quite insensitive in detecting periprosthetic soft tissue infection of implanted joints. [39]