Advances in the Management of Bacterial Septic Arthritis

Myo M Lynn; Catherine J Mathews


Int J Clin Rheumatol. 2012;7(3):335-342. 

In This Article

Diagnosis of Bacterial Septic Arthritis

The initial suspicion of the diagnosis of septic arthritis comes with the typical clinical presentation of a short, 1–2 week duration of an acutely hot and swollen joint (or joints).[23] On examination, the joint is often not only swollen but almost invariably has extreme limitation of range of movement. Although traditionally thought of as a monoarticular process, septic arthritis can be polyarticular in up to 22% of cases.[24] Polyarticular presentations can therefore mimic inflammatory arthritis but typically a septic joint in this context will be symptomatic to an extent that is out of proportion to the overall disease activity in the rest of the joints.

Although the diagnosis of septic arthritis rests primarily on clinical findings, there are laboratory investigations that can be helpful in guiding diagnosis.[25] But as the recent literature review by Carpenter et al. has demonstrated, there is still very little in the way of diagnostic tests that significantly and confidently alter the post-test probability of the diagnosis of septic arthritis over and above one's initial clinical hunch.[26] The identification of pathogens in the synovial fluid remains the crucial investigation in the diagnosis of septic arthritis.[16] Any acute hot swollen joint should always be aspirated before the initiation of antibiotics and sent for urgent Gram stain and culture. Warfarinization is not an absolute contraindication to joint aspiration[23] and nor is the presence of overlying cellulitis.[27] The only absolute contraindication to simple needle aspiration in suspected septic arthritis is if the joint is prosthetic in which case aspiration should always be performed under strict aseptic conditions in an operating theater.[16,23]

In a patient with septic arthritis due to hematogenous spread, cultures of extra-articular infective sources can provide invaluable information on the primary focus of infection and therefore guide antibiotic therapy.[23] Blood should always be cultured, and urine, sputum, skin and urethral discharge should be cultured based on the patient's clinical history and presentation irrespective of their body temperature,[23] as the absence of fever does not rule out septic arthritis.[1,2]

Routine serological tests may not be useful in the diagnosis of septic arthritis. No studies have demonstrated a significant level of sensitivity or diagnostic accuracy for the serum white cell count (WCC) in septic arthritis. In addition, the C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) are not reliably raised in cases of joint sepsis. Moreover, it is not always easy to differentiate between inflammatory and septic arthritis on the basis of serum inflammatory markers.[26] Markers such as C-reactive protein and ESR may well not be useful in distinguishing between the two. Serological tests including the total WCC, neutrophil count and inflammatory markers such as C-reactive protein and ESR are probably more reliable for monitoring purposes than as diagnostic tools.[26]

Serum procalcitonin is an inflammatory marker that rises significantly in response to bacterial infection.[28] Detectable levels of procalcitonin can rise to 100 ng/ml (compared with levels less than 10 pg/ml in the healthy population) in severe infections and procalcitonin has a half-life of 25–30 h. Higher levels of serum procalcitonin are associated more with Gram-negative than Gram-positive bacteremias and higher levels are seen when the source of joint sepsis is systemic rather than local infection. However, the use of procalcitonin as a marker for septic arthritis is limited by its poor sensitivity.[28,29]

Interestingly, among the synovial markers, synovial lactate dehydrogenase could be a useful diagnostic tool for differentiating septic arthritis and inflammatory arthritis. There is some evidence to suggest that a threshold of >10 mmol/l could be of diagnostic utility. However, the effectiveness of synovial lactate dehydrogenase in the diagnosis of septic arthritis in the acute setting remains to be established.[30,31] The utility of other synovial inflammatory markers, such as glucose and synovial procalcitonin, in diagnosis of septic arthritis is still controversial.

A raised synovial WCC (sWCC) is not an uncommon finding in bacterial septic arthritis. This synovial marker has been regarded as a potentially useful discriminator between bacterial septic arthritis and other causes of joint inflammation. In 2004, Trampuz et al. showed that prosthetic knee septic arthritis produces a lower sWCC than native joint bacterial septic arthritis.[32] Two studies in 2007 suggested that a threshold of 50,000 cells/µl might be discriminatory.[33,34] In 2008, Ghanem et al. suggested that the sWCC, as well as the synovial white cell differential counts, could be useful adjuncts to blood inflammatory markers with cutoff values of a sWCC of 1100 cells/10−3 cm3 and a neutrophil percentage of greater than 64% perhaps being diagnostic of septic arthritis.[35] However, there are also studies that have suggested that the sWCC is not a useful diagnostic marker if the clinician is trying to differentiate between crystal and bacterial septic arthritis.[36–38] The recent review by Carpenter et al. concluded that a sWCC of greater than 90% has no significant effect if one is attempting to calculate the probability of septic arthritis (sensitivity: 60%, specificity: 78%, + likelihood ratio: 2.7, - likelihood ratio: 0.51).[26]

It is possible that there may be a role for PCR pathogen-specific probes in the diagnosis of bacterial septic arthritis. A recent study demonstrated the use of a real-time broad-based PCR assay in the acute clinical setting for the identification of microbes by targeting a 16S rRNA gene.[39] The ability to identify organisms within 3 h suggested that a PCR assay might have a high clinical performance in detecting pathogenic microbes compared with traditional culture techniques. Two previous studies, however, have shown that PCR has no advantage over traditional synovial culture in the identification of common bacterial infections (staphylococcus and streptococcus) in the standard laboratory setting.[40,41] However, this nonculture-based PCR technique may be very useful in identifying slow-growing organisms, such as Mycobacterium tuberculosis, and fastidious organisms that require specialized environments due to their complex nutritional requirements.

Turning to imaging, there is no imaging technique, which has been shown to reliably diagnose septic arthritis.[23] MRI findings in septic arthritis are nonspecific and it is impossible to differentiate between septic arthritis, inflammatory- and crystal-induced arthritis with imaging methods.[42] However, MRI can be very useful in identifying associated complications of septic arthritis, such as osteomyelitis, surrounding abscesses, soft tissue infections and the presence of foreign bodies.[23,42]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.