Bone and joint infections are an important cause of morbidity and mortality worldwide, particularly in those with inflammatory arthritis. In the US, the prevalence of septic arthritis is 2–10/100,000 but rises to 30–70/100,000 in patients with rheumatoid arthritis. Patients with underlying rheumatological conditions are at an increased risk of infection due to existing abnormal joints (in the case of inflammatory arthritis), as well as immunosuppressive therapies. The British Biologics Register data shows that the risk of patients with rheumatoid arthritis developing septic arthritis doubles in patients treated with biologic therapy compared with non-biologic therapy.1
The classical signs and symptoms of bone and joint infection are joint pain, fever and swelling.
Presentation of bone and joint infections in patients with rheumatological conditions can be variable and more indolent. Correct diagnosis and management can therefore require a higher index of suspicion.
There are two major challenges: overlapping symptoms and an immunosuppressed state.
A history of pain, swelling and fever as well as classical symptoms of infection could also be due to the patient’s underlying inflammatory arthritis. This can lead to a delay in diagnosis. Even in patients without underlying rheumatological disease, the delay in diagnosis can range from two weeks to nine months.2
A rheumatological flare may affect several joints; however, if only a sterile joint is aspirated for microscopy and culture this could then lead to a missed infected joint elsewhere.
The other consideration is that both the use of immunosuppressive agents and the underlying inflammatory arthritis can alter the response of the immune system to infection. This makes patients more prone to acquiring an infection; but also once an infection has occurred, can make the symptoms and signs more insidious.
In addition, local administration of corticosteroids, can also be a risk factor for bone and joint infection, particularly when performed in primary care and on patients with rheumatoid arthritis.3 However, the overall incidence of this is low.4
Taking both of these factors into consideration, a thorough history and examination of new joint swelling, joint or back pain and fever, in the case of septic arthritis, and new back pain and fever in the setting of vertebral osteomyelitis (VOM) and a high index of suspicion on the part of the clinician, is required to appropriately investigate and manage the patient.
Obtaining a microbiological diagnosis is the cornerstone of investigation of bone and joint infection.
The main pathogens causing septic arthritis and osteomyelitis are bacterial, with Staphylococcus aureus infections in around three quarters of patients being the most prevalent. Pathogens and their risk factors are listed below.
For pyogenic bone and joint infections, appropriate samples to take include fluid aspirate for microscopy, Gram stain and culture, as well as blood cultures. It is important to highlight that Gram stain is positive in only 50% of patients, so for those with a high clinical index of suspicion, antimicrobials should be initiated/continued even with a negative Gram stain result until culture results are available. Intraoperatively, fluid samples in septic arthritis or tissue samples in VOM should also be sent to guide future antimicrobial therapy.
In the setting of culture negative bacterial bone and joint infection, 16sPCR can be used to amplify and identify any bacterial DNA on a sample.5 It is important to understand the limitations of molecular diagnostic tests – for example, a positive sample may still be a contaminant, and there are no commercially available techniques at present to predict sensitivity patterns even if a pathogen is isolated. However, PCR may still be useful if an organism is difficult to culture or if antimicrobials were started before culture was taken.
If tuberculosis is suspected, samples should be sent for Mycobacterial culture, and consideration of PCR for MTB complex (where available). A careful history and Imaging should be performed to elicit any evidence of disseminated tuberculosis. In patients with with suspected Lyme arthritis (that is, risk factors such as a tick bite from an endemic area), investigations should include Lyme serology on serum. Fluid aspirate samples can also be sent for PCR if locally available to aid detection of Borrelia burgdorferi.
Fungal infection is a rare cause of septic arthritis. If fungal infection is suspected, usually in the setting of a highly immunosuppressed patient or intravenous drug user, fluid should be sent for fungal culture. If this culture is negative – further tests could include 18sPCR, panfungal, and fungal markers such as galactamman and beta glucan in blood may be useful. In the setting of appropriate travel history to North and South America, serology for the diamorphic fungi, for example, Histioplasma, Coccoides and Paracoccoides, can also be useful.
C-reactive protein (CRP) has been the mainstay for detecting and monitoring inflammation and infection for many years. However, there are well described controversies with differentiating between the inflammation of rheumatological conditions and that of infection. There may also be issues with falsely low CRPs in the setting of a patient on DMARDS,6 Cox-2 inhibitors and corticosteroids.
Procalcitonin is a precursor to the hormone calcitonin and rises in response to bacterial infection. Interest in procalcitonin as a specific marker for bacterial infection has increased and there is some evidence to show that it is not affected by corticosteroids.7 A meta-analysis looking at serum procalcitonin in differentiating between septic and inflammatory arthritis showed a specificity of 95% (CI 87–98%) and a sensitivity of 54% (38–62%) compared with serum CRP and a sensitivity of 45% (CI 35–55%) and specificity of 7.9% (CI 2.1–2.5%).8
Unfortunately there is little published data on the effects of DMARDs on other serological biomarkers, for example, IL-1 or PTX-3, and their ability to distinguish between rheumatological inflammation and infection. However, this may be an interesting area for future research, as these markers are not yet commercially available.
Joint fluid biomarkers are also being developed as another method to try to differentiate between inflammation and infection, particularly in the setting of prosthetic joint infection. One systemic review and meta-analysis comparing 13 different tests determined that alpha defensin was the optimal synovial marker and that there may be scope to use others in combination. However the effect of immunosuppressive therapies on this is unknown.9
With regards to the radiological diagnosis of VOM, the IDSA10 recommends MRI spine as first-line modality followed by technicium gallium scan CT or PET CT if there are contraindications to MR. MRI has a sensitivity
of 97% and specificity of 93%; however, it is well known that radiological changes can be absent early on in infection and there can be artefact effect in the presence of prosthesis. Gallium scan combined with a bone scan is reported as having a sensitivity of 91% and specificity of 90%.
CT PET is also useful for the diagnosis of osteomyelitis and detecting metastatic infection; however, it would be difficult to differentiate between increased uptake due to inflammation, malignancy or infection. Nevertheless, there is evidence to show that CT PET is more sensitive and specific in the setting of chronic vertebral osteomyelitis than labelled WCC. CT PET has also been evaluated and found to be useful in serial scans in both the post-surgical setting and in the general follow up of patients, whereas with MRI, it is known that radiological change can lag behind clinical cure.
Unfortunately data evaluating this modality in the setting of the rheumatological patient are lacking, and the obvious concern would that be the areas of increased uptake could be due to either the underlying inflammatory disease or infection.
Septic arthritis of a native joint can be life-threatening and cause permanent joint damage if left untreated. In most cases, surgical washout is required, as well as appropriate antimicrobials, which might need to be initiated before a joint aspirate is taken if there is going to be a significant delay. Empirical antimicrobials are based on local sensitivities, patient risk factors (for example, allergy) and pharmacokinetics and dynamics (such as bone penetration and patient’s renal function). In general terms, empirical coverage is aimed at Staphylococcus aureus and Streptococcus species as these are the most common causes of infection. After appropriate source control the duration of antimicrobials of a native joint is usually two to four weeks.
If clinical suspicion of infection remains high, but initial bacterial culture is negative, cases should be discussed with an infection specialist to guide ongoing testing. This is particularly important for those with risk factors for more unusual pathogens, as listed in Table 1.
For pyogenic vertebral osteomyelitis, if the patient is not septic and there is no neurological compromise (either clinically or radiologically), the initiation of antimicrobials can be deferred until appropriate cultures are taken and a microbiological diagnosis is confirmed.10 Cultures to be taken include blood cultures and, where amenable, a biopsy. There are no clinical studies regarding duration of antimicrobials in patients with rheumatoid arthritis but a recent study has shown non-inferiority of six weeks’ duration of antimicrobials compared with 12 weeks in pyogenic infection.11
The type of antimicrobial depends on the pathogen and sensitivities. However, in patients with inflammatory arthritis who may be on a number of medications, a careful assessment of drug interactions and their risk versus benefit should be assessed. Common interactions are listed in Table 2. This is not an exhaustive list and a risk–benefit analysis needs to be made with a rheumatologist, infection specialist and the patient as to which chosen antimicrobial regimen is the most safe and most potent. Other considerations specific to the rheumatology patient in the face of an invasive infection is whether immunosuppressive medications should be reduced or stopped and, more importantly, when these should be restarted. Interestingly a recent meta-analysis in the paediatric, non-rheumatological population has shown favourable outcomes in corticosteroid use in septic arthritis, although it should be highlighted this might not be generalisable to the adult rheumatological population.12
The assessment, investigation and management of a patient with bone and joint infection and inflammatory arthritis is complex and requires a multidisciplinary approach. This should include rheumatology, infectious diseases, radiology and orthopaedics with expertise and special interest in managing the challenges with which this cohort of patients present.
More research is required into specific biomarkers and radiological modalities for this group of patients in order to deliver high quality and evidence-based care. It is likely that the prevalence of infection in rheumatological patients with complex infection will continue to rise with increasing use of immunosuppressives.
1 Galloway JB et al. Risk of septic arthritis in patients with rheumatoid arthritis and the effect of anti-TNF therapy: results from the British Society for Rheumatology Biologics Register. Ann Rheum Dis 2011;70(10):1810–14.
2 Beronius M et al. Vertebral osteomyelitis in Göteborg, Sweden: a retrospective study of patients during 1990–95. Scand J Infect Dis 2001;33:527–32.
3 Xu C et al. Risk factors and clinical characteristics of deep knee infection in patients with intra-articular injections: A matched retrospective cohort analysis. Semin Arthritis Rheum 2018;47:911–16.
4 Wang Q, Jiang X, Tian W. Does previous intra-articular steroid injection increase the risk of joint infection following total hip arthroplasty or total knee arthroplasty? A meta-analysis. Med Sci Monit 2014;20:1878–83.
5 Woo PC et al. Then and now: use of 16S rDNA gene sequencing for bacterial identification and discovery of novel bacteria in clinical microbiology laboratories. Clin Microbiol Infect 2008;14:908–34.
6 Mysler E et al. Influence of corticosteroids on C-reactive protein in patients with rheumatoid arthritis. Arthritis Res Ther 2004;6(Suppl 3)57:1–41.
7 Perren A et al. Influence of steroids on procalcitonin and C-reactive protein in patients with COPD and community-acquired pneumonia. Infection 2008;36(2):163–6.
8 Zhao J et al. Serum procalcitonin levels as a diagnostic marker for septic arthritis: A meta-analysis. Am J Emerg Med 2017;35(8):1166–71.
9 Lee YS et al. Synovial fluid biomarkers for the diagnosis of periprosthetic joint infection: A systematic review and meta-analysis. J Bone Joint Surg Am 2017:99:2077–84.
10 Berbari E et al. 2015 Infectious Diseases Society of America (IDSA) Clinical Practice guidelines for the diagnosis and treatment of native vertebral osteomyelitis in adults. Clin Infect Dis 2015;61(6):e26–e46.
11 Bernard L et al. Antibiotic treatment for 6 weeks versus 12 weeks in patients with pyogenic vertebral osteomyelitis: an open-label, non-inferiority, randomised, controlled trial. Lancet 2015;385(9971):875–82.
12 Farrow L. A systematic review and meta-analysis regarding the use of corticosteroids in septic arthritis. BMC Musculoskelet Disord 2015;16(241):1–8.