6.2.1 Risk Factors

Iatrogenic causes of septic arthritis range from 17 to 53% of cases, due to arthrocentesis in the majority of cases (43%) followed by open joint surgery (34%) and arthroscopy (23%). ^{1 , 3} Given that these cases frequently occur after an invasive procedure, it is important to have high suspicion for septic arthritis after any procedure that violates the joint. Several other risk factors for septic arthritis should be considered, and pre-existing joint disease is one of the most common, as 47% of patients with a septic joint have had prior joint problems. ⁴ A high index of suspicion should be had in patients with rheumatological conditions such as gout, pseudogout, systemic lupus erythematosus (SLE), and rheumatoid arthritis (RA). Patients with RA and SLE are at particularly high risk given that they often have pre-existing joint damage, require chronic immunosuppressive medication, and have poor skin conditions. Despite an increased risk in these patients, diagnosis is frequently difficult and can be delayed as clinical manifestations of RA flare can be similar to that of a septic joint. Particular vigilance is needed in patients with a monoarticular flare up, as immunosuppressive medications increase the risk of septic arthritis by fourfold. ⁴ Other comorbidities or conditions that should raise the suspicion of a septic joint include bacteremia, especially from an invasive procedure such as a colonoscopy that may result in hematogenous seeding, intravenous drug use, and other comorbidities that influence the immune system such as diabetes mellitus, renal failure, and immunosuppressive medications (Box 6.1).

6.2.2 Joint Involvement

The knee is the most commonly infected joint, as it makes up 45% of cases. ⁴ Other large joints that are infected include the hip (15%), ankle (9%), elbow (8%), wrist (6%), ⁴ and shoulder (5%) (▶ Fig. 6.1). Septic arthritis of cartilaginous joints of the axial skeleton are rare except in cases of bacteremia and intravenous drug users. In this population, the pubic symphysis, sternoclavicular, and sacroiliac joints may be affected. ⁴

6.2.3 Clinical Manifestations

A septic joint is one of the few surgical emergencies in orthopaedics, as prompt treatment can prevent further morbidity from joint destruction and even mortality. Timely diagnosis of a septic joint is thus crucial and relies on a combination of clinical and laboratory tests. The most common clinical manifestations include acute joint pain, joint effusion or swelling, erythema or warmth, joint immobility, and other constitutional symptoms such as fevers, chills, or rigors. It is important to note that sensitivity of fevers and chills are quite poor as a fever is present in only 58% of patients and chills in 25% of patients with a septic joint. ⁵ Although septic arthritis is usually monoarticular, it may involve multiple joints in up to 20% of cases, especially if bacteremia is present. ⁴ The knee is involved in 72% of cases in which there is polyarticular involvement.

Physical examination should be used to help determine if the swelling and inflammation are intra-articular versus periarticular, such as a prepatellar bursitis. In addition, erythema and warmth of the skin may be indicative of cellulitis. However, it is important to note that septic arthritis may still be present in the setting of cellulitis. Pain with passive range of motion is one test that may help distinguish septic arthritis from cellulitis or periarticular involvement such as a bursitis, as the latter two should not have pain with joint range of motion. Furthermore, joint immobility is often present in septic arthritis as the joint is often in a position to maximize intra-articular space. For example, the knee is often in an extended position and the hip is often abducted and externally rotated when infected to accommodate the maximum amount of joint fluid.

6.2.4 Serum Evaluation

While the clinical impression remains the mainstay of septic arthritis diagnosis, the diagnosis is often supplemented with laboratory data. Serum markers are often the first tests ordered and include white blood cell (WBC) count, polymorphonuclear cell count, C-reactive protein (CRP) levels, and erythrocyte sedimentation rate (ESR) (Box 6.2). In addition, serum procalcitonin and tumor necrosis factor alpha (TNF-α) have been demonstrated to help discriminate between septic versus inflammatory arthritis. ^{6 , 7 , 8} Interleukin-6 (IL-6) has not demonstrated strong diagnostic utility for a septic joint unlike in periprosthetic joint infection. ⁸ Blood cultures should also be obtained, especially when hematogenous spread or sepsis is suspected. In patients with bacteremia or sepsis, a serum lactate should be obtained as a surrogate for tissue perfusion as part of the sepsis guidelines. ⁹ These markers increase with inflammation and are nonspecific, since a variety of conditions can cause an increase in these markers. A CRP level of 10.5 mg/dL has demonstrated a high correlation with septic arthritis in native joints despite lacking specificity. ¹⁰ When ordering CRP, it is critical that the units are converted. Recently, highly sensitive CRP has been introduced in order to better quantify values in the normal range which has increased confusion. However, little difference has been found after unit conversion between the different CRPs. ¹¹ Furthermore, recent studies, ^{6 , 7 , 12} including one by Hügle et al, have also demonstrated that procalcitonin may outperform CRP in terms of specificity and sensitivity. ⁷ In addition, these markers may also be useful in order to monitor the therapeutic response as they may be a proxy for infection control.

6.2.5 Synovial Analysis

Aspiration of the involved joint should be performed in patients with a suspicion for infection or who have elevated serum inflammatory markers. The aspiration should be performed in a sterile manner using an alcohol or povidone-iodine preparation using an appropriate path that provides access to the joint while avoiding any areas of cellulitis. This may be performed under image guidance or by a radiologist. Synovial fluid should be sent for synovial WBC count with differential, crystal analysis, gram stain, and culture. While a gram stain has poor sensitivity (45%), it has high specificity in the literature. ¹³ Furthermore, synovial lactate and synovial IL-6 may also be ordered as they both demonstrate high accuracy in the literature. ^{14 , 15 , 16}

Traditionally, a cell count of greater than 50,000 cells/mm ³ and a polymorphonuclear cell count greater than 90% have been directly correlated with septic arthritis; however, there is significant overlap with crystalline arthroplasty. ¹⁰ Furthermore, the common mantra is that a cell count from 0 to 2,000 cells/mm ³ corresponds to a noninflammatory etiology and a cell count of 2,000 to 50,000 cells/mm ³ represents an inflammatory arthropathy. However, it is important to note that cell counts lower than 50,000 cells/mm ³ can occur in infectious arthritis. One investigation demonstrated this cutoff value as having a sensitivity of only 64%, as nearly one-third of patients had a cell count lower than 50,000 cells/mm ³ . ⁴ Patients with lower synovial fluid cell counts may occur in people with atypical or fastidious organisms, disseminated gonococcal disease, and immunosuppression. Furthermore, it is important to note that septic arthritis can occur in the setting of crystalline arthropathy, as the presence of crystals does not rule out a septic joint. In some reports, up to 5% of patients with proven crystalline arthritis have concomitant septic arthritis. ¹⁷ Besides synovial WBC count, leukocyte esterase (LE) may be a useful test, as a prospective study by Gautam et al reported a 100% sensitivity of LE for acute septic arthritis with a positivity predictive value of 94%. ^{4 , 5 , 18} Leukocyte esterase testing is performed using synovial fluid from the joint. It is first spun down using a centrifuge in patients with a bloody aspiration and a drop is then placed on a urine analysis stick.

While the fluid is often sent for cultures, over 20% of cases may have negative cultures. ¹⁸ Several possible explanations include premature antimicrobial therapy, an insufficient volume of fluid, inadequate incubation duration, or fastidious growth requirements. A study by Hindle et al demonstrated that premature antibiotic administration decreased the yield of culture from 79 to 28%, suggesting that administration of antibiotics should be avoided when feasible until a culture is isolated. ¹⁹

In some cases, cellulitis may be present in the setting of possible septic arthritis. In patients with surrounding cellulitis, there is concern that an aspiration through involved skin cellulitis may seed the joint. Thus, it is advisable to aspirate the joint through normal appearing skin when possible. In these patients, we are very careful to ensure that the procedure is performed in a sterile fashion with an alcohol or povidone-iodine solution. In addition, we take precautions to avoid touching the affected skin during the procedure. Furthermore, we have a lower threshold to have a musculoskeletal radiologist perform the aspiration, especially when the aspiration sites most commonly used have overlying cellulitis. However, if there is high clinical suspicion for a native septic joint, an aspiration may be performed through cellulitic skin, given that a missed or delayed diagnosis may outweigh the risk of inoculating an uninvolved joint with bacteria.

Unfortunately, there may be instances when the diagnosis is not clear, particularly when premature antibiotic therapy or an impaired immune system is present. Patients may thus have a nonconfirmatory cell count but a clinical concern for infection. In these instances, some options include repeat clinical examination, aspiration, and waiting for cultures. However, if high clinical concern remains, surgical treatment is likely warranted given the morbidly and mortality associated with delayed diagnosis and treatment.