Introduction
The etiology of new or persistent pain or poor function after total knee arthroplasty (TKA) often presents a diagnostic dilemma. Determining the cause of pain starts with the patient’s history, physical examination, and plain radiographic studies but often requires further diagnostic testing. Although there are many causes of pain or failure, a high index of suspicion must be maintained for periprosthetic joint infection (PJI). Because treatment of a deep infection is fundamentally different from treatment for other causes of failure, early and accurate diagnosis of PJI is critical. PJI is also the most common cause for revision TKA in the United States. The diagnosis of PJI is often challenging, because patients frequently present with only vague complaints of knee pain or dysfunction and without the classic symptoms of infection such as fever, chills, sweats, wound drainage, or erythema around the incision.
This chapter reviews the use of serologic tests and joint aspiration to rule in or rule out PJI after TKA. In general, the approach includes measurement of the serum erythrocyte sedimentation rate (ESR) and the level of C-reactive protein (CRP) and, if these are abnormal, aspiration of the knee joint. The fluid obtained from the aspiration is then sent for several tests, including a synovial fluid white blood cell (WBC) count and differential, to determine the percentage of polymorphonuclear cells (%PMNs) in the sample.
Serologic Testing
Serologic tests for systemic markers of inflammation (specifically ESR and CRP) are excellent screening tests to evaluate for PJI. They should be performed before all revision procedures and should be included in the initial evaluation of the painful TKA, as was suggested by the American Academy of Orthopaedic Surgeons’ 2011 clinical practice guideline for the diagnosis of PJI. These tests involve a peripheral venous blood draw, are generally available at all hospitals as part of a standard battery of tests, and are relatively inexpensive. The ESR is a general measure of systemic inflammation and is determined by the rate at which erythrocytes fall and collect (sediment) at the bottom of an upright tube, measured in millimeters per hour. Proinflammatory factors (e.g., fibrinogen) are elaborated by the immune system in the peripheral blood at the time of an immune response and attach to erythrocytes, resulting in the cells’ being attracted to one another. This “stickiness” leads to greater accumulation of cells at the bottom of the tube during the test and hence a higher sedimentation rate at the time of a systemic immune response. CRP is a protein that is produced by the liver in response to an acute inflammatory episode. It is thought to be triggered by release of interleukin-6 (IL-6) from macrophages and attaches to the offending substance to help initiate the complement cascade.
These tests are quite sensitive and therefore useful to rule out infection if the results are normal ( Tables 3.1 and 3.2 ). However, they are not specific for PJI, so if the results are abnormal, aspiration of the joint is suggested, particularly if there is any clinical suspicion of infection. In clinical practice, many use the cut-off values suggested by their local laboratory to define an “abnormal” result and need for further testing, although several studies have used receiver operator curves to determine an optimal cut-off point that balances both sensitivity and specificity. The serum WBC count traditionally has been useful for diagnosis of infection in general, but this test has been shown to be not helpful in diagnosing PJI and should not be routinely performed for this purpose unless otherwise indicated by systemic symptoms of sepsis.
| Author | ESR Cut-off Value (mm/hr) | N | Sensitivity | Specficity | Positive Likelihood Ratio | Negative Likelihood Ratio |
|---|---|---|---|---|---|---|
| Greidanus et al, 2007 | >22.5 | 151 | 0.93 | 0.83 | 5.5 | 0.08 |
| Greidanus et al, 2007 | >30 | 151 | 0.82 | 0.88 | 6.7 | 0.2 |
| Della Valle et al, 2007 | >30 | 94 | 0.9 | 0.66 | 2.66 | 0.15 |
| Bottner et al, 2007 | >32 | 78 ∗ | 0.81 | 0.89 | 7.69 | 0.21 |
| Author | CRP Cut-off Value (mg/dL) | N | Sensitivity | Specficity | Positive Likelihood Ratio | Negative Likelihood Ratio |
|---|---|---|---|---|---|---|
| Greidanus et al, 2007 | >1.0 | 151 | 0.93 | 0.83 | 5.5 | 0.08 |
| Greidanus et al, 2007 | >1.35 | 151 | 0.91 | 0.87 | 6.9 | 0.1 |
| Fink et al, 2008 | >1.35 | 145 | 0.73 | 0.81 | 3.81 | 0.34 |
| Della Valle et al, 2007 | >1 | 94 | 0.95 | 0.75 | 3.88 | 0.06 |
| Bottner et al, 2007 | >1.5 | 78 ∗ | 0.95 | 0.91 | 10.86 | 0.05 |
| Bottner et al, 2007 | >3.2 | 78 ∗ | 0.95 | 0.96 | 27.14 | 0.005 |
Acute Postoperative Infection
In the early postoperative period after the insult of the surgical procedure, systemic markers of inflammation are expected to be elevated as the body responds to the trauma of surgery. Because the ESR and CRP are typically elevated shortly after surgery, their utility as a screening test for infection at that time is questionable. Bedair and colleagues, in a review of 146 TKAs evaluated for infection within 6 weeks of the index surgery, reported that the ESR was not a useful test to screen for infection but the CRP did appear to possess diagnostic utility. With a CRP threshold of 95 mg/L (normal value, <8 mg/L), the negative predictive value of this test was 91%, making it an excellent rule-out test for infection in the early postoperative period. In practice, if there is any question regarding an acute postoperative infection, we order a CRP, and if the value is greater than 95 mg/L, an aspiration of the joint is performed.
Chronic Infection
Most authors have used previously established thresholds for ESR (>30 mm/hr) and CRP (>10 mg/L) to compare infected and noninfected TKAs. Trampuz and co-workers investigated these two tests in 207 knees and found that they were good tools to screen for infection. Ghanem and colleagues demonstrated similar findings in a multicenter study of 429 knees. Della Valle and associates found that these two tests were useful as independent predictors of infection after TKA in 94 knees examined and that their diagnostic powers increased when they were considered together. In this series, there was only 1 patient out of 41 with an infected TKA who had both a normal ESR and a normal CRP result (see Tables 3.1 and 3.2 ).
Future Serologic Tests
As more molecular markers of inflammation and infection are identified, serologic tests other than ESR and CRP may prove to be valuable aids in the attempt to accurately diagnose PJI. One of the most promising of these markers is serum IL-6. IL-6 is a cytokine released by T cells and macrophages in response to an infectious insult that is believed to help modulate the inflammatory response. Di Cesare and associates found that serum IL-6 possessed a higher diagnostic accuracy than ESR or CRP in patients with periprosthetic infection of the hip or knee. Bottner and colleagues showed that the presence of elevated values of both IL-6 and CRP identified all 78 patients in their study with periprosthetic infection. Although other molecular markers can be measured in the serum and may provide some diagnostic accuracy for sepsis, including tumor necrosis factor-α (TNF-α) and procalcitonin, currently none appear as promising as IL-6. However, widespread use of this test has been limited because it is not usually part of the standard battery of tests available at most centers and therefore requires transport to a specialty center that performs the test. This makes it less attractive in terms of increased cost and delay in results compared with the commonly available ESR and CRP tests.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
