Diagnosing the Septic TKA
The authors outline their stepwise approach to evaluating and diagnosing a suspected infection of a total knee arthroplasty.
Nicholas Ting, MD, and Craig Della Valle, MD
Infection is the most common reason for revision in total knee arthroplasty (TKA), comprising 25.2% of all revisions performed.  Infection, therefore, must always be ruled out even in the setting of obvious mechanical failure, as patients can often have concomitant diagnoses (eg, instability and infection), and a septic etiology dictates a very different treatment algorithm.
Many host factors can predispose patients to periprosthetic infection, including: [2,3]
- Prior surgery
- Inflammatory arthritis
- Steroid dependence
- Renal insufficiency
- Skin disorders
- Poor nutrition
- Immunocompromised state
Numerous modalities are used in evaluating the painful or failed TKA; however, no gold standard has been established for diagnosing periprosthetic joint infection (PJI). Given the implications for treatment, it is important to understand the basics for evaluating the painful and potentially infected TKA. Here, we outline our stepwise approach to evaluation and diagnosis of the infected TKA.
History and Physical
While frequently overlooked, the patient’s history often provides clues to a septic etiology, which helps steer a more focused and aggressive work-up. In patients with a history or risk factors that are highly suggestive of PJI, more assertive evaluation is performed to more thoroughly rule out PJI.
Specifically, patients should be asked about:
- Problems with wound healing around the time of the index arthroplasty
- Extended antibiotic administration
- Early returns to the operating room
- Persistent pain since the time of surgery (particularly if it is different from preoperative pain) and pain present at night or at rest (not activity related)
- Failure of an otherwise well-done knee replacement within the first 2 to 3 years postoperatively
The physical exam is important from diagnostic and treatment perspectives. While peri-incisional erythema, warmth, effusion and painful range of motion suggest infection, pay close attention to the surgical incision itself. Carefully inspect it for:
- Signs of poor wound healing, including actively draining or previously healed sinus tracts
- Multiple incisions, which suggest multiple prior procedures
Careful inspection of the wound plays an important role in surgical planning. Because wound closure can be challenging, a preoperative plastic surgery consultation may be helpful in case in which specialized wound closure techniques or flap coverage may be required.
Laboratory Tests and Imaging
Together , the erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) level are excellent screening tools with high sensitivities. They are rarely normal in the face of infection. [4,5]
These tests are easily obtained, inexpensive, and widely available. They should be done prior to revision TKA and when a patient presents with a failed or otherwise painful TKA.
If the ESR and/or CRP are elevated or if there is a high clinical suspicion of periprosthetic joint infection, further evaluation is required.
Radiographs are rarely helpful in diagnosing an infected TKA; they do not have a high level of sensitivity as a screening tool in this situation. Although radiographs should not be routinely relied on for diagnosis of an infection, they are critical for overall evaluation of the painful TKA.
Radiographic signs of infection include:
- Early implant loosening (within the first few years postoperatively, particularly if a cemented implant was utilized)
- Early osteolysis (do not assume it is from the bearing surface within the first decade)
- Periosteal reaction, although rare
Joint aspiration is the most useful tool for diagnosing an infected TKA. The procedure provides three points for evaluation:
- Synovial fluid white blood cell (WBC) count
Several studies have shown that the synovial WBC may be the single most accurate marker of infection, indicative of PJI when greater than 3,000 cells/uL. It is important to note, however, that while using a cut-off value is clinically useful, it is an oversimplification in a number of ways:
- These values represent a continuum – for example, 2,999 WBC/uL does not necessarily suggest the absence of PJI, but 3,001 does.
- Variations in these values can occur between labs and vary based on blood in the sample, the duration of time between collection and processing, and the equipment or techniques utilized by the lab.
Different studies demonstrate varying cut-offs for the differential indicative of infection, ranging from 60% to 80%. In general, deep periprosthetic infection should be strongly considered if the differential is >90%.
As Barrack et al demonstrated,  it is important to keep patients off antibiotics for at least 2 weeks prior to obtaining this aspirate. With some fastidious organisms that are more difficult to culture, a much longer period of time may be necessary to obtain a positive culture.
Leukocyte Esterase Reagent Strip Test
Recent work suggests that leukocyte esterase strips (the same ones used to diagnose urinary tract infections) can be used for the rapid diagnosis of periprosthetic joint infections. [7,8]
Specifically, synovial fluid can be placed directly on the strip and a colorimetric change (the box turns purple) suggests the presence of infection with high sensitivity. However, blood or other debris present in the sample makes these unreadable in about one third of cases.
The leukocyte esterase reagent strip has the advantages of providing real-time results and being simple and inexpensive, although the colorimetric nature of the test makes interpretation of the results somewhat subjective.
Nuclear Medicine Studies
In general, nuclear medicine studies have fallen out of favor as diagnostic modalities for infection. They are considered second-line tools to be utilized when infection is suspected, but for technical reasons, an aspirate cannot be obtained for synovial white blood cell count and cultures.
When using nuclear medicine studies, we have found that an indium scan combined with a sulfur colloid marrow scan is the most useful.  A negative scan can reliably rule out infection; however, a positive scan can still be falsely positive.
Intraoperative appearance of the prosthesis can be unreliable. Debris around the prosthesis can mimic the appearance of infection, as has been seen with metal-on metal bearings. Similarly, periprosthetic infection can be present even in the absence of gross purulence.
Intraoperative gram stain is also an inadequate diagnostic modality. Several studies to date have shown these gram stains to have poor sensitivity (~15%) and to be subject to false positive results. [10,11]
Therefore, neither intraoperative appearance nor gram stain should be used as the only diagnostic measures for infection.
Intraoperative frozen sections can be diagnostically helpful if the institution has an experienced musculoskeletal pathologist, but the cut-off criteria for infection remain controversial, and the test can be subject to sampling error.
In general, we view an average >10 polymorphonuclear leukocytes per high-power field in the five most cellular fields as positive for active infection. It is important to recognize that cells must be in tissue, not fibirin, to be counted.
Intraoperative cultures have long been considered the gold standard for diagnosing periprosthetic joint infection. However, they are prone to both false positive and negative results
In general, 3 to 5 culture specimens should be obtained from separate areas of the joint, sampling the most suspicious appearing areas (adjacent to the prosthesis if possible).
Tissue cultures are preferable to swabs and can be simply placed into a sterile cup and sent to the lab. Unless there is a suspicion for culture-negative periprosthetic joint infection, only standard aerobic and anaerobic tissues are sent with an incubation time of three to five days.
Ideally, fresh instruments should be used to harvest the samples to avoid contamination, and the tissue should be cut sharply with a knife as opposed to using electrocautery. The samples should be immediately transported to the lab for plating.
Unfortunately, it seems as if there is an ever-increasing number of cases in which serum serology and the synovial fluid WBC count and differential are consistent with infection, but the cultures are negative.
The main risk factor for a so-called “culture negative infection” is the use of antibiotics in cases of suspected infection prior to aspiration of the joint. Therefore, the temptation to administer antibiotics should be resisted until an adequate sample can be obtained from the joint for culture.
Appropriate cultures are necessary to assist with confirmation of the diagnosis. A culture-negative infection is far more difficult to treat, as antibiotic selection becomes empiric instead of specifically directed toward an organism type and susceptibility profile.
If a culture-negative periprosthetic joint infection is suspected, the first step is to perform a repeat aspiration of the joint. This will allow for a confirmation of the synovial fluid WBC count and differential, and provide more fluid for culture.
In general, we only send fluid for aerobic and anaerobic cultures; however, in cases in which a culture-negative periprosthetic infection is suspected, fluid obtained is also sent for fungal and acid-fast bacilli cultures.
At this point, we also take the fluid obtained and inoculate it into blood culture bottles, as this may increase the yield of bacterial recovery. 
It is often helpful at this juncture to also obtain a consult from an infectious disease specialist and discuss the case with the microbiology lab to ask them to hold the cultures for an extended incubation period (up to 2 weeks) in an attempt to culture more fastidious organisms.
If the above steps still do not yield an organism, an infectious disease consultation can be very helpful at this time (if not already obtained), as can a second opinion from another orthopaedic surgeon to confirm the diagnosis of periprosthetic joint infection. It can be hard for the patient to accept the diagnosis without a positive culture.
Repeating the aspiration a third time is also helpful to once again attempt to recover an organism.
Diagnosis of Infection in the Early Postoperative Period
Diagnosing a periprosthetic joint infection in the early postoperative period can be particularly challenging: The normal cues to infection can be difficult to interpret secondary to normal postoperative pain and inflammation around the incision.
Interestingly, the serum CRP has been shown in two reports [14,15] to be an excellent screening test for periprosthetic joint infection, with a cut-off of approximately 100 mg/L (normal range < 10 mg/L). In clinical practice, we use this to perform selective aspiration of the joint in cases in which there are concerns over the appearance of the wound and suspicion for early postoperative periprosthetic joint infection.
Likewise, two studies [14,15] have suggested that the synovial WBC count can be helpful as a diagnostic marker of infection, but with different thresholds than those used for chronic periprosthetic joint infection.
With a cutoff of 27,800 cells/uL, synovial WBC count can predict infection within 6 weeks of primary TKA, with positive and negative predictive values of 94% and 98%, respectively.  Synovial WBC <10,000 cells/uL in the acute setting indicates the absence of infection.
In the same study, the optimal diagnostic cutoff for the differential was 89% polymorphonuclear cells. 
Diagnosis in Patients with Rheumatoid Arthritis
Although intuitively one would expect that commonly used tests for evaluation periprosthetic joint infection would be inaccurate in patients with inflammatory arthritis, recent work suggests serum and synovial tests performance is similar in patients with non-inflammatory and in patients with inflammatory arthritis.
In a study by Cipriano et al,  the optimal cut-offs in patients with non-inflammatory and inflammatory arthritis were:
- 32 and 30 mm/hr, respectively, for the ESR
- 15 and 17 mg/L, respectively, for the CRP
- 3450/μL and 3444/μL, respectively, for the synovial fluid WBC count
- 78% and 75%, respectively, for the differential
In addition, the overall testing performance was similar between both groups, although the rate of periprosthetic joint infection was significantly higher following procedures in patients with inflammatory arthritis.
Therefore, elevated ESR and CRP levels should not be assumed to be secondary to inflammatory arthropathy. Rather, these markers may indicate periprosthetic joint infection; further evaluation is warranted.
Diagnosis in Unicompartmental Knee Arthroplasty
One study has evaluated commonly used tests for diagnosing periprosthetic joint infection in patients who underwent unicompartmental knee arthroplasty (UKA). 
In this study of 259 knees, the optimal cut-off values in UKA were similar to those used for TKA, including:
- 27 mm/h for the ESR
- 14 mg/L for the CRP
- 6200/μL for the synovial fluid WBC count
- 60% for the differential
Theses markers, therefore, are useful for diagnosing periprosthetic joint infection after UKA, with optimal cut-off values comparable to those used for TKA. However, the optimal synovial WBC count was found to be slightly higher, which may be due to the unresurfaced compartments.
In general, using a combination of a screening ESR and CRP followed by selective aspiration of the joint (if these values are abnormal or if the clinical suspicion for infection is high) should help the clinician rule in or out periprosthetic joint infection in most cases. Synovial fluid obtained should be sent for a synovial fluid WBC count and differential, with threshold values of approximately 3,000 WBC/uL and approximately 80% respectively.
Diagnosis of infection in the early postoperative period utilizes the same tests, but with different optimal threshold values:
- Serum CRP of approximately 100 mg/L (10 being normal)
- Synovial fluid WBC count of approximately 10,000 WBC/uL
- Differential of 90%
When interpreting these values, it is important to realize they will change over time within the first few weeks postoperatively.
In general, avoid the indiscriminate use of antibiotics prior to making the diagnosis to limit the risk of a culture-negative infection.
Nicholas Ting, MD, and Craig Della Valle, MD, are from Rush University Medical Center, Chicago, Illinois.
References and Additional Reading
- Bozic KJ, Kurtz SM, Lau E, et al. The epidemiology of revision total knee arthroplasty in the United States. Clinical orthopaedics and related research 2010;468:45-51.
- Kurtz SM, Ong KL, Lau E, Bozic KJ, Berry D, Parvizi J. Prosthetic joint infection risk after TKA in the Medicare population. Clinical orthopaedics and related research 2010;468:52-6.
- Parvizi J, Della Valle CJ. AAOS Clinical Practice Guideline: diagnosis and treatment of periprosthetic joint infections of the hip and knee. The Journal of the American Academy of Orthopaedic Surgeons 2010;18:771-2.
- Della Valle CJ, Sporer SM, Jacobs JJ, Berger RA, Rosenberg AG, Paprosky WG. Preoperative testing for sepsis before revision total knee arthroplasty. The Journal of arthroplasty 2007;22:90-3.
- Greidanus NV, Masri BA, Garbuz DS, et al. Use of erythrocyte sedimentation rate and C-reactive protein level to diagnose infection before revision total knee arthroplasty. A prospective evaluation. The Journal of bone and joint surgery American volume 2007;89:1409-16.
- Barrack RL, Jennings RW, Wolfe MW, Bertot AJ. The Coventry Award. The value of preoperative aspiration before total knee revision. Clinical orthopaedics and related research 1997:8-16.
- Wetters NG, Berend KR, Lombardi AV, Morris MJ, Tucker TL, Della Valle CJ. Leukocyte esterase reagent strips for the rapid diagnosis of periprosthetic joint infection. The Journal of arthroplasty 2012;27:8-11.
- Parvizi J, Jacovides C, Antoci V, Ghanem E. Diagnosis of periprosthetic joint infection: the utility of a simple yet unappreciated enzyme. The Journal of bone and joint surgery American volume 2011;93:2242-8.
- El Espera I, Blondet C, Moullart V, et al. The usefulness of 99mTc sulfur colloid bone marrow scintigraphy combined with 111In leucocyte scintigraphy in prosthetic joint infection. Nuclear medicine communications 2004;25:171-5.
- Della Valle CJ, Scher DM, Kim YH, et al. The role of intraoperative Gram stain in revision total joint arthroplasty. The Journal of arthroplasty 1999;14:500-4.
- Morgan PM, Sharkey P, Ghanem E, et al. The value of intraoperative Gram stain in revision total knee arthroplasty. The Journal of bone and joint surgery American volume 2009;91:2124-9.
- Lonner JH, Desai P, Dicesare PE, Steiner G, Zuckerman JD. The reliability of analysis of intraoperative frozen sections for identifying active infection during revision hip or knee arthroplasty. The Journal of bone and joint surgery American volume 1996;78:1553-8.
- Font-Vizcarra L, Garcia S, Martinez-Pastor JC, Sierra JM, Soriano A. Blood culture flasks for culturing synovial fluid in prosthetic joint infections. Clinical orthopaedics and related research 2010;468:2238-43.
- Bedair H, Ting N, Jacovides C, et al. The Mark Coventry Award: diagnosis of early postoperative TKA infection using synovial fluid analysis. Clinical orthopaedics and related research 2011;469:34-40.
- Yi PB CM, Moric M, Sporer SM, Berger RA, Della Valle CJ. Diagnosis of Infection in the Early Postoperative Period Following Total Hip Arthroplasty. 2013;In press.
- Cipriano CA, Brown NM, Michael AM, Moric M, Sporer SM, Della Valle CJ. Serum and synovial fluid analysis for diagnosing chronic periprosthetic infection in patients with inflammatory arthritis. The Journal of Bone and Joint Surgery American volume 2012;94:594-600.
- Society of Unicondylar R, Continuing E. Diagnosis of periprosthetic joint infection after unicompartmental knee arthroplasty. The Journal of Arthroplasty 2012;27:46-50.