Infection rates of primary hip and knee joint arthroplasty have been reduced to 0.3% to 2% with modern aseptic techniques, but this rate may reach 20% in some revision procedures.1,2 Periprosthetic joint infection (PJI) can be disastrous, and a prompt diagnosis of the offending organism is important for accurate and effective treatment protocols to be initiated. Although technology has improved, we are still faced with challenges relating to the diagnosis of PJI.
Imaging plays a critical role in the diagnosis and management of patients with musculoskeletal infections and PJI. It is important to keep the clinical question of concern in mind, as well as the advantages and disadvantages of each imaging modality, when requesting imaging of suspected areas of infection. In many cases, imaging aids in the early diagnosis of infection, which is critical in limiting morbidity and infection-related complications. Multiple imaging modalities may be useful in diagnosing and evaluating musculoskeletal infection, including conventional radiography, MRI, CT, ultrasound, and nuclear medicine imaging.
Radiography is generally considered the initial study of choice by the American College of Radiology Appropriateness Criteria, and is almost always the first step in imaging areas of suspected infection.3 The main advantages of plain radiography are its ability to survey large areas of the body, widespread availability, and reproducibility; these features allow for initial baseline and follow-up evaluation. The primary drawback of plain radiography is its limited sensitivity and specificity for diagnosing osteomyelitis because a significant amount of bone loss (30% to 50%) is necessary to result in visible changes; radiographic findings may lag behind the clinical onset of disease by 1 to 2 weeks.4 In many cases, plain radiography is performed in conjunction with alternate imaging modalities to help establish the diagnosis, and the plain film result should not preclude additional imaging studies.3
Magnetic Resonance Imaging
MRI has become a critical imaging modality in detecting musculoskeletal infections and determining the extent of disease. It is more sensitive than other imaging modalities for evaluating the extent of infection5 because of its superior soft-tissue contrast and high-resolution multiplanar technique. As a result, MRI is considered the modality of choice for detecting osteomyelitis and associated soft-tissue infections,6,7 and is complementary to plain radiography according to the American College of Radiology Appropriateness Criteria.3 The drawbacks of MRI are the relatively long acquisition times required to obtain images and the limited field of view, which may preclude the survey of large areas of the body in acutely ill patients. Additionally, MRI may be of limited value in the setting of PJI because of metal artifact, or in patients with contraindicated implantable devices, such as various pacemakers and cochlear implants.
MRI protocols vary in the setting of osteomyelitis, and the diagnosis can usually be achieved with the use of multiplanar T1-weighted nonfat-suppressed sequences and fluid-sensitive sequences. Fat-suppressed, fluid-sensitive sequences (short T1 inversion recovery or T2 with fat suppression) are the most sensitive and may show increased signal in the setting of marrow and/or periosteal edema, soft-tissue edema, and fluid collections. T1 images are most specific for the diagnosis of osteomyelitis,8 and demonstrate confluent marrow replacement with decreased signal intensity that is darker than skeletal muscle, often with associated cortical destruction or erosion. Intravenous gadolinium may not be necessary for the diagnosis of osteomyelitis9 with adequate T1 and T2 noncontrast images. However, contrast-enhanced sequences may be helpful in evaluating soft-tissue infections,10 specifically in the setting of drainable rim-enhancing fluid collections, areas of soft-tissue devitalization, or identification of sinus tracts. As with any imaging technique, MRI has its limitations; in some cases, its specificity for infection is diminished. Marrow signal changes on MRI can be seen in instances of noninfectious inflammatory arthropathy. Noninfected bone marrow signal abnormalities that are seen adjacent to areas of active soft-tissue infection may actually represent reactive osteitis and be confused with an active bone infection.11
CT has a role in some cases of musculoskeletal infection, and its findings often overlap with those seen on plain radiography. The primary advantages of CT are high spatial resolution, the ability to survey a large area with a short scan time, multiplanar technique, and superior evaluation of soft-tissue architecture compared with plain radiography.12 Contrast-enhanced CT may be helpful to depict fluid collections, joint effusions, or areas of soft-tissue inflammation.
Dual-energy CT (DECT) is a relatively new imaging technique that has shown promise in musculoskeletal imaging applications. DECT acquires images at two different energies, typically 80 kVp and 140 kVp. Because various materials behave differently at each energy level, postprocessing techniques can be used to create virtual images that detect materials of differing atomic numbers and therefore demonstrate the chemical composition of certain structures. In musculoskeletal imaging, noncontrast DECT has the ability to discriminate uric acid deposits from calcium; these substances can be difficult to differentiate by plain radiography or conventional CT. This may be helpful in identifying cases of gout13,14 because an acute gouty flare can present in a similar fashion to infection. Also, when DECT is performed with IV contrast, postprocessing techniques allow for generation of iodine maps to assess the distribution of iodine atoms within the contrast medium and distinguish true enhancement from calcium, which may be a clue to the presence or absence of bone infection.
With ultrasound’s inability to penetrate bone, its utility to diagnose osteomyelitis is somewhat limited, but it can be useful in evaluating soft-tissue infections, particularly when there is concern for fluid collections or effusions amenable to percutaneous drainage. Thus, ultrasound is typically used for evaluation of infection confined to the soft tissues, as an adjunct to other imaging modalities in cases of osteomyelitis, or as a tool for image-guided intervention. Deep soft-tissue swelling on ultrasound is an early finding in osteomyelitis, followed by periosteal elevation and a thin layer of periosteal fluid that can progress to abscess formation.15 Ultrasound-guided aspiration may be useful for imaging around metal implants when there is concern for PJI, and it has been shown to have a higher sensitivity and specificity for predicting septic loosening of a hip prosthesis compared with fluoroscopic-guided aspiration.16
Nuclear medicine imaging techniques, including bone scintigraphy, leukocyte-labeled white blood cells (technitium-99m or indium In 111), and (18F) fluorodeoxyglucose (FDG) positron emission tomography (PET), can been used to diagnose infection. However, these modalities are becoming less common with the advent of widespread MRI and are typically considered only when MRI is not available or is contraindicated.3 In most cases, nuclear medicine imaging demonstrates poor spatial resolution compared with MRI, and it does not have the same diagnostic yield for infection.17
Three-phase bone scans may be helpful in the setting of suspected PJI because it is often difficult to distinguish PJI clinically from aseptic loosening. A negative three-phase bone scan effectively excludes both infection and aseptic loosening. Both loosening and infection demonstrate abnormal uptake around the prosthesis; however, the pattern of uptake is different. Loosening demonstrates focal uptake around the tip of the prosthesis and infection shows more generalized uptake. Leukocyte scanning with labeled white blood cells (WBC) can be achieved with either indium In 111 or 99mTc-labeled hexamethylpropylene amine oxime with a reported sensitivity and specificity between 80% to 90% for diagnosis of osteomyelitis.18,19 However, chronic infections, infections of the spine, and previous antibiotic therapy have been shown to reduce sensitivity; thus, tagged WBC studies are less useful in these circumstances.20 FDG-PET offers some advantages over other radionuclide methods because of more precise anatomic localization when performed in conjunction with CT, as well as quantitative analysis of the lesion being evaluated. FDG-PET has a reported sensitivity of >95% and a specificity of 75% to 99% for diagnosis of acute and subacute soft-tissue infection and bone infection.21 Additionally, because of FDG accumulation in macrophages, FDG-PET is useful for diagnosis of chronic and low-grade infection and has been shown to have a sensitivity of 100% and a specificity of 88% to 92% in these instances.
Traditional Cultures: When and How
Improving the microbiologic diagnosis of PJI remains a challenge. New diagnostic methods, such as multiplex polymerase chain reaction (PCR) and mass spectrometry, are under development, but the utility of these highly sensitive diagnostic techniques with unknown specificity should be interpreted with caution.22 Such methods also fail to determine the antibiotic susceptibility of the identified organisms, thus making traditional microbiology the mainstay for identifying organisms causing PJI and determining antimicrobial susceptibility. Only by knowing the antibiotic susceptibility of the infecting organism can administration of appropriate antibiotic therapy be reasonably ensured. Unfortunately, rates of negative intraoperative cultures range from 10% to 30%, making management of these infections difficult.23 Attempts need to be made to optimize the yield of traditional cultures.
It has been recommended that antibiotic treatment be delayed in patients with suspected PJI until after cultures from the joint have been obtained, and it is widely accepted by orthopaedic surgeons that antibiotics be withheld until aspiration has been performed to increase the odds of identifying an organism.24 In one study, the sensitivity of microscopy in all patients dropped from 58% to 12% when patients received antibiotics before aspiration (native knees, 46% to 0%, prosthetic knees, 72% to 27%).25 It has also been suggested that antibiotics for surgical prophylaxis be given only after intraoperative sample collection, but more recent data suggest that a single dose of preoperative antibiotics does not significantly decrease the yield of culture results.26 Because of the known positive effects of preoperative antibiotic surgical prophylaxis in routine primary arthroplasty procedures, the authors recommend preoperative antibiotics be given before skin incision in the setting of suspected PJI.26
Three to six tissue samples should be obtained whenever possible and submitted for aerobic and anaerobic culture. Tissue swabs should be avoided because of their low sensitivity, specificity, and negative and positive predictive values compared with tissue sampling.24,27 In one retrospective study of 73 patients undergoing 77 revision arthroplasties, obtaining only one intraoperative culture was compared with obtaining five separate specimens. 28 Obtaining multiple cultures changed the microbiological diagnosis in 26 of 77 cases (34%) and the antibiotic treatment in 23 of 77 cases (30%). This analysis demonstrates that obtaining multiple cultures can significantly change patient care, and it suggests that one or two cultures are insufficient.28 Culture of synovial fluid inoculated into blood culture vials may be more sensitive than intraoperative swab and tissue cultures. Font-Vizcarra et al29 reported that synovial fluid cultured in blood culture vials had higher sensitivity, specificity, and positive and negative predictive values for diagnosis of PJI compared with standard tissue and swab samples. However, the usefulness of this technique was less in chronic infections than that in acute infections. In addition, evidence supports the use of sonication of prosthetic implants to improve microbiologic yield, especially in patients on prior antibiotics, but many microbiology laboratories have been slow to adopt this practice routinely.30
Cultures of sinus tracts and the surface of wounds should be avoided because they likely reflect skin colonization. Tetreault et al31 assessed the utility of culturing draining wounds or sinuses in patients with suspected PJI. Fifty-five patients with a draining wound or sinus after total joint arthroplasty (28 knees, 27 hips) who had not received antibiotics for at least 2 weeks were prospectively studied. Superficial wound cultures were compared with intra-articular cultures to determine the accuracy in isolating the infecting organism or organisms. The superficial cultures were concordant with deep cultures in only 26 of 55 cases (47.3%) and were more likely to yield multiple bacteria. In 23 cases (41.8%), the superficial cultures would have led to a change in antibiotic regimen. Given the potential to adversely affect patient care, we recommend against obtaining superficial cultures in patients with a draining wound or sinus following hip or knee arthroplasty.
Intraoperative Frozen Section
Few laboratory tests are available that can be completed quickly enough to be of practical use to an orthopaedic surgeon during surgery. Although many laboratories still perform Gram stains on fluid obtained from patients with suspected PJI, it has been found to be of limited value with a sensitivity of only 27%.32 Peri-implant tissue sampling can be an effective diagnostic tool; the turnaround time for interpreting a single frozen section is approximately 20 minutes from the time it is received. Frozen sections have the advantage of being available relatively quickly, but the freezing process induces artifacts not seen in formalin-fixed tissue. For this reason, the frozen tissue blocks and other tissue from the same case are routinely processed in formalin for reinterpretation the next day (sometimes called permanent sections). Despite this limitation, experience suggests that frozen sections may be helpful to rule in or rule out infection at the time of revision arthroplasty.33
Critical to interpreting biopsies of peri-implant tissues is the recognition that different mechanisms of arthroplasty failure may produce different patterns of tissue reaction and/or inflammation. For example, frozen sections of tissue around an implant that has failed as a result of particle-induced osteolysis usually contain a high concentration of macrophages. The foamy quality of macrophage cytoplasm is caused by countless submicron particles that have been phagocytized. Perivascular lymphocytes are commonly found around many failed implants and are of unknown significance. Occasionally, cases show more diffusely distributed lymphocytes and plasma cells, but acute inflammation, characterized by neutrophils (polymorphonuclear leukocytes [PMNs]), is not a feature of particle-induced osteolysis.34 Implants that are grossly loose may induce the formation of a highly vascular membrane (granulation tissue); neutrophils are commonly seen within the capillaries and postcapillary venules of granulation tissue, but implant motion and/or fluid pressure alone are not thought to induce the movement of neutrophils out of blood vessels into the membrane itself. It follows that the absence of neutrophils on a frozen section of peri-implant tissue helps support the clinical impression of aseptic loosening.
Some implant membranes show marked acute inflammation in which the high concentration of neutrophils is essentially diagnostic of infection; however, biopsies that contain a low concentration of neutrophils can be more difficult to interpret. Many studies have attempted to define an optimum cut-off threshold for the tissue concentration of neutrophils to support the diagnosis of infection. A meta-analysis by the AAOS Clinical Practice Guidelines Committee24 identified eight studies of high enough quality to support either of two different diagnostic thresholds: (1) 10 neutrophils in each of five 400× high-power fields, or (2) 5 neutrophils in each of five high-power fields as a cut-off for diagnosing infection. The meta-analysis suggested that either level would be a good diagnostic test, but as expected, requiring fewer neutrophils per unit area slightly reduced specificity and increased sensitivity. Morawietz et al35 quantified neutrophils in 147 periprosthetic membranes from cases of aseptic loosening as well as infection, and correlated the morphologic results with the results of microbiologic cultures and clinical diagnoses. Using receiver-operating characteristic curves (ROCC), the authors suggested an optimum threshold of a total of 23 neutrophils in 10 high-power fields. This threshold yielded a sensitivity of 73% and a specificity of 95% using microbiologic cultures as the benchmark, and a sensitivity of 77% and a specificity of 97% using clinical impression as the reference standard for infection.
Several caveats should be kept in mind when interpreting frozen sections to diagnose infection during revision arthroplasty. First, joint capsules and peri-implant membranes of patients with an underlying inflammatory arthropathy (eg, rheumatoid arthritis) have not been carefully studied, thus morphologic criteria for diagnosing infection in those patients have not been established. Many of the studies that have attempted to establish a threshold of neutrophil concentrations for diagnosing periprosthetic infections have excluded patients known to have underlying inflammatory arthropathies. Also, inflammatory cells can be difficult to identify in cauterized tissue samples; thus surgeons are encouraged to use as little cautery as possible when submitting these samples for frozen section. Because surgeons often ream the femoral or tibial canal to obtain a sample, pathologists need to recognize that neutrophils within hematopoietic bone marrow do not necessarily reflect an infection.
Lymphocytes and plasma cells are commonly present in biopsies of patients who have been treated with antibiotics for infection, but these cells are nonspecific and are not predictive of active infection. Some specimens obtained at the second stage of a two-stage revision arthroplasty for known infection contain extensive surface necrosis, thick fibrous membranes, sheets of lymphocytes and plasma cells, and marked perivascular lymphocytes. These morphologic features can be similar to the inflammation that has been attributed to metal ions from failed metal-on-metal total hip prostheses,36 now sometimes referred to as an inflammatory pseudotumor.37 When this inflammatory reaction develops around an antibiotic-containing cement spacer, however, the inflammatory reaction is unlikely to reflect an immune reaction to metal ions. Finally, acute inflammation can accompany a recent periprosthetic fracture; thus a frozen section to rule out infection in the setting of a recent fracture could be interpreted as a false-positive.
With respect to the use of frozen sections to help diagnose periprosthetic infection, if there are morphologic differences between infections caused by virulent versus less virulent organisms, the question becomes whether threshold criteria should be site-specific.
The most common bacterial species associated with periprosthetic infections in the hip and knee is Staphylococcus aureus, and no evidence at this time suggests different morphologic thresholds should be used for those two sites. However, Propionibacterium acnes is thought to be a common pathogen associated with failed shoulder arthroplasty, and it is an example of an organism of relatively low virulence.
Grosso et al38 retrospectively reviewed laboratory test results of 45 patients who underwent revision shoulder arthroplasty. Multiple observations were used to define the four final clinical diagnoses of not infected, possibly infected, probably infected, or definitely infected. Routine laboratory tests, cultures, and frozen sections were obtained in all cases. Histology was reviewed, and four potential cut-off thresholds to diagnose infection were applied. The actual neutrophil counts were also obtained; therefore, a ROCC could be used to calculate an optimum threshold. There were no false-positive frozen section diagnoses, but the diagnostic sensitivity of all thresholds was lower than that usually reported for frozen sections from the hip and knee. The ROCC analysis indicated that 7 to 10 total neutrophils in 10 high-power fields would yield 72% sensitivity and 100% specificity. In general, these findings support the concept that infections associated with lower virulent bacteria may induce less pyogenic inflammation, and thus may be more difficult to recognize using the criteria generally applied to infections around the hip and knee.
Regarding whether the same criteria should be used at the second stage of a two-stage revision for known infection, we are aware of only one study that has specifically addressed this issue. Bori et al39 reviewed 21 cases of second stage reimplantation, seven of which were found to be culture-positive. The authors reviewed frozen section slides using several different cut-off thresholds and concluded the probability of infection is high when at least five PMNs per high-power field are found in the tissue, but it is not possible to rule out infection when the number of PMNs is less than five. In general, this supports one of the thresholds advocated by the AAOS Clinical Practice Guidelines committee.24
With regard to the frequency of discrepancies between the frozen section and additional samples submitted for permanent histology, and what represents optimum sampling for frozen sections during surgery, few studies have evaluated discrepancies between frozen and permanent sections in this context. Stroh et al40 reviewed frozen sections, permanent histology, and other tests in 76 patients who underwent staged revision knee arthroplasty for apparent infection. Of 304 frozen sections, 297 (97.7%) were concordant with permanent sections. There were seven false-negative frozen section interpretations; however, in each case, additional samples had been obtained during the surgery, so patient management was unaffected. In our experience, no studies provide quantitative data to support an optimum sampling protocol for frozen section diagnosis of periprosthetic infection. Additionally, there currently are no studies with adequate follow-up to determine the clinical consequences of an unexpected positive frozen section in a patient who has no other clinical or laboratory features of periprosthetic infection.
When using the recommended diagnostic thresholds, frozen sections of periprosthetic tissues around the hip and knee should have a sensitivity of 70% to 80%, with a specificity >90%. Frozen sections are less sensitive in detecting organisms of lower virulence, so a lower sensitivity is expected when hip and knee diagnostic thresholds are applied to periprosthetic tissues around shoulder implants. False-positive frozen section interpretations are rare (<2%), and are sometimes associated with tissue that has been distorted by thermal artifact. The clinical consequences of an unexpected positive frozen section in the absence of other clinical or laboratory features of infection are still unknown.
Propionibacterium acnes Orthopaedic Infection
With improved awareness of its presence, P acnes is now recognized as the most common cause of shoulder infection.41P acnes is a slow-growing, fastidious gram-positive bacillus that resides in the pilosebaceous glands common in the axilla, especially in male patients.42,43 These glands produce more sebum in response to high-testosterone environments (eg, pubescent males); the glands can then become clogged, creating comedones and the dermatologic manifestations of acne vulgaris. The slow growth and absence of a suppurative host response makes P acnes very difficult to detect, and the bacterium was generally considered nonvirulent until recently.44
Clinical findings are frequently underwhelming. Erythema, purulence, and wound drainage are rarely observed and pain is the only consistent, although entirely nonspecific, clinical finding.43,45,46 No commercially available preoperative or intraoperative test reliably predicts subsequent P acnes growth in culture.47 Laboratory studies consisting of erythrocyte sedimentation rate, C-reactive protein level, and WBC count with differential are not helpful,43 and serum interleukin-6 (IL-6) is ineffective for diagnosis of P acnes.48 Because cultures are the only means of diagnosis, one might expect preoperative aspiration to be helpful; however, large effusions are uncommon with P acnes, thus making aspiration difficult. Fluid cultures obtained from routine aspiration have not been helpful diagnostically, and synovial leukocyte count is rarely positive.46,47 Frozen sections for acute inflammation do not correlate with ultimate culture growth.49
The current standard for diagnosis of P acnes is long-hold cultures of tissue obtained with arthroscopic biopsy,50 open biopsy, or at the time of revision surgery.47 The mean time to growth of this organism ranges from 7 to 13 days but can take up to 3 weeks, thus adding to the diagnostic challenge.46,51,52 Unfortunately, these longer hold times may contribute to growth of contaminants and potential false-positives.51 Without a widely available confirmatory test or clinical findings, the clinician must decide when to ignore and when to act on growth in only a small percentage of cultures, or growth that occurs after many days of incubation. We recommend holding cultures for 14 days, and taking at least five cultures to improve diagnostic accuracy.
Given the absence of preoperative or intraoperative findings, revisions for presumed mechanical loosening occasionally will have cultures that are unexpectedly positive. In this setting, the clinician must decide whether a positive culture represents a true infection. Patients with unexpectedly positive P acnes cultures during revision shoulder arthroplasty have good outcomes, with only 6% to 10% exhibiting persistent, demonstrable infection.53,54 However, more infections may be detected with further follow-up because this organism is slow growing.
Research has focused on identifying the true pathogenicity of P acnes infections. Some shoulder infections are clearly caused by P acnes, but the significance of P acnes detected in deep shoulder tissue or fluid may be ambiguous. Without a reliable confirmatory test, the choice of response may be confusing when a minority of cultures is positive. It is unclear whether the positive cultures are true pathogens or contaminants. Some clinicians have hypothesized there are pathogenic and nonpathogenic strains of P acnes.42,55,56 PCR has shown no genomic difference between P acnes found in implants known to be infected and those strains found in implants felt to be uninfected.56 In a small study by Nodzo et al,55 the hemolytic phenotype of P acnes on brucella blood agar was shown to be a potential marker of pathogenicity, and the authors suggest that hemolytic strains of P acnes be treated as true pathogens.
Molecular techniques currently at the investigational stage, such as PCR, hold the potential of providing a confirmatory test to definitively confirm or refute the presence of P acnes, in addition to providing information about particular strains and sensitivities that may be clinically relevant. However, these tools may be exquisitely sensitive, so much so that P acnes DNA may be detected when it is introduced to tissue from the wound margin during surgery or to the synovial fluid during arthrocentesis. Quantification of molecular results may assist in disentangling the signal-to-noise ratio, but this remains to be investigated.
Although significant progress has been made in the past decade regarding the recognition and diagnosis of P acnes orthopaedic infection, much information needs to be determined, such as periprocedural strategies that minimize the risk of introducing P acnes into the deep shoulder tissue during primary arthroplasty, reconstruction, or arthrodesis; molecular diagnostic techniques that assist with the interpretation of positive culture results in revision arthroplasty; and the optimum antibiotic regimen for treatment of P acnes detected in deep shoulder specimens that leads to improved outcomes in the long term.57
Synovial Fluid Markers
Synovial WBC count analysis is an invaluable test for diagnosing PJI. It has a high sensitivity and specificity and, in some instances, it may be a superior diagnostic test to the benchmark of fluid culture. This test is included as a central component of the diagnostic criteria for PJI by the Musculoskeletal Infection Society and the International Consensus on Periprosthetic Joint Infection diagnostic criteria workgroup.24 One of the main advantages of this test is that a synovial fluid sample is relatively easy to obtain, and its analysis is conducted on standard hematologic measurement instruments available at hospitals worldwide. The results are generally obtained within an hour and can be immediately interpreted by the treating clinician to help establish or rule out the diagnosis of infection.
In recent years, thresholds have been established to aid in the diagnosis of PJI in many different clinical populations that include patients with chronic PJI, patients with acute postoperative infections, those with unicondylar arthroplasty, and patients with inflammatory arthropathies.58,59 Additionally, the literature suggests the synovial WBC count is similar in patients with and without systemic inflammatory conditions.58 However, this test has certain limitations. The synovial WBC count provides an analysis of the inflammatory status of the intra-articular environment without providing specific detail as to the cause of inflammation. In certain clinical situations, such as crystalline arthropathy-mediated inflammation, the inflammation could be confused with infection. Moreover, certain patients with PJI who are systemically immunosuppressed may have an artificially low WBC count that could be confused with an aseptic process. In addition, the diagnostic thresholds for infection can vary based on the clinical situation and thus require a certain baseline knowledge of which thresholds should be used with each clinical scenario. At this time, the synovial WBC count analysis provides an accurate, rapid, and easily accessible test to aid in the diagnosis of PJI; however, advancements in molecular testing may eventually supplant synovial WBC counts by providing a more specific etiology for the intra-articular inflammation beyond just demonstrating the presence of an inflammatory process.
Synovial fluid IL-6 is a potential marker in assisting with the diagnosis of PJI. Lenski et al60 evaluated synovial fluid IL-6 in 31 infected and 38 aseptic total joint arthroplasty patients. The authors found IL-6 to have a 90% sensitivity and a 94.7% specificity using ROCC. Randau et al61 found synovial IL-6 to have a specificity of 100% but a sensitivity of only 50% with levels above 9,000 pg/mL. The data support the potential use of IL-6 as a diagnostic marker, especially if threshold levels are met, but further research is necessary for routine use.
Another test demonstrated to have a high accuracy in diagnosing PJI is the synovial fluid α-defensin test.62 The α-defensin protein is an antimicrobial peptide that is released by neutrophils responding to pathogens; it concentrates in the synovial fluid, thus indicating the presence of infection. The α-defensin test has been demonstrated to closely match the results of the more complex Musculoskeletal Infection Society criteria for PJI, attaining a sensitivity of 97% and a specificity of 96% when combined with synovial fluid CRP levels.63 These results included patients on antibiotics and patients with systemic inflammatory diseases. The α-defensin testing has also been shown to outperform the leukocyte esterase test, achieving a higher accuracy in diagnosing PJI.64 The cost-effectiveness, rapid results, and analytic nature of the α-defensin test for PJI make it an excellent method to aid in the workup of patients with a painful arthroplasty.
As the state of molecular and point-of-care diagnostics expands, the potential for a more rapid and accurate diagnosis of PJI will emerge. Intraoperative cultures and joint aspirations will remain imperative for the diagnosis of the offending organism, as well as determining antibiotic susceptibilities for treatment protocols. Frozen sections can be used to assist the surgeon intraoperatively in making real-time decisions about active acute inflammation, whereas various synovial fluid markers, such as α-defensin and IL-6, are emerging as excellent indicators of PJI. Future research should be focused on improving the diagnosis of hard to detect and slow-growing organisms such as P acnes and on establishing new, more sensitive and more specific criteria for diagnosing PJI. A definitive test for the diagnosis of PJI remains elusive, and the recommendations established by the AAOS and the International Consensus on Periprosthetic Joint Infection should be used to guide the clinician in everyday practice.
References printed in bold type are those published within the past 5 years.
1. Lentino JR: Prosthetic joint infections: Bane of orthopedists, challenge for infectious disease specialists. Clin Infect Dis 2003;36(9):1157–1161.
2. Joseph TN, Chen AL, Di Cesare PE: Use of antibiotic-impregnated cement in total joint arthroplasty. J Am Acad Orthop Surg 2003;11(1):38–47.
3. Schweitzer ME, Daffner RH, Weissman BN, et al.: ACR Appropriateness Criteria on suspected osteomyelitis in patients with diabetes mellitus. J Am Coll Radiol 2008;5(8):881–886.
4. Game FL: Osteomyelitis in the diabetic foot: Diagnosis and management. Med Clin North Am 2013;97(5):947–956.
5. Kapoor A, Page S, Lavalley M, Gale DR, Felson DT: Magnetic resonance imaging for diagnosing foot osteomyelitis: A meta-analysis. Arch Intern Med 2007;167(2):125–132.
6. Liu C, Bayer A, Cosgrove SE, et al.: Clinical practice guidelines by the infectious diseases society of america for the treatment of methicillin-resistant Staphylococcus aureus infections in adults and children: Executive summary. Clin Infect Dis 2011;52(3):285–292.
7. Turecki MB, Taljanovic MS, Stubbs AY, et al.: Imaging of musculoskeletal soft tissue infections. Skeletal Radiol 2010;39(10):957–971.
8. Johnson PW, Collins MS, Wenger DE: Diagnostic utility of T1-weighted MRI characteristics in evaluation of osteomyelitis of the foot. AJR Am J Roentgenol 2009;192(1):96–100.
9. Craig JG, Amin MB, Wu K, et al.: Osteomyelitis of the diabetic foot: MR imaging-pathologic correlation. Radiology 1997;203(3):849–855.
10. Kan JH, Young RS, Yu C, Hernanz-Schulman M: Clinical impact of gadolinium in the MRI diagnosis of musculoskeletal infection in children. Pediatr Radiol 2010;40(7):1197–1205.
11. Rozzanigo U, Tagliani A, Vittorini E, Pacchioni R, Brivio LR, Caudana R: Role of magnetic resonance imaging in the evaluation of diabetic foot with suspected osteomyelitis. Radiol Med 2009;114(1):121–132.
12. Fayad LM, Carrino JA, Fishman EK: Musculoskeletal infection: Role of CT in the emergency department. Radiographics 2007;27(6):1723–1736.
13. Nicolaou S, Yong-Hing CJ, Galea-Soler S, Hou DJ, Louis L, Munk P: Dual-energy CT as a potential new diagnostic tool in the management of gout in the acute setting. AJR Am J Roentgenol 2010;194(4):1072–1078.
14. Choi HK, Burns LC, Shojania K, et al.: Dual energy CT in gout: A prospective validation study. Ann Rheum Dis 2012;71(9):1466–1471.
15. Mah ET, LeQuesne GW, Gent RJ, Paterson DC: Ultrasonic features of acute osteomyelitis in children. J Bone Joint Surg Br 1994;76(6):969–974.
16. Battaglia M, Vannini F, Guaraldi F, Rossi G, Biondi F, Sudanese A: Validity of preoperative ultrasound-guided aspiration in the revision of hip prosthesis. Ultrasound Med Biol 2011;37(12):1977–1983.
17. Morbach H, Schneider P, Schwarz T, et al.: Comparison of magnetic resonance imaging and 99mTechnetium-labelled methylene diphosphonate bone scintigraphy in the initial assessment of chronic non-bacterial osteomyelitis of childhood and adolescents. Clin Exp Rheumatol 2012;30(4):578–582.
18. Schauwecker DS: Osteomyelitis: Diagnosis with In-111-labeled leukocytes. Radiology 1989;171(1):141–146.
19. Guze BH, Webber MM, Hawkins RA, Sinha K: Indium-111 white blood cell scans: Sensitivity, specificity, accuracy, and normal patterns of distribution. Clin Nucl Med 1990;15(1):8–10.
20. El-Maghraby TA, Moustafa HM, Pauwels EK: Nuclear medicine methods for evaluation of skeletal infection among other diagnostic modalities. Q J Nucl Med Mol Imaging 2006;50(3):167–192.
21. Palestro CJ: FDG-PET in musculoskeletal infections. Semin Nucl Med 2013;43(5):367–376.
22. Rak M, Barlič-Maganja D, Kavčič M, Trebše R, Cőr A: Comparison of molecular and culture method in diagnosis of prosthetic joint infection. FEMS Microbiol Lett 2013;343(1):42–48.
23. Shanmugasundaram S, Ricciardi BF, Briggs TW, Sussmann PS, Bostrom MP: Evaluation and Management of Periprosthetic Joint Infection-an International, Multicenter Study. HSS J 2014;10(1):36–44.
24. Della Valle C, Parvizi J, Bauer TW, et al.; American Academy of Orthopaedic Surgeons: American Academy of Orthopaedic Surgeons clinical practice guideline on: The diagnosis of periprosthetic joint infections of the hip and knee. J Bone Joint Surg Am 2011;93(14):1355–1357.
25. Hindle P, Davidson E, Biant LC: Septic arthritis of the knee: The use and effect of antibiotics prior to diagnostic aspiration. Ann R Coll Surg Engl 2012;94(5):351–355.
26. Tetreault MW, Wetters NG, Aggarwal V, Mont M, Parvizi J, Della Valle CJ: The Chitranjan Ranawat Award: Should prophylactic antibiotics be withheld before revision surgery to obtain appropriate cultures? Clin Orthop Relat Res 2014;472(1):52–56.
27. Aggarwal VK, Higuera C, Deirmengian G, Parvizi J, Austin MS: Swab cultures are not as effective as tissue cultures for diagnosis of periprosthetic joint infection. Clin Orthop Relat Res 2013;471(10):3196–3203.
28. DeHaan A, Huff T, Schabel K, Doung YC, Hayden J, Barnes P: Multiple cultures and extended incubation for hip and knee arthroplasty revision: Impact on clinical care. J Arthroplasty 2013;28(suppl 8)59–65.
29. Font-Vizcarra L, García S, Martínez-Pastor JC, Sierra JM, Soriano A: Blood culture flasks for culturing synovial fluid in prosthetic joint infections. Clin Orthop Relat Res 2010;468(8):2238–2243.
30. Trampuz A, Piper KE, Jacobson MJ, et al.: Sonication of removed hip and knee prostheses for diagnosis of infection. N Engl J Med 2007;357(7):654–663.
31. Tetreault MW, Wetters NG, Aggarwal VK, et al.: Should draining wounds and sinuses associated with hip and knee arthroplasties be cultured? J Arthroplasty 2013;28(suppl 8)133–136.
32. Morgan PM, Sharkey P, Ghanem E, et al.: The value of intraoperative Gram stain in revision total knee arthroplasty. J Bone Joint Surg Am 2009;91(9):2124–2129.
33. Bauer TW, Hayashi R: The role of the pathologist in diagnosing periprosthetic infection. Surgical Pathology 2012;5:67–77.
34. Fujishiro T, Moojen DJ, Kobayashi N, Dhert WJ, Bauer TW: Perivascular and diffuse lymphocytic inflammation are not specific for failed metal-on-metal hip implants. Clin Orthop Relat Res 2011;469(4):1127–1133.
35. Morawietz L, Tiddens O, Mueller M, et al.: Twenty-three neutrophil granulocytes in 10 high-power fields is the best histopathological threshold to differentiate between aseptic and septic endoprosthesis loosening. Histopathology 2009;54(7):847–853.
36. Davies AP, Willert HG, Campbell PA, Learmonth ID, Case CP: An unusual lymphocytic perivascular infiltration in tissues around contemporary metal-on-metal joint replacements. J Bone Joint Surg Am 2005;87(1):18–27.
37. Campbell P, Ebramzadeh E, Nelson S, Takamura K, De Smet K, Amstutz HC: Histological features of pseudotumor-like tissues from metal-on-metal hips. Clin Orthop Relat Res 2010;468(9):2321–2327.
38. Grosso MJ, Frangiamore SJ, Ricchetti ET, Bauer TW, Iannotti JP: Sensitivity of frozen section histology for identifying Propionibacterium acnes infections in revision shoulder arthroplasty. J Bone Joint Surg Am 2014;96(6):442–447.
39. Bori G, Soriano A, García S, Mallofré C, Riba J, Mensa J: Usefulness of histological analysis for predicting the presence of microorganisms at the time of reimplantation after hip resection arthroplasty for the treatment of infection. J Bone Joint Surg Am 2007;89(6):1232–1237.
40. Stroh DA, Johnson AJ, Naziri Q, Mont MA: Discrepancies between frozen and paraffin tissue sections have little effect on outcome of staged total knee arthroplasty revision for infection. J Bone Joint Surg Am 2012;94(18):1662–1667.
41. Singh JA, Sperling JW, Schleck C, Harmsen WS, Cofield RH: Periprosthetic infections after total shoulder arthroplasty: A 33-year perspective. J Shoulder Elbow Surg 2012;21(11):1534–1541.
42. Patel A, Calfee RP, Plante M, Fischer SA, Green A: Propionibacterium acnes colonization of the human shoulder. J Shoulder Elbow Surg 2009;18(6):897–902.
43. Pottinger P, Butler-Wu S, Neradilek MB, et al.: Prognostic factors for bacterial cultures positive for Propionibacterium acnes and other organisms in a large series of revision shoulder arthroplasties performed for stiffness, pain, or loosening. J Bone Joint Surg Am 2012;94(22):2075–2083.
44. Bojar RA, Holland KT: Acne and Propionibacterium acnes. Clin Dermatol 2004;22(5):375–379.
45. Dodson CC, Craig EV, Cordasco FA, et al.: Propionibacterium acnes infection after shoulder arthroplasty: A diagnostic challenge. J Shoulder Elbow Surg 2010;19(2):303–307.
46. Millett PJ, Yen YM, Price CS, Horan MP, van der Meijden OA, Elser F: Propionibacterium acnes infection as an occult cause of postoperative shoulder pain: A case series. Clin Orthop Relat Res 2011;469(10):2824–2830.
47. Kelly JD II, Hobgood ER: Positive culture rate in revision shoulder arthroplasty. Clin Orthop Relat Res 2009;467(9):2343–2348.
48. Grosso MJ, Frangiamore SJ, Saleh A, et al.: Poor utility of serum interleukin-6 levels to predict indolent periprosthetic shoulder infections. J Shoulder Elbow Surg 2014;23(9):1277–1281.
49. Topolski MS, Chin PY, Sperling JW, Cofield RH: Revision shoulder arthroplasty with positive intraoperative cultures: The value of preoperative studies and intraoperative histology. J Shoulder Elbow Surg 2006;15(4):402–406.
50. Morman M, Fowler RL, Sanofsky B, Higgins LD: Arthroscopic tissue biopsy for evaluation of infection before revision arthroplasty. J Shoulder Elbow Surg 2011;20(3):e15–e22.
51. Butler-Wu SM, Burns EM, Pottinger PS, et al.: Optimization of periprosthetic culture for diagnosis of Propionibacterium acnes prosthetic joint infection. J Clin Microbiol 2011;49(7):2490–2495.
52. Lutz MF, Berthelot P, Fresard A, et al.: Arthroplastic and osteosynthetic infections due to Propionibacterium acnes: A retrospective study of 52 cases, 1995-2002. Eur J Clin Microbiol Infect Dis 2005;24(11):739–744.
53. Foruria AM, Fox TJ, Sperling JW, Cofield RH: Clinical meaning of unexpected positive cultures (UPC) in revision shoulder arthroplasty. J Shoulder Elbow Surg 2013;22(5):620–627.
54. Grosso MJ, Sabesan VJ, Ho JC, Ricchetti ET, Iannotti JP: Reinfection rates after 1-stage revision shoulder arthroplasty for patients with unexpected positive intraoperative cultures. J Shoulder Elbow Surg 2012;21(6):754–758.
55. Nodzo SR, Hohman DW, Crane JK, Duquin TR: Hemolysis as a clinical marker for propionibacterium acnes orthopedic infection. Am J Orthop (Belle Mead NJ) 2014;43(5):E93–E97.
56. Sampedro MF, Piper KE, McDowell A, et al.: Species of Propionibacterium and Propionibacterium acnes phylotypes associated with orthopedic implants. Diagn Microbiol Infect Dis 2009;64(2):138–145.
57. Crane JK, Hohman DW, Nodzo SR, Duquin TR: Antimicrobial susceptibility of Propionibacterium acnes isolates from shoulder surgery. Antimicrob Agents Chemother 2013;57(7):3424–3426.
58. 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. J Bone Joint Surg Am 2012;94(7):594–600.
59. Dinneen A, Guyot A, Clements J, Bradley N: Synovial fluid white cell and differential count in the diagnosis or exclusion of prosthetic joint infection. Bone Joint J 2013;95(4):554–557.
60. Lenski M, Scherer MA: Synovial IL-6 as inflammatory marker in periprosthetic joint infections. J Arthroplasty 2014;29(6):1105–1109.
61. Randau TM, Friedrich MJ, Wimmer MD, et al.: Interleukin-6 in serum and in synovial fluid enhances the differentiation between periprosthetic joint infection and aseptic loosening. PLoS One 2014;9(2):e89045.
62. Bingham J, Clarke H, Spangehl M, Schwartz A, Beauchamp C, Goldberg B: The alpha defensin-1 biomarker assay can be used to evaluate the potentially infected total joint arthroplasty. Clin Orthop Relat Res 2014;472(12):4006–4009.
63. Deirmengian C, Kardos K, Kilmartin P, Cameron A, Schiller K, Parvizi J: Combined measurement of synovial fluid α-Defensin and C-reactive protein levels: Highly accurate for diagnosing periprosthetic joint infection. J Bone Joint Surg Am 2014;96(17):1439–1445.
64. Deirmengian C, Kardos K, Kilmartin P, et al.: The Alpha-defensin test for periprosthetic joint infection outperforms the leukocyte esterase test strip. Clin Orthop Relat Res 2015;473(1):198–203.