wamp5如何启动apache:美国骨科医师学会：髋膝关节置换术后假体周围感染诊断指南

美国骨科医师学会：髋膝关节置换术后假体周围感染诊断指南

2011-11-02 18:11 来源：丁香园 作者：hotstone

Abstract No preferred test for diagnosis of periprosthetic joint infection exists, and the algorithm for the workup of patients suspected of infection remains unclear. The work group evaluated the available literature to determine the role of each diagnostic modality and devise a practical algorithm that allows physicians to reach diagnosis of periprosthetic joint infection. Ten of the 15 recommendations have strong or moderate evidence in support. These include matters involving erythrocyte sedimentation rate and C-reactive protein level testing, knee and hip aspiration, and stopping the use of antibiotics prior to obtaining intra-articular cultures. The group recommends against the use of intraoperative Gram stain but does recommend the use of frozen sections of peri-implant tissues in reoperation patients in whom infection has not been established, as well as multiple cultures in reoperation patients being assessed for infection. The group recommends against initiating antibiotic treatment in patients with suspected infection until after joint cultures have been obtained, but recommends that prophylactic preoperative antibiotics not be withheld in patients at lower probability for infection.  Previous SectionNext SectionOverview and Rationale This clinical practice guideline was approved by the American Academy of Orthopaedic Surgeons (AAOS) on June 18, 2010. It is based on a systematic review of published studies on the diagnosis of periprosthetic hip and knee joints in patients who have undergone total knee and hip arthroplasty. In addition to providing practice recommendations, this guideline highlights gaps in the literature and areas that require future research.  The purpose of this clinical practice guideline is to help improve treatment based on the current best evidence. Current evidence-based practice standards demand that physicians use the best available evidence in their clinical decision making. To assist in this, the clinical practice guideline consists of a series of systematic reviews of the available literature regarding the modalities and tests that may be used to reach a diagnosis of periprosthetic joint infection (PJI). These systematic reviews were conducted on studies written in English and published during or after 1970. For MRI, CT, and positron emission tomography, only studies published after 2000 were included. For other nuclear imaging modalities, only studies published after 1975 were considered. The review was conducted on articles that were full manuscripts, published in peer-reviewed journals, and of the highest level of available evidence. AAOS staff and the Periprosthetic Joint Infection Diagnosis Work Group systematically reviewed the available literature and subsequently wrote the following recommendations based on a rigorous, standardized process.  Musculoskeletal care is provided in many different settings by many different providers. We created this guideline as an educational tool to guide qualified physicians through a series of treatment decisions in an effort to improve the quality and efficiency of care. This guideline should not be construed as including all proper methods of care or as excluding methods of care reasonably directed to obtaining the same results. The ultimate judgment regarding any specific procedure or treatment must be made in light of all circumstances presented by the patient and the needs and resources particular to the locality or institution. This guideline represents a cross-sectional view of current diagnosis and may become outdated as new evidence becomes available. The AAOS may revise this guideline in accordance with new evidence, changing practice, rapidly emerging treatment options, or new technology. This guideline will be updated or withdrawn in 5 years in accordance with the standards of the National Guideline Clearinghouse.  Previous SectionNext SectionMethodsThe methods used to develop this clinical practice guideline were designed to combat bias, enhance transparency, and promote reproducibility. Their purpose is to allow interested readers the ability to inspect all of the information the work group used to reach all of its decisions and to verify that these decisions are in accord with the best available evidence. The draft of this guideline was subject to peer review and public commentary, and it was approved by the AAOS Evidence-Based Practice Committee; Guidelines and Technology Oversight Committee; Council on Research, Quality Assessment and Technology; and the Board of Directors.  All tables, figures, and appendices, as well as details of the methods used to prepare this guideline, are included in the full clinical practice guideline, which is available at http://www.aaos.org/research/guidelines/PJIguideline.asp.  Previous SectionNext SectionRecommendations Recommendation 1 In the absence of reliable evidence about risk stratification of patients with a potential PJI, it is the opinion of the work group that testing strategies be planned according to whether there is a higher or lower probability that a patient has a hip or knee periprosthetic infection.  Strength of Recommendation: Consensus Rationale: Pretest probability is weighted heavily in the performance of any diagnostic test. Thus, when initiating the diagnostic evaluation, the identification of a patient as having a higher or lower probability of PJI is important (Figures 1 and 2). Further, a determination of how likely or unlikely a diagnosis may be—based on patient symptoms, risk factors, comorbidities, and physical examination—is typically an integral part of the evaluation that guides clinical decisions. Thus, despite the paucity of high-level data that support the specific identification of such factors as they relate to the probability of PJI, it is important to outline factors that can assist the clinician in identifying patients with a higher probability of periprosthetic infection and who therefore warrant a more extensive diagnostic evaluation. Similarly, in patients deemed to have a lower probability of periprosthetic infection, a less vigorous evaluation may be warranted. Clinicians often make such assessments of probability routinely in their practice, which leads to more prudent selections among testing modalities based on specific individual patient characteristics. This routine assessment is of limited cost and low morbidity to the patient. The work group felt that stratification of patients based on their risk for PJI allows implementation of appropriate tests to reach or refute the diagnosis of PJI.  We included four level II and two level III prognostic studies, one level IV prognostic study, and one level V prognostic study of hip patients (or mixed hip and knee patients, with most being hip patients).1-8 The results are summarized in Table 5. (This and all other tables mentioned in the text are available at http://www.aaos.org/research/guidelines/PJIguideline.asp.) We included five level II and two level III prognostic studies, and one level IV prognostic study, of knee patients (or mixed hip and knee patients, with most being knee patients).3,8-14 The results are summarized in Table 6. We also included 13 randomized trials investigating the use of drains and 1 randomized trial investigating the use of anticoagulation.15-28 The results are summarized in Table 20.  We included seven diagnostic studies for this recommendation. Two studies of other diagnostic modalities reported data on patients with pain;29,30 two reported on warmth, redness, or swelling;29,31 one reported on sinus tracts;29 and one reported on early implant failure.32 Two prevalence studies addressed metachronous PJI,33,34 and one prevalence study addressed recent septicemia.35 With the exception of the study that reported on early implant failure, the data from these studies are of low quality and therefore not reliable. The results are summarized in Table 7.  Recommendation 2We recommend erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) level testing for patients assessed for PJI.  Strength of Recommendation: Strong Rationale: The use of screening inflammatory markers (ie, ESR, CRP level) is the starting point in the workup investigations for the diagnosis or exclusion of PJI. Our systematic review of the literature found strong evidence for using both ESR and CRP level testing. We recommend use of this combination of tests to rule out infection.  When both the ESR and CRP level are negative, periprosthetic infection is unlikely (negative likelihood ratio [LR?], 0 to 0.06). When both tests are positive, periprosthetic infection must be considered (positive likelihood ratio [LR+], 4.3 to 12.1), and this result warrants further investigation. However, the clinician also needs to be aware of other inflammatory conditions (eg, rheumatoid arthritis, neoplasms, coronary artery disease, polymyalgia rheumatica, inflammatory bowel disease) that can lead to the elevation of inflammatory markers. The use of either test alone is less reliable for ruling out or ruling in infection than when both tests are combined. A negative ESR is better at ruling out infection than a positive result is for ruling in infection (ie, better negative likelihood ratio than positive likelihood ratio). Similarly, the CRP level is a better test for ruling out infection but is somewhat better than the ESR at ruling in infection. The evidence strongly supports obtaining an ESR and CRP level in all patients being assessed for periprosthetic infection.  Six level I studies evaluating the ESR,36-41 six level I studies evaluating the CRP level,36-38,40-42 and two level I studies evaluating the combined use of ESR and CRP level38,41 were used for analysis.  Recommendation 3We recommend joint aspiration of patients being assessed for periprosthetic knee infections who have abnormal ESR and/or CRP level results. We recommend that the aspirated fluid be sent for microbiologic culture, synovial fluid white blood cell count, and differential white blood cell count.  Strength of Recommendation: Strong Rationale: Our systematic review of the literature suggests that ESR/CRP level testing is valuable for screening of (ruling out) periprosthetic infection with extremely high sensitivity (>90%). These tests are not, however, specific for diagnosis of periprosthetic infection and may be elevated with any type of infection or inflammation. Hence, for the patient with abnormal ESR/CRP level who is being investigated for periprosthetic infection of the knee, the most appropriate next test is aspiration of the knee joint.  We recommend that the fluid obtained from the joint be sent for analysis of synovial fluid white blood cell count, percentage of neutrophils, and culture for aerobic and anaerobic organisms. Studies suggest that either a synovial fluid white blood cell count >1,700 cells/μL (range, 1,100 to 3,000 cells/μL) or a neutrophil percentage >65% (range, 64% to 80%) is highly suggestive of chronic periprosthetic infection. However, the threshold for cell count and neutrophil percentage indicative of acute PJI (within 6 weeks of index arthroplasty) is yet to be determined, and the values and ranges reported above may not be applicable in diagnosing acute periprosthetic infections.  Two studies with reliable data addressed the diagnostic efficacy of aspiration cultures among patients being assessed for periprosthetic knee infections.37,42 Three studies with reliable data addressed synovial fluid white blood cell count and differential among patients being assessed for periprosthetic knee infections.37,43,44 Recommendation 4We recommend a selective approach to aspiration of the hip based on the patient’s probability of PJI and the results of the ESR and CRP level. We recommend that the aspirated fluid be sent for microbiologic culture, synovial fluid white blood cell count, and differential white blood cell count.  Strength of Recommendation: Strong Rationale: AAOS conducted a systematic review that identified six level I hip studies32,45-49 and one hip-and-knee study50 on the diagnostic performance of hip aspiration and culture. Given the potential problems with instituting treatment and missing the diagnosis of infection, higher probability patients with an abnormal ESR and/or CRP level without planned reoperation should also receive hip aspiration.  There is no reliable evidence, however, on the diagnostic performance of hip aspirations in patients who are not to undergo reoperation. Possible harms include the possibility of false-positive results, the possibility of the introduction of bacteria into the joint during the procedure, and patient pain and/or discomfort while undergoing the procedure. Therefore, universal hip aspiration does not seem indicated. Aspiration is indicated for lower probability hip patients without planned reoperation only when both the ESR and CRP level are abnormal. Lower probability hip arthroplasty patients without planned reoperation who have an abnormal ESR or CRP level are addressed in Recommendation 6.  We do not recommend that higher or lower probability patients with normal ESR and CRP level have hip aspiration before planned reoperation.  Recommendation 5We suggest a repeat hip aspiration when a discrepancy exists between the probability of PJI and the initial aspiration culture result.  Strength of Recommendation: Moderate Rationale: A repeat hip aspiration is suggested when a discrepancy exists between the clinical probability of infection and the result of the initial aspiration culture. The results of additional tests will raise or lower the probability of periprosthetic infection. The study on which this recommendation is based predated the more routine use and acceptance of synovial fluid white blood cell count and differential white blood cell count. Thus, depending on the results of the synovial fluid white blood cell count and differential, as well as the ESR and CRP level, the diagnosis or exclusion of periprosthetic infection may be apparent, and repeat aspiration may not be necessary.  One study with reliable data performed repeat aspiration in 28 patients because the initial aspiration results contradicted clinical suspicions.45 The results are presented in Table 43.  Recommendation 6 In the absence of reliable evidence, it is the opinion of the work group that lower probability hip arthroplasty patients without planned reoperation who have abnormal ESR or abnormal CRP levels be reevaluated for infection within 3 months. We are unable to recommend specific diagnostic tests at the time of this follow-up.  Strength of Recommendation: Consensus Rationale: In the opinion of the work group, the possible harms associated with performing hip aspiration need to be weighed when conflicting ESRs or abnormal CRP levels are found in the patient. Lower probability hip arthroplasty patients without planned reoperation who have an abnormal ESR or CRP level are to be reevaluated within 3 months. If the patient is asymptomatic, further testing may not be necessary.  Hip aspiration is a useful test to diagnose periprosthetic hip infection in patients who have a higher probability of infection and abnormal ESR or CRP level, as referenced in Recommendation 4. However, possible harms of joint aspiration include the possibility of false-positive results, the possibility of the introduction of bacteria into the joint during the procedure, and patient pain and/or discomfort while undergoing the procedure. There are also concerns about the cost of the procedure. Lower probability hip arthroplasty patients without planned reoperation who have an abnormal ESR or CRP level are to be reevaluated within 3 months.  Insufficient evidence exists to address whether further observation alone, diagnostic testing, or both can be recommended based on lack of available evidence to address the recommendation.  Recommendation 7 In the absence of reliable evidence, it is the opinion of the work group that a repeat knee aspiration be performed when a discrepancy exists between the probability of PJI and the initial aspiration culture result.  Strength of Recommendation: Consensus Rationale: It was the consensus of the work group that data be extrapolated from Recommendation 5 for periprosthetic knee infections even though there was no reliable evidence pertaining to the utility of a repeat knee aspiration.  It is the opinion of the work group that repeat aspiration be performed when (1) the clinical probability of infection is low but the initial aspiration culture result is positive or (2) the clinical probability of infection is high and the initial aspiration culture result is negative.  The final decision whether to perform a repeat knee aspiration may also be influenced by other factors. The results of additional tests will raise or lower the probability of periprosthetic infection. Thus, depending on the results of ESR and CRP level, as well as the synovial fluid white blood cell count and differential, the diagnosis or exclusion of periprosthetic infection may be more apparent, and repeat aspiration may not be necessary.  Recommendation 8 We suggest patients be off antibiotics for a minimum of 2 weeks before obtaining intra-articular culture.  Strength of Recommendation: Moderate Rationale: Culture “yield” (ie, isolation of infecting organism) is lower in patients who are receiving or have recently (within 2 weeks) received antibiotics. This may be because elution of antibiotics into the joint fluid interferes with isolating the infecting organism by culture. Although the exact antibiotic-free time needed to allow washout of antibiotics from systemic circulation and the joint is not known, the work group has accepted 2 weeks as the minimum time required. One study showed that the false-negative rate of cultures was significantly higher in patients who received antibiotics within 2 weeks of aspiration of a joint.51 Recommendation 9 Nuclear imaging (labeled-leukocyte imaging combined with bone or bone marrow imaging, F-18 fluorodeoxyglucose–positron emission tomography [FDG-PET] imaging, gallium imaging, or labeled-leukocyte imaging) is an option in patients in whom diagnosis of PJI has not been established and who are not scheduled for reoperation.  Strength of Recommendation: Weak Rationale: The information presented throughout this guideline shows that no single preoperative investigation can reliably diagnose an infection. Diagnosis usually depends on the combined use of numerous tests. Recognizing subclinical infections before revision surgery is, therefore, a major concern. Patients in whom diagnosis of infection has not been established include those with a higher probability of infection who have abnormal ESR and/or CRP levels but whose aspiration results are inconclusive (ie, fluid cannot be obtained from the joint; culture and cell count results disagree).  The studies included for this recommendation, although of level I and level II quality, were not specific to patients not scheduled for reoperation; thus, the work group downgraded the strength of the recommendation from moderate to weak. In addition, imaging studies may require special expertise and consultation with the imaging provider.  Technetium-labeled leukocyte imaging: We included four studies: one study of mixed hip and knee patients with reliable data (could separate out according to site, but <25 knee patients), two studies of hip patients with moderately reliable data, and one study of mixed patients with moderately reliable data.40,50,52,53 Indium-labeled leukocyte imaging: We included five studies—one of hip patients, one of knee patients, and three of mixed patients.29,50,54-56 Combined labeled-leukocyte/bone imaging: We included three technetium Tc-99m hydroxymethane diphosphate/indium In-111 leukocyte studies of mixed patients (could separate hip from knee patients in one), with reliable data in one study and moderately reliable data in the other two.31,54,57 Gallium imaging: We included two studies with moderately reliable data.58,59 FDG-PET: We included one study of hip patients with reliable data and two studies of mixed hip and knee patients with moderately reliable data.60-62 Technetium Tc-99m bone imaging: We included two studies of hip patients (one with reliable data, one with moderately reliable data) and two studies of mixed hip and knee patients (both with reliable data).52,63-65 Recommendation 10 We are unable to recommend for or against CT or MRI as a diagnostic test for PJI.  Strength of Recommendation: Inconclusive Rationale: One included study indicated that some CT findings are possibly useful in the diagnosis of periprosthetic infection but that more study is needed (range of LR+, 0.3 to 46; range of LR?, 0.04 to 1.3).66 The evidence indicates that CT may be a good rule-in test but not, in comparison, as good a rule-out test. This evidence is insufficient and does not allow a recommendation for or against the use of this modality.  There is insufficient evidence to address the diagnostic efficacy of MRI for diagnosing periprosthetic infections. Based on the lack of evidence, we cannot recommend for or against the use of MRI.  Recommendation 11 We recommend against the use of intraoperative Gram stain to rule out PJI. Strength of Recommendation: Strong Rationale: Our systematic review included three studies of Gram stain in the diagnosis of PJI.67-69 Negative likelihood ratios suggest Gram stain is not a good rule-out test (LR? values >0.5). Comparatively, negative likelihood ratios for synovial fluid white blood cell count and differential white blood cell tests (Recommendation 4) are much lower (LR? values <0.1). Thus, Gram stain is not a useful test for ruling out periprosthetic infection.  Recommendation 12 We recommend the use of frozen sections of peri-implant tissues in patients who are undergoing reoperation for whom the diagnosis of PJI has not been established or excluded.  Strength of Recommendation: Strong Rationale: Eight level I studies were identified that evaluated the use of frozen sections to aid in the diagnosis of peri-implant infection.37,41,67,70-74 Three studies excluded patients with known underlying inflammatory arthropathy.41,70,72 All available studies based the histologic diagnosis of probable infection on the tissue concentration of acute inflammatory cells, usually defined by two variables: (1) the number of neutrophils in a high-magnification (400×) microscopic field, and (2) the minimum number of fields containing that concentration of neutrophils.  Four studies used 10 or more neutrophils per high-power field, and three of the four studies required 10 more neutrophils in at least five fields. The remaining study required 10 or more neutrophils in any given area. Our meta-analysis of these studies indicated that frozen section is a very good rule-in test (ie, a positive result has a high likelihood of infection; LR+, 23) but is a relatively low-value rule-out test (ie, a negative result does not have a high likelihood of absent infection; LR?, 0.23).  Four other studies used five or more neutrophils per high-power field; several of these specified at least five microscopic fields, but the other studies did not. Meta-analysis indicated that, compared to a criterion of 10 polymorphonuclear neutrophils per high-power field, this lower inflammation threshold may have similar sensitivity but also slightly lower specificity (ie, a higher frequency of false-positive results).  Insufficient information exists to distinguish 5 from 10 neutrophils per high-power field as the best threshold needed for diagnosis. Insufficient information is available to determine the efficacy of frozen sections in patients with an underlying inflammatory arthropathy.  Recommendation 13We recommend that multiple cultures be obtained at the time of reoperation in patients being assessed for PJI.  Strength of Recommendation: Strong Rationale: Four level I studies compared the effects of using one as opposed to more than one positive culture as a threshold for diagnosing infection, with histology of peri-implant tissue as the gold standard.51,75-77 Two of these studies obtained at least three cultures per patient, and two of the studies obtained at least two cultures per patient. Compared to histology, more than one positive culture result had a higher positive likelihood ratio for diagnosing infection than did a single positive culture result.  Assessing infection using multiple cultures increases the chances of identifying infection. Cultures are easily performed during the procedure and provide reliable results, as indicated by the positive-likelihood ratio. Obtaining more than one culture decreases the likelihood of a false-negative result and may assist the clinician in clarifying a result that may be deemed a false positive based on the results of other tests.  Recommendation 14 We recommend against initiating antibiotic treatment in patients with suspected PJI until after cultures from the joint have been obtained.  Strength of Recommendation: Strong Rationale: One level I study addressed whether administration of antibiotic therapy before obtaining cultures from the joint affected the sensitivity of the cultures in diagnosing periprosthetic infections.51 The study found a false-negative rate of 55% in patients receiving antibiotics within the previous 14 days compared with 23% in patients not receiving antibiotics during the same time period. The difference was statistically significant. Hence, there is a concern that antibiotics can interfere with isolation of the infecting organism, leading to confusion regarding the diagnosis or inability to use organism-specific antibiotics subsequently when infection is confirmed. Thus, administration of oral or intravenous antibiotics to patients with suspected diagnosis of PJI is discouraged until aspiration of the joint is performed or samples for culture are obtained.  The finding of only one level I study supporting this recommendation would be evaluated as moderate strength. However, because of the severity of the potential harm to the patient in getting a false-negative culture result, the strength of the recommendation was elevated to strong.  Recommendation 15 We suggest that prophylactic preoperative antibiotics not be withheld in patients at lower probability for periprosthetic infection and those with an established diagnosis of PJI who are undergoing reoperation.  Strength of Recommendation: Moderate Rationale: Two level II studies and two joint registry studies found that the administration of perioperative antibiotics significantly decreased the risk of postoperative infection.12,78-80 These studies, however, do not address whether the administration of preoperative antibiotics interferes with intraoperative cultures. One level III study did not find a significant difference in the false-negative rate of intraoperative knee cultures in patients receiving perioperative antibiotic prophylaxis (12.5% versus 8.1%; P = 0.34).  Patients who contract periprosthetic infection face catastrophic complications, including additional surgery, increased cost, and prolonged treatment, as well as possible permanent loss of the prosthesis. The important role of antibiotic prophylaxis in helping to prevent these catastrophic events outweighs the potential risks (Figures 1 and 2).  Previous SectionNext SectionFuture Research Although many of the recommendations in this guideline are supported by high-quality evidence, certain areas are still lacking in high-quality evidence. Several diagnostic tests exist, and no single test has yet been identified that consistently allows the clinician to rule in or rule out the presence of a periprosthetic infection in all patients. Given the potential problems with instituting treatment and missing this diagnosis, future research in the area of diagnosing a PJI is warranted.  Specifically, further studies that characterize facets of the patient’s history and physical examination, as well as the presence of comorbidities that may affect the probability of a PJI, are needed to help guide diagnostic algorithms.  Molecular techniques to aid in the diagnosis of PJI are currently under investigation, including the use of microarrays, polymerase chain reaction, proteomics, and enzyme-linked immunosorbent assay for measuring various cytokines, enzymes, interleukin-6, and neutrophil enzymes. The routine role of sonification of removed implants and cement spacers also needs to be clarified. At the time of publication of these guidelines, there are inconclusive results to recommend the use or cost-effectiveness of these tests in clinical practice. Thus, further studies are needed to determine their utility.  Although there was strong evidence to support the use of an ESR and CRP level as initial screening tests for infection, the value of these tests for patients with inflammatory arthritis (eg, rheumatoid arthritis) or other systemic diseases associated with increased ESR and CRP level is unclear.  There is currently inadequate evidence to support or refute the role of wound swab in the management of patients with suspected PJI. The efficacy of various tests at the second stage of a two-stage operation for known periprosthetic infection also requires clarification.  Patients who present with a painful total hip or total knee arthroplasty and who do not otherwise have a clear indication for reoperation or revision also present a diagnostic challenge because cultures from the joint may be difficult or impossible to obtain. Further, analysis of the value of noninvasive testing modalities that do not rely on direct sampling from around the prosthesis (eg, advance imaging studies) would be of value.  Finally, the effect that antibiotic treatment has on the utility of many of the commonly used tests for diagnosing a PJI is incompletely understood. Many patients suspected of a PJI may have received prior treatment with antibiotics, and the temporal effect of such treatment is unclear. Although the work group issued a strong recommendation against the administration of antibiotics before obtaining a culture from the joint, many patients unfortunately have had such treatment by the time they are evaluated by an orthopaedic surgeon, thus adding to the diagnostic challenge. Although the benefits of prophylactic antibiotics before the surgical incision for preventing surgical site infections is clear, the benefit of withholding such antibiotic prophylaxis before obtaining deep cultures at the time of reoperation or revision is unclear, and future studies that examine the effect of a single prophylactic dose of antibiotics on the sensitivity of operative cultures are warranted.  Previous SectionNext SectionAcknowledgments The authors wish to acknowledge the assistance of Sara Anderson, MPH, Kevin Boyer, and Laura Raymond, MA.  Previous SectionNext SectionFiguresFigure 1 Algorithm for patients with a higher probability of hip or knee periprosthetic joint infection. a Perform repeat aspiration when a discrepancy exists between the probability of infection and the result of the initial aspiration cultureb Perform frozen section when the diagnosis has not been established at the time of surgery; synovial fluid white blood cell count and differential may also be obtained intraoperativelyc Nuclear imaging modalities: labeled-leukocyte imaging combined with bone or bone marrow imaging, F-18 fluorodeoxyglucose–positron emission tomography, gallium imaging, or label-leukocyte imaging CRP = C-reactive protein, ESR = erythrocyte sedimentation rate    Figure 2  Algorithm for patients with a lower probability of hip or knee periprosthetic joint infection.a Perform repeat aspiration when a discrepancy exists between the probability of infection and the result of the initial aspiration cultureb Perform frozen section when the diagnosis has not been established at the time of surgery; synovial fluid white blood cell count and differential may also be obtained intraoperativelyCRP = C-reactive protein, ESR = erythrocyte sedimentation rate   全文下载：PJIguideline.pdf