Background: Infection is a catastrophic complication after prosthetic joint replacement. In the United States, infection is the most common cause of revision after total knee arthroplasty (TKA) and the 3rd most common cause of revision after total hip arthroplasty (THA), with an overall incidence between 1% and 3%. Once small amounts of bacteria are introduced after arthroplasty due to circulation or via local wounds, they can easily colonize and grow on the surface of the artificial joint and spread within the joint cavity. Postoperative infections are usually classified as acute, subacute (hematogenous), or chronic, depending on the speed of progression and the pathogenesis of the disease, and treatment protocols are usually developed based on this classification. Currently, the increase in costs incurred for the management of postoperative infections has significantly outpaced the increase in costs required for the prevention of infections. An analysis of our database, which is based on a database of cases over the past several decades and which anticipates the future of prosthetic joint replacement, shows that 3,308 patients underwent revision surgery after initial replacement, and that a total of 821 of the revision cases were due to postoperative infection. The study of these cases can better guide us in the management of postoperative arthroplasty, especially in the prevention, diagnosis, and management of postoperative infections. In this review, we will mention the best diagnostic and therapeutic approaches available. In terms of depth, this review only touches on the part of post-arthroplasty infection that is repeatedly mentioned, namely the management of post-operative infection. However, it will provide the reader with a complete understanding of the disease, further benefiting the patient. Current Diagnostic Approach: There is no clinical consensus on a “gold standard” for post-arthroplasty infection because there is no highly specific diagnostic method for post-arthroplasty infection. The diagnosis of the disease currently relies on a combination of clinical suspicion, serologic testing, bacterial culture, histologic examination, and some basic molecular testing techniques. To a large extent, however, current diagnostic protocols do not provide the necessary and accurate information to demonstrate the presence and pathogenicity of septic bacteria in the infected joint. Current protocols used to diagnose infection combine serologic testing (ESR, CRP), histologic examination, appearance of the diseased joint, specimen culture, results of preoperative joint aspirate testing (including bacterial culture), and white blood cell count and classification. There are now many references to these methods, while the AAOS provides appropriate guidance for each of these tests in the relevant guidelines. The application of these methods is constantly updating the definition of post-arthroplasty infection, and new methods for its diagnosis will be discussed here. Current recommended diagnostic methods and AAOS protocols When infection is suspected, ESR and CRP levels are usually checked first, and they are both sensitive indicators of infection. However, because they are significantly elevated in the presence of either inflammatory lesion, they have low specificity in diagnosing infection after prosthetic arthroplasty. The presence of infection is usually considered when ESR is higher than 30 mm/hr and CRP is higher than 10 mg/dl. At our institution, however, receiver operating characteristic curve (ROC) analysis revealed that more reasonable criteria for ESR and CRP would be 31 mm/hr and 2 mg/dl, respectively, and that the sensitivity and specificity of the diagnosis of periprosthetic infection were 96% and 59%, respectively, when ESR and CRP levels were higher than these criteria. If the ESR or CRP is above the standard values and there are no clinical signs of postoperative infection, further testing is required. A more effective method is to perform an arthrocentesis to extract joint fluid as a specimen for analysis. When the serologic parameters are abnormal, we usually use the results of joint fluid as a decisive reference indicator. Therefore, the AAOS guidelines recommend joint fluid examination as a method of further investigation, and this method is quite inexpensive. Culture of the joint fluid taken by puncture may yield pathogenic bacteria and make further targeted antibiotic treatment. The problem with this method, however, is that it can produce erroneous results, which we will continue to discuss below. Over the past few decades, an uninterrupted study has shown that white blood cell counts suggest the presence of postoperative infection. This study concluded that a knee puncture fluid leukocyte count greater than 1700 cells/μl or a polymorphonuclear granulocyte (PMN) ratio greater than 65% was indicative of infection. For the hip, these criteria were greater than 4200/μl and greater than 80%, respectively. However, it is difficult to determine the presence of infection in the early postoperative period even if the joint aspirate is abnormal, because inflammatory markers are normally elevated in the early postoperative period, and Bedair et al. suggested that the criteria for arthrocentesis for early postoperative diagnosis of periprosthetic infection are a WBC count of >10,700/μl and a PMN ratio of >89%. If the presence of infection cannot be determined by multiple punctures of the joint fluid, imaging should be considered if the surgeon is not considering surgical intervention. We used 18F-labeled tests in patients with hip pain after THA to identify whether the lesion was bacterial or aseptic. The sensitivity and specificity of FDG-PET for the diagnosis of postoperative infection were found to be 85% and 93%, respectively, and Lovetal found a similar role for FDG-PT in postoperative infection after hip and knee arthroplasty. Other imaging methods, including leukocyte labeling as recommended in the AAOS guidelines and gallium imaging, are not supported by strong clinical evidence. There is no evidence at this time that MRI and CT have diagnostic value for postoperative infections. If these tests are not definitive, the only method available is surgical collection of periprosthetic tissue for frozen section and bacterial culture. Intraoperative finding of a sinus tract that communicates with the joint is a definite indication of infection and requires immediate surgical intervention. Although the false-positive rate of finding pus is low, on the other hand, the presence of infection is not completely certain even if pus is found intraoperatively, due to the fact that it has been found that some of the pus present after metal-on-metal arthroplasty may have pus-like pathological changes due to the patient’s hypersensitivity reaction to metal ions. In unpublished studies, we believe that the intraoperative finding of pus has a sensitivity of less than 50% for the diagnosis of infection. They concluded that postoperative infection cannot be diagnosed solely on the basis of pus found in the joint cavity. Although many joint surgeons support histologic examination in cases with high suspicion of infection, and the AAOS guidelines support this practice. However, we do not recommend intraoperative frozen section examination. Because cheaper and more reliable methods such as arthrocentesis are available; intraoperative frozen sectioning is too complex and subjective. If the unproven hypothesis that the percentage of PMN in the arthrocentesis fluid test is highly positively correlated with the neutrophil content of the frozen section holds true, intraoperative frozen section testing will have little diagnostic significance in the diagnosis of infection after arthroplasty. If arthrocentesis fluid testing does not lead to a positive conclusion, then frozen sections are hardly likely to confirm or deny the presence of infection; moreover, we and other investigators have found that Gram staining is not a very effective tool for diagnosing postoperative infection. Traditionally, culture isolation of periprosthetic tissue in solid medium is the “gold standard” for diagnosing periprosthetic infections, but it has been reported that this method does not detect the presence of the causative organism in approximately 2-18% of infections. The inability to identify the causative organism will complicate diagnosis and treatment. We have found that culture-negative infections are often indicative of treatment failure using methods such as debridement and irrigation. Based on this fact, we set out to investigate the mechanisms to improve the sensitivity of tissue samples in culture. The AAOS guidelines recommend that antimicrobial therapy be administered after specimen collection if infection is highly suspected. However, we found that prophylactic antibiotics did not affect the accuracy of the culture, and Schäfer et al. showed that extending the culture time increased the positive rate (63% for 1 week of culture and 77% for 2 weeks). Although extending the incubation time may again lead to false positive results due to the growth of contaminating bacteria, Schäfer et al. concluded that more than half of the contaminating bacteria were cultured within the first week of incubation. Notably, their analysis showed that those rare pathogenic bacteria usually started to grow significantly only during the 2nd week of culture. The false positive rate of tissue culture from contamination ranges from 5 to 37%, which can lead to unnecessary surgery or complicate the procedure. Definition of post-arthroplasty infection There are many ways in the medical community to identify post-arthroplasty infection from other diseases, but each of these methods has its own drawbacks. While the AAOS provides guidelines for the use of these tests when infection is suspected in patients, a definitive definition of infection is essential for both interstudy comparisons and clinical diagnosis. Therefore, many investigators have proposed their own definitions of infection after prosthetic arthroplasty based on the diagnostic methods described above. To date, these definitions have not included the number and classification of cells in the puncture aspirate. Therefore, our proposed new definition of post-arthroplasty infection includes the analysis of puncture aspirates and the definition of a positive tissue sample culture. Since no gold standard exists for the diagnosis of post-arthroplasty infection, our recommended new definition was derived by comparing the existing definitions with each other. It should be added that our new definition is also being compared. This study found that in 24% of cases, when infection was identified by one definition, it was diagnosed as a bacterial-free infection by another definition. This demonstrates that the diagnosis of post-arthroplasty infection often relies on an understanding of its definition, and that our recommended definition has a diagnostic accuracy of 53%-100% when compared with known definitions. As previously discussed, our definition of postoperative infection does not include histologic analysis. More importantly, the presence of pus in the joint is not only of little value for diagnosis, but may even lead to misdiagnosis. The Musculoskeletal Infection Society recently published its updated definition of postoperative infection (Figure 1), which the Society hopes will be used as the “gold standard” for infection. The definition of post-arthroplasty infection is becoming more standardized and accurate based on current methods, and other methods will continue to evolve. If their validity is confirmed, they will become one of the indicators for the diagnosis of post-arthroplasty infection and will further guide the clinical practice. Leukocyte esterase The level of leukocyte count in the joint fluid and the typing of neutrophils are highly sensitive and specific for infections after total knee arthroplasty. Therefore, we are able to assume that leukocyte esterase has the same role. A prospective study we organized showed that leukocyte esterase is a highly specific indicator for the diagnosis of postoperative infections. Currently, colorimetric assays for the detection of this enzyme are widely used for urinary tract infections. The results shown are divided into four mutually independent strata (reflecting the amount of leukocyte esterase in the sample) according to the change in color on the indicator bar. In this study, postoperative infections were analyzed by analysis of joint fluid taken by puncture as a standard. The highest leukocyte esterase group (++) had 81% sensitivity and 100% specificity. When both groups with the highest levels of leukocyte esterase (+ and ++) were suggestive of positivity, the results had 94% sensitivity and 87% specificity. There was a good correlation between leukocyte esterase and ESR, CRP, synovial fluid leukocyte count, and synovial fluid PMN ratio. Although research on leukocyte esterase is just beginning, the use of the leukocyte esterase colorimetric method will provide surgeons with an accurate method for diagnosing joint infections. The method also has the advantage of being inexpensive and providing rapid results. The ability to obtain results quickly is invaluable to the surgeon, and to date, other rapid tests have had limited significance in the diagnosis of postoperative infection. Other cytologic markers Leukocyte esterase has been shown to be an effective method for diagnosing post-operative infection after prosthetic joint replacement, and we will then supplement it with other molecular markers to further improve the diagnostic accuracy of post-operative infection. Therefore, we are also studying various factors in the inflammatory response. Abnormally increased factors in joint synovial fluid are expected to be as rapid a method of diagnosing postoperative infections as pregnancy and urinary tract infections. In one of our studies, with 74 patients with revision joints, 46 known inflammatory factors were used to assess the diagnostic value of infection. Of these, 31 cases were caused by infection and 43 were sterile. Proteomic analysis was used to measure the amount of inflammatory factors in each sample, and we used ROC curve analysis to establish optimal threshold values for each protein marker. This analysis revealed five inflammatory factors that can accurately diagnose postoperative infection: IL-6, IL-8, CRP, alpha-2 macroglobulin, and vascular endothelial growth factor. The results of the Deirmengian study came to the same conclusion. In both studies, IL-6 was the most accurate indicator of postoperative infection. This study needs to expand the sample size and the participation of more relevant institutions. Our ideal test would be able to diagnose postoperative infections quickly and accurately, complementing the existing diagnostic criteria. Leukocyte esterase testing has great potential to clarify the diagnosis, reduce costs, and guide treatment. Synovial CRP concentration Although CRP is an important test in cases of suspected postoperative infection, as mentioned above, plasma CRP has a low specificity for postoperative infection. As an indicator of response to inflammation, it has been hypothesized that it would be significantly elevated in periprosthetic tissue and a series of studies have been conducted on this subject. In 66 knee revisions performed over one year, again, the cause of the revision was classified as infectious or aseptic. the results of the ROC curve analysis showed that a CRP concentration of 3.7 mg/l in the joint fluid was diagnostic, whereas a plasma CRP concentration of 16.5 mg/l was diagnostic in the same group of cases. The sensitivity of joint fluid CRP concentration for diagnosis of postoperative infection was 84%, specificity was 97%, and accuracy was 96%, while plasma CRP sensitivity was 76%, specificity was 93%, and accuracy was 91%. These studies, although still in their infancy, provide a method for early diagnosis of postoperative infection and reduce the false positive rate in diagnosis. Based on these findings, it is hoped that synovial fluid CRP concentrations will become an indicator of postoperative infection. The advantages of synovial fluid CRP are that the results are easy to read and that the test can be performed in almost all hospital clinical laboratories without the need for new equipment and personnel. It should be added that the results of this index are not influenced by the subjective factors of the operator and are therefore suitable for a wide range of clinical applications. Current treatment of post-arthroplasty infections Accurate and timely diagnosis of post-arthroplasty infections is imperative. This is because its treatment is a matter of great urgency. Treatment varies greatly depending on the extent of the postoperative infection removal. Once a postoperative infection is diagnosed, the following aspects need to be clarified first: the duration of symptoms, the patient’s immune status and general health, the history of prosthetic joint infection in the diseased joint as well as in other joints, the status of any joint wounds, expectations for joint function, and the characteristics of the pathogenic organism. These will assist in the selection of surgical treatment options. If the patient is in poor health, serious adverse outcomes are expected with surgical treatment, and the causative organism is less virulent and sensitive to antibiotics, then antimicrobial therapy alone is a better choice. Although there is no conclusive evidence to support the use of antimicrobial therapy alone without surgical intervention, it is a reasonable option for those who cannot tolerate surgery. Long-term antibiotic therapy with subsequent surgical debridement is also an option if the patient refuses or is unable to tolerate subsequent surgery. This approach has had some success in controlling infection. Phase II Surgical Management In North America, phase II surgical management is the gold standard for treating post-arthroplasty infections, in which the infected joint is removed in phase I, an antibiotic cement spacer is placed, and a new joint is implanted in phase II after the infection has been effectively controlled. In cases of acute postoperative infection, if the artificial joint is well positioned and fixed, the causative organism is sensitive to antibiotics, and the periarticular soft tissue is well covered, the use of debridement and irrigation to preserve the artificial joint can be of great benefit in reducing disability and promoting recovery of limb function. We have studied the outcomes of this treatment approach, and we did not find a statistical difference in the use of debridement irrigation after postoperative infection in the treatment of acute infections, acute hematogenous infections, and chronic infections that occur after surgery. However, it is possible that these studies do not currently have the ability and cannot measure an accurate difference. These analyses demonstrate that staphylococcal infections are an independent risk factor for treatment failure using debridement and lavage to preserve the prosthetic joint. In addition, two independent studies have shown that debridement success rates for methicillin-resistant Staphylococcus aureus (MRSA) infection alone are low (16% and 37%). In another study, the debridement success rate for Streptococcal infections (which are traditionally considered to be well preserved by debridement and irrigation) was 65%, compared with 71% for other infections in general. These data suggest that the management of post-arthroplasty infections can be managed with less surgical treatment in favor of attempting to preserve the original artificial joint when appropriate. Our institutional treatment protocol suggests that the management of acute non-MRSA infections in the early postoperative period (within 2 weeks) after biologic THA is to replace the prosthetic joint in one stage, in which case we can use debridement and irrigation. Although the efficacy of this treatment is not known, it has the advantage of allowing for more thorough debridement and removal of the pathogenic tissue. The use of debridement and irrigation to preserve the artificial joint may seem like a stepping stone to second-stage surgery. However, we have found that the success rate of debridement is lower when the joint is first debrided and lavaged to preserve the prosthesis and then later replaced with a second stage compared to a direct second stage replacement. Moreover, we had a similar situation in cases of revision of the total knee joint due to infection and aseptic reasons. There is no method for predicting the outcome of postoperative infections with appropriate management, but efforts have been made to do so. Studies of cases of recurrent or persistent infection after second-stage revision management of postoperatively infected knees have found negative periprosthetic tissue cultures, methicillin-resistant pathogenic bacterial infections, and long reoperation times to be independent risk factors for recurrence of infection. In an independent study, the success rate of Gram-negative cases after second-stage revision (52%) was comparable to that of MRSA-infected cases treated in the same way (51%). In contrast, the success rate for methicillin-susceptible gram-positive infections was significantly higher (69%), and these recent data reveal that even the gold standard approach (second-stage surgical treatment) to treat postoperative infections has less favorable outcomes. Due to the high failure rate of postoperative infection treatment, patients are often at risk of reinfection with second-stage surgical management. In the event of a recurrence of infection after a second-stage revision, physicians are faced with the embarrassing situation of having very few management options available. In our experience, repeat second-stage replacement surgery provides patients with results that are within reasonable expectations. In our case study of repeat second-stage artificial knee surgery, 14 of 18 patients were infection-free for at least two years, and two of the four failed cases were cured by a third second-stage surgery. In an independent study of artificial hip replacements, only 8 of the 15 patients were cured, but 7 of these 8 patients were free of reinfection. For comparison, failure in 7 of 11 patients who underwent a second artificial hip replacement was reported in the Kalra case study. Salvage measures need to be considered if the patient is less likely to recover limb function, or if the patient is immunocompromised, or if the patient is too unwell to tolerate multiple surgeries. Salvage measures for the knee include salvage measures such as knee fusion and above-knee amputation. All of these methods may be effective in removing the source of infection and restoring some degree of function to the affected limb. After several analyses of these cases at our institution, we believe that suprapopliteal amputation is the next best surgical option. A cautionary note is that performing internal fixation during knee fusion may result in biofilm formation and even permanent infection. Salvage measures for hip lesions are limited to hip fusion, which requires the use of metal fixation, and surgeons should be prepared for long-term treatment with antibiotics. Discussion As the number of patients with post-arthroplasty infections is on the rise, more and more clinicians will encounter this disease. Overcoming this disease will require both continued improvement in the diagnostic and therapeutic capabilities of clinicians, as well as continued efforts by basic researchers to find better screening methods. This review broadly describes the most cutting-edge diagnostic and therapeutic approaches to postoperative infection. With the continued maturation of synovial fluid analysis, other biomarker technologies that may mature, and rapid test strip technology, the diagnostic tests are constantly evolving. Of course, there is still a considerable distance to go before the ideal test method and successful treatment are available. Combined with the knowledge already mentioned in this review and the future research that will be conducted, we believe that orthopedic surgeons and colleagues at the forefront of medicine will one day overcome this disease!