Infections after fracture are devastating problems. They can be challenging to manage, costly to the health care system, and result in significant morbidity to the patient. Treating these infections can result in functional deficits, time lost from work and lost wages, decreased quality of life, and may ultimately require amputation to obtain source control.1 The incidence of infection after fracture fixation surgery ranges from as low as 1% in closed low-risk fractures, to >30% in severe open injuries. Despite efforts to minimize infection and prevent injury with timely administration of antibiotics, early transfer to regional trauma centers, and multidisciplinary management for revascularization, fracture treatment, and early soft tissue coverage, there are still 100,000 cases of reported fracture-related infection (FRI) yearly in the United States.1–11
Because FRIs are typically caused by biofilm-forming organisms they are more difficult to eradicate than planktonic infections. Biofilm-related orthopedic infections are not unique to FRI. Extensive work around periprosthetic joint infections (PJI) has been adapted to the diagnosis and treatment of FRI. In addition, an evidence-based definition of PJI has evolved over a number of years to provide clinicians with a clear template for how to utilize the available diagnostic tests to aid in clinical decision making. Unfortunately, there are a number of important differences between these patient populations that make treating and diagnosing FRI even more challenging than treating PJI.
Differences between FRI and PJI patients include the lack of preoperative optimization in trauma patients, vascular injury due to trauma, contamination, and soft tissue compromise coverage concerns after open fractures. To further complicate the diagnosis in FRI, it is not typically possible to obtain diagnostic tissue samples before taking the patient to the operating room for debridement. In addition, fracture stability complicates FRI treatment when it has not healed and implants are necessary for skeletal stabilization.
When treating infections, the clinician strives to answer the following questions:
- Is the patient infected?
- Has the infected tissue been adequately debrided?
- What is the pathogen being treated?
The purpose of this manuscript will be to evaluate the current diagnostic tests available to answer each of the above questions, and the available evidence supporting their use.
IS THE PATIENT INFECTED?
Definitions of FRIs
“It is astonishing that in all papers in which infection is mentioned, the term ‘infection’ is not defined.12” Arens, 1996
“To our surprise, only 2% of…100 RCTs [that describe infectious complications after fracture fixation] used a validated definition. … In 28% the authors used a self-designed definition…70% of the RCTs did not give a definition at all.13” Metsemakers, 2018
Greater than 20 years have passed since Arens and colleagues observed that papers discussing infection as a complication after musculoskeletal trauma lacked a proper definition. Since then, there have been a variety of definitions used in the literature that reports on FRI.14 Some authors use the Centers for Disease Control (CDC) criteria. Others have used modified CDC criteria. Other authors provide no definition of what constitutes FRI, while still others created custom definitions that were not validated or previously used in the literature to define infection. Most recently, an international consensus group led by the Arbeitsgemeinschaft für Osteosynthesefragen (AO) Research Institute has made an initial attempt to present a unified definition of infection after fracture surgery.15,16
Adding to the confusion, there are multiple versions of the CDC criteria for infection that have been applied to defining surgical site infection after fracture surgery.17 The most recent version updated in 2016 has specific diagnostic criteria for PJI,18 which mirrors the new definition of periprosthetic infection,19 but does not have a specific definition for FRI. The definition of surgical site infection after “open reduction of fracture” includes any of the nonspecific definitions of bone infection, deep incisional primary infection, joint or bursa infection, and superficial incisional primary infection. Table 1 summarizes these criteria.
There are concerns about using the CDC surveillance definitions of surgical site infection for FRI after fracture surgery. The first is the time restrictions used by the CDC for defining infections after fracture surgery. Deep infections are limited to 90 days after fracture surgery, and superficial infections are limited to 30 days. Although 50% of infections typically present within 3 months,20,21 the definition is not applicable to the other half of infections that may present later than this.22 Next is the heterogeneity of surgical locations. Many areas of fracture fixation have limited soft tissue coverage over the implants (eg, the ankle), making it difficult to distinguish between superficial and deep space infections. Another concern is the circular nature of the arguments used to define infection. Ideally, the diagnosis of infection would be confirmed before taking the patient to the operating room for surgical debridement; however, the CDC definition includes requirements of either positive tissue cultures, or “the lack of taking cultures” as part of the diagnostic criteria. Finally, the CDC definition continues to include the potentially subjective criteria of “diagnosis of surgical site infection by the attending” as a criteria for infection. This can be otherwise stated as: because a surgeon said the wound is infected, it is infected!
A systematic review by Metsemakers et al13 in 2018 evaluated high quality prospective randomized controlled trials (RCT) that reported infection as an outcome in the trial. They included all RCTs published before 2016 that reported on infection as a complication after fracture surgery, and evaluated the methods to determine whether the authors used a validated definition, a self-designed definition, or no definition. In 28 instances where the methods were ambiguous, RCT authors were contacted directly to ask for clarification. Only 7/28 (25%) responded to the survey, and of those 7, only 1 author clarified that the CDC guidelines were used. They concluded that only 2% of RCTs used validated infection definitions, which in both instances were those provided by the CDC. In 70% of the RCTs, there was no definition of infection described in the methods. The remaining 28% of the RCTs used a custom definition that was created by the authors. These custom definitions included a variety of clinical signs and symptoms, and tests. These included the presence of fever, rubor, swelling, pain, purulent drainage, need for surgical debridement, need for implant removal, “XR evidence” of infection, or positive wound cultures.
In response to the lack of a cohesive definition of infection after fracture surgery, the AO Foundation supported an international panel to generate an initial consensus definition of FRI. Prior consensus definitions have been created for PJI23–33 and diabetic foot ulcers,34 but are not applicable to the unique FRI patient population. Therefore, the AO foundation used a Delphi process to identify experts on fracture infection, who exchanged ideas about what criteria to discuss. After this, the second phase of the process included face-to-face work groups to agree on specific topics and vote on resolutions. The final result was the definition (Table 2) that includes confirmatory criteria and suggestive criteria for FRI. The suggestive criteria include clinical signs, radiologic signs, lack of culture-positive organism identification, or elevated inflammatory markers. These criteria are intended to guide the clinician to further investigate to determine whether there is infection present with the confirmatory criteria. Although this new definition is likely an important step, this new definition is different than what has been done previously so comparison to prior work is not possible. Furthermore, it has not been validated, and of course has its own potential limitations. More work is needed to determine the utility of these various potential definitions.
Clinical History: Search for Host Factors to Increase the Pretest Probability of Successful Diagnosis
The role of the clinical history should be to obtain information that can be useful to help risk-stratify the likelihood of whether the patient has a FRI. This includes a clinical history of the original injury, how many previous surgeries the patient has undergone, any concomitant injuries or infections at the time of injury, medical comorbidities and markers of immune disfunction, and how much disability the patient currently has as a function of their musculoskeletal injury.
It is important to obtain a detailed history of original injury. Certain anatomic locations have increased risk of postoperative infection. For example, the reported infection rate after closed pilons is around 11% compared with 1% or 2% after closed upper extremity fracture surgery.35–38 Open fractures and their classification correlates well with infection risk. Open tibia fractures have some of the worst reported infection rates, with open Type IIIB fractures typically having deep infection rates in the neighborhood of 30%.2,3,5–9 Other useful information may include the time from injury to administration of IV antibiotics, the time to transfer to definitive trauma center, or time to soft tissue coverage when needed.39,40 Large delays in these treatments may be correlated with higher infection rates.
It is also useful to take a history of the patient’s functional disability. Although pain and difficulty ambulating can be normal in the early postoperative course after fracture fixation, any new decline in functional ability or worsening pain with ambulation can be indicative of delayed healing or underlying infectious process.
Numerous host factors have been shown correlated with the risk of developing FRIs and these are discussed in detail elsewhere in this issue (Wise and colleagues). Host immunosuppression has been known to increase the risk of osteomyelitis for many decades,41 and host characteristics are a cornerstone of the Cierny-Mader classification of osteomyelitis.42–44 Any host factors that could contribute to the inability of the patient to combat contamination and infection, heal wounds, or tolerate the metabolic stress of injury and surgical treatment can predispose them to higher infection rates.41 Systemic factors include malnutrition, hepatic or renal functional impairment, alcohol use, acquired immune deficiency (eg, HIV), chronic hypoxia, malignancy, diabetes mellitus, corticosteroid use, or smoking.42,45–48 Any history of postoperative wound healing problems should be documented, and any duration of oral or IV antibiotic treatment. Other sources of infection, such as urinary tract infections, pneumonia, etc., should be documented as hematogenous seeding during even transient bacteremia can occur at any point after fracture fixation surgery.49
In a study by Paryavi et al,50 the authors identified patient-based risk factors for infection after surgical fixation of lower-extremity trauma. They used a database of tibial plateau, pilon, and calcaneus fractures treated at a single institution to determine a scoring system to risk stratify patients based on preoperative patient-specific factors. There were 217 patients (83 plateau, 91 pilon, and 61 calcaneus fractures) who had a median follow-up of 344 days (interquartile range, 153 to 573 d). The overall infection rate as defined by CDC criteria during the follow-up period was 14%. Age, sex, body mass index (BMI), ASA classification, fracture classification, smoking status, alcohol use, intravenous drug use, diagnosis of diabetes mellitus, and insulin use were evaluated. Factors included in the multivariate model for FRI were fracture classification (AO type C3), male sex, ASA class 3 or higher, BMI <30 kg/m2, and insulin use. After the final multivariate analysis, fracture classification, BMI, and ASA classification were predictive of postoperative infection.
Physical Examination: Clinical Signs of Infection
Whereas the clinical history provides clues as to which patient may be at risk for FRI, the physical examination provides clues to determine which patients actually have a FRI. There are some pathognomonic signs of infection, such as exposed hardware and copious purulent drainage, and more subtle signs that may be suggestive of an underlying infection that merit further investigation. Multiple classification schemes exist for temporally categorizing infection after fracture fixation. They can be roughly broken down into early infection (<2 wk after fracture surgery), delayed infection (2 to 10 wk), and late infection (>10 wk).49,51 The signs of presentation can vary between these groups. Although these signs are suggestive guidelines to identifying infection, there is little prospective data to validate these nonspecific signs seen on physical examination.
There are several obvious signs of infection that can make the diagnosis of FRI easier. These include wound breakdown with visible hardware or bone, purulence draining from the incision, fever >101°F, or a sinus tract that communicates with the deep tissue space. There may be various reasons that lead to wound breakdown, such as vascular insufficiency, traumatic wound healing problems, or host factors, once hardware or necrotic bone is exposed to the external environment it is colonized with skin flora and should be treated with surgical debridement and hardware removal or exchange, with soft tissue coverage if necessary. Wound breakdown that secondarily leads to infected or colonized bone and hardware can be very difficult to differentiate from deep infections that lead to wound breakdown as these 2 situations may present to clinicians with very similar clinical appearances. It is unclear whether the distinction is clinically important as the treatment is typically similar in both cases.
Sometimes the diagnostic signs on physical examination are not obvious. These suggestive signs can be nonspecific and subtle. They include localized swelling, erythema, superficial wound healing issues. When the patient had an intra-articular fracture, or a femoral nail placed, adjacent joints can present with infections. New or worsening effusions, especially weeks after the immediate injury, should be further evaluated with an aspiration to determine that the patient does not have septic arthritis.
Preoperative Serum Tests for FRI
Once a patient is suspected to have an infection after fracture fixation, it is current practice to perform laboratory analysis. The laboratory tests are generally markers of systemic inflammation, and frequently are nonspecific. Their results should be interpreted with caution, as none of them are specific to biofilm-related infections, and should be used merely as markers for the body’s response to inflammation. Low markers are not always sensitive for ruling out infection, and elevated serum markers may not be specific to accurately diagnose an implant-related infection.
The 3 standard laboratory tests used in orthopedic trauma include serum leukocyte count (WBC), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP). The evidence to support their use as effective screening and diagnostic tests in fracture patients is limited. In PJI infections around hip and knee replacement, there is other data evaluating the use of serum biomarkers such as ineterleukin-6, procalcitonin,52 but these have not been rigorously evaluated in trauma patients.
Positive results do not necessarily indicate a FRI. The markers can be elevated after systemic trauma,53–56 obesity,53–56 and in patients who have systemic inflammatory arthropathies. In a study by Neumaier and Scherer,57 the systemic response of CRP was evaluated in 1418 fracture patients after surgical treatment. Patients were excluded from the study if they had any evidence of postoperative complications, unrelated infections (eg, pneumonia, urinary tract infections, etc.), deep vein thrombosis, or did not have CRP values drawn. This left 787 patients for review. CRP values were taken on admission, on postoperative day 1, and at least 3 more instances between 1 and 12 days after surgery. Regardless of fracture location, the CRP values reached a peak at 2 days postoperatively, and had returned to normal by 12 days after surgery. There were 17 patients who went on to develop deep wound infections confirmed by microbiologic diagnosis. In all 17 patients they experienced a secondary rise in CRP, and they concluded that by 4 days postoperatively an elevated CRP >96 mg/L was 92% specific and 93% specific for a diagnosis of FRI.
The role of ESR and CRP is further complicated by nonunion, which may be present when evaluating patients for chronic infection. Indolent infections have been implicated as a cause for nonunion58–60 and some authors have shown that upwards of 80% of nonunions may have some evidence of bacterial contamination or colonization detected via molecular diagnostic techniques.61,62 Stucken et al58 investigated the correlation of serum inflammatory markers to culture-positive infections from nonunion surgery. They reported on 95 nonunions that had WBC, ESR, and CRP drawn as part of a preoperative workup. Individually, only ESR and CRP were statistically significant, with odds ratios of 5.2 and 4.3, respectively. However, when combined, their diagnostic efficacy increased. When all 3 serum tests were positive, they reported a 100% positive predictive value; 2 of 3 positive tests dropped this to 56%. Conversely, using negative test values to rule out infection only provided an 82% negative predictive value if all the 3 tests were negative.
Various imaging modalities are available for use as diagnostic aids. Most commonly, serial radiographs (XR) are used to evaluate the postinjury course of fracture patients. Advanced imaging, such as computed tomography (CT) and magnetic resonance imaging (MRI) can be used. Nuclear imaging has also been evaluated as an adjunct to help diagnose infections. Despite the lack of rigorous validation, many of these modalities are included as “Suggestive Criteria” in the AO FRI definition.
Serial radiographs are routinely performed on all fracture patients. These can provide a variety of clues regarding the stages of fracture healing, possible implant failure, limb alignment, arthrosis of adjacent joints, and bone quality. Serial images should be carefully compared for signs of soft tissue swelling, adjacent joint effusions, lytic bone destruction, implant loosening, or periosteal reactions.63 A normal appearing XR does not necessarily preclude infection. However, there are some radiographic signs of chronic osteomyelitis. These include sequestration, cortical irregularities, bone resorption over time, periosteal reaction, and lack of callus formation.47,64,65 Many of these changes are only apparent in chronic infection, as 50% to 75% of the bone matrix must be destroyed before lytic changes appear on plain XRs.66,67
Advanced imaging modalities include CT scans and MRI. CT scans provide useful information when considering whether a patient has a FRI. They can assess fracture healing and hardware integrity which helps with preoperative planning. Furthermore, they can identify the extent of infection, especially when abscess is present. In addition to identifying abscesses, it is important to use CT scans to identify cortical bone reactions, bony sequestration indicating the presence of osteomyelitis, or the presence of an intraosseous fistula.22,47,65 MRI scans are frequently not used after fracture surgery. Although MRI is useful when evaluating for osteomyelitis, the increased signal due to metal artifact can mimic infection when there is actually none present.68 There are numerous other postsurgical and postfracture physiological states that can mimic infection on MRI. These include fibrous scar tissue, callus formation, and reactive bone marrow edema that are indistinguishable from infection. For this reason, MRI is typically not used after fracture surgery.
There is mixed evidence regarding the use of nuclear medicine scans for diagnosing FRIs. There are several types of nuclear medicine scans, which use various different radiopharmaceuticals to aid in visualization of physiological changes. These typically involve fracture healing, bone remodeling, and the inflammatory response to infection. Bone scintigraphy can be performed with 3-phase technetium-99m-diphosphonates (99mTc), Indium-111 white blood cell-tagged (WBC) scintigraphy, or 18F-fluorodeoxyglucose positron emission tomography (18FDG-PET). Standard 3-phase bone scans are not useful for diagnosing infection because uptake is increased in any condition of increased bone metabolism, including healing fractures, postoperative changes, nonunion, or infection.69–73 The specificity for infection ranges from 0% to 10% because these changes have been shown to continue up to 2 years after fracture.72,73 Tagged WBC scintigraphy has reasonable sensitivity and specificity against infections after fracture surgery. However, these studies are expensive and time consuming. The patient must have autologous WBCs collected, labeled ex vivo, and then reinjected. The overall sensitivity ranges from 50 to 100%, and specificity from 40% to 97%.74–77 Because of the technical difficulty with performing and interpreting the examination, we do not typically perform WBC-tagged bone scan. PET/CT scan has also been reported in the literature as an imaging adjunct for diagnosing FRI.77–80 In the largest study of PET/CT scans, Wenter et al78 report on its accuracy in 215 patients with suspected chronic osteomyelitis and implant-associated infections. The scan was performed on patients who had increased pain in the absence of clinical or laboratory signs of infection. Unfortunately, the authors define infection as the presence of culture-positive bone biopsy or “clinically uneventful follow-up…for at least 1 year.” Only 89 patients had culture-positive infections, and only 66 patients had orthopedic implants. Of the 66 patients who had implants, the positive predictive value of PET/CT was 66%, and the negative predictive value was 79%. Overall, nuclear medicine studies are expensive, time consuming imaging modalities that have mixed reported usefulness and have not been rigorously evaluated for how best to incorporate their use in the diagnosis of FRI.
HAS THE INFECTED TISSUE BEEN ADEQUATELY DEBRIDED?
When used in combination, physical examination, serum inflammatory markers, and imaging studies can be useful for determining whether or not the patient is infected. Although these can help identify abscesses or the extent of osteomyelitis, they do not provide insight into what pathogen is causing the infection. Once the decision has been made to treat the patient surgically, more information must be obtained to guide intraoperative decision making regarding sufficient debridement. In addition, this debrided tissue can be sent for acute histology in the form of frozen section, and culture or molecular diagnosis. Finally, debrided tissue can include skin, subcutaneous tissue, muscle, and bone. This can result in soft tissue and skeletal stability dilemmas, which create challenges of their own to manage.
Intraoperative Tissue Sampling
Intraoperative sampling should be performed in a standardized manner. The use of deep tissue swabs has been shown to have low efficacy and has fallen out of favor.81 Some authors recommend obtaining between 3 and 5 unique tissue cultures.49 Ideally these should be taken with clean, separate instruments that have not been previously handled to minimize the risk of contamination. Samples should be obtained from sites where infection is concerning: prior fracture, necrotic tissues, implant bed, etc. These can be sent for histologic, microbiologic, and molecular diagnostic testing. Molecular diagnostic testing is discussed in detail in a related article by Natoli and colleagues. There are mixed reports in the literature concerning whether or not preoperative antibiotics should be held or administered. Ideally, the patient should not have been on prolonged antibiotics preoperatively, as this has been linked with culture negative results at least in PJI.82 Perioperative antibiotics, however, have not been associated with increased false negative rates in PJI.83 In a RCT of known PJI in 65 patients (confirmed by preoperative culture from aspiration), patients were randomized to receive standard preoperative antibiotic prophylaxis, or to hold prophylaxis until intraoperative tissue samples were obtained. In the patients who received antibiotics before incision 28/34 (82%) patients had positive cultures, compared with 25/31 (81%) of patients.
Intraoperative Findings and Debridement
The only current real-time test available to clinicians to help guide the intraoperative diagnosis of infection is frozen pathology. Unfortunately, in the setting of FRI there are many confounders that give histology low yield particularly in the setting of a fracture that has not healed.58 Stucken and colleagues reported on intraoperative Gram stain and pathology for nonunion surgery. The Gram stain had a particularly poor sensitivity (25%), but was useful if positive (100% specificity). Using 3 leukocytes per high powered field (HPF) as a positive frozen section, pathology had a 40% sensitivity and 81% specificity. Frozen pathology is operator dependent and there is no well-defined criteria for “acute inflammation.” Pathologists must screen numerous areas of the tissue sample for signs of acute inflammation, which is characterized by abundant neutrophils, edema, and bone necrosis. If tissue is obtained in the acute setting after trauma, necrosis and inflammation is part of the normal physiological response to trauma. Later, callus formation and healing may further mimic acute inflammation.49 Authors have tried to quantify the number of neutrophils per HPF in the setting of PJI. However, even in PJI the routine use of histology has been abandoned.23–33 The most useful finding is the rare instance in which a microorganism is visualized during microscopic evaluation. This is rare, and still can be complicated by contamination as the source of bacteria. A recent paper on the use of histopathology for chronic infections (eg, nonunions), the authors evaluated intraoperative frozen sections in 156 surgically treated nonunions. These included 64 confirmed infections, 66 aseptic nonunions, and 26 “possibly” infected nonunions. They showed that >5 neutrophils per HPF had 100% specificity and positive predictive value for FRI. In addition, the complete absence of neutrophils had high specificity (98%) and negative predictive value (98%) for predicting aseptic nonunion.84 These data may be useful to aid decision making when performing nonunion surgery.
In the absence of other immediate tests that can provide definitive evidence of bacterial infection, surgeons are left with their assessment of intraoperative findings to determine whether sufficient bone and soft tissue has been debrided. Signs that are suggestive of infection include gross purulence, necrotic tissue and bone, and implant loosening. The interpretation of these signs is not always obvious, and may also be operator dependent. However, while their presence may help confirm suspected infection, their absence does not implicate absence of infection.
The determination of adequate site control is particularly difficult with bone infections for 2 reasons. The first is that currently dogma states that necrotic bone without a blood supply must be removed to prevent recurrence of infection as this material is thought to provide safe harbor for the bacteria to hide until systemic antibiotics are completed. Second, the bacteria burden in the wound in general could guide when the general debridement is sufficient to lower the bacterial burden to a level that further surgeries are unnecessary. Currently we neither know the level of bacteria that is low enough to allow infection clearance without further surgeries, nor do we have any currently available tools for measuring bacterial volume in the wounds. These are real limitations of our ability to ensure site control and also ensure that we are not performing unnecessary additional surgeries to clear the infection.
Hardware and Fracture Management
Fractures may be in various stages of healing when infections present. Therefore, the surgeon must decide how to handle limb stabilization at the time of infection surgery, which can be challenging. Options include retaining the existing hardware, revising it, or removing it.85
There is little published data to guide this decision-making process. If intramedullary fixation was used and the fracture is not healed, serial debridements with antibiotic nail placement can help with fracture stabilization and antibiotic delivery. Hardware removal may be an option if the fracture is healed. However, there are instances where hardware removal may not be feasible or may compromise the ability to maintain articular reduction or limb stability. In these instances, early debridement with hardware retention has been reportedly successful. In a paper by Berkes et al,86 the authors report on the surgical treatment of FRI with hardware retention. Of 123 patients treated with debridement within 1 to 6 weeks after index surgery, 87 (70%) had successful fracture union without needing hardware removal or exchange.
WHAT PATHOGEN TO TREAT?
It is currently not possible to determine without surgical tissue biopsy what pathogen is causing FRI. Consequently, it is difficult to determine what organism to target with antibiotic therapy. Intraoperative point-of-care tests do not currently exist to provide the surgeon with real-time information regarding the causative organism. The current gold standard is tissue cultures. However, culture data can also be unreliable and have been abandoned outside of clinical medicine in favor of more modern techniques in other applications of molecular diagnostics. Other methods of obtaining culture data have been used, such as sonication of explanted hardware or tissue samples. Molecular diagnostic tests and immunologic tests are being developed, but are currently expensive, have not been extensively studied in FRI, and are not readily available for commercial use.
Microbiologic Culture From Tissue Samples
Positive tissue cultures remain the most used data for targeted antibiotic therapy and confirmation that infection is present. Advantages of tissue cultures include organism identification and antibiotic sensitivity. However, there are a number of issues with tissue cultures that have led investigators in the field of microbiology to abandon cultures as the gold standard for microorganism identification. These include false positive results, tissue sampling errors, and failure to grow organisms that may not thrive in laboratory conditions. Most of the research concerning culture-negative periprosthetic infections have been conducted in the arthroplasty literature, where they have reported up to 25%.87–95 Culture negative infections have been reported in 7% to 9% of studies on FRIs.20,21,96 To minimize the risk of culture negative infections, it is recommended that multiple tissue samples be obtained as described above. In the arthroplasty literature, Kheir et al97 published the results of 2676 unique culture specimens in 711 PJIs that met Musculoskeletal Infection Society criteria for PJI. They specifically investigated the optimal number of cultures to provide at least 2 positive results, and reviewed the typical time to a positive culture result. They concluded that 4 specimens were optimal to have a minimum of 2 positive results with the same organism, and that the cultures should be held for a minimum of 8 to 10 days to maximize the likelihood of growing organisms such as p acnes and gram-negative species, some of which took up to 25 days to grow in culture media.97 There is no analogous study on the number of cultures to optimize positive results in postfracture patients.
To increase the culture yield, some investigators applied low-intensity ultrasound (sonication) to retrieved implants. This process theoretically dislodges the biofilm and its organisms into the sonication fluid with a higher positive culture result. Most of these studies have been performed in the arthroplasty literature, and a meta-analysis concluded based on 12 studies the sonicated cultures had a sensitivity of 0.80, and specificity of 0.95.98 Yano et al99 evaluated culture results after implant sonication for implants retrieved from FRI. They sonicated implants from 180 patients regardless of reason for implant removal. These reasons included symptomatic hardware, nonunion/malunion, or infection. They defined infection as a wound with exposed bone or hardware, gross purulence intraoperatively, draining sinus tract, or acute inflammation on histology. A sonication culture was considered pathologic if there were at least 2 separate cultures with identical organisms. Of the 180 patients, 125 (69%) were considered infected, and 55 (31%) were considered aseptic. When compared with tissue culture, there were 113 positive sonication cultures compared with 71 patients with positive tissue cultures, and sonication was more likely to identify polymicrobial infections. From a diagnostic standpoint, sonication offered an improvement in sensitivity (90% for sonication vs. 57% for tissue culture).99 Despite this marginal improvement in sensitivity, sonication is not frequently used due to its technical complexity and the fact that it is not readily available outside of research laboratories.
In the absence of positive cultures, the natural history and timing of FRI can be useful to help correlate empiric therapy. Torbert et al20 reviewed the common bacterial speciation and presence of antibiotic resistance in deep infections after surgery for extremity and pelvic fractures over a 4-year period from 2006 to 2010. They reported on 214 deep infections in 211 patients who had an average age of 45 years (range, 16 to 95 y) and who had their surgical treatment of FRI within 12 months of index fracture surgery. Of the 214 infections, 198 (93%) grew an organism in culture. The most common organism was Staphylococcus aureus, followed by gram negative rods (GNR), and then enterococcus. There were 79/249 (32%) cultures that had clinically relevant resistance to antibiotics. The rate of GNR infection was similar between open (32%) and closed (33%) fractures, but was higher in pelvic, acetabular, and proximal femoral fractures (63%; P=0.0002). The average time to presentation was 11 weeks after surgery (range, 3 d to 51 wk). Infections from GNR tended to present earlier (62% of GNR presented by week 2, compared with only 25% of S. aureus infections). These findings may be site specific and may change over time as demonstrated by this same group published that their local bacteriology seemed to change over time (Montalvo).
Molecular Diagnostic Tools
Molecular diagnostic tools utilize a known database of bacterial DNA sequences to identify the presence of specific bacterial DNA or RNA from a sample and are discussed in detail elsewhere in the study by Natoli and colleagues. The most common technique is polymerase chain reaction (PCR). PCR can identify fragments of genetic material from bacterial ribosomal 16S subunits. Samples can be tissue samples, fluid samples (eg, synovial fluid), or sonication fluid from removed implants. Another technique used to identify organisms is a PCR-based test in combination with mass spectrometry (Ibis, Abbott Laboratories, Lake Bluff, IL). The Ibis technology has the ability to identify polymicrobial infections due to multiple organisms, and the presence of specific antibiotic resistance genes. Each of these technologies has advantages and disadvantages.
Proponents of the use of molecular methods to identify causative organisms in implant-related infections argue that it can identify difficult-to-culture bacteria, is effective after administration of systemic antibiotics, and can detect small amounts of bacteria. Furthermore, small amounts of bacterial DNA can be identified.100–102 Problems with PCR include its cost, time to diagnosis, inability to distinguish between living and dead bacterial genetic material, and the fact that it is not currently not used in routine clinical diagnosis. The fact that even small amounts of bacterial DNA can be identified has led some to question whether these techniques may be too sensitive to differentiate between contaminants and pathogens.103–105
Renz et al106 evaluated the use of PCR to identify bacteria in patients who had hardware removed for any reason. There were 51 patients who had a mean age of 61 years (range, 21 to 86 y). There were 38 patients who had diagnosed infections using modified Infectious Disease Society of America criteria, and 13 who had “aseptic” hardware failure as the reason for removal. Only 92% (35/38) of the septic hardware failures had a positive culture from either tissue or sonication fluid, and 6 infections were polymicrobial. There were 17 discordant pairs when they compared the PCR and culture. There was 1 false positive in the group of aseptic hardware removals. There were 9 false negative PCR results in the septic patients who had positive cultures. There were 3 infected patients who had different organisms identified by PCR from what was identified by culture, and 4 patients had a negative culture but positive PCR. There were 2 pathogens for which the primer was not included in the PCR analysis, so it was not possible for them to be identified.
It is difficult to currently identify how best to use this technology. Although it is exciting that it can identify difficult to culture bacteria, it is impossible to discern how to interpret the “discordant” results. What is the definitive causative bacteria? Are the false negatives truly not infected? Or is the database of known organisms inadequate? What is the true gold standard when one technology identifies the presence of different organisms when compared with another technology. More work needs to be performed before this technology is used on a regular basis. However, it is a useful tool to consider when conventional methods fail.
Diagnosing infections can be challenging. Clinicians must have a high index of suspicion when cases are not obvious. To further complicate the ability to diagnose, there is no validated definition of what constitutes an infection after fracture surgery. Physical examination findings may be helpful when overwhelming infection is present. Sometimes the signs can be as subtle as persistent pain. Most imaging modalities are nonspecific, and culture results are far from perfect and tissue samples are not possible to obtain without surgery. We have attempted to present the reader with a stepwise algorithm to help when diagnosing infection. Future technologies will hopefully focus on less invasive, fast, inexpensive, and accurate modalities to help determine whether the patient has an infection, and help guide infection treatments by identifying the causative pathogen and any antibiotic resistance. Future studies should continue to evaluate how to optimize the use of existing diagnostics, and how to incorporate these new technologies to improve patient care.
1. MacKenzie EJ, Bosse MJ, Kellam JF, et al. Factors influencing the decision to amputate or reconstruct after high-energy lower extremity trauma. J Trauma. 2002;52:641–649.
2. Bhandari M, Tornetta P III, Sprague S, et al. Predictors of reoperation following operative management of fractures of the tibial shaft. J Orthop Trauma. 2003;17:353–361.
3. Petrisor B, Anderson S, Court-Brown CM. Infection after reamed intramedullary nailing of the tibia: a case series review. J Orthop Trauma. 2005;19:437–441.
4. Valenziano CP, Chattar-Cora D, O’Neill A, et al. Efficacy of primary wound cultures in long bone open extremity fractures: are they of any value? Arch Orthop Trauma Surg. 2002;122:259–261.
5. Benson DR, Riggins RS, Lawrence RM, et al. Treatment of open fractures
: a prospective study. J Trauma. 1983;23:25–30.
6. Caudle RJ, Stern PJ. Severe open fractures
of the tibia. J Bone Joint Surg Am. 1987;69:801–807.
7. Cullen MC, Roy DR, Crawford AH, et al. Open fracture of the tibia in children. J Bone Joint Surg Am. 1996;78:1039–1047.
8. DeLong WG Jr, Born CT, Wei SY, et al. Aggressive treatment of 119 open fracture wounds. J Trauma. 1999;46:1049–1054.
9. Hope PG, Cole WG. Open fractures
of the tibia in children. J Bone Joint Surg Br. 1992;74:546–553.
10. Metsemakers WJ, Handojo K, Reynders P, et al. Individual risk factors for deep infection and compromised fracture healing after intramedullary nailing of tibial shaft fractures: a single centre experience of 480 patients. Injury. 2015;46:740–745.
11. Ktistakis I, Giannoudi M, Giannoudis PV. Infection rates after open tibial fractures: are they decreasing? Injury. 2014;45:1025–1027.
12. Arens S, Hansis M, Schlegel U, et al. Infection after open reduction and internal fixation with dynamic compression plates: clinical and experimental data. Injury. 1996;27(suppl 3):SC27–SC33.
13. Metsemakers WJ, Kortram K, Morgenstern M, et al. Definition of infection after fracture fixation: a systematic review of randomized controlled trials to evaluate current practice. Injury. 2018;49:497–504.
14. Govaert GAM, Kuehl R, Atkins BL, et al. Diagnosing fracture-related infection
: current concepts and recommendations. J Orthop Trauma. 2020;34:8–17.
15. Metsemakers WJ, Morgenstern M, McNally MA, et al. Fracture-related infection
: a consensus on definition from an international expert group. Injury. 2018;49:505–510.
16. Obremskey WT, Metsemakers WJ, Schlatterer DR, et al. Musculoskeletal infection in orthopaedic trauma: assessment of the 2018 International Consensus Meeting on musculoskeletal infection. J Bone Joint Surg Am. 2020. [Epub ahead of print].
17. Mangram AJ, Horan TC, Pearson ML, et al. Guideline for prevention of surgical site infection, 1999. Hospital Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol. 1999;20:250–278; quiz 279-280.
18. Berrios-Torres SI, Umscheid CA, Bratzler DW, et al. Centers for Disease Control and Prevention Guideline for the Prevention of Surgical Site Infection, 2017. JAMA Surg. 2017;152:784–791.
19. Parvizi J, Zmistowski B, Berbari EF, et al. New definition for periprosthetic joint infection: from the Workgroup of the Musculoskeletal Infection Society. Clin Orthop Relat Res. 2011;469:2992–2994.
20. Torbert JT, Joshi M, Moraff A, et al. Current bacterial speciation and antibiotic resistance in deep infections after operative fixation of fractures. J Orthop Trauma. 2015;29:7–17.
21. Montalvo RN, Natoli RM, O’Hara NN, et al. Variations in the organisms causing deep surgical site infections in fracture patients at a level I trauma center (2006-2015). J Orthop Trauma. 2018;32:e475–e481.
22. Metsemakers WJ, Kuehl R, Moriarty TF, et al. Infection after fracture fixation: current surgical and microbiological concepts. Injury. 2018;49:511–522.
23. Villa JM, Pannu TS, Piuzzi N, et al. Evolution of diagnostic definitions for periprosthetic joint infection in total hip and knee arthroplasty. J Arthroplasty. 2020;35(3S):S9–S13.
24. Amanatullah D, Dennis D, Oltra EG, et al. Hip and knee section, diagnosis
, definitions: Proceedings of International Consensus on Orthopedic Infections. J Arthroplasty. 2019;34(2S):S329–S337.
25. Shohat N, Bauer T, Buttaro M, et al. Hip and knee section, what is the definition of a periprosthetic joint infection (PJI) of the knee and the hip? Can the same criteria be used for both joints?: Proceedings of International Consensus on Orthopedic Infections. J Arthroplasty. 2019;34(2S):S325–S327.
26. Parvizi J, Tan TL, Goswami K, et al. The 2018 definition of periprosthetic hip and knee infection: an evidence-based and validated criteria. J Arthroplasty. 2018;33:1309.e2–1314.e2.
27. Klement MR, Siddiqi A, Rock JM, et al. Are all periprosthetic joint infections the same? Evaluating major vs minor criteria. J Arthroplasty. 2018;33:1515–1519.
28. Honkanen M, Jamsen E, Karppelin M, et al. Concordance between the old and new diagnostic criteria for periprosthetic joint infection. Infection. 2017;45:637–643.
29. Tansey R, Mirza Y, Sukeik M, et al. Definition of periprosthetic hip and knee joint infections and the economic burden. Open Orthop J. 2016;10:662–668.
30. Metsemakers WJ, Moriarty TF, Morgenstern M, et al. Letter to the Editor: new definition for periprosthetic joint infection: from the Workgroup of the Musculoskeletal Infection Society. Clin Orthop Relat Res. 2016;474:2726–2727.
31. Parvizi J, Gehrke T. International Consensus Group on Periprosthetic Joint I. Definition of periprosthetic joint infection. J Arthroplasty. 2014;29:1331.
32. Workgroup Convened by the Musculoskeletal Infection Society. New definition for periprosthetic joint infection. J Arthroplasty. 2011;26:1136–1138.
33. Parvizi J, Jacovides C, Zmistowski B, et al. Definition of periprosthetic joint infection: is there a consensus? Clin Orthop Relat Res. 2011;469:3022–3030.
34. Bakker K, Apelqvist J, Lipsky BA, et al. International Working Group on the Diabetic Foot. The 2015 IWGDF guidance documents on prevention and management of foot problems in diabetes: development of an evidence-based global consensus. Diabetes Metab Res Rev. 2016;32(suppl 1):2–6.
35. Blauth M, Bastian L, Krettek C, et al. Surgical options for the treatment of severe tibial pilon fractures: a study of three techniques. J Orthop Trauma. 2001;15:153–160.
36. Patterson MJ, Cole JD. Two-staged delayed open reduction and internal fixation of severe pilon fractures. J Orthop Trauma. 1999;13:85–91.
37. Pugh KJ, Wolinsky PR, McAndrew MP, et al. Tibial pilon fractures: a comparison of treatment methods. J Trauma. 1999;47:937–941.
38. Sirkin M, Sanders R, DiPasquale T, et al. A staged protocol for soft tissue management in the treatment of complex pilon fractures. J Orthop Trauma. 1999;13:78–84.
39. Pollak AN, Jones AL, Castillo RC, et al. The relationship between time to surgical debridement and incidence of infection after open high-energy lower extremity trauma. J Bone Joint Surg Am. 2010;92:7–15.
40. D’Alleyrand JC, Manson TT, Dancy L, et al. Is time to flap coverage of open tibial fractures an independent predictor of flap-related complications? J Orthop Trauma. 2014;28:288–293.
41. Cierny G III, Mader JT. Approach to adult osteomyelitis
. Orthop Rev. 1987;16:259–270.
42. Cierny G III, Mader JT, Penninck JJ. A clinical staging system for adult osteomyelitis
. Clin Orthop Relat Res. 2003;414:7–24.
43. Cierny GI II. Surgical treatment of osteomyelitis
. Plast Reconstr Surg. 2011;127(suppl 1):190S–204S.
44. Mader JT, Shirtliff M, Calhoun JH. Staging and staging application in osteomyelitis
. Clin Infect Dis. 1997;25:1303–1309.
45. DeWall M, Henderson CE, McKinley TO, et al. Percutaneous reduction and fixation of displaced intra-articular calcaneus fractures. J Orthop Trauma. 2010;24:466–472.
46. Metsemakers WJ, Smeets B, Nijs S, et al. Infection after fracture fixation of the tibia: analysis of healthcare utilization and related costs. Injury. 2017;48:1204–1210.
47. Patzakis MJ, Zalavras CG. Chronic posttraumatic osteomyelitis
and infected nonunion of the tibia: current management concepts. J Am Acad Orthop Surg. 2005;13:417–427.
48. Willey M, Karam M. Impact of infection on fracture fixation. Orthop Clin North Am. 2016;47:357–364.
49. Morgenstern M, Kuhl R, Eckardt H, et al. Diagnostic challenges and future perspectives in fracture-related infection
. Injury. 2018;49(suppl 1):S83–S90.
50. Paryavi E, Stall A, Gupta R, et al. Predictive model for surgical site infection risk after surgery for high-energy lower-extremity fractures: development of the risk of infection in orthopedic trauma surgery score. J Trauma Acute Care Surg. 2013;74:1521–1527.
51. Trampuz A, Zimmerli W. Diagnosis
and treatment of infections associated with fracture-fixation devices. Injury. 2006;37(suppl 2):S59–S66.
52. Ettinger M, Calliess T, Kielstein JT, et al. Circulating biomarkers for discrimination between aseptic joint failure, low-grade infection, and high-grade septic failure. Clin Infect Dis. 2015;61:332–341.
53. Lindstrom T, Gullichsen E, Heinonen O, et al. Group II phospholipase A2 in serum after knee surgery and intramedullary nailing of tibial shaft fracture. Injury. 1997;28:169–171.
54. Easton R, Balogh ZJ. Peri-operative changes in serum immune markers after trauma: a systematic review. Injury. 2014;45:934–941.
55. Garnavos C, Xirou ST, Nikolatos A, et al. Alteration of body temperature, erythrocyte sedimentation rate, and C-reactive protein after reamed intramedullary nailing: a prospective study. J Orthop Trauma. 2005;19:323–328.
56. Giannoudis PV, Smith RM, Bellamy MC, et al. Stimulation of the inflammatory system by reamed and unreamed nailing of femoral fractures. An analysis of the second hit. J Bone Joint Surg Br. 1999;81:356–361.
57. Neumaier M, Scherer MA. C-reactive protein levels for early detection of postoperative infection after fracture surgery in 787 patients. Acta Orthop. 2008;79:428–432.
58. Stucken C, Olszewski DC, Creevy WR, et al. Preoperative diagnosis
of infection in patients with nonunions. J Bone Joint Surg Am. 2013;95:1409–1412.
59. Motsitsi NS. Management of infected nonunion of long bones: the last decade (1996-2006). Injury. 2008;39:155–160.
60. Meyer S, Weiland AJ, Willenegger H. The treatment of infected non-union of fractures of long bones. Study of sixty-four cases with a five to twenty-one-year follow-up. J Bone Joint Surg Am. 1975;57:836–842.
61. Palmer MP, Altman DT, Altman GT, et al. Can we trust intraoperative culture results in nonunions? J Orthop Trauma. 2014;28:384–390.
62. Natoli RM, Marinos D, Montalvo R, et al. Are all nonunions infected? Comparison of culture versus bacterial DNA presence. Orthopaedic Trauma Association Annual Meeting; 2018.
63. Firoozabadi R, Alton T, Wenke J. Novel strategies for the diagnosis
of posttraumatic infections in orthopaedic trauma patients. J Am Acad Orthop Surg. 2015;23:443–451.
64. Tumeh SS, Aliabadi P, Weissman BN, et al. Disease activity in osteomyelitis
: role of radiography. Radiology. 1987;165:781–784.
65. Gross T, Kaim AH, Regazzoni P, et al. Current concepts in posttraumatic osteomyelitis
: a diagnostic challenge with new imaging options. J Trauma. 2002;52:1210–1219.
66. Butt WP. The radiology of infection. Clin Orthop Relat Res. 1973;96:20–30.
67. Butt WP. Radiology of the infected joint. Clin Orthop Relat Res. 1973;96:136–149.
68. Ledermann HP, Kaim A, Bongartz G, et al. Pitfalls and limitations of magnetic resonance imaging in chronic posttraumatic osteomyelitis
. Eur Radiol. 2000;10:1815–1823.
69. Lee YJ, Sadigh S, Mankad K, et al. The imaging of osteomyelitis
. Quant Imaging Med Surg. 2016;6:184–198.
70. Pineda C, Espinosa R, Pena A. Radiographic imaging in osteomyelitis
: the role of plain radiography, computed tomography, ultrasonography, magnetic resonance imaging, and scintigraphy. Semin Plast Surg. 2009;23:80–89.
71. Buhne KH, Bohndorf K. Imaging of posttraumatic osteomyelitis
. Semin Musculoskelet Radiol. 2004;8:199–204.
72. Govaert GA, IJpma FF, McNally M, et al. Accuracy of diagnostic imaging modalities for peripheral post-traumatic osteomyelitis
: a systematic review of the recent literature. Eur J Nucl Med Mol Imaging. 2017;44:1393–1407.
73. Govaert GAM, Glaudemans A. Nuclear medicine imaging of posttraumatic osteomyelitis
. Eur J Trauma Emerg Surg. 2016;42:397–410.
74. Ballani NS, Al-Huda FA, Khan HA, et al. The value of quantitative uptake of (99m)Tc-MDP and (99m)Tc-HMPAO white blood cells in detecting osteomyelitis
in violated peripheral bones. J Nucl Med Technol. 2007;35:91–95.
75. Glaudemans AW, de Vries EF, Vermeulen LE, et al. A large retrospective single-centre study to define the best image acquisition protocols and interpretation criteria for white blood cell scintigraphy with (9)(9)mTc-HMPAO-labelled leucocytes in musculoskeletal infections. Eur J Nucl Med Mol Imaging. 2013;40:1760–1769.
76. Kaim A, Ledermann HP, Bongartz G, et al. Chronic post-traumatic osteomyelitis
of the lower extremity: comparison of magnetic resonance imaging and combined bone scintigraphy/immunoscintigraphy with radiolabelled monoclonal antigranulocyte antibodies. Skeletal Radiol. 2000;29:378–386.
77. Meller J, Koster G, Liersch T, et al. Chronic bacterial osteomyelitis
: prospective comparison of (18)F-FDG imaging with a dual-head coincidence camera and (111)In-labelled autologous leucocyte scintigraphy. Eur J Nucl Med Mol Imaging. 2002;29:53–60.
78. Wenter V, Muller JP, Albert NL, et al. The diagnostic value of [(18)F]FDG PET for the detection of chronic osteomyelitis
and implant-associated infection. Eur J Nucl Med Mol Imaging. 2016;43:749–761.
79. Schiesser M, Stumpe KD, Trentz O, et al. Detection of metallic implant-associated infections with FDG PET in patients with trauma: correlation with microbiologic results. Radiology. 2003;226:391–398.
80. Shemesh S, Kosashvili Y, Groshar D, et al. The value of 18-FDG PET/CT in the diagnosis
and management of implant-related infections of the tibia: a case series. Injury. 2015;46:1377–1382.
81. Aggarwal VK, Higuera C, Deirmengian G, et al. Swab cultures are not as effective as tissue cultures for diagnosis
of periprosthetic joint infection. Clin Orthop Relat Res. 2013;471:3196–3203.
82. Malekzadeh D, Osmon DR, Lahr BD, et al. Prior use of antimicrobial therapy is a risk factor for culture-negative prosthetic joint infection. Clin Orthop Relat Res. 2010;468:2039–2045.
83. Tetreault MW, Wetters NG, Aggarwal V, et al. The Chitranjan Ranawat Award: should prophylactic antibiotics be withheld before revision surgery to obtain appropriate cultures? Clin Orthop Relat Res. 2014;472:52–56.
84. Morgenstern M, Athanasou NA, Ferguson JY, et al. The value of quantitative histology in the diagnosis
of fracture-related infection
. Bone Joint J. 2018;100-B:966–972.
85. Metsemakers WJ, Morgenstern M, Senneville E, et al. General treatment principles for fracture-related infection
: recommendations from an international expert group. Arch Orthop Trauma Surg. 2019. [Epub ahead of print].
86. Berkes M, Obremskey WT, Scannell B, et al. Maintenance of hardware after early postoperative infection following fracture internal fixation. J Bone Joint Surg Am. 2010;92:823–828.
87. Choi HR, Kwon YM, Freiberg AA, et al. Periprosthetic joint infection with negative culture results: clinical characteristics and treatment outcome. J Arthroplasty. 2013;28:899–903.
88. Zimmerli W, Trampuz A, Ochsner PE. Prosthetic-joint infections. N Engl J Med. 2004;351:1645–1654.
89. Morris AJ, Wilson SJ, Marx CE, et al. Clinical impact of bacteria and fungi recovered only from broth cultures. J Clin Microbiol. 1995;33:161–165.
90. Berbari EF, Marculescu C, Sia I, et al. Culture-negative prosthetic joint infection. Clin Infect Dis. 2007;45:1113–1119.
91. Masri BA, Panagiotopoulos KP, Greidanus NV, et al. Cementless two-stage exchange arthroplasty for infection after total hip arthroplasty. J Arthroplasty. 2007;22:72–78.
92. Ure KJ, Amstutz HC, Nasser S, et al. Direct-exchange arthroplasty for the treatment of infection after total hip replacement. An average ten-year follow-up. J Bone Joint Surg Am. 1998;80:961–968.
93. Font-Vizcarra L, Garcia S, Martinez-Pastor JC, et al. Blood culture flasks for culturing synovial fluid in prosthetic joint infections. Clin Orthop Relat Res. 2010;468:2238–2243.
94. Duff GP, Lachiewicz PF, Kelley SS. Aspiration of the knee joint before revision arthroplasty. Clin Orthop Relat Res. 1996;331:132–139.
95. Pandey R, Berendt AR, Athanasou NA. Histological and microbiological findings in non-infected and infected revision arthroplasty tissues. The OSIRIS Collaborative Study Group. Oxford Skeletal Infection Research and Intervention Service. Arch Orthop Trauma Surg. 2000;120:570–574.
96. Gitajn IL, Heng M, Weaver MJ, et al. Culture-negative infection after operative fixation of fractures. J Orthop Trauma. 2016;30:538–544.
97. Kheir MM, Tan TL, Ackerman CT, et al. Culturing periprosthetic joint infection: number of samples, growth duration, and organisms. J Arthroplasty. 2018;33:3531.e1–3536.e1.
98. Liu K, Fu J, Yu B, et al. Meta-analysis of sonication prosthetic fluid PCR for diagnosing periprosthetic joint infection. PloS One. 2018;13:e0196418.
99. Yano MH, Klautau GB, da Silva CB, et al. Improved diagnosis
of infection associated with osteosynthesis by use of sonication of fracture fixation implants. J Clin Microbiol. 2014;52:4176–4182.
100. Bergin PF, Doppelt JD, Hamilton WG, et al. Detection of periprosthetic infections with use of ribosomal RNA-based polymerase chain reaction. J Bone Joint Surg Am. 2010;92:654–663.
101. Gomez E, Cazanave C, Cunningham SA, et al. Prosthetic joint infection diagnosis
using broad-range PCR of biofilms dislodged from knee and hip arthroplasty surfaces using sonication. J Clin Microbiol. 2012;50:3501–3508.
102. Greenwood-Quaintance KE, Uhl JR, Hanssen AD, et al. Diagnosis
of prosthetic joint infection by use of PCR-electrospray ionization mass spectrometry. J Clin Microbiol. 2014;52:642–649.
103. Clarke MT, Roberts CP, Lee PT, et al. Polymerase chain reaction can detect bacterial DNA in aseptically loose total hip arthroplasties. Clin Orthop Relat Res. 2004;427:132–137.
104. Panousis K, Grigoris P, Butcher I, et al. Poor predictive value of broad-range PCR for the detection of arthroplasty infection in 92 cases. Acta Orthop. 2005;76:341–346.
105. Tsuru A, Setoguchi T, Kawabata N, et al. Enrichment of bacteria samples by centrifugation improves the diagnosis
of orthopaedics-related infections via real-time PCR amplification of the bacterial methicillin-resistance gene. BMC Res Notes. 2015;8:288.
106. Renz N, Cabric S, Morgenstern C, et al. Value of PCR in sonication fluid for the diagnosis
of orthopedic hardware-associated infections: has the molecular era arrived? Injury. 2018;49:806–811.