Reports of deep infection indicate that all types of replacement arthroplasty can be complicated by infection. The incidence is small, indicating relative safety of the procedure. Diabetes and prior surgery are the two main associations noted, and skin necrosis postoperatively is a potential causative factor.
Profile of high-risk patients is part of the prevention.
Operating room precautions include inclusive gowns, laminar flow, and limiting traffic in and out of the operating room.
Early diagnosis: imaging studies, erythrocyte sedimentation rate, and C-reactive protein serologies.
Intraoperative cultures include polymerase chain reaction for common organisms.
Reoperation: very few patients, good general principles, and local use of antibiotics.
Salvage of an infection may depend on favorable organism: Streptococcus and methicillin-sensitive Staphylococcus aureus .
Optimize preoperative conditions: nutrition important, thorough elimination of skin pathogens, nasal cultures, and possible mupiricin for patient
Chlorhexidine skin prep
Antibiotic cement spacer after removal of infected prosthesis
Revision with tibiotalocalcaneal fusion if severe loss of bone
HISTORY/INTRODUCTION/SCOPE OF THE PROBLEM
Total ankle arthroplasty (TAA) continues to emerge and refine itself as a viable treatment for end-stage ankle arthritis. Despite a small body of literature on the complications of modern ankle joint replacement, it is intuitive that as ankle replacement surgery gains more acceptance within the orthopedic community, the volume of surgery will increase, as will the number of complications. With increasing acceptance and favorable results, there has also been an increase in the number of different ankle replacement systems on the market. Each system has unique characteristics that can result in problems and challenges specific to that prosthesis.
Certain complications, however, are common to all TAAs. One of the most troublesome and serious is deep periprosthetic infection. Deep periprosthetic infection following total joint arthroplasty is considered a major complication. Although it occurs in only a small percentage of patients, it results in substantial morbidity and a decline in functional outcome. The presence of infection can be devastating to both the patient and the surgeon. For that reason, it is essential to thoroughly understand the concepts of infection by focusing on its prevention, accurate early diagnosis, and sound management.
The challenge is that there is a paucity of literature regarding infection in TAA. Much of what is written in this chapter comes from general orthopedic knowledge, a review of current hip and knee arthroplasy literature, and our 10-year experience performing TAA. The benefit of this approach is that there is a large volume of literature written on infection in total hip and knee arthroplasty and current recommendations for other joint replacements are an extension of this body of knowledge. However, there are significant limitations with this approach. The most significant is that the patient populations, underlying diseases, and surgical challenges are very different between hip/knee replacements and ankle replacements. These differences can and do have a major impact on the treatment of infection following TAA.
INCIDENCE AND PREDISPOSING FACTORS
Deep infection is a potential complication of any joint replacement surgery; ankle arthroplasty is no exception. The risk of infection is generally quoted as being 1%. A review of literature has shown the incidence of infection in primary TKA ranging from 0.5% to 2%. It is quoted as 7% during revision surgery. This range also applies to TAA in our experience. The rate of infection depends on multiple factors, which include host factors, operating room environment, wound issues, and surgical technique.
The host is always an important factor when discussing infection. When comparing TAA patients with total hip and knee arthroplasty patients, it is clear that there are significant differences in the hosts. Patients with ankle arthritis almost always have their disease as a result of trauma, whereas hip/knee arthritis patients almost always have degenerative osteoarthrtitis. As a result, patients with ankle arthritis are substantially younger than patients undergoing hip/knee replacements. Comorbidities, at this point in history, seem similar. However, what is unclear is whether certain systemic diseases, which are generally similar in both groups, affect each group in the same way. For example, a slightly obese 60-year-old patient with a 10-year history of non–insulin-dependent diabetes mellitus controlled by diet and oral hypoglycemic agents and maintaining an HbA 1c level of 6.5 with no clinically detectable distal neuropathy is most likely a candidate for either total ankle and total knee replacement. The question arises whether the systemic condition affects the microenvironment around the knee and ankle equally. If so, is it a significant difference in terms of their propensity for, or ability to, handle infection?
Several other general host factors have been identified as risk factors for developing infection in patients undergoing arthroplasty. The inability of the host to prevent infection may be one of the most significant factors, as evidenced by the increased risk of infection in the immunocompromised patient. In patients with rheumatoid arthritis the rate of infection has been reported to be 2 to 3 times higher compared with other patients. Diabetes is also a risk factor for infection. In the total knee arthroplasty literature, the incidence is reported to range from 3% to 7%. Other host factors, such as morbid obesity, oral steroid use, and concurrent infection at other sites (e.g., dental abscess, urinary tract infection, or distal foot infection), have been associated with increased rate of infection. Advanced age, poor nutrition, chronic renal failure, chronic alcoholism, smoking, and malignancy may also play roles in the development of infection.
Similarly, the local and regional soft tissue environments found around the hip/knee and ankle are very different. The primary hip/knee surgeon operates in an area with thick skin, usually without significant scars, and then cuts through bone compromised only by age and chronic arthritic changes. The primary total ankle surgeon faces thin skin commonly scarred by previous surgical approaches and makes cuts through traumatically damaged bone further compromised by age and arthritic changes. Previous operations on the ankle have been associated with increased risk of wound healing and infection. The local wound environment is often suboptimum in patients who have a history of multiple operative procedures, extensive scarring, or wound infection. A practical corollary of this is that the anterior ankle incision may not heal as well as the anterior knee incision, allowing for a greater chance of postoperative infection. Careful preoperative assessment may allow modification of the operative approach and technique to minimize the risk of infection. The utilization of existing scars to avoid postoperative skin necrosis is especially essential about the ankle joint, although this is limited in applicability since most total ankle systems use a single unchangeable surgical approach ( Fig. 18-1 ).
Although the prevalence of postoperative deep wound infection after total joint arthroplasty has decreased over the past few decades, it remains a devastating complication and, despite improved outcomes of treatment for established deep periprosthetic infection, prevention remains a meaningful objective. A thorough appreciation of the different factors that may contribute to an infection is essential in the development of an overall approach to prevention. Infection depends on the number and virulence of the bacteria introduced into a wound, the host’s ability to eliminate these bacteria, and the status or viability of the wound environment.
Screening patients who may be colonized with bacteria that can lead to a postoperative infection is emerging as an important concept in an era of antibiotic-resistant bacteria. Staphylococcus aureus is a ubiquitous organism responsible for the majority of total joint infections. For over 50 years, there have been studies to suggest the relationship of nasal carriage of S. aureus and surgical site infections. In 1999, the Centers for Disease Control and Prevention listed nasal carriage of S. aureus as a risk factor for surgical site infection, with studies showing that these patients are 2 to 3 times more likely to develop surgical site infections than are noncarriers. As many as 25% to 30% of the general population may carry methicillin-sensitive S. aureus in their nose. More recently, drug-resistant organisms are emerging as a greater threat for infection. Methicillin-resistant S. aureus (MRSA) is one such organism. From 1975 to 2002, the incidence of reported MRSA isolates increased from 2% to 57%, and 46 of every 1000 patients were either infected or colonized with MRSA. Colonization may also occur in the axilla, a chronic wound or decubitus ulcer, perineum, sputum, or around a tracheostomy or gastrostomy site. As many as 5% of patients on orthopedic and surgical wards may be colonized with MRSA. Current guidelines for screening patients for S. aureus infection include recent or prolonged hospitalizations within the past 24 months, advanced age, nursing home admission within the past 12 months, antibiotic exposure within the past 12 months, intravenous drug use, a history of an invasive procedure, or close contact with someone who is positive for any situation in this list.
In high-risk patients, a typical screening protocol consists of a nasal swab culture taken more than 2 weeks before surgery. If the results are negative, then no treatment is offered; if positive, then a decolonization protocol is instituted. A follow-up nasal swab culture performed more than 1 week before surgery is useful since decolonization may not be successful in all cases.
Multiple treatment protocols have been proposed for decolonization. Mupirocin ointment (Bactroban Nasal, GlaxoSmithKline) has been shown to be highly effective in eliminating 97% of S. aureus from the nares. The recommended regimen includes the application of 2% mupirocin ointment to the anterior nares 2 or 3 times per day for 5 days. Chlorhexidine (Hibiclens, GC America) is bacteriocidal on contact and can be used as baths to eliminate S. aureus organisms on the skin, particularly in the axilla and perineum. A typical regimen is topical daily wash for 3 to 7 days. Oral antibiotics have limited use in eliminating the carrier state since most have inconsistent ability to penetrate the nasal mucosa. Rifampin, tetracyclines, and trimethoprim–sulfamethoxazole have been used with varying results.
The value of eliminating the MRSA carrier state to decrease the morbidity associated with subsequent infections is of questionable value. A study in the New England Journal of Medicine showed mupirocin did not significantly reduce the rate of surgical site infection from S. aureus. Another study, however, concluded that preoperative mupirocin decreases the incidence of infection in orthopedic surgery. With a paucity of supporting literature, it would seem reasonable to consider decolonization in patients at high risk. A study in 1996 estimated that $16,000 would be saved for each surgical site infection prevented when intranasal mupirocin was used before surgery. Whether it is prudent to screen and treat all patients undergoing ankle replacement is unanswered.
The overall operating room environment does play an important role on the incidence of infection. The considerations include the number and dress of personnel, use of a laminar airflow, and time of procedure. Operating room personnel are the major source of the bacteria. In one study, microbiological counts in an unoccupied operating room increased significantly ( P < .05) when the door was left open to the hallway, and the addition of five operating room personnel further increased the microbiological counts by more than sixfold. Some type of an environmental control, such as laminar airflow or ultraviolet light, is helpful with greater than 90% reduction of airborne bacteria at the wound and 60% reduction of airborne bacteria in the operating room. Laminar air flow results in a statistically significant reduction in airborne bacterial colony-forming units (CFUs), but a decrease in infection rates with statistical significance has not been shown, and no uniform opinion about laminar flow efficacy has developed. Laminar air flow rooms are expensive to install and are not found in many institutions where TAA may be performed. The use of inclusive gowns, such as hooded body exhaust, is also helpful. Several studies have shown a reduced infection rate in orthopedic implant surgeries performed in ultra clean air facilities and with the use of body exhaust suits. Body exhaust suits are portable and relatively inexpensive, making their use more practical. However, all operating room personnel, including anesthesia personnel, circulating nurses, visitors, and the operating room team, must wear inclusive gowns. Face masks and head covers offer no environmental protection. Therefore, to reduce environmental bacterial contamination, the number of personnel in the operating room and movement in and out of the room must be limited.
Prophylactic administration of antibiotics is probably the single most effective method for reducing the prevalence of postoperative wound infection. In the total hip replacement literature, it is reported that without preoperative antibiotics there is a sevenfold increased risk of deep infection. Most clinicians routinely use systemic antibiotic prophylaxis at the time of total joint arthroplasty. The current controversies with regard to such prophylaxis include identification of the optimum antibiotic and the appropriate timing and duration of antimicrobial prophylaxis.
The ideal prophylactic antimicrobial agent would have excellent in vitro activity against staphylococci and streptococci, penetrate tissue well, have a relatively long serum half-life to provide coverage for the duration of the entire operative procedure, be relatively nontoxic, and be inexpensive. Cefazolin, a first-generation cephalosporin that has been studied extensively, has a long serum half-life (1.8 hours), is relatively nontoxic, and is inexpensive compared with other agents; thus, it is a good choice for antimicrobial prophylaxis. For patients who have a type I hypersensitivity reaction to penicillin, vancomycin or clindamycin is a excellent alternative for antimicrobial prophylaxis.
The appropriate time to administer antimicrobials prophylactically is within 30 to 60 minutes before the skin incision is made and at least 5 to 10 minutes before inflation of a tourniquet, to allow proper penetration of the tissues by the antibiotic. Data from surveillance programs designed to monitor hospital acquired infection show that consistent administration of prophylactic antibiotics is sometimes performed outside of the above guidelines. Surgeons should review their hospital’s procedures for the administration of perioperative antibiotics and be sure that a routine inquiry is made as to whether antibiotic coverage was given before incision. This can effectively be done as part of the “time-out” performed by the operating room personnel before incision.
For prolonged procedures exceeding one to two times the half-life of the antibiotic or for procedures associated with extensive blood loss, an additional intraoperative dose of antibiotic is advised. Most authors recommend a single preoperative dose of antibiotics followed by only two or three postoperative doses to reduce the possibility of side effect, prevent antibiotic resistance, and minimize the expense associated with antimicrobial prophylaxis. There is no good literature to suggest that extending the postoperative course of antibiotic treatment beyond 24 hours is beneficial in primary joint replacement.
Remote sites of infection, such as the oral cavity, the genitourinary tract, the pulmonary system, and skin ulcers, increase the risk of postoperative infection. Abrasions or follicular infection at the intended operative site, venous stasis ulcers, skin breaks in the web spaces of the toes, infections under toenails, poor dentition, and infections of the pulmonary system or urinary tract should be identified and eliminated before the operation whenever possible. Removing hair from the operative site with an electric razor rather than a safety razor is another way of reducing operative site infection. Specific to TAA is that the foot is included in the operative field. Studies have shown that interdigital spaces are highly contaminated and standard skin preparation may not eliminate bacterial flora from these areas. Chlorhexidine has been demonstrated to be superior to iodine containing skin prep for the foot. The use of a chorhexidine prep solution along with careful draping to isolate the forefoot from the anterior ankle incision by carefully wrapping the forefoot with Steri-Drape is an intuitive precaution to take during TAA.
Operative sites are invariably contaminated with bacteria to some degree, despite fastidious attempts to minimize the bacteria during the operative procedure. Copious irrigation with a syringe, pulsatile lavage, application of antiseptic agents, and use of antibiotic irrigating solutions are all reasonable methods of decreasing wound decontamination. Pulsatile lavage has the potential to remove as many as 99% of wound contaminants, but high-pressure lavage may damage tissue.
The extensive dissection and long operative time necessary for total joint arthroplasty, combined with the implantation of a large foreign body, create a fertile environment for infection. The length of time for the actual surgery should be reduced because wound contamination occurs first by direct fallout from the environment and second by contaminated equipment and gloved hands that initially were contaminated by the environment. Prolonged operating time influences the rate of deep periprosthetic infection. TAA has a steep learning curve. Experience is key to avoiding complications and reducing operating time. Appropriately placed incisions avoid the need for excessive tissue retraction. Gentle handling of tissue and avoidance of devitalization of tissue are essential. Dissection of subcutaneous tissue from the underlying fascia devitalizes the subcutaneous tissue layer, and awareness of this tissue fragility suggests that retractors should be placed carefully along fascial planes rather than in the subcutaneous tissue plane. Prolonged use of self-retaining retractors may also produce large areas of devitalized tissue. The anterior incision used in TAA seems particularly vulnerable to wound-healing difficulties. Increasing the length of the incision, which reduces the need for forceful retraction, has helped decrease wound healing problems. During retraction, only blunt retractors like Army-Navy retractors are used. The assistants should retract the side where dissection is occurring while avoiding retraction force on the contralateral side. Keeping the skin edges moist during the procedure may lessen wound edge desiccation.
Since the incidence of anterior ankle wound-healing problems is significant, we recommend a layered closure be performed. Controversy exists as to whether to excise or retain the anterior capsule in TAA procedures. Proponents of excision point to its scarred condition and, therefore, it being a possible source of postoperative discomfort. However, if the capsule can be closed, it may act as a barrier to prevent superficial contamination resulting from minor anterior wound dehiscence from spreading deep to the prosthesis, so every effort should be made to close the capsule well. In many TAA cases, we have used platelet-rich gel (Symphony) to coat the prosthesis before implantation. The nonplatelet component can be used during skin closure to possibly enhance healing. Finally, care must be taken during application of the dressing following TAA. There is a natural tendency to apply the dressing circumferentially around the lower leg, ankle, and foot and then dorsiflex the ankle and apply the splint. This must be avoided as it can put pressure on the anterior aspect of the ankle, resulting in skin slough. Any sterile cotton dressing should be applied with the ankle in neutral and then the bandage cut anteriorly and additional noncircumferential strips of cotton dressing applied before application of the splint. In most cases, we prefer to inspect the incisions about 1 week from surgery so that we can proactively manage any wound issues.
Periprosthetic infections can present at various times following surgery. Acute postoperative infections present within 6 weeks after surgery with acute onset of pain, local signs of infection (erythema, cellulites, drainage), and/or systemic signs such as fever, chills, or night sweats. Late infections are most commonly develop after the 6-week period. These late infections can be due to indolent organisms that require prolonged time to produce clinical infection or hematogenously spread from dental, genitourinary, or gastrointestinal procedures. Additionally, infections can be superficial (skin and subcutaneous tissue) or deep (subfascial) periprosthetic infection. Differentiating between superficial and deep infection can be challenging. Generally, any chronic superficial infection strongly suggests deep involvement. The diagnosis of periprosthetic infection requires a high index of clinical suspicion. There is no standard single reliable test, but the diagnosis is based on clinical evaluation, serologic investigations, diagnostic imaging, and microbiological analysis. History and physical examination alone can be used to accurately diagnose infection in 25% of cases. Accurate and early diagnosis is important because a delay in diagnosis may decrease the likelihood of successful eradication of infection and limit treatment options.
Assessment of the white blood cell count is of limited benefit as it is frequently normal. The erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) are key indicators in the patient who has no other reasons for elevation such as rheumatic diseases or other inflammatory conditions. ESR peaks on postoperative days 5 to 7 and returns to baseline usually by 3 months. An ESR higher than 30 mm/hr has been shown to have a sensitivity of 82%, a specificity of 85%, a positive predictive value of 58%, and a negative predictive value of 95%. 41 The CRP is a better indicator of infection as it is more sensitive. It peaks at 48 hours after surgery and should return to normal within the first 3 weeks after operation, compared with the ESR, which can take up to 3 months to 1 year to become normal. A CRP value greater than 1 mg/dL (10 mg/L) has been associated with a 96% sensitivity, a 92% specificity, a 74% positive predictive value, and a 99% negative predictive value. The surgeon should check the CRP reference value for their laboratory as this can vary among laboratories. If both the ESR and CRP are elevated, the probability of infection has been noted to be 83%, and when both are negative, infection may be reliably excluded. In patients with systemic inflammatory disease, where there are other reasons for elevated serologic tests, ESR higher than 30 mm/hr and CRP greater than 1 mg/dL are equivalent in diagnosing periprosthetic infection in revision total knee arthroplasty.
A newer test to measure interleukin 6 (IL-6) is showing promising results. IL-6 is a cytokine produced by monocytes and macrophages, and it induces the production of acute-phase proteins. It peaks during the first 12 hours after surgery, returning to baseline within 3 days. At a threshold level of 12 pg/mL, it has a sensitivity of 100%, specificity of 95%, positive predictive value of 89%, and negative predictive value of 100% in prosthetic infection. Its use is not widespread at this point in time.
Note should be made of the relatively low positive predictive value of serological testing. As will be shown later in this chapter, reoperations on suspected total ankle infections do not carry the same chance of success as would be obtained in a similar total knee arthroplasty case, so great caution must be taken in deciding whether a total ankle is infected.
A growing body of knowledge is defining the role of biofilms in orthopedic hardware infections. Bacterial biofilms are complex organized communities of adhered bacteria entrenched in an extracellular matrix, specifically in an exopolysaccharide, or “slime” layer. Biofilm aggregates may consist of single or multiple species and display many of the hallmarks of multicellular organisms, including small molecule intercellular communication systems, specialized phenotypes, and differentiated metabolism. Confocal laser scanning microscopy has demonstrated the three-dimensional structure of bacterial biofilms, revealing matrix-enclosed cellular towers separated by open water channels that act as a sort of primitive circulatory system for the delivery of nutrients and elimination of waste products.
As a consequence of their quiescent metabolic state, biofilm bacteria are difficult to culture by standard hospital culturing techniques. A study of samples from 22 total joints explanted for presumed septic loosening were evaluated for bacterial presence using standard hospital culturing techniques (5 days), extended culturing time (7 days), and scanning microscopy. The respective positive findings were 9 of 22, 14 of 22, and 19 of 22, indicating that standard hospital culturing techniques, the gold standard used by the orthopedic community for confirming infection, is probably inadequate. The number of bacteria grown by standard culturing techniques, if any, is often severely underestimated.
Many methods of imaging are used in the assessment of periprosthetic infection.
These are of limited value in acute infection because of the absence of reliable diagnostic features. However, they should be still used in evaluation as a vital step in excluding other diagnoses. Chronic infection can cause radiographic changes, including periostitis, osteopenia, endosteal reaction, and rapid progressive loosening or osteolysis. Osteolysis and loosening may have other causes, but the possibility of infection must always be considered when these processes are rapid, particularly when there are no indicators of a mechanical cause. Because most total ankle replacements are implanted for posttraumatic arthritis, many patients have significant preoperative warmth and swelling of their ankle. This can persist to some degree indefinitely postoperatively, making clinical assessment for infection unreliable. Couple this with a significant incidence of normal periprosthetic lucency around TAA components, and the use of plain radiographs to make the diagnosis of deep infection is very unreliable.
Radionuclide Bone Scan
A 99m-technetium ( 99m Tc) isotope bone scan is often performed in the assessment of a failed total hip arthroplasty. Although it has a high sensitivity, the low specificity for infection limits its use. Since it is a measure of bone metabolic activity and remodeling, the technetium bone scan can remain positive for extended periods of time around asymptomatic total ankle components. Its use as an isolated test for deep infection is not recommended. Indium-111–labeled white cell scans have a much higher sensitivity in infection, which has been found to be 77%, with specificity of 86%, a positive predictive value of 54%, and a negative predictive value of 95%. The test is performed by withdrawing blood from the patient, labeling them with 111 In, and reinjecting them back into the patient. However, this test is expensive and time consuming. More important, it can demonstrate slightly increased uptake around prosthetic components where there is high metabolic activity as demonstrated by a positive technetium bone scan even in the face of no infection. Therefore, our assessment generally requires the use of both tests. First, the indium scan is performed and, if negative, then no further testing is performed. If the indium scan is positive, then a technetium bone scan is performed. If the technetium scan is highly positive but the indium scan is only slightly positive (disconcordant scan results), we believe that this demonstrates a low probability that the prosthesis is infected. If both are strongly positive, then we believe that there is a higher likelihood of infection. Other isotopes have been investigated, but none has demonstrated clinically useful sensitivity or specificity. The use of radioactive immunoglobulin G has also been described but has not become common, as its sensitivity and specificity were similar to those of standard laboratory investigations. 99m Tc sulfur colloid imaging has the advantage that osteomyelitis stimulates leukocyte accumulation but inhibits sulfur colloid and may have use in detecting infection.
Positron Emission Tomography
Positron emission tomography using fluorine-19-fluoro-2-deoxy- d -glucose has been used to detect sites of increased metabolic activity, suggestive of infection. The reported sensitivity of this method was 91.7%, with 96.6% specificity. However, areas that showed a nonspecific increase in uptake were seen up to an average of 71 months after surgery, even in uninfected total hip arthroplasty. The authors concluded that the area of increased uptake may be more important than the intensity. Although these results are encouraging, further study is required.
Computed Tomography/Magnetic Resonance Imaging
There does not seem to be much information on the use of computed tomography scanning in diagnosing infected prostheses. Newer magnetic resonance imaging software is making it possible to subtract out the prosthesis from the bone to assess for periprosthetic bone changes that may indicate infection. We have no experience with computed tomography or magnetic resonance imaging in the diagnosis of implant infections.
The organisms most commonly isolated in infected total joint arthroplasty are S. aureus and Staphylococcus epidermidis , followed by gram-negative bacteria. A recent review of the results of culture noted that coagulase-negative staphylococci were increasing in prevalence and that gram-negative infections were decreasing. Vancomycin resistance, in particular with enterococci, has been reported in as many as 23% of enterococcal periprosthetic infections. This emphasizes the importance of identifying the pathogen before initiating antibiotic treatment.
Preoperative Aspiration, Culture, and Sensitivity
When there is a high index of suspicion of periprosthetic infection, the ankle can be aspirated and the culture and sensitivities determined. The patient should not have been treated with antibiotics for a minimum of 2 weeks before aspiration. Such investigations have revealed a sensitivity between 50% and 92% and a specificity ranging from 88% to 97%. Routine total hip and knee aspirations are commonly performed in clinical practice. Following ankle arthroplasty, the potential spaces between components become filled with scar, including the anterior ankle recess. Placing a needle into these potential spaces would be difficult enough but, in practice, these spaces are occupied with soft tissue, so obtaining a meaningful aspiration is usually impractical in suspected TAA cases.
If fluid can be aspirated from an ankle, it can be inspected by Gram stain and analyzed for cell counts and differentials. Gram stain has a sensitivity of only 15% to 30% and is ineffective in diagnosing infection. Criteria used for infected nonreplaced ankles are probably not accurate in diagnosing periprosthetic infections. A normal ankle aspirate with a white blood cell count of 50,000 cells/μL and 75% polymorphonuclear neutrophils is considered diagnostic of infection. These values are too high and have no value in infected total joints. In a study of revision total knee arthroplasties, an aspirate with 2500 cells/μL and 60% polymorphonuclear neutrophils had a sensitivity of 98% and a specificity of 95% for infection. Clinically, a white blood cell count of more than 2000 cells/μL with greater than 70% polymorphonuclear neutrophils is considered very suspicious for deep infection. Patients with suspected deep infection but a negative initial aspiration should be reaspirated as the sensitivity of the test is increased by repeating the procedure.
Molecular Biological Investigation
Polymerase chain reaction (PCR) has been evaluated in the diagnosis of infected joint arthroplasty. PCR is highly sensitive and produces fairly accurate bacterial count within 24 hours. However, the regular PCR is limited by its inability to distinguish between viable and dead bacteria. The reverse transcription (RT)-PCR, which detects messenger RNA, has been found to produce no false-negative results and could detect all viable bacteria in about 5 hours.
Intraoperative Frozen Section
Studies have shown that greater than 5 polymorphonuclear leukocytes per high-power field in more than 5 high-power fields is suggestive of infection, with a sensitivity of 84% and a specificity of 99%. Between 5 and 10 polymorphonuclear leukocytes per high-power field is equivocal and the physician must weigh other clinical factors in making the diagnosis of infection. In cases of acute deep infection with cloudy fluid or frank pus surrounding the prosthesis, we obtain swab cultures and sometimes obtain quantitative tissue cultures. Specificity of 98% is obtained when using the criteria of greater than 10 polymorphonuclear leukocytes per high-power field in more than 5 high-power fields. Less than 5 polymorphonuclear leukocytes per high-power field is considered not indicative of infection. In cases of loose prostheses, where we have a low suspicion of deep infection but are considering a one-stage debridement and reimplantation, we would consider an intraoperative frozen section. Frozen section analysis cannot be performed on bone tissue, so care must be taken to obtain representative soft tissue to submit for analysis. A frozen section can also be used to assess eradication of infection during reimplantation in a two-stage revision. Taking three or more samples has been suggested as having good sensitivity and specificity for infection. If an organism is isolated from three or more specimens, infection can be confirmed in 89% of cases.