Periprosthetic joint infection (PJI) continues to be a devastating problem in the field of total joint arthroplasty. There are a number of surgical options to decide from and management decisions are based upon the interplay between host, pathogen, and surgeon characteristics. The goal of management is to maximize function, prevent systemic complications, and eradicate infection. Throughout this chapter we will discuss the most relevant recent literature and guiding theories to assist the treating orthopaedic surgeon in the surgical decision-making process.
7 Management of Periprosthetic Joint Infection
Management options of periprosthetic joint infection (PJI) are based on the chronicity of infection (acute versus chronic PJI).
Acute PJI can be managed successfully with debridement, antibiotics, irrigation, and component retention (DAIR).
Appropriate debridement should consist of both mechanical and chemical debridement.
Chronic PJI requires irrigation and debridement, removal of prosthesis, and reimplantation in one or two surgeries.
If one-stage exchange arthroplasty is chosen, patients should be immunocompetent, be infected with a known nonvirulent organism, have a healthy soft tissue envelope and adequate bone stock to accept a prosthesis, and have the medical reserve to tolerate a lengthy procedure.
Periprosthetic joint infection (PJI) is a devastating problem affecting 0.5 to 2% of all hip and knee replacements and continues to be one of the leading causes of revision arthroplasty in the United States. 1 The surgical and nonsurgical management of PJI is complex and depends on the host, surgeon, and disease factors. In this chapter, we will discuss the most recent literature regarding surgical management of PJI with the goal of guiding the treating orthopaedic surgeon to appropriately manage this difficult problem. Traditionally, PJI management entails the use of pathogen directed antimicrobials in combination with a surgical procedure to decrease the bacterial bioburden within the affected joint. However, there are a number of procedures to choose from, including debridement, antibiotics, irrigation, and component retention (DAIR), one-stage exchange arthroplasty, two-stage exchange arthroplasty, resection arthroplasty, fusion, and amputation (▶ Table 7.1). DAIR should be considered in patients with acute onset or late hematogenous PJI with a known organism, a well-fixed prosthesis, and a healthy soft tissue envelope. One-stage exchange arthroplasty should be considered in chronic infections in an immunocompetent host, with a known nonvirulent organism preoperatively, with both a healthy soft tissue envelope and adequate bone stock to accept a prosthesis, as well the medical reserve to tolerate a lengthy procedure. Two-stage exchange arthroplasty should be considered in medically comorbid or immunocompromised patients with polymicrobial, virulent, resistant, or unknown infecting pathogens, or bony or soft tissue compromise. Antibiotics suppression alone can be considered in patients with chronic PJI with well-fixed prosthesis components who are either too sick to tolerate surgery or who have exhausted their reconstructive options and do not wish to proceed with fusion, resection arthroplasty, or amputation. The goal of suppression is to prevent systemic symptoms of their local infection and maximize function. Renal and hepatic functions must be assessed in order to ensure patients can tolerate extended antibiosis. Lastly, fusion, amputation, and resection arthroplasty are the three mainstay salvage options available for patients with persistent or recurrent PJI who are no longer candidates for successful prosthetic reimplantation. Choosing between these options is individualized for each patient and depends on the overall clinical status, the local soft tissue and bony environment in conjunction with patient’s preference. Treating orthopaedic surgeons should have each of these surgical options in their armamentarium and apply them in the appropriately selected patient and clinical context. Throughout this chapter, we will discuss the most relevant literature and guiding theories so that the most appropriate management options can be chosen for each individual patient presenting with PJI.
7.2 Chronicity of Infection
In order for one to understand the varying surgical treatment options for PJI, it is vital for the managing surgeon to understand the appropriate classification of PJI for both total hip arthroplasty (THA) and total knee arthroplasty (TKA). A large majority of current guidelines for PJI diagnosis differentiate PJI based on timing of infectious symptoms relative to prosthesis implantation. The reason duration of symptoms and/or time of symptoms since implantation is the first step in determining the PJI treatment algorithm is because the development of bacterial biofilm on implanted devices is thought to be a time-dependent process. 2 Thus, the overall successful treatment of PJI is thought to be due to the appropriate reduction in biofilm. Thus, it makes sense that international guidelines on the treatment of PJI differentiate infections into two overall categories: acute (early onset) or chronic (delayed or late onset) infections (▶ Table 7.2). 3 , 4 , 5 An early or acute PJI is thought to occur within either 3 weeks or <30 days from implantation, or in the case of late acute hematogenous infection, presentation within 3 weeks of the development of infectious symptoms. 6 Any infection developing thereafter is then considered late or chronic by certain classification systems. 3 The distinction between acute and chronic infection was originally based on the assumption that biofilm develops within 3 weeks on the surface of components, and thus DAIR alone would not suffice in reducing the bioburden of infection. 7 However, this distinction, while still used clinically, may need to be revisited as studies have shown that the development of biofilm occur within hours to days after inoculation. 2 , 8 It is widely accepted that acute infections may be initially managed with DAIR and late or chronic infections require implant resection in either one or two stages before reimplantation of a new prosthesis. 9
7.3 Antibiotic Management Alone
Treatment with antibiotics alone for patients with PJI is a rarely utilized treatment strategy, as suppression of the infection may only limit systemic effects. When antibiotic management is used in isolation, there is little hope for infection eradication and should only be considered after a lengthy discussion regarding the goals of care with the patient. Patients who undergo this treatment strategy are either compromised hosts that cannot tolerate surgery or hosts that have exhausted their surgical options for reimplantation and do not wish to proceed with fusion, amputation, or resection arthroplasty procedures.
Predictably, suppression is more successful in highly sensitive organisms. Antibiotic suppression alone is only indicated when the components are well-fixed without signs of instability, the antibiotic chosen is safe to administer orally for prolonged periods of time, the patient has adequate renal and liver functions, and the patient has the ability to undergo regular testing to monitor for the safety and effectiveness of prolonged antibiotic use. Contraindications to long-term antibiotic suppression are usually radiological signs of loosening or any signs of osteomyelitis. 10 If this is the case, surgical salvage options should be considered that will be discussed later in this chapter.
7.4 Irrigation and Debridement
Irrigation and debridement (I&D) continues to be a primary tool for the treatment of acute PJI. Historically, I&D consisted of three techniques: arthroscopic debridement, open I&D without modular component exchange, and open I&D with the exchange of modular components, more recently termed DAIR. However, studies have shown that I&D either arthroscopic or open without the exchange of modular components leads to an unacceptably high failure rate and leads to worse outcomes in subsequent revision surgeries for persistent infection; thus, DAIR has emerged as the recommended treatment modality for acute onset and late hematogenous PJI. 11 Overall, one can consider DAIR for PJI in the setting of early postoperative infections that occur within 30 days of the index procedure. Additional indications for DAIR include patients with late acute hematogenous PJI that occurred within 3 weeks of an inciting event with less than 3 weeks of presenting symptoms. 12 The success of these procedures is variable and ranges anywhere from 0 to 89%. 13
7.4.1 Irrigation and Debridement without Polyethylene Exchange
Historically, there was a question of whether the removal of modular components (i.e., polyethylene) was necessary when attempting an I&D for appropriately selected PJI. Overall, there is little evidence in the literature specifically addressing the need for modular component exchange. Changing modular components during DAIR incurs added expenses of new components, increased surgical time, and potential increased morbidity. However, the overwhelming dogma in arthroplasty surgery is that removal of polyethylene and other modular components is necessary for a successful debridement to get access to all compartments. 12 , 14 , 15 , 16 With the information stated above, it is unsurprising that arthroscopy has strikingly worse outcomes for PJI than open I&D and should play no role in the treatment algorithm for PJI. 14 , 17 , 18 Overall, I&D without removal of polyethylene should play a minimal role in the overall management of PJI, and modular components should be exchanged whenever possible.
The standard of care for acute onset PJI is an extensive open I&D with the removal and exchange of modular components along with directed antibiotic management. This treatment strategy can be considered in the setting of acute onset PJI (i.e., within 30 days of prosthesis implantation) with a well-fixed prosthesis without the evidence of soft tissue compromise. 3 , 19 Additionally, DAIR is also used in the setting of late hematogenous PJI with the prerequisites of a well-fixed prosthesis without a sinus tract or soft tissue compromise. 3 , 20 , 21 , 22 The removal of modular components allows access to otherwise inaccessible areas of the joint, which is especially true for access to the posterior joint capsule in TKA. 6 , 23 , 24 Additionally, polyethylene removal usually reveals a viscous fluid collection on the tibial tray underneath the polyethylene insert, and removal of the polyethylene allows debridement of this fluid layer and decreased bacterial bioburden.
DAIR with modular component exchange in conjunction with parenteral antibiotics helps to prevent infection recurrence in up to 71% of acute-onset infections. 14 , 25 , 26 Many believe that in order for DAIR with modular component exchange to successfully eradicate infection, the pathogen must be known and appropriately susceptible to oral antimicrobial agents. 3 As expected, success rates are higher in those hosts with less comorbidities, less virulent organisms, and shorter duration of symptoms. 27 Sinus tracts, soft tissue envelope compromise, and loose prostheses are all contraindications for DAIR. 27 Furthermore, even in the setting of acute onset or late hematogenous PJI, if the patient is demonstrating signs of sepsis with hemodynamic compromise, then DAIR should be abandoned and all components should be resected if the patient can medically tolerate the procedure.
DAIR may be slightly delayed in order to appropriately optimize the patient prior to the procedure. All efforts should be made to correct any immediately reversible medical conditions and organ dysfunction, including coagulopathy, anemia, and hyperglycemia. Nutritional status should be checked and nutritional supplementation should be provided, as needed. 27 Once optimized, the patient should be taken into the operating room, and DAIR with polyethylene exchange should be performed in a meticulous manner. In the operating room, the skin is prepped with solutions that combine ethyl alcohol with iodophores (DuraPrepTM) or chlorhexidine gluconate (ChloroPrepTM). These preparations are used as they are resistant to removal and inactivation by blood or irrigant solutions. Additionally, these agents have been shown in several studies to be more effective at reducing skin bacteria counts than traditional iodine “paint.” 28 , 29 , 30 , 31 There have been studies showing that the chlorhexidine formulation was more effective than the iodophore formulation in reducing bacterial counts of the skin in shoulder and ankle surgery. 32 , 33 However, chlorhexidine formulation can erase the surgical site markings and interfere with drape adhesion. 34 After skin preparation, the surgical site should be draped with an iodine-impregnated adhesive drape to prevent bacterial recolonization. 30 Once prepped, the joint should be accessed through the same incision as the primary procedure, even in the presence of surrounding erythema. Scar excision is not routinely implemented. The fascia should be opened to clean out the deep wound space, and it is recommended to take at least 5 representative tissue and fluid samples from the periprosthetic region to help guide antibiotic treatment. 3 The areas should be sampled from the most macroscopically infected appearing region based on surgeon’s decision-making. Areas should include superficial, deep, periprosthetic layers and interfaces between modular components. These samples should be submitted for aerobic and anaerobic cultures. 35 Prophylactic antibiotics do not need to be held in cases of proven preoperative acute PJI. 36 Once cultures are obtained, a thorough debridement ensues, including the removal of necrotic soft tissue, debris, hematoma, or collections of pus from around the prosthesis. At that point, all modular components are removed, and further debridement continued, especially the posterior capsule of the knee. Next, the prosthesis is assessed at both the cement–bone and implant–bone interfaces, and one can proceed with implant retention if the components are well-fixed. Once modular components are removed, then mechanical debridement of all metallic surfaces should ensue. This can be done with a sterile toothbrush, 37 or sterile betadine or chlorhexidine brush. 38 Although newer advances such as devices that perform hydrosurgery (i.e., VERSAJET, Smith and Nephew, Memphis, TN, USA) may be useful, there are no large long-term studies in the arthroplasty literature assessing their efficacy. 39 Once a successful mechanical debridement is undertaken, all modular components are removed, and if the implant is felt to be biomechanically stable, then the surgeon can proceed to copiously irrigate and chemically debride the wound with 9 L of saline solution via low-pressure pulse lavage. 14 , 27 At this stage, most surgeons recommend the addition of a chemical agent to enhance bacterial neutralization. However, the addition of bacitracin to irrigation is not recommended based on recent Centers for Disease Control and Prevention (CDC) and World Health Organization (WHO) guidelines, as there appears to be no added benefit based on the literature but may lead to increased antibiotic resistance. 40 , 41 , 42 Additionally, in vitro studies have shown that bacitracin added no increased bactericidal efficacy when added to irrigation solutions. 43 , 44 , 45 The goal is to dislodge any nonviable tissue while simultaneously diluting the bacterial bioburden. 44 , 46 There have been studies demonstrating that aqueous chlorhexidine gluconate acts as both an antiseptic and detergent, and has a greater ability than dilute povidone-iodine and castile soap to decrease bacterial bioburden in biofilm-forming organisms. 24 , 47 , 48 , 49 However, other solutions such as dilute povidone-iodine and acetic acid (vinegar) have also demonstrated benefit. When compared to five commercially available solutions, dilute povidone-iodine showed the most optimal combination of being bactericidal while maintaining host cell viability. 50 Other clinical studies have demonstrated significant decreases in infection rate with the use of dilute betadine. 51 , 52 The argument against betadine use is that it is not used in a manner that allows it to reach its full bactericidal potential via drying and desiccation and is deactivated by blood, 53 making chlorhexidine theoretically more advantageous. When used as adjuvant chemical debridement in PJI, 3% acetic acid has also been shown to be safe and effective, but larger comparative studies are needed before formal recommendations are made. 54 , 55 Nonetheless, the optimal chemical irrigation solution is still unknown and should be chosen at the surgeon’s discretion.
After I&D, the surgeon should inspect the tissues again, and if the wound appears clean and free of necrotic tissue, new drapes, gloves, gowns, and instruments should be used, 56 and modular components should be trialed and reimplanted. The wound is then closed in layered fashion using nonbraided suture, such as polydiaxonone suture (PDS) and monocryl. Drains are used at the discretion of the treating surgeon. Currently, there are no recommendations for the role of either catheter infused intra-articular antibiotic infusions after DAIR, vancomycin powder, or the use of resorbable impregnated pellets. 27 All of these augmentations have been described in the literature with encouraging results, 57 , 58 , 59 , 60 , 61 but further research with large comparative studies are needed before formal recommendations can be made (Video 7.1).
Overall, the success rates for DAIR vary widely in the literature and range from 16 to 83%. 12 , 14 , 23 , 62 , 63 , 64 , 65 , 66 , 67 , 68 , 69 The largest series of combined THA and TKA PJI showed a success rate of 51.8%, with more recent series showing success rate of 65% at an average of 38-month follow-up. 62 Studies have tried to assess what factors can be used to predict success of DAIR with modular component exchange. Patients who have a high risk of failure of DAIR include those with significantly elevated erythrocyte sedimentation rate (ESR) preoperatively, methicillin-resistant Staphylococcus aureus (MRSA) or methicillin-sensitive Staphylococcus aureus (MSSA) PJI, vancomycin-resistant Enterococcus (VRE), as well as symptom duration greater than 21 days. 23 , 64 , 70
It is recommended that 2 to 6 weeks of pathogen-specific parenteral antibiotics should be used after DAIR, with transition to oral antibiotics thereafter. Recommendations for oral antibiotics after DAIR are 3 months after THA and 6 months after TKA in conjunction with rifampin for biofilm penetration for Staphylococcus infections. 3 Unfortunately, two-stage exchange arthroplasty after DAIR failure leads to worse clinical outcomes. 69 A recent study evaluating 216 cases of DAIR had a failure rate of 57% after 4 years, with nearly 20% 5-year mortality. Of those that failed DAIR, 54.1% went on to two-stage exchange, 11.1% required amputation, and 6.4% underwent a fusion procedure. The other 28.4% of patients that failed initial DAIR required multiple I&Ds for successful eradication. 71 Nonetheless, DAIR remains a useful technique and should be implemented in the appropriate clinical setting.
7.6 One-Stage Exchange Arthroplasty
It is widely accepted that the standard of care for chronic and late onset PJI (i.e., greater than 30 days) often requires explantation of the infected implanted prosthesis to achieve infection eradication. This has been described in either one or two stages. 72 Two-stage exchange arthroplasty is considered by most to be the “gold standard” treatment for chronic PJI, and is the preferred procedure in the United States, with one-stage exchange arthroplasty gaining considerably more popularity in Europe as the first line of treatment for late onset and chronic PJI. 73 The concept of one-stage exchange arthroplasty involves performing two procedures in a single trip to the operating room. The first procedure in both one- and two-stage exchange arthroplasties is an extensive synovectomy and debridement that culminates in explantation of the infected prosthesis. For one-stage exchange arthroplasty, this is followed by prosthesis reimplantation with antibiotic-impregnated cement within the same anesthetic time period. 3 Overall, the appropriate candidate for one-stage exchange is controversial and often debated within the literature. 74 Despite that, there are recent published reports with inclusion and exclusion criteria that can guide treatment decision-making. 75 , 76
Patients can be considered for one-stage exchange if they are immunocompetent hosts without signs of sepsis or hemodynamic compromise. There must be a healthy soft tissue envelope that can be closed postoperatively, and the extent of debridement needed should not compromise soft tissue closure. Postdebridement bone stock needs to be adequate to accept new components. The organism(s) must be known preoperatively and should have low virulence with available antibiotic sensitivities known prior to surgery for organism-directed antibiotic management. Conversely, the medically comorbid, immunocompromised host infected with virulent, resistant, or unidentified organisms, with significant soft tissue and bony compromise, or decompensated septic patients are not candidates for one-stage exchange arthroplasty. 75 , 76 , 77 , 78 Whether or not the presence of sinus tract or fistula communicating with the joint in question is a contraindication to one-stage exchange is controversial. 77 Sinus tracts are considered a relative contraindication to one-stage exchange given the widely held belief that a chronically draining sinus is a poor prognostic sign of PJI eradication. Despite this belief, there are still reports of patients who presented with chronically draining sinuses and were treated with a one-stage exchange with resolution of their infection. 79 , 80 , 81 Lastly, and perhaps most importantly, the patient needs to have the physiologic reserve to undergo a prolonged revision procedure and tolerate general anesthesia, 82 as one-stage exchange arthroplasty can be lengthy and have considerable blood loss.
In the operating room, patients are positioned appropriately with hair already clipped. The skin is then preliminary washed with a sterile 0.5% povidone-iodine or chlorhexidine surgical brush combined with water to remove any dead skin or gross necrotic tissue from the surgical site. The solution is left on the skin for at least 3 minutes for optimal effect. 83 Next, the skin is prepped twice with a preoperative skin preparation containing alcohol (i.e., 2% chlorhexidine gluconate [CHG]/70% isopropyl alcohol [IPA] formulation). Drapes are then placed in standard sterile fashion, and the incision is marked with a sterile marking pen. Every effort should be made to use the same surgical incision as the previous procedure. The prepared skin is then enclosed in antimicrobial incision drapes, with the intention to circumferentially seal the entire extremity. Prophylactic preoperative antibiotics should also be administered based on previous synovial analysis and pathogen sensitivities in conjunction with infectious disease consultation. 84 This can be done prior to obtaining samples, as the organism should already be identified.