Treatment of the Infected Hip Replacement

Abstract

Selection of the suitable treatment method for an infected hip arthroplasty requires careful assessment of patient-related variables and expected treatment goals. The basic treatment options include antibiotic suppression, open debridement, resection arthroplasty, arthrodesis, reimplantation of another prosthesis, and amputation. Successful treatment of infection requires complete debridement of all infected and foreign material and appropriate antimicrobial therapy. When possible, the preferred treatment approach is insertion of another prosthesis with a delayed reconstructive treatment technique. Patients now are presenting with an increasing incidence of resistant organisms and severe bone loss, which increases the difficulty of treatment.

Although surgical techniques and treatment methods have improved significantly, rapid diagnosis and expeditious use of established treatment principles are essential to the successful outcome of an infected hip arthroplasty. Because many aspects of treatment are influenced by the accurate identification of the causitive microorganism, the indiscriminate use of antibiotics that can mask the diagnosis and adversely affect culture techniques, must be discouraged.

Among the various therapeutic approaches available, common objectives have been eradication of infection, alleviation of pain, and restoration of function (Table 1). Selection of the suitable treatment method requires careful assessment of patient-related variables and expected treatment goals (Table 1). The six basic treatment options include: antibiotic suppression, open debridement, resection arthroplasty, arthrodesis, reimplantation of another prosthesis, and amputation (Table 1). With the exception of chronic antibiotic suppression, which does not eliminate infection, the cornerstone treatment principles include thorough surgical debridement combined with appropriate use of antibiotics.

Table 1: Treatment of the Infected Hip Replacement

Traditionally, the starting point in the decision-making process has been to establish and classify the infection according to the onset and duration. The original classification of Coventry¹⁶ of deep infection after hip replacement, which separated the clinical presentation of symptoms into three stages, recently has been expanded to four categories and modified according to current treatment guidelines.^25,74 This system describes four different infection categories: (1) positive intraoperative culture; (2) early postoperative infection; (3) acute hematogenous infection; and (4) late chronic infection. The majority of infections are either Type 2 positive intraoperative culture or Type 4 late chronic infections (Table 2).

Table 2: Classification of Deep Periprosthetic Infection

TREATMENT WITH ANTIBIOTICS ONLY

Antibiotic Suppression

Occasionally, in elderly or frail individuals, suppressive antibiotic therapy may be selected even though the infection will not be eliminated. This method of treatment should be used only when the following criteria are met: (1) prosthesis removal is not feasible (usually because of a medical condition that precludes an operative procedure); (2) the microorganism has low virulence; (3) the microorganism is susceptible to an oral antibiotic; (4) the antibiotic can be tolerated without serious toxicity; and (5) the prosthesis is not loose.^30,75 Combining limited clinical data reveal that antibiotic suppression was successful, defined only as retention of the prosthesis, in nine (31%) of 29 hips.^27,30,36,75

Despite the fact that most patients fail to meet these selection criteria, antibiotic suppression commonly is attempted, a practice that unfortunately prolongs the presence of infection and often complicates subsequent treatment attempts. In particular, in younger, healthy patients this approach is to be condemned because suppressive therapy can convert a relatively localized infectious process into an extensive and recalcitrant condition, making eradication of the infection extremely difficult.

TREATMENT WITH SURGICAL INTERVENTION

General Principles

Many aspects of the surgical debridement are similar regardless of the selected treatment approach. Old incisions are used to expose the hip unless exposure will be compromised. Many incisions are invaginated and it often is advisable to excise the scar and any adjacent sinus tracts so that the skin and subcutaneous tissue layers can be opposed easily for wound closure. The choice of surgical approach generally is based on the surgeon’s philosophy. However, it usually is better to use the same approach used at the index arthroplasty to facilitate complete removal of any nonresorbable sutures and avoid devascularization of the proximal femur. The use of extensile surgical approaches, such as the extended trochanteric osteotomy, now are being used more frequently for removal of distally fixed porous or well-fixed cemented implants.

Antibiotics should be withheld until tissue specimens are obtained from the pseudocapsule and the bone-prosthetic interfaces of both components. Although all necrotic and foreign material should be excised, it is well known that the completeness of debridement is difficult to quantify and to assess. Retained bone cement has been shown to be associated with an increased risk of recurrent, infection after a delayed reconstruction technique.^54,74 The use of a thoracoscope inserted into the femoral canal is extremely helpful for localizing retained cement. Complete removal of intrapelvic cement may require a separate retroperitoneal exposure. If there is difficulty localizing or removing all bone cement, additional tests obtained after debridement, such as CT scanning, occasionally are helpful and then additional debridement can be done later.¹⁵

Traditionally, it was taught that several sequential surgical procedures were necessary to achieve adequate debridement.^25,54 In truth, the majority of infected hips can and should be treated with one definitive surgical debridement unless there is extensive soft tissue necrosis or other extenuating circumstances. Open wound treatment should be avoided because wound contracture often leads to a recalcitrant hip wound⁵⁵ and the wound is at risk for superinfection with multidrug resistant nosocomial organisms. The use of inflow suction irrigation systems is discouraged.

Debridement with Prosthesis Retention

Open debridement occasionally is indicated for acute, fulminant infection in the immediate postoperative period or for late hematogenous infection of a securely fixed and previously functional prosthesis. Suggested criteria for use of this treatment technique include: (1) a short duration of symptoms (< 3 to 4 weeks); (2) organisms susceptible to antibiotics that can be tolerated by the patient; (3) no prosthetic loosening; and (4) absence of excessive scar tissue from prior surgical procedures.^8,17,73 A relative contraindication for debridement and attempted salvage of the prosthesis includes the patient with multiple joint replacements or a prosthetic heart valve to avoid the risk of metachronous infection to these other prosthetic devices.⁵⁷

The reported results of debridement are difficult to assess because of extreme variability in time to treatment, microorganism differences, subsequent antibiotic treatment, quality of debridement, the status of implant fixation, the condition of the surrounding soft tissues, and the criteria for success. In a series of 41 infected THAs treated with surgical debridement, 35 hips were treated in the first month after arthroplasty and six hips were treated for acute hematogenous infection.⁷⁴ Debridement was successful in 26 hips (74%) with early postoperative infection and three hips (50%) with acute hematogenous infection. These results suggest that aggressive surgical debridement is reasonable and relatively successful if stringent patient selection criteria are used.

In a report of 42 infected hip arthroplasties, all with well-fixed components, only six patients (14%) were treated successfully by debridement with prosthesis retention after a mean followup of 6.3 years.¹⁷ Four of 19 patients with an early postoperative infection and two of four patients with an acute hematogenous infection were treated successfully. All 19 patients with a late chronic infection had failure of treatment. Debridement had been done at a mean of 6 days (range, 2–14 days) after the onset of symptoms for patients who were treated successfully. Debridement was done at a mean of 23 days (range, 3–94 days) in those patients for whom treatment had failed.

The window of opportunity for a successful debridement is even less when the offending organism is Staphylococcus aureus because prostheses debrided more than 2 days after onset of symptoms were associated with a higher probability of treatment failure than were those debrided within 2 days of onset (relative risk, 4.2; 95% confidence interval, 1.6–10.3).⁸ Expeditious debridement is essential and usually not successful when done more than 2 weeks after the onset of symptoms, and debridement with prosthesis retention for chronic infection universally fails and therefore should not be attempted.

Arthroscopic irrigation and debridement for late, acute periprosthetic infection in eight hips was successful in all patients at an average of 70 months followup.³⁹ These authors stressed that effective treatment requires early diagnosis, prompt arthroscopic debridement, well-fixed components, a sensitive microorganism, and patient tolerance to and compliance with antibiotic therapy. These patients were selected carefully and all were taking long-term suppressive antibiotic therapy. Until other reports support an arthroscopic approach, open debridement of the infected hip prosthesis should be considered the treatment of choice.

Resection Arthroplasty

Resection arthroplasty of the hip with excision of all foreign material is a highly successful method of eradicating infection.^11,13,31 Although resection arthroplasty often provides pain relief, most patients require use of ambulatory aids, have a Trendelenberg gait, fatigue easily, experience hip instability, and have a large limb-length discrepancy.^{6,31,44,64,71} Implantation of another prosthesis (reimplantation) for the failed, infected prosthesis provides the patient with markedly better functional recovery than a resection arthroplasty.^2,43,71 In patients who are not candidates for reimplantation, a resection arthroplasty should be done as the definitive procedure (Fig. 1). It is important to recognize that the choice of resection arthroplasty does not burn any bridges and if the patient desires and is an acceptable candidate for reimplantation later, an arthroplasty can be done at that time.

FIGURE 1.: An anteroposterior radiograph shows resection (Girdlestone) arthroplasty after removal of an infected hip prosthesis.

Arthrodesis

Unlike the knee, arthrodesis of the hip after the treatment of an infected THA rarely has been advocated.⁴⁷ In a series of 14 patients treated with hip arthrodesis for a failed hip arthroplasty, seven arthroplasties failed because of infection. The average age of the patients with an infected prosthesis was 39 years (range, 24–67 years) and many had high demand occupations. All hips were fused successfully using a modification of the AO technique using a laterally placed cobra plate, an anteriorly contoured Dynamic AO compression plate, and autogenous bone graft. Postoperatively, the patients wore a hip spica cast. Currently, arthrodesis of the hip in this setting rarely is indicated or accepted by the patient.

Amputation (Disarticulation)

Although hip disarticulation occasionally is required to control life-threatening infection, the need for this procedure for the infected hip is exceedingly rare.^9,24 Hip disarticulation was deemed necessary in 11 (1.3%) of 857 infected THAs for: (1) life-threatening infection; (2) severe loss of soft tissue and bone stock; and (3) vascular injury. Combining several series indicates that hip disarticulation occurred in 0.7% of 1682 infected THAs.^27,36

An alternative technique termed the tibia-hindfoot osteomusculocutaneous rotationplasty with calcaneopelvic arthrodesis has been developed in an effort to avoid hip disarticulation.⁶³ This procedure is indicated for patients who otherwise would require hip disarticulation for severe loss of bone stock but have a good limb distal to the knee. After resection of the remaining distal femur, the lower leg is prepared by removing the skin and subcutaneous tissue and disarticulating the foot distal to the calcaneus. The leg then is turned up and placed in the soft tissue envelope of the thigh so that the calcaneus is placed and fused within the acetabular defect. Therefore, the proximal tibia is positioned distally and serves as a weightbearing stump while the tibiotalar joint allows reasonable motion at the level of the previous hip. This procedure has successfully allowed patients to function at a low thigh amputation or knee disarticulation level, which is a marked improvement compared with the function achieved after hip disarticulation.

INSERTION OF ANOTHER PROSTHESIS (REIMPLANTATION)

Reimplantation generally is the most desirable method of treatment for most patients with an infected hip prosthesis.^27,53 The potential for improved functional outcome with another prosthesis must be balanced carefully against the disadvantage of a higher reinfection rate when compared with a definitive resection arthroplasty. Generally accepted contraindications for reimplantation include: persistent or recalcitrant infection, medical conditions that prevent multiple reconstructive procedures, and severe local soft tissue damage or systemic conditions that most likely will predispose toward reinfection.

There are numerous codependent treatment variables that affect the outcome of reimplantation of another prosthesis (Table 3). Many of these variables are codependent and analysis of these variables in many reports is hindered by small patient numbers. To help with these difficulties, a staging system is required to clearly delineate and stratify patients according to the infection type, host, and wound variables so that different treatment approaches and these different treatment variables can be analyzed effectively.

Table 3: Treatment Variables Affecting Outcome of Reimplantation

Antibiotic Therapy

The optimum duration of intravenous antibiotics after removal of the infected prosthesis has not been established definitively and review of available reports reveals the duration of antibiotic therapy to be extremely variable.^{11,14,29,38,77,81} Although general guidelines for the use of intravenous antibiotic therapy have emerged, there is no consensus on the proper use of oral antibiotics. Salvati et al⁶⁹ and Lieberman et al⁵¹ recommended a two-stage reimplantation protocol in selected patients consisting of resection arthroplasty, followed by 6 weeks of intravenous antibiotics in doses sufficient to achieve a postpeak serum bactericidal titer of at least 1:8 where the serum bactericidal titer represents the highest serum dilution killing 99% of the infecting organism. Antibiotic-loaded beads or spacers and antibiotic-loaded cement at reimplantation were not used routinely in their patients. It is possible that when local antibiotic-loaded beads or spacers are used it may be possible to decrease the duration of parenteral antibiotic therapy to 4 weeks rather than 6 weeks. Unfortunately, there are no clinical trials that compare the efficacy of differing time durations of parenteral antibiotic therapy and as such, the 6-week period has become the accepted duration.

There has been interest in the safety and efficacy of long-term treatment with oral antibiotics, particularly in combination with rifampin, for the treatment of infected orthopaedic implants.^{19,20,72,77,81} Some regimens have included a combination of initial intravenous therapy followed by oral antibiotics and have been done with or without initial surgical debridement. The addition of oral rifampin with intravenous antibiotics also seems to be effective in animal models⁴⁰ and in the clinical setting with traditional two-stage treatment protocols.^37,41 Clinical trials that assess the role of oral antibiotics and their role in reducing the duration of intravenous antibiotics clearly are needed.

Microorganism Characteristics

Currently, a higher proportion of patients are infected with multidrug-resistant organisms.^3,35,42,46 Infections caused by methicillin-resistant staphylococci now are commonplace and worrisome organisms such as vancomycin-resistant enterococci now are emerging. The primary problem with these resistant organisms is that the options for available antibiotics, particularly oral agents, are limited and thereby the treatment options of debridement with prosthesis retention or single-stage exchange are restricted severely.^35,42,46

Although many orthopaedic surgeons think that gram-negative infections are more difficult to treat,^{12,17,35,55,65} others have determined no difference in the incidence of recurrent infection with either gram-positive or gram-negative organisms.^{2,26,51,54,59} Two reports indicate that certain organisms, such as Staphylococcus aureus, are more virulent organisms than most gram-negative microorganisms and may produce a higher risk of failure with delayed reimplantation protocols.^7,29 The combination of emergence of resistance patterns in organisms that naturally are more virulent is likely to be an area of increasing concern.

Certain bacteria are capable of forming an exopolysaccharide-glycocalyx or slime often considered to be a biofilm permitting microbial growth while providing a protective shield from antibodies and antibiotics. Although tests are available to determine the production of glycocalyx in laboratory cultures, the association between glycocalyx production and recurrence of infection is not available and therefore these tests are not currently clinically relevant.

Others have suggested that antibiotic resistance is not necessarily related to the ability to produce slime but rather is associated with metabolic characteristics associated with surface colonization and these changes are dependent on the specific type of biomaterial substrate.⁵⁸ Consistently higher MIC levels were required for bacteria adherent to a biomaterial than when isolated in vitro and also were dependent on the type of biomaterial.⁵⁸ This concept of changing antibiotic susceptibility, based on a specific organism combined with a specific biomaterial, has significant bearing on many of the treatment concepts already discussed. For example, in vitro testing methods may not reflect the true antibiotic susceptibility of the infecting organism if retained pieces of bone cement or other foreign objects remain in the wound.

Antibiotic-Loaded Bone Cement

Local antibiotic delivery by addition of gentamicin to acrylic bone cement initially was proposed as a prophylaxis method and subsequently as a therapeutic application.⁹ Elution of multiple antibiotics from different acrylic bone cements has been studied under various conditions.^{18,21,28,45,48,61,62} Antibiotics leach in higher concentrations and for longer periods from Palacos bone cement than from Simplex-P, CMW., and Sulfix acrylic bone cements.⁶¹ The amount of antibiotic elution is highly dependent on the porosity of the bone cement and the concentration of antibiotics in the cement.^4,11,48 The addition of 25% dextran, which increases porosity, facilitates elution of antibiotics.⁴⁸ Combining two antibiotics in bone cement improves elution of both antibiotics and the minimum combination of 2.4 g tobramycin and 1.0 g vancomycin per 40-g packet of bone cement powder has been suggested for proper antibiotic elution.⁶¹

Antibiotic-loaded cement, as a local antibiotic-delivery system, can be used in the form of beads or spacers and when incorporated into bone cement used for prosthesis fixation. A distinct disadvantage with the use of beads about the hip is the extreme difficulty removing these beads after approximately 6 weeks. Large amounts of antibiotic powder (as much as 8 g to 9 g antibiotic per 40-g packet of bone cement) can be added to the cement for beads and spacers but most physicians use only 1 g to 2 g of antibiotic powder per batch of cement when used for prosthesis fixation to avoid weakening of the cement.¹⁸ The use of antibiotic-impregnated cement for prosthesis fixation has been almost universal when using a direct-exchange technique but has been controversial among surgeons favoring delayed reconstruction primarily because the desire to used prostheses that obtain fixation by biologic ingrowth.

INSERTION OF ANOTHER PROSTHESIS (REIMPLANTATION)

Direct-Exchange

One of the primary decisions when inserting another prosthesis is the selection of a direct-exchange technique or a delayed reconstruction. Advocates of the direct-exchange method cite lower morbidity for their patients who do not endure the period with a resection arthroplasty, lower costs because of the absence of a second hospitalization and surgical procedure, and avoidance of the technical difficulty associated with delayed reconstructive techniques.^{10,12,22,56,66–68,70,76} The factors associated with a successful direct exchange include absence of wound complications after the initial THR, good general health of the patient, methicillin-sensitive Staphylococcus epidermidis, Staphylococcus aureus, and Streptococcus species, and an organism that was sensitive to the antibiotic mixed into the bone cement.⁴² Factors associated with failure include polymicrobial infection, gram-negative organisms, especially Pseudomonas species, and certain gram-positive organisms such as methicillin-resistant Staphylococcus epidermidis and Group D Streptococcus.⁴²

Contrary to the generally accepted criteria for direct-exchange,^10,76 Raut et al^66–68 detailed their experience with one-stage revision of infected THA associated with a draining sinus or in the presence of a gram-negative organism. They reported an 86% cure rate in 57 hips with draining sinuses at an average followup of 7.4 years⁶⁷ and a cure rate of 93.4% for patients with gram-negative infections at an average followup of 8 years.⁶⁶ They attribute their success to meticulous surgical technique, preoperative parenteral antibiotics, and antibiotic-loaded cement. These results have yet to reproduced by other investigators.

In an effort to assess the traditional selection criteria for a direct-exchange procedure, the feasibility of published patient selection criteria for direct exchange arthroplasty were assessed in 37 consecutive infected hip arthroplasties.³⁵ These criteria included the requirement of a healthy patient with good soft tissues, minimal femoral bone loss, and an organism identified preoperatively as an antibiotic-sensitive gram-positive organism.⁷⁶ Only four patients (four hips) (11%) were deemed potential candidates for a direct exchange procedure. Exclusions included 14 patients (15 hips) with gram-negative or methicillin-resistant gram-positive organisms obtained from preoperative joint aspirations, 10 patients (10 hips) with moderate or severe femoral bone loss, four patients (four hips) who required a proximal femoral osteotomy for component removal, two patients (two hips) with multiple medical problems, and two patients (two hips) with severely compromised soft tissues. It was concluded that with the increasing emergence of antibiotic-resistant bacteria and an increased prevalence of revision arthroplasties with associated bone loss, the feasibility of direct exchange currently is limited.

If one chooses to do a direct exchange, the use of antibiotic-loaded cement for prosthesis fixation seems to be important. Direct-exchange using plain bone cement not impregnated with antibiotics has been successful in 40 (60%) of 67 infected THAs, whereas success was achieved in 1352 (83%) of 1630 hips treated with the use of antibiotic-impregnated cement.³⁶ Although the use of a direct-exchange technique is more common in Europe than in North America, if this approach seems desirable in an individual patient, careful preoperative selection of patients combined with the use of antibiotic-impregnated cement for prosthesis fixation is recommended. A philosophy of delayed reconstruction for the treatment of the infected hip prosthesis seems most appropriate in the current treatment of patients.

Delayed Reconstruction

Delayed reconstruction techniques allow the physician to observe the patient’s response to therapy and to assess for possible recurrence of infection after antibiotics are discontinued.

Disadvantages of this approach include the hardship experienced by patients with a resection arthroplasty, the attendant costs of a second surgical procedure, and the technical difficulties associated with prosthesis implantation on a delayed basis. Advocates of the delayed reconstructive approach have been interested in the determination of the shortest acceptable time delay between resection arthroplasty and reimplantation to minimize patient hardship, improve functional results, and decrease the difficulty of the revision procedure but also to maintain the lowest possible reinfection rate. This interest has been somewhat diminished with the introduction of the Prosthesis of Antibiotic Loaded Acrylic Cement (PROSTALAC).^32,79,80

PROSTALAC

The PROSTALAC was developed to reduce patient morbidity and to decrease the technical difficulties associated with delayed reconstruction.^79,80 This hip prosthesis facsimile has a thin PE acetabulum and modular stainless steel femoral endoskeleton, which are coated with antibiotic-loaded cement to act as a local antibiotic delivery system and to maintain limb length and anatomic relationships (Fig. 2). Patients are mobilized rapidly, have a shorter hospital stay, and a more comfortable existence as they await reimplantation. In 48 patients having a two-stage revision of an infected hip using a PROSTALAC who were followed up for an average of 43 months, three (7%) had an additional episode of infection.⁸⁰ Two patients had reinfections with different organisms and one patient had a reinfection with the same organism. Finally, treatment of an infection associated with extensive loss of the proximal part of the femur is a challenging problem that particularly is suited for the use of a PROSTALAC.^33,79 The prosthesis acts as an internal splint to maintain the length of the femur, allows flexibility for the interval period, facilitates safer and easier exposure at reimplantation, and allows the potential for allograft reconstruction at reimplantation.

FIGURE 2.: A, Photograph shows the antibiotic-loaded PROSTALAC in mold just before implantation. B, Anteroposterior radiograph shows the PROSTALAC used between removal of infected hip prosthesis and eventual insertion of a new prosthesis.

One of the primary disadvantages of the PROSTALAC is hip instability and this particularly is true primarily for patients with severe proximal femoral bone loss. In the presence of good acetabular bone stock, hip instability can be reduced with a snap-fit articulation between the femoral prosthesis and acetabular PE. In patients with poor acetabular bone stock it is difficult to use these prosthetic designs and hip stability remains a significant issue.

Uncemented Prostheses

Many surgeons prefer the use of uncemented femoral fixation and in some situations the femoral bone stock is not favorable for cement fixation.⁷⁸ A report of delayed reconstruction using uncemented prostheses detailed 34 patients followed up for an average of 4 years and was successful in 28 hips (82.3%).⁵⁹ Antibiotic-impregnated cement beads or spacers were not used in the interval between resection and reimplantation. Combining several series using an uncemented femoral component after use of antibiotic-loaded beads or spacers reveals successful eradication of infection in 160 (92%) of 174 patients.^{23,34,49,50,74} It would seem that the use of antibiotic-loaded beads or spacers during the interval between implant removal and subsequent reimplantation of an uncemented prosthesis is a highly successful technique comparable with implants fixed with antibiotic-loaded bone cement.

Use of Bone Graft

The use of morselized cancellous, small structural fragments, or massive structural bone grafts for reconstruction of a previously infected THA have not been associated with an increased risk of reinfection in several studies.^1,5,34,52,59 The need for these massive femoral allografts for delayed reconstruction for the treatment of the infected THA seems to be increasing. Additional studies are required to determine whether these patients have a higher rate of recurrent infection.

Recurrent Infection After Reimplantation

In 34 patients treated for an infected THA with removal of the prosthesis and implantation of another prosthesis, infection recurred an average of 2.2 years after reimplantation.⁶⁰ This occurrence seldom was compatible with a good functional outcome. Resection arthroplasty was reliable in eradicating reinfection but led to poor function and was associated with persistent pain. Reimplantation of a third prosthesis allowed three patients to achieve an excellent result but the eight hips that failed a third reimplantation attempt had the worst functional results. Those patients, in whom the same single microorganism can be identified from the failed primary THA and from the failed first reimplantation, may be reasonable candidates for another attempt at a two-stage reimplantation of a third prosthesis. This particularly is true when a deficiency in prior antibiotic therapy or surgical technique can be identified.

DISCUSSION

The treatment of the infected hip prosthesis has continued to evolve during the past 4 decades. The careful assessment of the patient with current classification systems has helped clarify the proper treatment approach. In the majority of patients, removal of the prosthesis is required for successful eradication of the infection. The delivery of high dose local antibiotics in the form of spacers or beads has facilitated a successful cure in more than 90% of patients when treated by a two-stage reimplantation protocol.³⁶ In the current era, the emergence of resistant organisms and severity of bone loss encountered in many patients precludes the use of a direct-exchange technique.³⁵ Although in carefully selected situations, a one-stage procedure can be effective, a delayed reconstruction technique is recommended for most patients.⁷⁶

Occasionally in elderly patients with medical problems, suppressive antimicrobial therapy can be attempted, however the selection of this treatment approach should be used sparingly and only as a last resort.⁷⁵ In patients with acute postoperative infection or a late hematogenous infection in an otherwise well functioning arthroplasty, debridement and 4 weeks of intravenous antibiotics occasionally can be effective if the debridement is done as soon as possible.⁷⁴ Retention of the prosthesis is rarely successful if debridement occurs more than 2 weeks after the onset of symptoms.^7,17

Future directions in the treatment of an infected hip prosthesis include enhancement of new technologies for the accurate diagnosis of infection such as use of polymerase chain reaction and immunuglobulin labeled scans. Refinement of staging systems for more accurate stratification of risk factors to help predict patient outcome with various treatment options should help improve the rate of success. Finally, improvement in antibiotics resulting in more oral treatment and less intravenous therapy and technological advances in local antibiotic delivery systems are active areas of investigation that hold great promise.