INFECTIONS OF PROSTHETIC JOINTS
Total hip and knee replacements are frequent orthopedic procedures, with about 450,000 arthroplasties performed each year in the United States. Infection is the most dreaded complication of joint replacement operations. For most series, the overall infection rate after total hip replacement is 0.5% to 2.0%. The rate of infection after knee replacement is higher, 0.6% to 3.9%. The rate of infection is still higher for certain types of prostheses, such as a metal-hinged knee (11%). Infection is more likely to develop following total joint arthroplasty in patients with rheumatoid arthritis or who have had prior surgery than in those who have underlying osteoarthritis or are undergoing a primary operation. Although the overall infection rate after hip replacement is low (approximately 1%), more than 1,000 infected patients are seen annually in the United States.
Prosthetic joint infections can be classified as early (within the first 3 months after replacement), delayed (within the first 2 years), and late (after 2 years). In the early category, most of the infections occur within the first month after operation. The majority (67%) of prosthetic infections are detected within the first 2 years after replacement, and about one third of patients present with late infections after 2 years. Underreporting of late infections may occur; moreover, the potential for late infections persists indefinitely. Infections that develop within the first 2 years are largely the result of contamination of the prosthetic joint or wound at the time of replacement or of other nosocomial events, such as undetected line-related bacteremias. Infections that occur after 2 years are the result of hematogenous seeding from an infected focus, such as a distant cellulitis, a urinary tract infection, or a transient bacteremia related to a minor dental or surgical procedure. Late infections can also develop from very low-grade infections initiated at the time of replacement or during the perioperative period. The proportion of late infections attributed to a hematogenous source compared with that of infections from other sources is debatable.
Early prosthetic joint infections often result from the contiguous spread of bacteria from a local wound infection to the prosthesis. The patient may have an obviously infected wound early in the postoperative period. Whether the infection is superficial and does not involve the prosthesis or is deep and does affect the joint replacement is often unclear. Surgical exploration of the wound may be required to establish the correct diagnosis. The patient may also have a painful hematoma that proves to be infected after it is aspirated. Frequently, joint pain is the prominent symptom, and the wound may appear entirely normal (no warmth, erythema, tenderness, or inflammation) or only slightly inflamed. Pain usually occurs with active and passive motion of the affected joint. Another clue to deep infection is persistent wound drainage. Fever is usually present.
The clinical presentation of delayed infection is more insidious, key findings being pain on bearing weight and on motion of the joint. Clues that usually indicate infection, such as inflammation, drainage, and fever, are often absent. Late infection is usually characterized by the acute onset of joint pain and fever after a long asymptomatic period following the initial surgery. The history may reveal an earlier distant focus of infection that may have been untreated. The diagnosis of an infected joint is frequently difficult to make and may be established only at reoperation. Pain, usually constant, is the hallmark of joint infection. The pain caused by mechanical loosening of the prosthesis is generally related to motion and weight bearing and is not present at rest.
A number of laboratory studies are used to support the clinical diagnosis of an infected prosthetic joint: WBC count, erythrocyte sedimentation rate (ESR), plain radiographs, sinogram, arthrogram, joint aspiration, technetium bone scan, and gallium and indium 111 scans. The WBC count in patients with infection is variable. An elevated ESR (>20 mm/h) suggests infection if no other explanation for the increase, such as underlying rheumatoid arthritis, can be found. A normal ESR makes infection unlikely (3% to 11% of cases) but does not exclude it entirely. The ESR usually falls to 20 mm/h at 6 months after an uncomplicated joint replacement. Serial plain radiographs are helpful in assessing the presence of infection. Deep infections may produce radiolucencies at the prosthesis-cement-bone interface. These changes of bone resorption may also result from mechanical loosening of the prosthesis. Arthrography and sinography may also provide useful information to establish the diagnosis of infection.
Aspiration of the joint is the most reliable test to diagnose infection. Fluoroscopy ensures proper placement of the needle. A Gram’s stain should be performed, and the aspirate should also be cultured aerobically and anaerobically, as well as fungi and mycobacteria in selected cases. Positive results on technetium bone, gallium, and indium 111 scans suggest infection; however, with mechanical loosening of the prosthesis, results of the gallium scan are normal, and the technetium bone scan result is abnormal.
Another approach to differentiate infection from mechanical loosening of the prosthesis is to obtain five biopsy samples for culture. Bacterial growth in one or two cultures indicates contamination, and growth in all five samples suggests infection. Growth in both solid and broth media indicates infection; growth only in broth medium indicates contamination.
Staphylococci are the most common cause of prosthetic joint infections, accounting for at least 30% to 40% of all infections. Previous studies showed that Staphylococcus aureus is the leading cause of infection in the early, delayed, and late-onset periods. In one report, however, Staphylococcus epidermidis was shown to cause 40% of the infections and S. aureus only 19%. In another report, staphylococci accounted for almost half of the infections, evenly divided between S. aureus and S. epidermidis. S. epidermidis infection usually appears as an indolent infection, causing minimal symptoms and signs. The symptoms and signs of S. aureus infection can vary from fulminant sepsis to an indolent infection. Because S. epidermidis is also a frequent cause of contamination of culture specimens, more than one isolate is needed to confirm this as the etiologic agent.
After staphylococci, gram-negative bacilli, such as Escherichia coli, Proteus species, Enterobacter species, and Pseudomonas aeruqinosa, account for 20% to 30% of infections. Gram-negative bacilli are more common in the early-onset period than in late-onset infections. Genitourinary and gastrointestinal tract procedures or infections are associated with gram-negative prosthetic infections. Other organisms that are implicated less often include streptococci of different groups, anaerobic streptococci, Propionibacterium acnes, and, rarely, fungi or mycobacteria. Anaerobes account for about 16% of cases, and in about 10% of patients, all cultures are negative despite clinical evidence of infection. Cultures can be negative because of prior administration of antimicrobials and, occasionally, inadequate bacteriologic techniques. Polymicrobial infections can also occur.
A number of studies address the problems of treating prosthetic joint infections. If the patient has an early-onset infection, the prosthesis can be left in place if during surgical procedures there is no evidence of loosening of the device. Adequate debridement should be performed, and appropriate antimicrobials, depending on the results of culture and sensitivity testing, must be given parenterally for 6 weeks. Most studies report better results with gram-positive than with gram-negative bacilli.
A loose hip prosthesis can be replaced with a new one even when infection is present. This procedure is limited to gram-positive infections, and a favorable outcome is reported in up to 85% of selected patients. Appropriate antimicrobials should then be continued for 6 weeks. Some authorities favor surgical debridement of the wound, removal of the prosthesis, a subsequent 6-week course of parenteral antimicrobials, and then reimplantation of another joint. If reimplantation is not possible, an excision arthroplasty and prolonged antimicrobial suppression can be attempted. In delayed infections, salvage of the prosthetic hip is rare, and the prosthesis should be removed at the time of the debridement. An attempt can be made to save the hip prosthesis in patients with acute hematogenous joint infection. Some patients (30%) can be cured with debridement and prolonged parenteral antimicrobials. In the majority of these patients, however, either removal and a reimplantation procedure or an excision arthroplasty is required.
The management of prosthetic knee infections is similar to that of hip infections. If the prosthesis is not loose and the etiologic agent is a gram-positive organism, then debridement and parenteral antimicrobials may be adequate. More often, however, the prosthesis must be removed.
Administering antimicrobials to prevent late-onset joint infection in patients who undergo procedures involving dental or genitourinary tract manipulation, which causes a transient bacteremia, is controversial. Most late-onset infections in patients with rheumatoid arthritis result from hematogenous seeding of a joint from established distant foci of infection. Examples of these infections are skin abscesses and urinary tract infections. Prompt treatment of these infections with appropriate antimicrobials prevents hematogenous seeding and subsequent joint infection. The risk for hematogenous seeding of a prosthetic joint from a transient bacteremia, such as that caused by a dental cleaning, appears to be extremely low. In one report, the risk associated with dental procedures was 0.05%, so low that the cost to prevent a single case of joint infection would be high. Legal issues must also be considered in such situations. Because most late-onset infections result from established infections, patients should be treated promptly when infections are present.
An advisory committee of the American Dental Association and American Academy of Orthopaedic Surgeons has formulated guidelines to determine the need for antibiotic prophylaxis to prevent hematogenous prosthetic joint infections in patients undergoing a procedure such as a dental extraction. The panel of experts has suggested that antibiotics are not necessary for patients undergoing a dental procedure if they have a pin, plate, or screw, or for most patients with a total joint replacement. Patients who are at increased risk for hematogenous joint infection and who should receive antibiotic prophylaxis include those who are immunosuppressed (e.g., rheumatoid arthritis or systemic lupus erythematosus), have had a joint replacement within the previous 2 years, have had previous prosthetic joint infections, or have diabetes and are receiving insulin. These patients should receive a single dose of amoxicillin, or a cephalosporin such as cephalexin or cefazolin, or clindamycin if they are penicillin-allergic. In addition to antibiotic prophylaxis to prevent hematogenous seeding of a prosethetic joint, maintenance of good oral health is critical. (N.M.G.)
Ainscow DAP, Denham RA. The risk of hematogenous infection in total joint replacement. J Bone Joint Surg Br 1984;66:580.
Patients with rheumatoid arthritis are at increased risk for development of hematogenous infection of a prosthetic joint.
Anonymous. Advisory statement. Antibiotic prophylaxis for dental patients with total joint replacements. American Dental Association; American Academy of Orthopaedic Surgeons. J Am Dent Assoc 1997;128:1004–1008.
Antibiotic prophylaxis is not indicated for patients with pins, plates, and screws or for most patients with total joint replacements who are undergoing a dental procedure, such as an extraction.
Booth RE Jr, Lotke PA. The results of spacer block technique in revision of infected total knee arthroplasty. Clin Orthop 1989;248:57.
Describes use of a tobramycin-impregnated polymethylmethacrylate spacer block to treat infection locally during the exchange interval.
Bose WJ, et al. Long-term outcome of 42 knees with chronic infection after total knee arthroplasty. Clin Orthop 1995;319:285–296.
Overall success rate was 74%. Use of a two-stage reimplantation procedure was most effective.
Brandt CM, et al. Staphylococcus aureus prosthetic joint infection treated with debridement and prosthesis retention. Clin Infect Dis 1997;24:914–919.
Debridement of an infected prosthetic joint without replacement has a high failure rate (69%).
Brause BD. Infected orthopedic prostheses. In: Bisno AL, Waldvogel FA, eds. Infections associated with indwelling medical devices. Washington, DC: American Society for Microbiology, 1989.
Cherney DL, Amstutz HC. Total hip replacement in the previously septic hip. J Bone Joint Surg Am 1983;65:1256.
The success rate was better for gram-positive prosthetic infections.
Deacon JM, et al. Prophylactic use of antibiotics for procedures after total joint replacement. J Bone Joint Surg Am 1996;78-A:1755–1770.
Prophylactic antibiotics are usually not indicated in patients with a prosthetic joint who are undergoing a procedure associated with transient bacteremia.
Fitzgerald RH, et al. Deep wound sepsis following total hip arthroplasty. J Bone Joint Surg Am 1977;59:847.
Hip infections were classified as acute fulminant infection (occurring within the initial 3 months), delayed sepsis (occurring in the initial 26 months), and late hematogenous infection.
Forster IW, Crawford R. Sedimentation rate in infected and uninfected total hip arthroplasty. Clin Orthop 1982;168:48.
An elevated sedimentation rate suggests infection rather than aseptic loosening of the prosthesis.
Gillespie WJ. Infection in total joint replacement. Infect Dis Clin North Am 1990; 4:465.
Gillespie WJ. Prevention and management of infection after total joint replacement. Clin Infect Dis 1997;25:1310–1317.
Review of management.
Goksan SB, Freeman MAR. One-stage reimplantation for infected total knee arthroplasty. J Bone Joint Surg Br 1992; 74-B:78.
In a small series of infected knee prostheses caused by gram-positive organisms, a one-stage exchange procedure was effective in eradicating the infection.
Goldman RT, Scuderi GR, Insall JN. Two-stage reimplantation for infected total knee replacement. Clin Orthop 1996;331:118–124.
Infected prosethetic knees responded (77%) to an approach with three phases: (a) removal of the prosthesis and debridement, (b) 6 weeks of antibiotics, and (c) reimplantation with a new knee.
Green SA, Ripley MJ. Chronic osteomyelitis in pin tracks. J Bone Joint Surg Am 1984; 66:1092.
Inman RD, et al. Clinical and microbial features of prosthetic joint infection. Am J Med 1984;77:47.
Presenting symptoms included pain (95%), fever (43%), swelling (38%), and drainage (32%).
Kamme C, Linderg L. Aerobic and anaerobic bacteria in deep infections after total hip arthroplasty. Clin Orthop 1981;154:201.
A normal sedimentation rate does not exclude infection.
Lachiewicz PF, Rogers GD, Thomason HC. Aspiration of the hip joint before revision total hip arthroplasty. Clinical and laboratory factors influencing attainment of a positive culture. J Bone Joint Surg Am 1996;78-A;749–754.
In patients with a high sedimentation rate (mean of 80 mm/h), aspiration of the hip joint was helpful in predicting infection (sensitivity of 92%).
McDonald DJ, Fitzgerald RH Jr, Ilstrup DM. Two-stage reconstruction of a total hip arthroplasty because of infection. J Bone Joint Surg Am 1989;71-A:828.
The rate of recurrence was lower for patients who had a reimplantation more than 1 year after the resection arthroplasty. For gram-negative bacilli and enterococci, antimicrobial therapy should be given for at least 28 days.
Mont MA, et al. Multiple irrigation, debridement, and retention of components in infected total knee arthroplasty. J Arthroplasty 1997;12:426–433.
In patients with an infected knee prosthesis less than 30 days after surgery and no radiographic signs of osteitis or a loose prosthesis, debridement and antibiotics produced a high cure rate (83%).
O’Neill DA, Harris WH. Failed total hip replacement: assessment by plain radiographs, arthrograms, and aspiration of the hip joint. J Bone Joint Surg Am 1984; 66:540.
Aspiration of the hip joint is useful to diagnose infection.
Oyen W, et al. Diagnosis of bone, joint, and joint prosthesis infections with In-111-labeled nonspecific human immunoglobulin G scintigraphy. Radiology 1992;182:195.
Use of IgG labeled with indium 111 was valuable in the diagnosis of infections involving prosthetic hips and knees.
Pagnano MW, Trousdale RT, Hanssen AD. Outcome after reinfection following reimplantation hip arthroplasty. Clin Orthop 1997;338:192–204.
Patients in whom infection develops after reimplantation of a new prosthesis are candidates for an attempt at a third prosthesis in which a two-stage approach is used (success rate of 27%).
Pellegrini VD, Jr. Management of the patient with an infected knee arthroplasty. Instr Course Lect 1997;46:215–219.
Review. Factors that predispose patients to infection include use of steroids, rheumatoid disease, prior knee surgery, and the presence of open skin lesions on the affected leg.
Powers KA, et al. Prosthetic joint infections in the elderly. Am J Med 1990;88:5–9N, 1990.
In the elderly, only 61% of prosthetic joint infections responded with removal of the device; none resolved without removal of the prosthesis.
Rand JA, Brown ML. The value of indium 111 leukocyte scanning in the evaluation of painful or infected total knee arthroplasties. Clin Orthop 1990;259:179.
Indium 111 had a sensitivity of 83% and a specificity of 85% in the diagnosis of infected prosthetic knees.
Rushton N, et al. The value of technetium and gallium scanning in assessing pain after total hip replacement. J Bone Joint Surg Br 1982;64:313.
Suggests the use of both technetium bone and gallium scans to diagnose prosthetic joint infection.
Salvati EA, et al. Infection rates after 3,175 total hip and knee replacements performed with and without a horizontal unidirectional filtered air-flow system. J Bone Joint Surg Am 1982;64:525.
Infection rates were 0.9% after total hip replacement and 3.9% after total knee replacement.
Schmalzried TP, et al. Etiology of deep sepsis in total hip arthroplasty. Clin Orthop 1992;280:200.
The incidence of infection after a total hip arthroplasty is 1.5%. The major source for late infections is the genitourinary tract.
Tsukayama DT, Estrada R, Gustilo RB. Infection after total hip arthroplasty. J Bone Joint Surg Am 1996;78:512–523.
Infections were classified into four groups—positive intraoperative culture (at least two specimens positive), early postoperative infection (within 1 month of surgery), late chronic infection, and hematogenous infection (documented or suspected bacteremia). Aerobic gram-positive cocci were isolated in 74%, gram-negative bacilli in 14%, and anaerobes in 8% of cases.
Wilde AH, Ruth JT. Two-stage reimplantation in infected total knee arthroplasty. Clin Orthop 1988;236:23.
When a two-stage reimplantation with an antimicrobial-impregnated cement spacer was used in the treatment of infected knee prostheses, the cure rate was 90%.
Wilson MG, Kelley K, Thornhill TS. Infection as a complication of total knee-replacement arthroplasty. J Bone Joint Surg Am 1990;72:878–883.
Infection occurred in 1.6% of patients after knee replacement arthroplasty. Risk factors associated with infection included underlying rheumatoid arthritis, presence of skin ulcers, and prior surgery.
Wininger DA, Fass RJ. Antibiotic-impregnated cement and beads for orthopedic infections. Antimicrob Agents Chemother 1996;40:2275–2279.
Review. (see also Duncan CP, Masri BA. The role of antibiotic-loaded cement in the treatment of infection after hip replacement. Instr Course Lect 1995;46:305–313).
Zimmerli W, et al. Role of rifampin for treatment of orthopedic implant-related staphylococcal infections: a randomized controlled trial. Foreign-Body Infection (FBI) Study Group. JAMA 1998;279:1537–1541.
In selected patients with staphylococcal infections, rifampin plus ciprofloxacin for 3 to 6 months after an initial debridement was effective without removal of the implant.