110 REVIEW PROSTHETIC JOINT INFECTIONS A Role for Prophylaxis WARREN D. BLACKBURN. JR. and GRACIELA S. ALARCON Total joint replacement has been clearly established as one of the major therapeutic advances in rheumatology (1). It is estimated that as many as 200,000 total joint arthroplasties were performed during 1988 in the US alone (2). The joint prostheses most frequently implanted are the hip and the knee (3). A successful operation may enable a patient to continue or to resume daily work and leisure-related activities; however, an unsuccessful operation may result in significant morbidity, with functional impairment and even death (3,4). Infection remains the most serious and the most dreaded complication with a prosthetic joint; infection is responsible for a significant proportion of the cases of prosthetic loosening and failure (2-8). Improvements in surgical techniques and prosthetic design, the perioperative use of antibiotics, and the recognition of the high-risk patient (2-12) have contributed significantly to the reduction in the occurrence of early prosthetic joint infections. Late prosthetic joint infections, their cause, and their prevention continue to be a subject of much discussion among internists, primary care physicians, orthopedic surgeons, rheumatologists, and dentists (13-32). Although general and specific guidelines for the prophylactic use of antibiFrom the Department of Medicine, Division of Clinical Immunology and Rheumatology, The University of Alabama at Birmingham, and the Birmingham Veterans Administration Medical Center, Birmingham, Alabama. Warren D. Blackburn, Jr., MD: Associate Professor of Medicine; Graciela S. Alarcbn, MD, MPH, Professor of Medicine. Address reprint requests to Graciela S . A l a r c h , MD, MPH, 603 MEB, UAB Station, The University of Alabama at Birmingham, Birmingham, AL 35294. Submitted for publication April 12, 1990; accepted in revised form June 15. 1990. Arthritis and Rheumatism, Vol. 34, No. 1 (January 1991) otics in the prevention of late prosthetic joint infection have been proposed (33), such a regimen is far from being accepted as the standard of care. The lack of data to prove the benefit of the prophylactic use of antibiotics in preventing prosthetic joint infection, the cost, the potential risk, and the possible legal implications of accepting such recommendations as the “standard of care” have been cited as reasons not to accept the general prophylactic use of antibiotics. We review here the available literature on the subject, present data on the practice patterns of antibiotic prophylaxis for patients with prosthetic joints among health care professionals in the state of Alabama, and give some general recommendations to the practicing clinician, based on the data gathered from the literature and that generated at our institution. Review of the literature Early prosthetic joint infection. Soon after prosthetic joint replacement began to be performed, it became apparent that early infections were a major problem. Fortunately, infection of the prosthesis during insertion has been greatly reduced during recent years. Initial studies from the 1960s, evaluating total hip replacements, reported infection rates approaching 10% (34,35). Patients with early prosthetic joint infection typically will have persistent joint pain. Fever, drainage, and swelling are present in the minority of patients. A number of studies have demonstrated that the patient’s skin, the airborne bacteria in the operating room, bacteria shed from operating room personnel, and contaminated instruments and fluids can be sources of these early prosthetic joint infections (35- ANTIBIOTIC PROPHYLAXIS AND PROSTHETIC JOINT INFECTION 37). Prolonged duration of surgery, increased numbers of persons in the operating room, and inexperience of the primary surgeon have been associated with higher rates of infection at the time of initial prosthetic joint placement (1 1,38). Previous surgery or prior infection in the joint appear to increase the likelihood of infection of the prosthetic joint. Patients with rheumatoid arthritis (RA) appear to be at higher risk of early prosthetic joint infection than are patients with osteoarthritis. Older patients and patients taking immunomodulatory medications also are at higher risk (1 1,38). Preliminary studies are inconclusive as to whether RA patients who are maintained on methotrexate through surgery have a higher rate of postoperative complications (39,40). Skin infections, local wound infection, skin breakdown, pneumonia, and urinary tract infections occurring perioperatively may lead to seeding of the prosthetic joint. Preoperative screening directed toward the detection and correction of dental and urogenital infection may limit the perioperative development of prosthetic joint infections. In the postoperative period, attention should be directed toward the early detection of the infections that may seed the joint, as noted above. Wound infections occurring in the absence of prosthetic joint infection are generally treated with systemic administration of antibiotics and local drainage. A number of changes in techniques and procedures have been associated with a decrease in the rate of these early infections. Most studies, though, have incorporated more than one change, making it difficult, if not impossible, to attribute the decreased rate of infections solely to a single modification. The perioperative use of antibiotics is one change that has been associated with a decrease in early infections (2,12,41-43). During surgery, tissues become devitalized, hematomas form, and cement polymerization interferes with the body’s natural defenses. Bacteria introduced into these areas may grow relatively unimpeded and remain sheltered from antibiotics that are subsequently administered. Because of these considerations, it is generally agreed that, to be effective, antibiotics should be administered prior to incision of the skin, so that bactericidal levels of antibiotics will be present in the surgical wound. Although antibiotic therapy is often continued for days after surgery, some studies suggest that once surgery is completed, antibiotics can be discontinued without increased rates of infection (4 1). Prolonged antibiotic therapy may promote the emergence of resistant or- 111 ganisms and increase the likelihood of allergic reactions, as well as raise the total cost of the surgery. The perioperative use of antibiotics has been associated with early infection rates of 2% or less (38,4143). First-generation cephalosporins are often recommended, but few studies have compared them with other antibiotics. It would seem reasonable to adjust, if necessary, antibiotic(s) used, based on the sensitivity of the bacteria at the individual center. Other approaches to limiting early prosthetic joint infection have been employed. Ultraclean filtered air, laminar air flow, and ultraviolet lights appear to be effective, according to the findings of some studies (12,42,43). It is not clear, however, that these modalities significantly decrease the rate of prosthetic joint infection when antibiotic prophylaxis is also employed. In a large study of the effects of laminar air flow on the incidence of infection in patients who underwent either total knee or total hip replacement surgery without antibiotic prophylaxis, those in rooms with laminar air flow actually showed an increased rate of early prosthetic joint infection, but only in those who had knee implants (42). It was suggested that when total knee replacements were performed, the operating room personnel were more likely to be upstream from the patients, and shed bacteria would seed the wound. When the operating room personnel wore exhaust-ventilated suits, rates of early prosthetic joint infection of both the hip and the knee were decreased, even in patients who were not treated with antibiotic prophylaxis. Antibiotic-impregnated cement is being used in some centers, with success, in patients with previous prosthetic joint infection (29,33). Concerns regarding prolonged toxicity and efficacy have fueled the controversy regarding this technique (2,ll). These observations indicate that appropriate preoperative screening, referral to an experienced surgical team, and prophylactic use of antibiotics at the time of surgery will result in an acceptably low rate of early prosthetic joint infection. Late prosthetic joint infection. Late prosthetic joint infections are usually due to hematogenous seeding of the joint. Most investigators have arbitrarily distinguished these infections from those introduced during surgery based on the time at which the infection is initially detected. Early in their course, late prosthetic joint infections may be difficult to differentiate from aseptic loosening of the implant. Pain is the predominant symptom. Few patients have systemic symptoms such as fever, and examination of the infected joint may provide few clues to the diagnosis 112 (44). The use of radionuclide scans to distinguish mechanical loosening from prosthetic joint infection has not been generally accepted. Other imaging techniques, including computed tomography and nuclear magnetic resonance, are of limited value or no value in the differential diagnosis of mechanical versus infectious loosening of the implant. Joint aspiration or bacteriologic examination at the time of replacement of the “painful joint” prosthesis may be required to establish the diagnosis. Once seeding of the prosthetic joint has occurred, the likelihood of saving the prosthesis is not very high (38,44). Patients with a longer duration of symptoms prior to the institution of antibiotics, those with inflammatory forms of arthritis such as RA, those taking corticosteroids, and those infected with gramnegative organisms, particularly Pseudomonas, may be more likely to require removal of the prosthesis (38,45). Although a second prosthesis can be inserted after removal of the infected prosthesis, the rate of success with re-implantation is considerably less than with initial implantation. Often, there is a delay in excess of 6 months between the time of removal of the first prosthesis and implantation of a replacement prosthesis (46). Unsuccessful re-implantation, which is not uncommon, may lead to joint fusion, creation of an unstable joint, and occasionally, chronic osteomyelitis. The mortality associated with prosthetic joint infection is not insignificant; rates as high as 20% have been reported by different investigators (46,47). Because of the devastating results associated with these late prosthetic joint infections, a number of health professionals have advocated the prophylactic use of antibiotics in patients with prosthetic joints who undergo procedures which might result in bacteremia (25,48,49). Others, citing the relative avascular nature of bone, the apparent requirement of large numbers of bacteria to seed the joint, and the relative infrequency of hematogenous seeding of prosthetic joints, have suggested that the actual cost and risk of adverse reactions related to the antibiotic outweigh any potential benefits (17,20,27,50). Clearly, to resolve such a dilemma, large, multicenter studies of long duration are required. Unfortunately, such studies have not been performed. Lacking this information, any attempt to determine the role of antibiotic prophylaxis requires addressing a number of considerations. These include the rate at which prosthetic joints become infected, whether certain patients are at higher risk of infection, the outcome of prosthetic joint infection, a clear definition of the BLACKBURN AND ALARCON circumstances associated with bacterial seeding of prosthetic joints, and the efficacy of antibiotic prophylaxis in preventing seeding of the prosthetic joint. These considerations have to be balanced against the cost and risk of antibiotic prophylaxis. The number of patients with joint prostheses who will develop late and, presumably, hematogenous-related infection is largely unknown. It seems reasonable to assume that once a prosthetic joint is implanted, the patient will be at risk of developing an infection in that joint for the rest of his or her life. It is also difficult to determine from most of the reported studies which of the procedures a patient may undergo will result in bacteremia of sufficient magnitude to seed a joint. In one study, in which antibiotic prophylaxis was not used, 1,000 patients with 1,112 prosthetic joints were followed for an average of 6 years; there were 11 late prosthetic joint infections, yielding an annual rate of 0.18% (18). Other investigators have suggested that the annual rate may be at least 3 times as high (0.60%) (51). Does the duration of time that a prosthesis is in place alter the incidence of infection? With time and wear, prosthetic joints tend to loosen. Loosening may be associated with inflammation, provoked by metal joint fragments and increased vascular supply. This may increase the likelihood of bacteremic seeding. There are no studies addressing this concern. Another consideration is whether certain patients are at higher risk for late hematogenous seeding of the prosthetic joint. Patients with RA have evidence of chemotactic and phagocytic dysfunction, are often taking immunomodulatory drugs, and may have more, and possibly longer, episodes of bacteremia. Several studies have demonstrated that both early and late prosthetic joint infection are more likely to occur in patients with RA. The risk of prosthetic joint infection in RA patients has been reported to be I S % , and can possibly be as high as 4.2 times the rate in patients without RA (51). In another study, though of relatively small numbers of patients, all who developed prosthetic joint infection more than 3 years after surgical implantation had RA (38). A third, and critical, consideration is whether the circumstances associated with late prosthetic joint infection can be identified. A number of reports which have included a list of bacteria isolated from the infected joint have shown that an established infection elsewhere in the body may seed the prosthetic joint (10,16,18,13). In cases of established pneumonia, skin infection, or genitourinary infection, it would seem ANTIBIOTIC PROPHYLAXIS AND PROSTHETIC JOINT INFECTION prudent to treat these disorders early and vigorously with antibiotics. Consideration of the bacteria isolated from late infection of the prosthetic joint gives some clues to the possible source. Late infections are often caused by Staphylococcus aureus or Staphylococcus epidermidis (44), with as many as 53% of late infections associated with these organisms (25). Bacterial isolates compatible with a dental source can be found in 6 1 3 % of the cases of late prosthetic joint infection (25,46,47,51). In as many as 25%, a source was not identified. In those situations, prevention would obviously be difficult. Can the prophylactic utilization of antibiotics prevent late prosthetic joint infection? If so, it remains to be determined which antibiotic should be used and for which types of procedures. Little and Jasper have argued that since staphylococci were responsible for most cases of late prosthetic joint infection, a cephalosporin would be the drug of choice for antibiotic prophylaxis (17,20). However, based on the organisms associated with late infection, they also stated that prosthetic joint seeding during “routine” dental work was extraordinarily rare and did not require antibiotic prophylaxis. It would seem that these arguments are mutually exclusive. Can “routine” dental work be separated a priori from more invasive dental work as being likely to seed a joint? Since the exact conditions responsible for late prosthetic joint infection have not been clearly identified, this seems unlikely. Furthermore, there is little information to support the concept that if antibiotic prophylaxis were to be used, then an agent geared toward eradication of organisms not generally found in the oral cavity should be employed. Even less is known about such prophylaxis for elective genitourinary or gastrointestinal tract procedures. Given this dearth of factual information, can any recommendations be made regarding prophylaxis for late prosthetic joint infection? From their analysis concentrating on fatal outcomes, Jacobson et a1 concluded that antibiotic prophylaxis for dental procedures was not justified (27). Their recommendations were based on the assumptions that the risk of hematogenous infection is only 0.7%, that only 6% of the infections are dentally related, and that the mortality associated with such an occurrence is approximately 6.5%. They estimated that for one million dental visits, there would be 4.86 deaths or amputations attributable to infection of the prosthetic joint. These estimates seem exceedingly low. The drug reactions that might occur with prolonged antibiotic use in such patients were not included in their analysis. However, assum- 113 ing that penicillin would prevent only 27% of dental visit-associated prosthetic joint infections, there would be an estimated 4.45 deaths or amputations. They assumed that by using cephalexin, 85% of these infections would be prevented and only 1.21 deaths or amputations would occur. Arguing that 400 and 200 “severe” reactions (not fatal, but yet, not defined by the authors) to penicillin and cephalexin, respectively, could be expected, they concluded that antibiotic prophylaxis for dental treatment was not an effective strategy. Not factored into this analysis was the pain and limitation associated with prosthetic joint infection. It is likely that the authors did not consider the probability that patients with prosthetic joints have previously received antibiotics and may therefore be at less risk of developing an antibiotic-associated fatal reaction. Finally, the rate of occurrence of an antibiotic reaction related to parenteral or prolonged use and the risk of serious reaction may have been overestimated. Norden, who examined the economic cost of late prosthetic joint infection (19), assumed an annual risk of prosthetic joint infections associated with dental procedures to be 0.03%. Considering a cost of $15 for antibiotic prophylaxis and assuming that 1 dental visit per year would require such prophylaxis, the annual cost to 100,000 patients would $15 million. At $30,000 per infection, the cost of therapy for the estimated 30 dental procedure-associated prosthetic joint infections would be $900,000. Calculations can also be performed based on arguments similar to those in the published reports, but with markedly different results. If one assumes that the risk of late prosthetic joint infection is 0.18% annually and that 12% of the infections are related to dental procedures, then 216 dental-associated prosthetic joint infections per million patient-years at risk would be expected. Assuming an 18% mortality rate, there would be 39 deaths. Estimating that therapy for 1 such infection would cost $50,000, the total cost would be $10.8 million. If prophylaxis was employed using 1 gm of cephalexin prior to a procedure and two 500-mg doses afterwards, and assuming that 85% of dental-associated prosthetic joint infections would be prevented, then there would be 32 infections and deaths (6 infection associated, 1 antibiotic associated). If, instead, 1 gm of penicillin V potassium followed by two 500-mg doses were used, and assuming that oral penicillin would provide protection comparable with that of cephalexin, a similar number of infections and infection-related deaths would be observed. BLACKBURN AND ALARCON 114 Table 1. Risk factors for prosthetic joint infections Early infection Late infection ~ Prolonged duration of surgery Number of operating room personnel Inexperienced primary surgeon Advanced age Rheumatoid arthritis Perioperative nonarticular infections ~~ Type of prosthesis Rheumatoid arthritis Nonarticular infections Duration of implant (?) Loosening of implant (?) Based on 1.5 dental visits per patient per year, fatal anaphylaxis (assuming 9 such reactions per 10 million events) might be expected in 1.35 patients. The cost of prophylaxis (medication plus treatment of unprevented prosthetic joint infection) would be $8.95 million dollars, based on costs at our institution ($4.90 per course of prophylaxis with either penicillin V potassium or cephalexin and $50,000 per prosthetic joint infection). The risk factors associated with both early and late prosthetic joint infection are shown in Table 1; the estimated cost of antibiotic prophylaxis is shown in Table 2. Practice patterns among health care professionals In many situations in which there are few data to make clear decisions about therapy, medical practice is often based upon local standards of care. We have recently surveyed rheumatologists, orthopedic surgeons, primary care practitioners, and dentists in the state of Alabama (32). Listings of these practitioners were obtained from the corresponding state professional associations. Since there were fewer than 100 rheumatologists and orthopedic surgeons who practiced in Alabama, all were surveyed. Because of the large number of primary care practitioners and dentists practicing in Alabama, a 10% random sample was believed to be adequate and representative of the entire Alabama membership for these 2 professional groups. A questionnaire including demographic and general practice patterns for these practitioners was developed. The questionnaire and a prepaid return envelope were mailed to the practitioners. Physicians and dentists were asked to give their recommendations for prophylaxis in 4 situations: prior to dental, genitourinary, and gastrointestinal procedures and prior to surgery of infected tissues. Overall, 163 professionals responded to the survey (response rate of 40.5%). One of every 3 practitioners had known or had treated a patient who had a prosthetic joint infection during the year before the survey. These same practitioners were also asked about their practice patterns for the same 4 clinical events in patients with prosthetic heart valves. Although this questionnaire was not formally tested for internal and external reliability, the questions were clearly formulated, and a “yes” or “no” answer was anticipated; thus, it very likely represents the actual practice patterns of these health professionals. The overall reported use of antibiotic prophylaxis for patients with prosthetic joints was 60.5%; the lowest was for gastrointestinal procedures (53.9%), and the highest was for surgery on contaminated tissues (68.1%). These rates also varied among the different practitioners: the lowest among the primary care practitioners (35.6%), and the highest among the orthopedic surgeons (87.8%). In contrast, the same practitioners would use antibiotic prophylaxis for dental procedures in patients with prosthetic heart valves 81-100% of the time, for genitourinary and gastrointestinal procedures 79.3-90% of the time, and for surgery on infected tissues 65.9-100% of the time. When the patterns of prophylactic use of antibiotics for these 4 clinical events were compared between patient groups, the differences were statistically significant (P < 0.001) among all groups of practitioners, except for the orthopedic surgeons, who used antibiotic prophylaxis with a similar frequency in both groups of patients. Because orthopedic surgeons are more likely to be the group who actually performs the joint replacement surgery, they are also more likely to be treating a patient with a prosthetic joint infection, and, thus, report a higher rate of prophylactic use of antibiotics. We also surveyed a random sample of RA Table 2. Estimated risk (per million patient-years) and annual cost (in millions of dollars) of infection or death in patients with prosthetic joints undergoing dental procedures* Antibiotic prophylaxis Risk of infection Risk of death cost ($1 Yes No 32 7 8.95 216 39 10.8 * Estimates are based on the following assumptions: annual risk = 0.18% (or 1,80O/year);dental-associated = 12.0% (or 216/year); cost of therapy for prosthetic joint infection = $50,000; cost of cephalexin or penicillin V potassium = $4.90; protection with antibiotics = 85%; dental visitdyear = 1.5; fatal anaphylaxis = 0.9/million. Costs are 1990 dollars for The University of Alabama Hospital. ANTIBIOTIC PROPHYLAXIS AND PROSTHETIC JOINT INFECTION patients who receive followup care at our institution. A questionnaire pertaining to the use of antibiotics for the same procedures as described in the questionnaire to the health practitioners was used. A trained research assistant conducted the interview by telephone. Of the 80 patients interviewed, only 10% indicated that they had been made aware of any recommendations for the prophylactic use of antibiotics (32). It should also be noted that 8 of these patients had experienced a joint infection. Three of these infections had occurred in prosthetic joints, and in 1 instance, it was clearly associated with a procedure known to be associated with bacteremia. Several other surveys inquiring into orthopedists’ recommendations regarding the prophylactic use of antibiotics have been performed. In one study, with 604 orthopedic surgeons responding (36% response rate), 41% concluded there was an association between bacteremia and prosthetic joint infection; 93% thought that antibiotic prophylaxis was indicated, and most recommended a cephalosporin (20). In still another survey of 177 orthopedic residency programs, from which there were 107 responses (60%), 89% recommended prophylactic use of antibiotics, and as in the other survey, a cephalosporin was favored (48). Even though our survey was conducted in Alabama, and in a strict sense, its results cannot be extrapolated to the entire US population, it nevertheless demonstrates a difference in practice patterns between orthopedic surgeons and other health professionals. The other surveys (20,48), showed that practice patterns were similar among orthopedic surgeons, at least in regard to antibiotic prophylaxis, which suggests that the results of our survey are not likely unique to the state of Alabama. Conclusions and recommendations Based primarily on the review of available information and, to a lesser extent, on our own experience, we give our recommendations for antibiotic usage in patients with prosthetic joints. Whereas some of the recommendations can clearly be supported by published reports, others are based on our interpretation of the literature and our clinical judgment. Patients being considered for joint replacement surgery should undergo preoperative screening for the presence of infection at potential sources of seedingdental, genitourinary, gastrointestinal, or cutaneous. Any such infections should be corrected prior to 115 surgery. Dental hygiene should be emphasized to limit postoperative problems. Medications with the potential to compromise normal host-bacteria interactions should be tapered to the minimal acceptable dosage or, preferably, discontinued prior to surgery. When antibiotics are used perioperatively, the initial doses should be timed so that bactericidal levels are present at the time of surgery. Other strategies (ultraclean air, small-weave surgical garb, etc .) may further reduce the risk of early prosthetic joint infection, but such techniques cannot be uniformly recommended at this time. Bacterial infections occurring in patients with a prosthetic joint should be rapidly diagnosed and appropriately treated, which would include surgery of infected tissue. As a corollary to this, the patient with a joint prosthesis should be informed of the need for this kind of surveillance and aggressive treatment. Patients at high risk of developing an infection in the prosthetic joint or those likely to have a poor outcome associated with such a development (i.e., patients with RA or other chronic debilitating disease, or those taking immunomodulatory drugs) should receive antibiotics before any procedure that is likely to result in bacteremia. The antibiotic to be used for prophylaxis depends on the type of procedure and the patient’s history of drug allergies. The prophylactic use of antibiotics in other patients should be determined by the physician after an analysis of the cost-benefit ratio of such a practice in the physician’s own community. Reasonable estimates of the cost and risk of a hematogenous infection of the prosthetic joint and of antibiotic prophylaxis can be obtained in a manner similar to that described in Table 2, but using local data. Until data from randomized clinical trials are published, these estimates, paired with sound clinical judgment, should guide a physician in dealing with the patient with a prosthetic joint who undergoes procedures that are likely to result in bacteremia. Based on our experience, we routinely use antibiotic prophylaxis for all our rheumatoid arthritis patients with prosthetic joints undergoing such procedures. ACKNOWLEDGMENTS We thank Janet Austin for conducting the patient interviews and Brenda K. Bunn and Donna R. Haffner for their expert secretarial assistance. 116 BLACKBURN AND ALARCON REFERENCES 1. Fries JF: Advancement in the management of rheumatic 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. diseases, 1965 to 1985. Arch Intern Med 149:1002-1011, 1989 Maguire JH: Advances in the control of peri-operative sepsis in total joint replacement. Rheum Dis Clin North Am 14519-535, 1988 Windsor RE, Insall JN: The knee, Textbook of Rheumatology. Third edition. Edited by WN Kelley, ED Harris Jr, S Ruddy, CB Sledge. Philadelphia, WB Saunders, 1989 Charnley J: Low Friction Arthroplasty of the Hip: Theory and Practice. New York, Springer-Verlag, 1979 Poss R, Sledge CB: The hip, Textbook of Rheumatology. Third edition. Edited by WN Kelley, ED Harris Jr, S Ruddy, CB Sledge. Philadelphia, WB Saunders, 1989 Hirshman HP, Schurman DJ: Deep infections following total hip replacement, Current Clinical Topics in Infectious Diseases. Vol. 3. Edited by SJ Remington, NM Swartz. New York, McGraw-Hill, 1980 Thornhill TS, Maguire JH: Infected total knee arthroplasty, Total Knee Revision Arthroplasty. Edited by NW Scott. Orlando, Grune & Stratton, 1987 Hayte RG, Mallory TH: Infections in total joint replacement. Ohio State Med J 82:131-134, 1986 Persson U, Montgomery F, Carlsson A, Lindgren B, Ahnfelt L: How far does prophylaxis against infection in total joint replacement offset its cost? Br Med J 296:99102, 1988 Maniloff G, Greenwald R, Laskin R, Singer C: Delayed postbacteremic prosthetic joint infection. Clin Orthop 223:194197, 1987 Nelson CL: The prevention of infection in total joint replacement surgery. Rev Infect Dis 9:613-618, 1987 Lidwell OM, Elson RA, Lowbury EJ, Whyte W, Blowers R, Stanley SJ, Lowe D: Ultraclean air and antibiotics for prevention of postoperative infection: a multicenter study of 8052 joint replacement operations. Acta Orthop Scand 58:4-13, 1987 Rubin R, Salvati EA, Lewis R: Infected total hip replacement after dental procedures. Oral Surg Oral Med Oral Pathol41:18-23, 1976 Jacobsen PL, Murray W: Prophylactic coverage of dental patients with artificial joints: a retrospective analysis of thirty-three infections in hip prostheses. Oral Surg Oral Med Oral Pathol 50:130-133, 1980 Mulligan R: Late infections in patients with prostheses for total replacement of joints: implications for the dental practitioner. J Am Dent Assoc 101:44-46, 1980 Thomas BJ, Moreland JR, Amstutz HC: Infection after total joint arthroplasty from distal extremity sepsis. Clin Orthop 181:121-125, 1983 Little JW: The need for antibiotic coverage for dental treatment of patients with joint replacements. Oral Surg Oral Med Oral Pathol55:20-23, 1983 18. Ainscow DAP, Denham RA: The risk of hematogenous infection in total joint replacements. J Bone Joint Surg 66B:580-582, 1984 19. Norden CW: Prevention of bone and joint infections. Am J Med 78:229-232, 1985 20. Jasper MT, Little JW: Prophylactic antibiotic coverage in patients with total arthroplasty: current practice. J Am Dent Assoc 11 1:943-948, 1985 21. Jacobson JJ, Matthews LS: Bacteria isolated from late prosthetic joint infections: dental treatment and chemoprophylaxis. Oral Surg Oral Med Oral Pathol 63:122126, 1987 22. Hirschmann JV: Controversies in antimicrobial prophylaxis. Chemioterapia 3:202-207, 1987 23. Ching DWT, Gibson PH, Gould IM, Rennie JAN: Prevention of hematogenous infection in prosthetic joints. Scott Med J 33:363-365, 1988 24. Cutler LS: Evaluation and management of the dental patient with a prosthetic joint replacement. J Conn State Dent Assoc 61:74-75, 1987 25. Cioffi GA, Terezhalmy GT, Taybos GM: Total joint replacement: a consideration for antimicrobial prophylaxis. Oral Surg Oral Med Oral Pathol 66:124-129, 1988 26. Hardie J: Prosthetic joints, antibiotic prophylaxis and dental treatment. Can Dent Assoc J 54:355-357, 1988 27. Jacobson JJ, Schweitzer S, DePorter DJ, Lee JJ: Chemoprophylaxis of dental patients with prosthetic joints: a simulation model. J Dent Educ 525994504, 1988 28. Eskenazi D, Rathburn W: Is there a need for antibiotic prophylaxis in dental patients with prosthetic joints? (editorial). Oral Surg Oral Med Oral Pathol66:430-431, 1988 29. Fitzgerald RH Jr: Problems associated with the infected total hip arthroplasty. Clin Rheum Dis 12537-554, 1986 30. Nelson JP, Fitzgerald RH, Jaspers MT, Little JW: Prophylactic antimicrobial coverage in arthroplasty patients. (editorial). J Bone Joint Surg 72A:1, 1990 31. Sullivan PM, Johnston RC, Kelley SS: Late infection after total hip replacement, caused by an oral organism after dental manipulation. J Bone Joint Surg 72A: 121123, 1990 32. Blackburn Jr WD, Tracy IC, Austin JS, Alarcon GS: Antibiotic prophylaxis, in patients with prosthetic joints: a survey of Alabama practitioners and patients (abstract). Arthritis Rheum 32 (suppl 4):R43, 1989 33. Brause BD: Infections associated with prosthetic joints. Clin Rheum Dis 12523-536, 1986 34. Wilson PD, Amstutz HC, Czerneick A, Salvati EA, Mendez DG: Total hip replacement with fixation by acrylic cement. J Bone Joint Surg 54A:207, 1972 35. Charnley J, Eftakhar N: Postoperative infections in total prosthetic arthroplasty of the hip joint with special ANTIBIOTIC PROPHYLAXIS AND PROSTHETIC JOINT INFECTION 36. 37. 38. 39. 40. 41. 42. 43. reference to bacterial content of the air of the operating room. Br J Surg 56:641-645, 1969 Charnley J: Low friction arthroplasty of the hip joint. J Bone Joint Surg 53B:149, 1971 Charnley J: Postoperative infection after total hip replacement with special reference to air contamination in the operating room. Clin Orthop 87:167, 1972 Poss R, Thornhill TS, Ewald FC, Thomas WH, Batte NJ, Sledge CB: Factors influencing the incidence and outcome of infection following total joint arthroplasty. Clin Orthop 182:117-126, 1984 Bridges S L Jr, Lopez-MendCz A, Tracy I, Alarcon GS: Should methotrexate be discontinued prior to total joint arthroplasty? (abstract). Arthritis Rheum 32 (suppl 4): S43, 1989 Perhala RS, Clough JD, Segal AM: Local infection complications following large joint replacement in rheumatoid arthritis patients treated with methotrexate (abstract). Arthritis Rheum 32 (suppl 4):S129, 1989 Heydemann JS, Nelson CL: Short term preventive antibiotics. Clin Orthop 205:186187, 1986 Salvati EA, Robinson RP, Zen0 SM, Koslin BL, Brause BD, Wilson Jr PD: Infection rates after 3175 total hip and total knee replacements performed with and without a horizontal unidirectional filtered air-flow system. J Bone Joint Surg 64A:525-535, 1982 Lidwell MO, Lowbury EJ, Whyte W, Blowers R, Stanley 44. 45. 46. 47. 48. 49. 50. 51. 117 SJ, Lowe D: Effect of ultraclean air in operating rooms on deep sepsis in the joint after total hip or knee replacement: a randomized study. Br Med J 285: 10-14, 1982 Inman RD, Gallegos KV, Brause BD, Redecha PB, Christian CK: Clinical and microbial features of prosthetic joint infections. Am J Med 77:47-53, 1984 Buchholz HW, Elson RA, Heiner K: Antibiotic loaded acrylic cement: current concepts. Clin Orthop 190:96108, 1984 Hunter G, Dandy D: The natural history of the patient with an infected total hip replacement. J Bone Joint Surg 59Bz293-297, 1977 Fitzgerald RH, Jones DR: Hip implant infection: treatment with resection arthroplasty and late total hip arthroplasty. Am J Med 78:225-232, 1985 Howell RM, Green JG: Prophylactic antibiotic coverage in dentistry: a survey of need for prosthetic joints. Gen Dent 33:320-323, 1985 Stinchfield FE, Biglian LU, Neu HC, Goss TP, Foster CR: Late hematogenous infection of total joint replacement. J Bone Joint Surg 62A:1345-1350, 1980 McGowan DA, Hendrey ML: Is antibiotic prophylaxis required for dental patients with joint replacement? Br Dent J 158:336338, 1985 Maderazo EG, Judson S, Pesternak H: Late infections of total joint prostheses: a review and recommendations for prevention. Clin Orthop 229:13 1-142, 1988 Reminder to Contributors A list of key words is provided in the January and July issues. Please refer to this list when choosing key words for an article. The Notes to Contributors includes the new requirement of subheads to be used in abstracts. Abstracts should now be divided into the following sections: Objective, Methods, Results, and Conclusion. Manuscripts for publication, letters, news, and notes should be sent to: Peter H. Schur, MD, Editor, Arthritis and Rheumatism, Arthritis and Rheumatism Editorial Off ice, Room 422, Richardson Fuller Building, 221 Longwood Avenue, Boston, MA 021 15.