* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Review Antimicrobial prophylaxis in orthopaedic surgery: the role of
Survey
Document related concepts
Trichinosis wikipedia , lookup
Traveler's diarrhea wikipedia , lookup
Sarcocystis wikipedia , lookup
Schistosomiasis wikipedia , lookup
Marburg virus disease wikipedia , lookup
Hepatitis C wikipedia , lookup
Dirofilaria immitis wikipedia , lookup
Staphylococcus aureus wikipedia , lookup
Antibiotics wikipedia , lookup
Human cytomegalovirus wikipedia , lookup
Hepatitis B wikipedia , lookup
Anaerobic infection wikipedia , lookup
Clostridium difficile infection wikipedia , lookup
Carbapenem-resistant enterobacteriaceae wikipedia , lookup
Oesophagostomum wikipedia , lookup
Transcript
JAC Journal of Antimicrobial Chemotherapy (1998) 41, 329–340 Review Antimicrobial prophylaxis in orthopaedic surgery: the role of teicoplanin Piero Periti*, Enrico Mini and Giorgio Mosconi Department of Preclinical and Clinical Pharmacology, University of Florence, Italy Orthopaedic joint replacement is generally considered ‘clean’ surgery characterized by a low incidence of infection. In recent years the use of a clean theatre environment, high local concentrations of antibiotic in the cement and systemic antibiotic prophylaxis have been recognized as important measures to reduce infection rates significantly, and this has been supported by clinical trials. Staphylococcus aureus and Staphylococcus epidermidis cause at least half of all orthopaedic surgical infections. Gram-negative bacilli are involved to a much lesser extent (10–30%). First- and second-generation cephalosporins are currently considered by most authors as standard prophylaxis in elective orthopaedic surgery. In the light of the increasing incidence of methicillin resistance in coagulase-positive and -negative staphylococci, it is becoming more important for antibiotics to act efficiently against such organisms if they are to be of value in prophylaxis in orthopaedic surgery. A combined, single-dose of vancomycin/gentamicin has been used successfully in an open, controlled study in patients undergoing total joint arthroplasty but, given the disadvantages associated with the use of vancomycin, teicoplanin may be an alternative choice in such procedures. This review analyses four comparative trials of the efficacy and safety of teicoplanin, two with cefamandole, one with cefuroxime and one with cephazolin, as prophylaxis in orthopaedic total joint replacement surgery. Introduction The replacement of diseased or damaged joints with prostheses is a major accomplishment of modern orthopaedic surgery. Currently, infection represents the most feared and, perhaps, the most serious complication of prosthetic joint surgery.1 Although it occurs in only a small proportion of patients, the large number of such procedures performed annually makes this problem significant, and when infection does occur the outcome can be devastating, resulting in total loss of joint function, amputation and, occasionally, death.2 Consequently, antimicrobial prophylaxis, usually reserved for contaminated orthopaedic surgery, is today also recommended for any clean procedure involving implantation of orthopaedic prostheses to prevent the severe clinical and economic consequences of sepsis.3,4 Infection in an orthopaedic prosthesis has been arbitrarily classified as ‘early’ if it occurs within 3 months of placement. Infection occurring within this time-frame is generally considered to be caused by the ingress of bacteria during the peri-operative period. Infection after 3 months is classified as ‘late’, and usually occurs within the first year.5 Such infections tend to be low grade, to develop slowly and to be difficult to diagnose, and can be the consequence of blood-borne infection, though the cause may also be from the few organisms seeded at the time of operation.6–12 In most orthopaedic surgical procedures the bacterial contamination and eventual infection are usually from an exogenous source; the organisms most commonly causing deep wound infection are Staphylococcus aureus and coagulase-negative staphylococci (particularly Staphylo coccus epidermidis). Other organisms isolated from infected prosthetic joints include aerobic and anaerobic streptococci and a variety of Gram-negative *Corresponding address: Viale G. B. Morgagni 65, 50134 Firenze, Italy. Tel: +39-55-431720; Fax: +39-55-4361613. 329 © 1998 The British Society for Antimicrobial Chemotherapy P. Periti et al. bacilli.1,2,11,13–16 Factors increasing the incidence of infection include the duration of the operation, the length of preoperative hospitalization, rheumatoid arthritis, previous surgery performed on the joint, and the anatomic site. The rate of infection in clean orthopaedic surgery varies according to the procedure performed. Reported rates of deep infection are 1.3–11% in total hip replacement, and 0.7–9% in knee arthroplasty.17 The infection risk in elective orthopaedic surgery involving prosthetic implants is, on average, significantly greater in total knee replacement than in total hip replacement.11 Avoiding prosthetic contamination upon insertion is mandatory, and the use of laminar flow has been shown to be an important preventive measure.18 There is now conclusive evidence that hypersterile operating conditions decrease the number of air-borne organisms, and can significantly reduce the infection rate compared with surgery performed in conventional operating theatres.7,18–20 Additional factors that have led to a decrease in the frequency of infection following total joint arthroplasty in recent years include improved surgical technique and the use of ultraviolet light, antibiotic-impregnated methylmethacrylate and systemic prophylactic antibiotics.21–25 Of the measures most commonly used to reduce the risk of infection in orthopaedic implants (i.e. ultraclean air in vertical unidirectional flow, and antibiotics used systemically or in cement), and those taken to avoid infection common to all surgery, the most powerful procedure is the use of antibiotics.26 These authors found that systemic prophylaxis reduced the infection rate of 3.4% that is found in conventionally ventilated theatres by four-fold, and that ultraclean air systems reduced it by another half. Protective clothing, such as body-exhaust suits, helped further reduce infection by 0.5%. Principles of antimicrobial prophylaxis in orthopaedic surgery Today, arguments centre around the timing of dosage, duration, route of administration and choice of prophylactic antibiotics rather than the need for their use; failure to use antibiotics has been estimated to result in a sevenfold increase in the rate of infection.7 Many studies have now established that timing is critical to the efficacy of surgical antimicrobial chemoprophylaxis; antibiotic treatment must be given before surgery in order to ensure high levels in the tissues at the time of contamination.16,27,28 However, to achieve optimum effects, the antibiotic should not be administered too early in the preoperative period.3,7,17 As in other branches of surgery, providing the kinetics and timing are correct, there is emerging evidence in orthopaedic surgery that single-dose prophylaxis is adequate and that additional peri-operative antibiotics should not be necessary.16,18,28–33 Optimum antiseptic efficacy is achieved when an antimicrobial agent is administered intravenously to provide at equal dose the highest blood concentrations, and thus the maximum and most rapid penetration into the extravascular compartment.34 The interval between intravenous dosing and the start of surgery should be reduced to 20–30 min to allow saturation of bone, joint soft tissue or haematoma.18 The half-life of the drug determines the number of perioperative doses needed to ensure maximum antiseptic efficacy using the minimum amount necessary to provide full antimicrobial cover. A single, pre-operative dose can provide optimum chemoprophylaxis when the half-life is longer than the surgery time.7,16,18,35–38 Tissue penetration of antibiotics active against Gram-positive bacteria In orthopaedic surgery the chemoprophylactic agent must reach the bone and surrounding soft tissues at adequate concentrations during the operation (i.e. at the time of potential bacterial contamination). Although data from clinical studies are scarce, adequate concentrations at the time of surgery should be well in excess of the mean MIC for the bacteria most often causing postoperative infections of the wound, if the outcome is potentially lifethreatening.18,32,37,39 Antibiotics that have been reported to penetrate bone well are clindamycin, gentamicin, flucloxacillin, cephazolin, cefamandole, cefuroxime, cefoxitin, ceftriaxone, ceftazidime, rifampicin, ofloxacin, pefloxacin, vancomycin and teicoplanin.35,40–57 Teicoplanin is widely distributed in a variety of tissues and fluids, except for cerebrospinal fluid. Concentrations in bone are 65% of those in serum, and the peak is reached within 0.5–6 h after rapid, intravenous administration of teicoplanin 400 mg. The results at the end of surgery suggest that levels in excess of the MIC90 of Gram-positive cocci, S. aureus and S. epidermidis as well as enterococci, will be obtained in serum and bone for 3 h.58 On the basis of the favourable pharmacokinetic characteristics described, as well as the antimicrobial properties (i.e. activity against Gram-positive bacteria) and the low toxicity, the first- and second-generation cephalosporins (such as cephazolin, cefamandole and cefuroxime) are considered the prophylactic antibiotics of choice in total hip or knee replacement.7,13,20,28,59–62 In particular, cephazolin has been the most often used and recommended agent in this setting.20,59–63 After an intravenous bolus dose of cephazolin 2 g, bone concentrations are well above the MBC for methicillin-susceptible staphylococci.62 However, there is a real need for alternative antibiotics, such as the fluoroquinolones and glycopeptides (vancomycin and teicoplanin). The pharmacokinetic data and activity of these agents against Gram-positive bacteria, including -lactamase-producing and methicillin- 330 Review resistant staphylococci, qualify them as effective agents to provide valid peri-operative prophylactic cover.46,64–70 Choice of antibiotic As most infections in joint prostheses are caused by staphylococci, antibiotic prophylaxis is generally directed against Gram-positive bacteria. Cephalosporins, semisynthetic penicillins, fluoroquinolones and glycopeptides have the desired spectrum of activity, safety and tissue penetration.13,35,39,62,64,65,67,71 The most widely used antibiotics in total joint replacement procedures are the firstand second-generation cephalosporins by virtue of their excellent activity against the pathogens commonly involved in joint prosthesis infections, and their favourable pharmacokinetics and safety.13,62,63,71 The preferred agents are cephazolin and cefamandole. Multiple doses (pre- and post-intervention) of cephazolin are today considered by most authors as standard prophylaxis during clean surgical procedures, including elective orthopaedic surgery, if methicillin-resistant staphylococci have not been isolated in the nosocomial environment.10,20,22,31,59,61,72–77 Pefloxacin has been the only fluoroquinolone to be studied as a prophylactic agent in hip or knee replacement surgery.78–82 It has been shown to be as effective and welltolerated as comparative cephalosporins (cefamandole, cephazolin, ceftazidime and ceftriaxone).79–82 The inhibitory activity of these drugs is, however, somewhat lower against staphylococci than Gram-negative bacteria, and staphylococcal MICs are generally close to the breakpoint of susceptibility. There is also evidence of increasing staphylococcal resistance, particularly in methicillinresistant strains, during treatment with fluoroquinolones. This has been associated with the extensive use of these antibiotics in the hospital setting.83–86 The glycopeptide antibiotics, teicoplanin and vancomycin, represent valid alternative for prevention of infection following orthopaedic surgery.13,64,71 Considering the growing incidence of methicillin-resistant coagulasepositive and -negative staphylococci in nosocomial infections, it is becoming more important for antibiotics to act efficiently against these bacteria if they are to be of value for prophylaxis in orthopaedic surgery.84,87,88 A combined, single-dose vancomycin/gentamicin regimen has been used successfully in an open, controlled study in patients undergoing unilateral or bilateral total joint arthroplasty.89 Recommendations for the use of vancomycin alone as an alternative peri-operative antimicrobial prophylactic agent in orthopaedic prosthetic surgical procedures have been made.31,32,71,90–92 However, vancomycin is more difficult to administer (by long-term intravenous infusion only) and has a shorter half-life than teicoplanin. Results of a recent meta-analysis of 11 randomized, clinical trials conducted in patients with either febrile neutropenia or severe Gram-positive infections, including those with methicillin-resistant S. aureus, suggest that vancomycin is as effective as teicoplanin, but is tolerated significantly less well.93 Perioperative antibiotic prophylaxis with vancomycin is also characterized by anaphylactoid reactions, such as significant hypotension and the ‘red man’ syndrome.92,94,95 Teicoplanin may be an optimal choice in this setting as it is active against staphylococci, including methicillin-resistant strains.64,96,97 Teicoplanin has a long half-life, low toxicity and good tissue penetration, 64,98 which allows it to achieve therapeutically valid concentrations in bone and the surrounding soft tissue (references 51,57,65, and Garaud, J. J., Cavenaghi, L. & Girat, T., data on file, Merrell Dow Research Institute). The recent recognition of increasing resistance to glycopeptides among Gram-positive bacteria, however,99–101 has caused concern that glycopeptideresistant strains of S. aureus and coagulase-negative staphylococci will assume clinical importance.102 This makes it prudent to limit the use of these agents for shortterm antimicrobial prophylaxis in orthopaedic surgical procedures in institutions in which methicillin-resistant S. aureus or S. epidermidis are important wound pathogens,103 or in patients allergic to -lactam antibiotics. Clinical trials with teicoplanin Studies Four comparative trials of the efficacy and safety of teicoplanin as antimicrobial prophylaxis in orthopaedic surgery have been performed. Two were carried out in the UK, and compared teicoplanin with cefuroxime,104 or cefamandole,72 while the other two were conducted in Italy and compared teicoplanin with cefamandole or cephazolin.73–74,77 The first pivotal trial was an unblinded, randomized study comparing a single dose of teicoplanin against four cefamandole doses over a 24 h period in patients undergoing prosthetic joint surgery (hip or knee arthroplasty) at the Musgrave Park Hospital, Belfast.72 A total of 899 patients was entered into the study; of these 19 did not receive study medication and were not considered further. Patients were assessed for evaluability on an ‘intent-totreat’ (n 880) or ‘per protocol’ (n 765) basis (Table I). Of the 115 protocol violations, most resulted from inappropriate randomization. Patients were randomly allocated to two groups to receive either teicoplanin 400 mg iv at the induction of anaesthesia, or cefamandole 2 g iv at induction and 1 g every 6 h for 18 h after the operation. Most patients received gentamicin bone cement during their operations. The remainder received a cementless artificial joint. A primary failure was defined as a wound infection with proven bacteriology. Infections at all other sites were reported as secondary failures or complications. Efficacy was determined on the basis of post-operative 331 P. Periti et al. wound infection assessed at 8–10 days, 1 month and 6–18 months post-operation. The second pivotal clinical study was a multicentre, comparative, randomized study of a single dose of teicoplanin compared with five doses of cephazolin in patients undergoing hip or knee replacement in 12 centres in Italy. 69,73–74 Patients were randomized to receive either teicoplanin 400 mg iv at the induction of anaesthesia, or five doses of cephazolin 2 g at induction plus four 1 g iv doses spread over 24 h post-operatively. A total of 860 patients was enrolled, and 14 were excluded from data analysis due to pre-operative protocol violations (i.e. positive pre-operative urine culture or antibiotic therapy in the week before surgery). Twenty patients were not evaluable for efficacy as an unjustified antibiotic treatment had been administered post-operatively. Thus, 846 and 826 patients were evaluated for safety and efficacy, respectively (Table I). A primary failure was defined as a wound infection with or without proven bacteriology. Infections at other sites were reported as secondary failures or complications. Early assessments were performed in the post-operative stay and at 3 months, while a long-term follow-up was performed at 12 months. Two further clinical studies were conducted in the UK and Italy, and compared teicoplanin 400 mg iv at the induction of anaesthesia with three doses of cefuroxime and two doses of cefamandole, respectively.77,104 Cefuroxime was administered as three 750 mg iv doses, one at pre-medication, one at induction of anaesthesia and one 8 h post-operatively. Cefamandole 2 g iv was given before surgery and 1 g was given after surgery. In all, 146 patients were enrolled in the UK study (72 receiving teicoplanin and 74 receiving cefuroxime). The Italian trial recruited 520 patients, of whom 250 receiving teicoplanin and 246 receiving cefamandole were evaluable for efficacy (Table I). In both studies, efficacy was assessed by wound scores and wound infections during the post-operative hospital stay. Exclusion criteria common to all four studies were age (14 or 18 years), known allergy to study drugs, active infection or significant bacteriuria at time of operation, antibiotic therapy in the 2 weeks preceding operation, pregnancy or lactation, and renal impairment. The study methods used in the four trials considered here (including the selected patient populations and prophylactic regimens) did not differ substantially. The overall analysis of treatment effects was thus valid. The statistical methods appropriate for analysing data from retrospective randomized studies have been described in detail elsewhere.105,106 All analyses were based on individual data. Treatment effects on failure and adverse event rates were estimated initially within each trial, and then overall. These data were analysed using the Mantel–Haenszel 2 statistic.105 The results were examined in terms of odds ratios and their 95% confidence intervals (CIs). For a first approximation, an odds ratio can be 332 Review thought of as the relative risk of an event occurring in the teicoplanin group compared with the risk of its occurrence in the cephalosporin group. An odds ratio between 0 and 1 indicates a benefit of treatment with teicoplanin compared with cephalosporins, while an odds ratio of 1 indicates a benefit of cephalosporins over teicoplanin. The benefit is statistically significant when the 95% CI of the odds ratio does not include unity. Heterogeneity between trials was evaluated by the 2 test. Calculations were performed using appropriate statistical software (Statistica, Release 5.0, 1997 from Stata Corporation, College Station, TX, USA). Odds ratios and statistical results are reported in detail for primary failure data (Table II). For all other efficacy and safety data comparisons, only statistically significant differences or trends are indicated. Patient population The four studies were comparable with regard to major patient characteristics and risk factors. No clinically relevant differences were seen among the studies or between treatment groups for sex distribution, age, body weight or duration of surgery. Efficacy Wound infection. Table II summarizes the efficacy results, expressed as wound infections, for the four studies of orthopaedic prophylaxis. Early wound infection rates (i.e. those occurring within 3 months of the operation) were comparable for all the studies. In the study by Wall et al.,104 bacteriology was reported for all patients with suspected wound infection. However, the number of wound infections was not evaluated. In the study by Mollan et al.,72 in the ‘intent-to-treat’ analysis, nine early wound infections (1%) occurred, three (0.8%) in the teicoplanin group and six (1.2%) in the cefamandole group. The difference between study treatments was not significant. Similar results were obtained analysing data ‘per protocol’. Details are given in Table II. In the study by Periti et al.,69,73–74 19 early wound infections were recorded (2.3%), with no difference between the two treatment groups; nine (2.2%) in the teicoplanin group and ten (2.4%) in the cephazolin group. In the Suter et al.77 study the only infectious complications were seen in the cefamandole group (Table II). According to these authors, the difference between the groups was not statistically significant (P 0.059 by using Fisher’s exact test). Analysis of these data by the Mantel– Haenszel procedure, however, generated a P value of 0.043, suggesting a significant difference at this significance level. The overall wound infection rate seen in the study by Wall et al. (6.8%) was significantly higher than those obtained in the other three studies (0.8% in the Suter study, 1.0% in the Mollan study and 2.3% in the Periti study).69,72–74,77,104 There was no significant difference between the regimens in the development of wound infection (5.5% for teicoplanin and 8.1% for the comparator cephalosporin cited, i.e. cefuroxime). Data on long-term assessment at 6–18 or 24 months are available for the pivotal studies. In the Mollan study, of the 880 patients enrolled, 6–18-month follow-up reports are available for 842 patients (95.7%, ‘intent-to-treat’ analysis).72 Three late wound infections occurred; one (0.3%) in the teicoplanin group and two (0.4%) in the cefamandole group. In the Periti study, clinical data from 683 patients (79.4%) were available at the 12-month follow-up.69,73–74 One primary failure was seen in each of the teicoplanin (0.3%) and cephazolin (0.3%) groups. No late wound (24-month follow-up) infectious complications were seen in either treatment group in the Suter study (all 496 patients evaluable).77 No long-term assessment data are available for the study by Wall et al.104 The pathogens most commonly encountered were Gram-positive organisms (77.8%, Table III). S. aureus, S. epidermidis and other coagulase-negative staphylococci accounted for 53.3% of total isolates. In the study by Mollan et al.,72 66.7% of pathogens isolated from wound infections were S. aureus and coagulase-negative staphylococci. In the Periti study,69,73,74 15 of 21 wound infections were bacteriologically proven, of which eight (53.3%) were caused by staphylococci. For the supporting studies by Suter et al.77 and Wall et al.,104 staphylococci accounted for 16.7% and 58.3%, respectively, of pathogens isolated from positive wound cultures. Table IV gives a breakdown of overall wound infection rates by the type of joint replacement for the major orthopaedic surgery studies. It has been reported that knee replacement surgery involves a significantly higher risk of infection than hip replacement.11 The Mollan study supports this observation since the overall wound infection rates on an intent-to-treat basis were 3.2% and 0.8% for knee and hip replacement, respectively.72 The results of Periti’s study show an inverse trend, even though the higher wound infection rate seen in hip compared with knee replacement surgery (1.8% and 0.7%, respectively) was not significant.69,73–74 No statistically significant differences were seen between treatment arms in either of these studies, according to type of joint replacement. Secondary infections. The incidence of secondary infections (i.e. infections at sites remote from the wound) shows agreement between treatment in the three studies in which this parameter has been evaluated (Table V).72,74,77 Although the overall incidence of wound infection was the same in the Mollan study, the incidence of urinary tract infection (UTI) was significantly higher in the teicoplanin group, with almost twice as many Gram-negative infections occurring, while the number of lower 333 P. Periti et al. 334 Review Table III. Wound microbiology in the randomized studies on orthopaedic surgery with teicoplanin or cephalosporins72,74,77,104 Species Teicoplanin Cephalosporin Total Gram-positive bacteria Staphylococcus aureus coagulase-negative staphylococci Streptococcus spp. Enterococcus spp. other Gram-positive rods total 4 5 0 0 2 11 (73.3%) 9 6 2 3 4 24 (80%) 13 11 2a 3 6b 35 (77.7%) Gram-negative bacteria Escherichia coli and coliforms Serratia marcescens Enterobacter cloacae Pseudomonas aeruginosa unspecified Gram-negative rods total 2 1 0 0 1 4 (26.7%) Unspecified isolates Negative cultures Total number of isolates 0 3 15 2 0 1 2 0 5 (16.7%) 1 (3.3%) 4 30 4 1 1 2 1 9 (17.8%) 1 (4.4%) 7 45 a One Streptococcus sanguis, and one unnamed -haemolytic streptococcus. Two Clostridium perfringens, one diphtheroid, one Peptococcus sp., one Micrococcus sp. and one unspecified Gram-positive anaerobe. b Table IV. Wound infection rates associated with different types of joint replacement Knee replacement Hip replacement Study teicoplanin cephalosporin teicoplanin cephalosporin Mollan et al.72 Periti et al.74 2/92 (2.2%) 1/72 (1.4%) 5/126 (4.0%) 0/74 (0%) 2/294 (0.7%) 5/338 (1.5%) 3/368 (0.8%) 7/342 (2.0%) respiratory tract infections (LRTIs) was significantly higher in the cefamandole arm.72 In the Periti and Suter studies, Gram-negative organisms were the predominant causative agents of UTI and LRTI in both treatment groups.74,77 However, it is recognized that antimicrobial prophylaxis is only beneficial in the prevention of infections at the surgical site. It has no role in the prevention of secondary post-operative infections, such as LRTI and UTI.107 Safety Tables VI and VII summarize the adverse events seen in the four studies as total adverse events and adverse events related to the study drug, as judged by the investigator. The incidence of adverse events was similar for the two treatment groups within each study, and was generally low. In the study by Periti et al.,74 a higher incidence of adverse events was observed in the cephazolin group (2.1% vs 0.7% in the teicoplanin group), though this difference did not reach statistical significance. Also, of the nine patients with adverse events in the cephazolin group, only two had events related to the study drug. Of 1140 patients receiving teicoplanin who were evaluable for safety, the overall incidence of adverse events was 5.1% (58 patients) of whom 1.0% (11 patients) had events attributed to the study drug. In comparison, 1252 patients received cephalosporins, of whom 6.5% (82 patients) experienced adverse events, with 1.0% (12 patients) being related to the study drug. Analyses of laboratory data in the studies by Mollan et al.72 and Periti et al.74 showed marked changes from baseline for haematological and biochemical parameters.69,72–74 However, these were comparable between the treatment groups and considered to be related to the effects of operation. 335 P. Periti et al. Table V. Summary of incidence of secondary infections for orthopaedic surgery prophylaxis with teicoplanin or cephalosporins in randomized trials Study Drug Mollan et al.72 teicoplanin cefamandole Periti et al.74 teicoplanin cephazolin Suter et al.77 teicoplanin cefamandole a P = 0.0027; bP 0.048; cP Fever Thrombo- Asymptomatic phlebitis bacteriuria UTI LRTI Antibiotic Other therapy infections 1 (0.3%) 4 (0.8%) 36 (8.8%) 41 (9.8%) 72 (27.7%) 71 (27.3%) – – 2 (0.5%) 2 (0.5%) – – 0 5 (1.0%)b 4 (1.0%) 1 (0.2%) 7 (2.7%) 7 (2.7%) 21 (5.4%) 39 (7.9%) 7 (1.7%) 3 (0.7%) – – – – 3 (0.7%) 9 (2.2%)c – – 18 (4.7%) 10 (2.0%)a 4 (1.0%) 2 (0.5%) 12 (4.6%) 13 (5.0%) 1 (0.3%) 0 2 (0.5%) 0 1 (3.8%) 1 (3.8%) 0.086. Table VI. Summary of adverse events in randomized studies for orthopaedic surgery prophylaxis with teicoplanin or cephalosporins Patients with adverse events Patients with adverse events related to the study drug Study Drug n % n % Mollan et al.72 teicoplanin cefamandole teicoplanin cefazolin teicoplanin cefamandole teicoplanin cefuroxime teicoplanin cephalosporins 43 61 13 19 11 12 11 10 58/1140 82/1252 11.1 12.3 10.7 12.1a 10.4 10.8 15.3 13.5 15.1 16.5 14 12 10 12 11 12 16 16 11/1140 12/1252 1 0.4 0 0.5 0.4 0.8 8.3 8.1 1.0 1.0 Periti et al.74 Suter et al.77 Wall et al.104 Total a P 0.083 Table VII. Adverse events possibly related to study drugs Teicoplanin Cephalosporin Study adverse event (n) total adverse event (n) total Mollan et al.72 allergic reaction (2) transient hypotension (1) deep vein thrombosis (1) 0 4 (1%) allergic reactions (2) 2 (0.4%) 0 itch (1) diarrhoea (1) allergic reactions (2) hypotension (1) deep vein thrombosis (4) intestinal obstruction (1) 2 (0.5%) Periti et al.74 Suter et al.77 Wall et al.104 allergic reactions (1) nausea (1) deep vein thrombosis (2) pyrexia (1) suspected pulmonary embolism (1) swollen ankle (1) 1 (0.4%) 6 (8.3%) 336 2 (0.8%) 6 (8.1%) Review Conclusions It can be concluded, on the basis of results of the four randomized studies, that single-dose teicoplanin has efficacy and safety equivalent to multiple-dose cephem antibiotics (cephazolin, cefamandole or cefuroxime) in the prevention of post-prosthetic joint infection. The number of patients required to show a statistical difference is large. For example, to demonstrate a reduction in infection rate from 2% to 1% with a power of 90% at the 95% CI, a study would need over 3000 patients per group. The practicalities of producing a trial in this area with such large patient numbers makes demonstrating such a difference unlikely. The studies presented here included 2406 patients, of whom 2233 were evaluable for efficacy on a ‘per protocol’ and 2348 on an ‘intent-to-treat’ basis. The two major studies together represent 1740 patients, of whom 1591 were evaluable. The infection rates seen in the two pivotal studies were 0.5–1.5% for teicoplanin and 0.4–1.7% for the comparator cephalosporin at the early assessment within 3 months of operation. At the long-term assessments, conducted 6–18 months post-operation, wound infection rates were 0.3% in both studies for teicoplanin and 0.3–0.4% for the comparator cephalosporins. These rates are comparable to published studies of antimicrobial prophylaxis in orthopaedic surgery, where rates of 0.5–4% have been reported following prophylaxis with cephalosporins, such as cephazolin, cefamandole and cephaloridine. Some of these studies used up to 5 day dosing regimens of the prophylactic antibiotic. No significant difference in the incidence of adverse drug events between teicoplanin and the comparator cephalosporins was observed. The microbiological results of the four studies confirm that Gram-positive bacteria are the primary pathogens responsible for wound infection in prosthetic joint surgery, accounting for approximately 78% of isolates. S. aureus and coagulase-negative staphylococci were predominant (about 53%). The results from these studies demonstrate that a single, peri-operative dose of teicoplanin can represent a valid alternative to multiple administration of antimicrobial drugs with a short half-life in preventing infection following total hip and total knee replacement. Teicoplanin has a good spectrum of antimicrobial activity against primary pathogens responsible for wound infection in orthopaedic surgery. In particular, it is highly active against staphylococci, both methicillin-sensitive and methicillin-resistant strains, which are the most common pathogens in prosthetic orthopaedic surgery. Teicoplanin also has excellent tissue penetration and low toxicity. Its elimination half-life is exceptionally long, outlasting the mean operating times in orthopaedic implant surgery, making it suitable for peri-operative prophylaxis. Teicoplanin as a single 400 mg dose by rapid intravenous route at induction of anaesthesia may represent a reasonable choice for use in antimicrobial prophylaxis in all orthopaedic surgical procedure associated with a high risk of infection due to Gram-positive bacteria, particularly methicillin-resistant strains of S. aureus and S. epidermidis. Acknowledgements The authors wish to thank both Ian Harding of Micron Research Ltd (Marmont Priory Farm, Wisbech, UK) for his contribution in providing clinical data from the three studies monitored by Marion Merrell Dow, and Professor Vieri Boddi of the Istituto di Patologia Generale, University of Florence, for his contribution to the statistical analysis of the data. References 1. Young, E. J. & Sugarman, B. (1988). Infections in prosthetic devices. Surgical Clinics of North America 68, 167–80. 2. Young, E. J. (1990). Prevention of infection in prosthetic joints. Current Opinion in Infectious Diseases 3, 666–9. 3. Blackburn, W. D. & Alarcón, G. S. (1991). Prosthetic joint infections. A role for prophylaxis. Arthritis and Rheumatism 34, 110–7. 4. Bengtson, S., Borgquist, L. & Lidgren, L. (1989). Cost analysis of prophylaxis with antibiotics to prevent infected knee arthroplasty. British Medical Journal 299, 719–20. 5. Ching, D. W. T., Gould, I. M., Rennie, J. A. N. & Gibson, P. H. (1989). Prevention of late haematogenous infection in major prosthetic joints. Journal of Antimicrobial Chemotherapy 23, 676–80. 6. Bengtson, S. & Knutson, K. (1991). The infected knee arthroplasty. A 6-year follow-up of 357 cases. Acta Orthopaedica Scandinavica 62, 301–11. 7. Court-Brown, C. M. (1990). Antibiotic prophylaxis in orthopaedic surgery. Scandinavian Journal of Infectious Diseases, Suppl. 70, 74–9. 8. Heath, A. F. (1991). Antimicrobial prophylaxis for arthroplasty and total joint replacement: discussion and review of published clinical trials. Pharmacotherapy 11, 157–63. 9. Inman, R. D., Gallegos, K. V., Brause, B. D, Redecha, P. B. & Christian, C. L. (1984). Clinical and microbial features of prosthetic joint infection. American Journal of Medicine 77, 47–53. 10. Nelson, J. P., Fitzgerald, R. H., Jaspers, M. T. & Little, J. W. (1990). Prophylactic antimicrobial coverage in arthroplastic patients. Journal of Bone and Joint Surgery 72, 1. 11. Taylor, G. J. S., Bannister, G. C. & Calder, S. (1990). Perioperative wound infection in elective orthopaedic surgery. Journal of Hospital Infection 16, 241–7. 12. Wilson, M. G., Kelley, K. & Thornhill, T. S. (1990). Infection as a complication of total knee-replacement arthroplasty. Risk factors and treatment in sixty-seven cases. Journal of Bone and Joint Surgery 72, 878–83. 13. Eron, L. J. (1985). Prevention of infection following orthopaedic surgery. Antibiotics and Chemotherapy 33, 140–64. 337 P. Periti et al. 14. Mollan, R. A. B., Haddock, M. & Webb, C. H. (1992). Teicoplanin vs cephamandole for antimicrobial prophylaxis in prosthetic joint implant surgery (preliminary results). European Journal of Surgery, Suppl. 567, 19–21. 15. Nelson, C. L. (1987). Prevention of sepsis. Clinical Orthopaedics and Related Research 222, 66–72. 16. Periti, P. & Mazzei, T. (1987). Principles of antimicrobial chemoprophylaxis in surgery. Chemioterapia 6, 196–201. 17. Chung, R. & Bivins, B. A. (1991). Antimicrobial prophylaxis in surgery. A synopsis. Infectious Disease Newsletter 10, 1–4. 18. Strachan, C. J. L. (1993). Antibiotic prophylaxis in peripheral vascular and orthopaedic prosthetic surgery. Journal of Antimicrobial Chemotherapy 31, Suppl. B, 65–78. Infectious Diseases, 4th edn (Mandell, G. L., Bennett, J. E. & Dolin, R., Eds), pp. 2742–56. Churchill Livingstone, New York, NY. 32. Martin, C., The French Study Group on Antimicrobial Prophylaxis in Surgery & The French Society of Anaesthesia and Intensive Care. (1994). Antimicrobial prophylaxis in surgery: general concepts and clinical guidelines. Infection Control and Hospital Epidemiology 15, 463–71. 33. Wymenga, A. B., Hekster, Y. A., Theeuwes, A., Muytjens, H. L., Van Horn, J. R. & Slooff, T. J. (1991). Antibiotic use after cefuroxime prophylaxis in hip and knee joint replacement. Clinical Pharmacology and Therapeutics 50, 215–20. 34. Bergamini, T. M. & Polk, H. C. (1989). The importance of tissue antibiotic activity in the prevention of operative wound infection. Journal of Antimicrobial Chemotherapy 23, 301–13. 19. Lidwell, O. M. (1993). Sir John Charnley, Surgeon (1911–82): the control of infection after total joint replacement. Journal of Hospital Infection 23, 5–15. 35. Mader, J. T. & Cierny, G. (1984). The principles of the use of preventive antibiotics. Clinical Orthopaedics 190, 75–82. 20. Norden, C. W. (1991). Antibiotic prophylaxis in orthopedic surgery. Review of Infectious Diseases 13, Suppl. 10, S842–6. 36. Paradisi, F. & Corti, G. (1992). Which prophylactic regimen for which surgical procedure? American Journal of Surgery 164, Suppl. 4A, 2S–5S. 21. Charnley, J. (1972). Post-operative infection after total hip replacement with special reference to air contamination in the operating room. Clinical Orthopaedics 87, 167–87. 22. Hill, C., Flamant, R., Mazas, F. & Evrard, J. (1981). Prophylactic cefazolin versus placebo in total hip replacement. Report of a multicentre double-blind randomized trial. Lancet 224, 795–6. 23. Ivey, F. M., Hicks, C. A., Calhoun, J. H. & Mader, J. T. (1990). Treatment options for infected knee arthroplasties: clinical reports. Review of Infectious Diseases 12, 468–78. 24. Lowell, J. D., Kundsin, R. B., Schwartz, C. M. & Pozin, D. (1980). Ultraviolet radiation and reduction of deep wound infection following hip and knee arthroplasty. Annals of the New York Academy of Science 353, 285–93. 25. Trippel, S. B. (1986). Antibiotic-impregnated cement in total joint arthroplasty. Journal of Bone and Joint Surgery 68, 1297–302. 26. Lidwell, O. M., Lowbury, E. J. L., Whyte, W., Blowers, R., Stanley, S. J & Lowe, D. (1984). Infection and sepsis after operations for total hip or knee-joint replacement: influence of ultraclean air, prophylactic antibiotics and other factors. Journal of Hygiene 93, 505–29. 27. DiPiro, J. T., Record, K. E., Schanzenbach, K. S. & Bivins, B. A. (1981). Antimicrobial prophylaxis in surgery: part 2. American Journal of Hospital Pharmacy 38, 487–94. 28. Ritter, M. A., Campbell, E., Keating, E. M. & Faris, P. M. (1989). Comparison of intraoperative versus at 24 hour antibiotic prophylaxis in total joint replacement. A controlled prospective study. Orthopedic Review 18, 694–6. 29. Classen, D. C., Evans, R. S., Pestotnik, S. L., Horn, S. D., Menlove, R. L. & Burke, J. P. (1992). The timing of prophylactic administration of antibiotics and the risk of surgical wound infection. New England Journal of Medicine 326, 281–6. 30. Gatell, J. M., Garcia, S., Lozano, L., Soriano, E., Ramon, R. & Garcia SanMiguel, J. (1987). Perioperative cephamandole prophylaxis against infections. Journal of Bone and Joint Surgery 69, 1189–93. 31. Kernodle, D. S. & Kaiser, A. B. (1995). Postoperative infections and antimicrobial prophylaxis. In Principles and Practice of 37. Reeves, D. S. & Lewis, D. A. (1993). Pharmacokinetic aspects of antibacterial prophylaxis. Journal of Antimicrobial Chemotherapy 31, Suppl. B, 11–21. 38. Sanderson, P. J. (1993). Antimicrobial prophylaxis in surgery: microbiological factors. Journal of Antimicrobial Chemotherapy 31, Suppl. B, 1–9. 39. Cunha, B. A. (1984). The use of penicillins in orthopaedic surgery. Clinical Orthopaedics and Related Research 190, 36–49. 40. Bressolle, F., Gonçalves, F., Gouby, A. & Galtier, M. (1994). Pefloxacin clinical pharmacokinetics. Clinical Pharmacokinetics 27, 418–46. 41. Coignard, S., Renard, C. & Lortat-Jacob, A. (1986). Diffusion de la pefloxacine dans le tissu osseux humain. Medicine et Maladies Infectieuses 16, 471–4. 42. Dalhoff, A. (1989). A review of quinolone tissue pharmacokinetics. In Quinolones (Fernandes, P. H., Ed.), pp. 277–312. JR Prous Science Publishers, Barcelona. 43. Dellamonica, P., Bernard, E., Etesse, H. & Garraffo, R. (1986). The diffusion of pefloxacin into bone and the treatment of osteomyelitis. Journal of Antimicrobial Chemotherapy 17, Suppl. B, 93–102. 44. Etesse-Carsenti, H., Garraffo, R., Giaume, F., Barbarin, A., Leveaux, G., Bernard, P. et al. (1989). Diffusion kinetics of pefloxacin in human bone following a single infusion. Reviews of Infectious Diseases 11, Suppl. 5, 1265–6. 45. Etesse-Carsenti, H., Giaume, F., Barbarin, A., Argeson, C. & Mondain, V. (1991). Kinetics of bone diffusion of ofloxacin after a single six milligram/kilogram dose infusion. European Journal of Clinical Microbiology and Infectious Diseases, Special Issue, 258–9. 46. Graziani, A. L., Lawson, L. A., Gibson, G. A., Steinberg, M. A. & MacGregor, R. R. (1988). Vancomycin concentrations in infected and non infected human bone. Antimicrobial Agents and Chemotherapy 32, 1320–2. 47. Kondell, P. A., Nord, C. E. & Nordenram, A. (1982). Concentrations of cloxacillin, dicloxacillin and flucloxacillin in dental alveolar serum and mandibular bone. International Journal of Oral Surgery 11, 40–3. 338 Review 48. Leigh, D. A., Marriner, J., Nisbet, D., Powell, H. D. W., Church, J. C. T. & Wise, K. (1982). Bone concentrations of cefuroxime and cephamandole in the femoral head in 96 patients undergoing total hip replacement surgery. Journal of Antimicrobial Chemotherapy 9, 303–11. 49. Leigh, D. A. (1989). Determination of serum and bone concentrations of cephradine and cefuroxime by HPLC in patients undergoing hip and knee joint replacement surgery. Journal of Antimicrobial Chemotherapy 23, 877–83. 50. Roncoroni, A. J., Manuel, C., Nedjar, C., Bauchet, J. & Mariani, D. (1981). Cefamandole bone diffusion in patients undergoing total hip replacement. Chemotherapy 27, 166–72. 51. Rosina, R., Avai, A., Suter, F., Bombelli, M., Gerundini, M., Danese, A. et al. (1989). Pharmacokinetics of teicoplanin in human bone. In Program and Abstracts of the 16th International Congress of Chemotherapy (Rubinstein, E. & Adams, D., Eds), Abstract 249. Lewis Epstein, Jerusalem. 52. Saul, T., Wittman, D. H., Fock, R. & Laufs, R. (1983). Serum, bone and tissue fluid concentrations of ceftriaxone. In Proceedings of the 13th International Congress of Chemotherapy (Spitzy, K. H. & Karrer, K., Eds), pp. 4.2/3–10. Egermann Druckereigesellschaft, Vienna. 53. Schurman, D. J., Hirshman, H. P., Kajiyma, G., Moser, K. & Burton, D. S. (1978). Cefazolin concentrations in bone and synovial fluid. Journal of Bone and Joint Surgery 60, 359–62. 54. Schurman, D. J., Hirshman, H. P. & Burton, D. S. (1980). Cephalotin and cefamandole penetration into bone, synovial fluid, and wound drainage fluid. Journal of Bone and Joint Surgery 62, 981–5. 55. Sirot, J., Prive, L., Lopitaux, R. & Glanddier, Y. (1983). Etude de la diffusion de la rifampicine dans le tissu osseux spongieux et compact au cours de protheses totales de hanches. Pathologie Biologie 31, 438–41. 56. Soudry, B., Sirot, J., Lopitaux, R., Dumont, C., Delisle, J. J. & Parenton, M. (1986). Diffusion de la ceftriaxone dans le tissu osseux humain. Pathologie Biologie 34, 859–62. 57. Wilson, A. P. R., Taylor, B., Treasure, T., Grüneberg, R. N., Patton, K., Felmingham, D. et al. (1988). Antibiotic prophylaxis in cardiac surgery: serum and tissue levels of teicoplanin, flucloxacillin, and tobramycin. Journal of Antimicrobial Chemotherapy 21, 201–12. 58. Rowland, M. (1990). Clinical pharmacokinetics of teicoplanin. Clinical Pharmacokinetics 18, 184–209. 59. Bryan, C. S., Morgan, S. L., Caton, R. J. & Lunceford, E. M. (1988). Cefazolin versus cefamandole for prophylaxis during total joint arthroplasty. Clinical Orthopaedics and Related Research 228, 117–22. 60. De Benedictis, K. J., Rowan, N. M. & Boyer, B. L. (1984). A double-blind study comparing cefonicid with cefazolin as prophylaxis in patients undergoing total hip or knee replacement. Review of Infectious Diseases 6, Suppl. 4, 901–4. 61. Evrard, J., Doyon, F., Acar, J. F., Salord, J. C., Mazas, F. & Flamant, R. (1988). Two-day cefamandole versus five-day cephazolin prophylaxis in 965 total hip replacements. Report of a multicentre double blind randomized trial. International Orthopaedics 12, 69–73. 62. Neu, H. C. (1984). Cephalosporin antibiotics as applied in surgery of bones and joints. Clinical Orthopaedics and Related Research 190, 50–64. 63. Oishi, C. S., Carrion, W. V. & Hoaglund, F. T. (1993). Use of parenteral prophylactic antibiotics in clean orthopaedic surgery. A review of the literature. Clinical Orthopaedics and Related Research 296, 249–55. 64. Brogden, R. N. & Peters, D. H. (1994). Teicoplanin. A reappraisal of its antimicrobial activity, pharmacokinetic properties and therapeutic efficacy. Drugs 47, 823–54. 65. Campoli-Richards, D. M., Brodgen, R. N. & Faulds, D. (1990). Teicoplanin. A review of its antibacterial activity, pharmacokinetic properties and therapeutic potential. Drugs 40, 449–86. 66. Cooper, G. L. & Given, D. B. (1986). Vancomycin: a Comprehensive Review of 30 Years of Clinical Experience. John Wiley, Park Row. 67. Dellamonica, P. & Bernard, E. (1993). Fluoroquinolones and surgical prophylaxis. Drugs 45, Suppl. 3, 102–13. 68. Matzke, G. R., Zhanel, G. G. & Guay, D. R. P. (1986). Clinical pharmacokinetics of vancomycin. Clinical Pharmacokinetics 11, 257–82. 69. Periti, P., LoRusso, D. & Mazzei, T. (1992). Ruolo della vancomicina nella chemioprofilassi antimicrobica perioperatoria in chirurgia. Farmacia and Terapia 9, 85–106. 70. Wilheim, M. P. (1991). Vancomycin. Mayo Clinic Proceedings 66, 1165–70. 71. Nichols, R. C. (1994). Antibiotic prophylaxis in surgery. Current Opinion in Infectious Diseases 7, 647–52. 72. Mollan, R. A. B., Webb, C. H. & Haddock, M. (1995). A comparison of teicoplanin vs cefamandole in orthopaedic surgical prophylaxis. In Program and Abstracts of the 7th European Congress of Clinical Microbiology and Infectious Diseases, Vienna, 1995. Abstract 758, p. 147. Excerpta Medica, Vienna. 73. Periti, P., Stringa, G., Donati, L., Mazzei, T., Mini, E., Novelli, A. & Participants of the Italian Study Groups for Antimicrobial Prophylaxis in Orthopedic Surgery and Burns. (1992). Teicoplanin: its role as systemic therapy for burns infections and prophylaxis for orthopaedic surgery. European Journal of Surgery, Suppl 567, 3–8. 74. Periti, P., Stringa, G., Mazzei, T., Mini, E. & Dentico, P. (1994). Antimicrobial prophylaxis in prosthetic joint implant surgery: a comparative multicentre trial between teicoplanin and cefazolin. In Abstracts of the Thirty-Fourth Interscience Conference on Antimicrobial Agents and Chemotherapy, Orlando, 1994. Abstract K4, p. 5. American Society for Microbiology, Washington, DC. 75. Sanderson, P. J. (1989). Preventing infection in orthopaedic implants. Journal of Antimicrobial Chemotherapy 24, 277–80. 76. Sanderson P. J. (1991) Infection in orthopaedic implants. Journal of Hospital Infection 18, Suppl. A, 367–75. 77. Suter, F., Avai, A., Fusco, U., Gerundini, M., Caprioli, S. & Maggiolo, F. (1994). Teicoplanin versus cefamandole in the prevention of infection in total hip replacement. European Journal of Clinical Microbiology and Infectious Diseases 13, 793–6. 78. Cartillier, J. C., Balay, B., Charlet, C., Machenaud, A. & Page, G. (1991). Pefloxacin as a prophylactic treatment against infection in total hip replacement. Comparison of two dosage schedules. In Program of the 3rd International Symposium on New Quinolones, Vancouver, 1990. Abstract 208, p. 261. 79. Hoogland, T. (1988). Comparative study of pefloxacin and cefamandole in the antimicrobial prophylaxis of orthopaedic surgery. In Program of the 2nd International symposium on new quinolones, Geneva, 1988. p. 240. 339 P. Periti et al. 80. Fihlo, J. L., Trigueiro, G. & Turibio, F. (1991). Pefloxacin versus cefazolin in the prophylaxis of postoperative infections after total hip or knee replacement. European Journal of Clinical Microbiology and Infectious Diseases, Special Issue, 637–8. 81. Guyot, A., Martens, M., Gillet, P., Desbarox, D. & Lemaire, R. (1991). Pefloxacin and ceftazidime in the perioperative prophylaxis of patients undergoing endoprosthetic hip or knee replacement. European Journal of Clinical Microbiology and Infectious Diseases, Special Issue, 638–9. 82. Periti, P., Stringa, G., Mini, E., Novelli, A., Servello, C. & other participants in the Italian Study Group for Antimicrobial Prophylaxis in Orthopaedic Surgery. (1990). Antimicrobial prophylactic activity of pefloxacin in orthopaedic surgery: preliminary results of a controlled multicentre study. In Second World Work of Professional Updating in Surgery and in Surgical and Oncological Disciplines of the University of Milan (Montorsi, M. & Zennaro, F., Eds), pp. 869–71. Monduzzi Editore, Bologna. 83. Cafferkey, M. T. (1991). Therapy for staphylococcal infection. Current Opinion in Infectious Diseases 4, 757–63. 84. Mulligan, M. E., Murray-Leisure, K. A., Ribner, B. S., Standiford, H. C., John, J. F., Korvick, J. A. et al. (1993). Methicillinresistant Staphylococcus aureus: a consensus review of the microbiology, pathogenesis and epidemiology with implications for prevention and management. American Journal of Medicine 94, 313–28. 85. Turnidge, J. & Grayson, M. L. (1993). Optimum treatment of staphylococcal infections. Drugs 45, 353–66. 86. Wilson, A. P. R. & Grüneberg, R. N. (1994). Use of teicoplanin in community medicine. European Journal of Clinical Microbiology and Infectious Diseases 13, 701–10. 87. Kloos, W. E. & Bannerman, T. L. (1994). Update on clinical significance of coagulase-negative staphylococci. Clinical Microbiology Reviews 7, 117–40. 88. Thornsberry, C. (1988). The development of antimicrobial resistance in staphylococci. Journal of Antimicrobial Chemotherapy 21, Suppl. C, 9–17. 89. Ritter, M. A., Barzilauskas, C. D., Faris, P. M. & Keating, E. M. (1989). Vancomycin prophylaxis and elective total joint arthroplasty. Orthopaedics 12, 1333–6. 90. American Society of Hospital Pharmacists’ Commission on Therapeutics. (1992). ASHP therapeutic guidelines on antimicrobial prophylaxis in surgery. Clinical Pharmacy 11, 483–513. 91. Page, C. P., Bohnen, J. M., Fletcher, J. R., McManus, A. T., Solomkin, J. S. & Wittmann, D. H. (1993). Antimicrobial prophylaxis for surgical wounds. Guidelines for clinical care. Archives of Surgery 128, 79–88. 92. Anonymous. (1995). Antimicrobial prophylaxis in surgery. Medical Letter on Drugs and Therapeutics 37, 79–82. 93. Wood, M. J. (1996). The comparative efficacy and safety of teicoplanin and vancomycin. Journal of Antimicrobial Chemotherapy 37, 209–22. 94. Southorn, P. A., Plevak, D. J., Wright, A. J. & Wilson, W. R. (1986). Adverse effects of vancomycin administered in the perioperative period. Mayo Clinic Proceedings 61, 721–4. 95. Valero, R., Gomar, C., Fita, G., González, M., Pacheco, M., Mulet, J. et al. (1991). Adverse reactions to vancomycin prophylaxis in cardiac surgery. Journal of Cardiothoracic and Vascular Anesthesia 5, 574–6. 96. Greenwood, D., Bidgood, K. & Turner, M. (1987). A comparison of the responses of staphylococci and streptococci to teicoplanin and vancomycin. Journal of Antimicrobial Chemotherapy 20, 155–64. 97. Hackbarth, C. J. & Chambers, H. F. (1989). Methicillinresistant staphylococci: detection methods and treatment of infections. Antimicrobial Agents and Chemotherapy 33, 995–9. 98. Davey, P. G. & Williams, A. H. (1991). A review of the safety profile of teicoplanin. Journal of Antimicrobial Chemotherapy 27, Suppl. B, 69–73. 99. Baiocchi, P., Capone, A. & Venditti, M. (1996). Changes in susceptibilities to teicoplanin, vancomycin and other antibiotics among Staphylococcus aureus isolates in a tertiary care university hospital. International Journal of Antimicrobial Agents 7, 93–6. 100. Johnson, A.P., Uttley, A.H.C., Woodford, N. & George, R.C. (1990). Resistance to vancomycin and teicoplanin: an emerging clinical problem. Clinical Microbiology Reviews 3, 280–91. 101. Shlaes, D. M. & Shlaes, J.H. (1995). Teicoplanin selects for Staphylococcus aureus that is resistant to vancomycin. Clinical Infectious Diseases 20, 1071–3. 102. Edmond, M.B., Wenzel, R.P. & Pasculle, A.W. (1996). Vancomycin-resistant Staphylococcus aureus: perspective on measures needed for control. Annals of Internal Medicine 124, 329–34. 103. The Hospital Infection Control Practice Advisory Committee (1995). Recommendations for preventing the spread of vancomycin resistance: recommendations from the Hospital Infection Control Practice Advisory Committee (HICPAC). American Journal of Infection Control 23, 87–94. 104. Wall, R., Klenerman, L., McCullough, C. & Fyfe, I. (1988). A comparison of teicoplanin and cefuroxime as prophylaxis for orthopaedic implant surgery: a preliminary report. Journal of Antimicrobial Chemotherapy 21, Suppl. A, 141–6. 105. Mantel, H. & Haenszel, W. (1959). Statistical agents of the analysis of data from retrospective studies of disease. Journal of the National Cancer Institute 22, 719–48. 106. Ioannidis, J.P.A., Cappelleri, J.C., Skolnik, P.R., Lau, J. & Sacks, H.S. (1996). A meta-analysis of the relative efficacy and toxicity of Pneumocystis carinii prophylactic regimens. Archives of Internal Medicine 156, 177–88. 107. Burnakis, T.G. (1984). Surgical antimicrobial prophylaxis: principles and guidelines. Pharmacotherapy 4, 248–71. Received 16 April 1997; returned 24 July 1997; revised 6 November 1997; accepted 24 November 1997 340