|
|
Staphylococcus aureus Updated June, 2007 Professor of Pathology, Pediatrics and Molecular Biosciences Chief, Division of Laboratory Medicine Women's and Children's Hospital Adelaide, South Australia, Australia Fax: 61 8 8161 6189 Phone: 61 8 8161 7417 Email: john.turnidge@cywhs.sa.gov.au
Clinical Professor of Medicine and Orthopaedics University of Pittsburgh School of Medicine Chief, Infectious Diseases Medical Director, Infection Control University of Pittsburgh Medical Center-Shadyside 5230 Centre Avenue Pittsburgh, Pennsylvania 15232 Phone: 412-623-1649 E-mail: raon@msx.upmc.edu
Associate Professor of Medicine National Defence Medical Center Chief, Division of Infectious Diseases and Tropical Medicine Tri-Service General Hospital Taipei, Taiwan Email: fychang@ndmctsgh.edu.tw
Assistant Professor of Medicine Department of Medicine Division of Infectious Diseases Duke University Medical Center Box 3824 Durham, N. Carolina 27710 Fax: (919) 684-8902; Phone: (919) 684-4335 Email: fowle007@mc.duke.edu
GENERAL DESCRIPTION Microbiology Guided Medline Search Staphylococcus aureus is facultative anaerobic gram-positive cocci which occur singly, in pairs, and irregulular clusters. S. aureus is nonmotile, non-spore forming, catalase and coagulase positive. Typical colonies are yellow to golden yellow in color, smooth, entire, slightly raised, and hemolytic on 5% sheep blood agar. However, many strains may appear dirty white and nonhemolytic. It also gives a positive mannitol fermentation and deoxyribonuclease test. The major criterion for identification is the organism’s ability to clot plasma. There are three coagulase-positive Staphylococcal species; S. aureus in humans and animals, and S. intermedius, and S. hyicus in animals. The presence of the enzyme coagulase separates the virulent pathogen, S. aureus, from the less virulent coagulase-negative Staphylococci species. There are two different tests that can be performed to detect the presence of coagulase: a tube test to detect free coagulase and a slide test to detect bound coagulase. The slide test is a rapid test; however, a small percentage of S. aureus strains may yield a negative result. If the organism is suspected as S. aureus, negative slide tests should be followed up with a tube test. MRSA/VISA/GISA/VRSA: Since the 1970s, S. aureus strains have emerged resistant to the penicillinase-stable penicillins (cloxacillin, dicloxacillin, methicillin, nafcillin, and oxacillin). The resistance is the result of a supplemental penicillin binding protein (PBP 2a) encoded by the chromosomal mecA gene. These strains historically are termed methicillin resistant S. aureus (MRSA) and are resistant to all beta-lactam agents. Laboratory confirmation of these strains can be problematic. The resistant strains are often heteroresistant. That is, two populations coexist, on susceptible and the other resistant. Each cell has the genetic information for resistance, but only a very small number express this resistance in vitro (1 in 104 to 1 in 108). Successful detection of MRSA largely depends on promoting the growth of the resistant population. This is done by lowering the incubation temperature to 35oC, using a 0.5 McFarland suspension directly from the colonies, supplementing with 2% sodium chloride, and incubating for a full 24 hours in ambient air. Even with these refinements, the heterogeneous expression of some isolates may be interpreted as susceptible. The oxacillin-salt screening plate which supplemented with 4% sodium chloride and 6 µg/ml of oxacillin can be used to improve detection of these strains. The growth of more than one colony indicates resistance. Most isolates of S. aureus are susceptible to vancomycin. The MIC is typically between 0.5 and 2 micrograms/mL (μg/mL). In contrast, S. aureus isolates for which vancomycin MICs are 8-16 μg/mL are classified as vancomycin intermediate (VISA), and isolates for which vancomycin MICs are ≥32 μg/mL are classified as vancomycin-resistant (VRSA). The glyopeptide class of antibiotics includes both vancomycin and teicoplanin. And, some of the original VISA strains were also intermediate for teicoplanin, hence the name, GISA. However, not all strains of VISA show intermediate susceptibility to teicoplanin. CLSI (formerly NCCLS) lists only susceptible disk diffusion interpretive criteria (in mm) for vancomycin and Staphylococcus spp. Insufficient numbers of non-susceptible isolates from patients have been isolated to develop resistant and intermediate breakpoints. Organisms for which the vancomycin zone diameters are ≥15mm are considered susceptible, although this breakpoint is unreliable for detecting VISA strains. As of November 2005, only four patients infected by vancomycin-resistant S. aureus (VRSA) have been confirmed by the U.S. Centers for Disease Control. Automated susceptibility systems have not been reliable in detecting these strains. When using MIC methods that have not been validated to detect VRSA, BHI vancomycin agar screening plates containing 6 μg/mL of vancomycin can be used to enhance the sensitivity of detecting vancomycin-resistant strains. [Abigail Orenstein: The Discovery and Naming of Staphylococcus aureus]
Epidemiology Guided Medline Search It is widely distributed in nature and carried by 25-33% of normal individuals in the anterior nares and skin. It can colonize and infect both healthy, immunologically competent people in the community and hospitalized patients with decreased host defences. S. aureus is one of the commonest and most important Gram-positive hospital-acquired organisms. It has a high propensity to colonize abnormal skin surfaces and open wounds, where it may merely reside rather than cause active infection. [Article: Menace in the Locker Room, Sports Illustrated , February 28, 2005.] [Article: CDC. MRSA infecting the St. Louis Rams football team. N Eng J Med February 3, 2005] [Article: CDC. MRSA in Horses and Horse Personnel, 2000-2002. Emerg Infect Dis, March 2005.] [Review Article: Rybak MJ, LaPlante KL. Community-Associated Methicillin-Resistant Staphylococcus aureus: A Review. Pharmacotherapy 2005;25(1):74-85.]
Clinical Manifestations Guided Medline Search Infections caused by S. aureus range from minor skin disorders such as wound infections, furuncles and carbuncles, and bullous impetigo, through locally invasive diseases such as cellulitis, osteomyelitis, sinusitis, and pneumonia, to major life-threatening septicemia and meningitis. It is also a frequent cause of medical device-related infections such as intravascular line sepsis and prosthetic joint infections. Although minor skin infections may resolve naturally without antibiotic intervention, once S. aureus invades deeper structures, it often spreads hematogenously to other organ systems, leading to metastatic infection. Endocarditis and septicemia have significant morbidity and mortality despite aggressive antimicrobial therapy. Toxin mediated disease include the toxic shock syndrome that presents with profound hypotension and a generalized erythematous rash. While TSS was commonly associated with menstruation and the use of hyperabsorbable tampons, the nonmenstrual form is commonly associated with wounds from different surgical procedures. Staphylococcal food poisoning occurs with a short incubation period of 2-6 hours and is characterized by nausea and vomiting, that is followed by abdominal cramps and diarrhea, which can be hemorrhagic. It is mediated by enterotoxin B and occurs due to ingestion of food contaminated with preformed toxins.
Laboratory Diagnosis Guided Medline Search The definitive diagnosis of disease is made by isolation and identification of the species of Staphylococcus. Depending in the conditions being investigated, samples of sputum, purulent material, blood and urine should be obtained. Several sets of blood cultures are required to make a diagnosis of S. aureus septicemia or endocarditis. Echocardiography, especially transesophageal, is crucial for confirming the diagnosis of endocarditis.
Pathogenesis Guided Medline Search Staphylococcus aureus produces and secretes a number of enzymes and toxins that account for its pathogenesis. The most important enzymes are coagulase, catalase, hyaluronidase and clumping factor. Among the toxins, the alpha-, beta-, and delta-hemolysins exert their effect by enzymatic action, while others such as toxic shock toxins and enterotoxins, act as superantigens and activate subset of T lymphocytes to release cytokines that cause potent systemic effects. Panton-Valentine leukocidin has been linked to virulence in skin and soft tissue infections and pneumonia. [Lina G, Vandenesch F, Etienne J. A brief history of Staphylococcus aureus Panton Valentine leucocidin] [Review Article: Rybak MJ, LaPlante KL. Community-Associated Methicillin-Resistant Staphylococcus aureus: A Review. Pharmacotherapy 2005;25(1):74-85.]
SUSCEPTIBILITY In VITRO AND IN VIVO Guided Medline Search Evolution of Resistance for Staphylococcus aureus Following its introduction into clinical practice in the 1940s, penicillin G became the treatment of choice for infections caused by S. aureus. However, S. aureus resistant to penicillin through the production of a beta-lactamase (penicillinase) rapidly emerged (274). High levels of resistance to penicillin (80-95%) are standard for community strains in almost all countries now. Since the 1970s, S. aureus strains resistant to the penicillinase-resistant penicillins (represented by the original member of the class, methicillin) have gradually emerged worldwide (11). Until recently, these strains have generally been multiresistant, exhibiting resistance to macrolides and lincosamides, and usually to tetracyclines and gentamicin. Resistance to trimethoprim and sulfonamides is also prevalent in some countries. This type of methicillin-resistant S. aureus (MRSA) is now a common cause of nosocomial infection in many countries in both the developed and the developing world. The rapidity with which MRSA developed in Europe after the introduction of methicillin in 1959 and its subsequent spread throughout the world have created therapeutic problems for physicians, in part because of the high propensity of MRSA to acquire new resistances. Methicillin resistance in MRSA isolates is chromosomally mediated and results, at least in part, from the presence of an additional and modified penicillin-binding protein (PBP-2a), which has reduced affinity for methicillin and other beta-lactams, and hence retains critical functions necessary for cell survival (40,54). PBP-2a is encoded by the mec A gene located on the staphylococcal chromosome within a discrete region called the staphylococcal cassette chromosome (SCCmec), for which a range of types have been described (295). Resistance to all beta-lactams including cephalosporins, penicillinase-resistant penicillins, beta-lactamase inhibitor combinations and carbapenems should be assumed once methicillin resistance has been demonstrated (monobactams have no anti-staphylococcal activity). Recently different types of MRSA has been described with origins in the community in several countries, including Australia, United States, Canada, New Zealand, Saudi Arabia, Finland and Taiwan (88,237). Resistance to penicillin and methicillin but susceptibility to most or all other drug classes characterizes these types of MRSA. It appears to be principally a community-acquired organism (320,337), but hospital outbreaks have been described (294). Methicillin resistance is also encoded by mecA in these strains, and they have different SCCmec types to those of classical hospital-acquired strains (295,117). In England and Wales, a further pattern of MRSA has emerged since 1990. Two types of MRSA called EMRSA-15 and EMRSA-16 have became widespread across hospitals (87, 276). Unlike classical hospital-acquired MRSA from other countries, EMRSA-15 and-16 harbors a few additional resistances beyond penicillin and methicillin, mainly erythromycin and ciprofloxacin (336, 449,276). The emergence of different types of MRSA with a reservoir in the community rather than the hospital has created two kinds of problems. The first relates to recognition and nomenclature. Given their different epidemiologies and rates of resistance to antibiotics other than beta-lactams, it is useful to have terminology to differentiate two types, as it has potential impact in choosing treatment alternatives to beta-lactams. In this chapter, the term haMRSA (hospital-acquired multiresistant oxacillin-resistant Staphylococcus aureus) is used to describe classical hospital-acquired MRSA, the term caMRSA (community-acquired oxacillin-resistant Staphylococcus aureus) to describe the new, predominantly community-acquired, strains, and the term EMRSA-15/16 the new epidemic-hospital acquired strains in England and Wales. These three types can be distinguished, at least for the present, by the number and types of resistances harbored in addition to beta-lactams: haMRSA strains are resistant to 3 or more of the following: erythromycin, clindamycin, rifampicin, ciprofloxacin, gentamicin, tetracycline, trimethoprim-sulfamethoxazole or chloramphenicol (414). EMRSA-15/16 are usually resistant to erythromycin and ciprofloxacin in addition to beta-lactams, and susceptible to others classes. For clarity the acronym MSSA is used for penicillin-resistant, methicillin-susceptible strains. Strains with increased MICs to vancomycin were first described in Japan in 1996 (175). More recently, similar strains have been detected in the United States, France, the United Kingdom, Spain, Hong Kong, Italy Germany, India, South Korea, the Philippines, Singapore, Thailand, Vietnam, Australia, the Netherlands and Poland (29, 77, 194, 246, 397, 377, 47, 176, 204, 212). These strains have been termed VISA (vancomycin-intermediate S. aureus) or GISA (glycopeptide-intermediate S. aureus). MICs are typically in the range of 6-16mg/L. Resistance may be heterogeneous or homogeneous. Heterogeneous strains are more difficult to detect and a reliable screening test is still being sought. The clinical significance of heterogeneous resistance is uncertain, but failures of vancomycin treatment in homogeneously resistant strains are now well documented (397,140; 176, 238) and failures with heterogeneously resistant strains are be reported (67). All VISA strains described to date have been MRSA. Antimicrobial activity of the different antibiotics for both methicillin-susceptible S. aureus (MSSA) and MRSA is listed in Table 1. There have now been three reports, all from the United States, of methicillin-resistant S. aureus clinical isolates harbouring the vanA gene complex previously found in enterococci (50, 51, 52, 67, 427, 428). These strains have MICs to vancomycin well into the resistant range. [Review Article: Hiramatsu K. Vancomycin-resistant Staphylococcus aureus: a new model of antibiotic resistance. Lancet Infect Dis. 2001 Oct;1(3):147-55.] Single Drugs Penicillins: Most staphylococci are resistant to penicillin (G, benzylpenicillin) and other beta-lactamase labile penicillins such as the aminopenicillins, carboxypenicillins and the ureidopenicillins through the production of specific beta-lactamase which was originally called penicillinase. All formulations of penicillin itself are affected: benzylpenicillin, procaine penicillin, benzathine penicillin, phenoxymethyl-and phenoxyethylpenicillin. A small proportion ( 5 to 10% ) of community-acquired strains of S. aureus do not produce a beta-lactamase and remain susceptible to penicillin G. beta-lactamase-mediated resistance results from production of an enzyme that is excreted extracellularly by S. aureus that inactivates the antibiotic by opening the beta-lactam ring. Penicillinase-Resistant Penicillins: Currently, the majority of community-acquired S. aureus strains remain susceptible to the antistaphylococcal, semisynthetic penicillins and first-generation cephalosporins (Table 1), although this is changing rapidly in some countries. Agents classified as penicillinase-resistant penicillins include methicillin and nafcillin, and the isoxazoyl penicillins oxacillin, cloxacillin, dicloxacillin, and flucloxacillin. Although there is minor variation in their relative beta-lactamase stability, this does not appear to be of clinical importance. Differences in the in vitro potency of the penicillin-resistant penicillins are small, with modal minimum inhibitory concentrations (MICs) in the range of 0.125 to 0.5 ug/mL (213). Cephalosporins: Cephalosporins show variable stability to staphylococcal beta-lactamase, depending on their chemical structure. Cephalothin is relatively resistant, while cefazolin is more sensitive to staphylococcal beta-lactamase degradation (130,328). Although this in vitro phenomenon has not been clearly demonstrated to be clinically significant, some prefer cephalothin for the treatment of life-threatening S. aureus infections (314). Data from animal studies suggest that cephalosporins are probably less effective than the penicillinase-resistant penicillins for treatment of serious staphylococcal infections. Cefazolin and cephalothin were less effective than nafcillin in the rabbit model of endocarditis (49,381). Compared with first-generation cephalosporins, the second-and third-generation cephalosporins in general have inferior in vitro activity against S. aureus. With the exception of cefamandole, cefuroxime, and possibly cefaclor, cephalosporins of later generations generally have lower activity against staphylococci and offer no advantages over first-generation cephalosporins when they need to be used in the management of staphylococcal infection. However, almost all cephalosporins have sufficient activity to provide initial coverage pending the results of laboratory investigations. Cephalosporins are not active against MRSA strains in vivo, despite the fact that some strains may appear susceptible in routine laboratory tests. Exceptions to this rule have been found in new cephalosporin molecules under development, such as ceftobiprole, LB11058 and RWJ-33341 (150, 425, 62). Penicillin/ beta-Lactamase Inhibitor Combinations: Staphylococcal beta-lactamase is readily inhibited by the currently available beta-lactamase inhibitors clavulanic acid, sulbactam, and tazobactam. Thus, combination of these inhibitors with beta-lactamase-labile penicillins restores activity against penicillin-resistant, methicillin-susceptible staphylococci (Table 1). Like cephalosporins, they are not active against MRSA strains. Carbapenems: Imipenem and meropenem have a very broad spectrum of activity that includes S. aureus. However, MRSA is also resistant to imipenem and meropenem (Table 1). Although some strains of MRSA appear susceptible to carbapenems in vitro, they are not susceptible in vivo (28,60). Macrolides: In general, the macrolides show a fairly uniform activity against staphylococci (16, 20, 443). haMRSA is always resistant to the macrolides (167) while many caMRSA are susceptible, although resistance can range up to 25%. Resistance to erythromycin is also prevalent worldwide in community-acquired MSSA strains. The newer macrolides -- dirithromycin, roxithromycin, clarithromycin, and azithromycin -- have activity similar to that of erythromycin against staphylococci (Table 1). Strains resistant to erythromycin are also resistant to these newer macrolides. A new class of agents that are derivatives of macrolides, the ketolides, are active against erythromycin-susceptible staphylococci, but not against resistant strains of S. aureus (165). Lincosamides (Clindamycin and Lincomycin): S. aureus resistant to erythromycin demonstrate two types of resistance: inducible and constitutive. Inducibly-resistant strains test as susceptible to lincosamides in vitro when tested alone, but as resistant when tested in the presence of erythromycin. Strains that are constitutively resistant test as resistant to lincosamides (222). The clinical significance of inducible resistance in unknown, as formal prospective studies have not been conducted to determine whether inducibly-resistant strains respond to treatment with lincosamides. Recent in vitro evidence shows that they have a higher rate of mutation to lincosamide resistance than erythromycin-susceptible strains (297), There is anecdotal evidence that they respond clinically, but that selection of resistance during treatment and relapse are common (107,138,230,256, 323,370). A consensus is emerging that infections caused by strains with inducible resistance should not be treated with clindamycin unless the infection is minor (234). MSSA is usually susceptible to clindamycin with an MIC range, < 0.06 ug/mL-0.125 ug/mL (Table 1). haMRSA is resistant to clindamycin with a consistent MIC above 256 ug/mL (Table 1). caMRSA strains generally test as susceptible to lincosamides. In a rabbit model of endocarditis, clindamycin is associated with a relatively slow rate of eradication of organisms from the infected vegetation, and relapse was more likely in rabbits given clindamycin than in those treated with penicillin or vancomycin (352). Fluoroquinolones: Fluoroquinolones are DNA gyrase inhibitors that are active in vitro and in vivo against S. aureus, including some MRSA strains (48, 148, 387). In terms of gram-positive activity, and activity against staphylococci, the older quinolones such as ciprofloxacin, ofloxacin and levofloxacin are less potent than the new generation agents such as trovafloxacin, moxifloxacin, gatifloxacin and garenoxacin. The MICs of ciprofloxacin are typically between 0.5 and 1 µg/mL for MSSA strains (Table 1), and two to four-fold lower for the new generation agents. Most strains of MORSA are now resistant to fluoroquinolones (92, 366, 412), including the new generation. Resistance to fluoroquinolones has been found in MSSA and MRSA strains (192,275). Both altered gyrase and energy-dependent efflux mechanisms account for the development of resistance to fluoroquinolones (192). Newer quinolones such as levofloxacin and trovafloxacin have been shown in animal models to be as effective as vancomycin in the treatment of S. aureus endocarditis, including that caused by methicillin-resistant strains (24, 59, 119,188). A new quinolone (TG-873870) with excellent anti-MRSA activity and anti-anaerobic activity is undergoing phase 2 trials (Biotechnology Co., Ltd., Taipei, Taiwan). In animal model studies at least, 24-hour AUC/MIC ratios of 100 for quinolones appear to be optimum for favourable outcome (6). Glycopeptides: Until recently, MRSA strains were invariably susceptible to vancomycin and teicoplanin (Table 1). MICs range between 0.25 and 4µg/mL (Table 1). However, vancomycin is slowly and incompletely bactericidal against MSSA in vitro compared with nafcillin (23,163,197). Newer glycopeptides under development (oritavancin, dalbavancin, telavancin) show potential advantages over other glycopeptides (317, 417, 386). They are rapidly bactericidal, and is active against VISA (172, other refs). Oritavancin shows efficacy equivalent to vancomycin in the rabbit endocarditis model (189). Teicoplanin (not available in the United States) is another glycopeptide with a spectrum of activity similar to that of vancomycin but which may be less active against some coagulase-negative staphylococci. In animal models, teicoplanin has activity equivalent to that of vancomycin for treatment of experimental endocarditis caused by both MSSA and MRSA (9,58). Teicoplanin MICs for VISA are usually increased more than vancomycin (397). Tetracyclines: Resistance to tetracyclines is common in community-acquired strains of S. aureus. Some tetracycline-resistant strains appear susceptible to minocycline and doxycycline. The incidence of tetracycline resistance is over 90% in haMRSA (114). In vitro susceptibility of S. aureus to minocycline has been documented for a number of years (262). There is recently evidence of efficacy against tetracycline-susceptible MRSA. Minocycline and vancomycin are equally effective in reducing bacterial density in infected vegetations in a rabbit model of MRSA endocarditis (285). Glycylcyclines: Glycylcyclines are chemical modifications of tetracyclines (458). The only agent in this class so far to have reached and advanced stage of development is tigecycline. Like tetracycline, tigecycline has quite a broad spectrum, and has the advantage of having activity against almost all tetracycline-resistant strains. Its potential as an important antistaphylococcal agent has not been explored, but phase III studies in an array of infections, including complicated skin and skin structure infection (310), have shown efficacy when S. aureus was the cause. Trimethoprim-sulfamethoxazole (TMP-SMX): Most strains of MSSA are susceptible to TMP-SMX. In vitro susceptibility of haMRSA to TMP-SMZ varies around the world. In one study, up to 95% of haMRSA strains were susceptible to TMP-SMX (116), while in other areas almost all strains are resistant (412). TMP-SMX was strikingly inferior to vancomycin for infection caused by either an MSSA or an MRSA strain in the rabbit model of aortic valve endocarditis (96). The numbers of bacteria in the vegetations of rabbits treated for 3 days with TMP-SMX were minimally reduced, compared with untreated controls, and no infection was sterilized, versus a rate of 70 to 80% for vancomycin. Mupirocin: Mupirocin is used as a topical agent for the treatment of superficial skin infections and S. aureus carriage. Some 414 nasal and 586 non-nasal S. aureus isolates, both methicillin resistant and methicillin susceptible, showed similar MIC 90 s and a susceptibility of 99.1% to the topical antimicrobial agent mupirocin (Table 1) (416) However, there is evidence of resistance to mupirocin emerging to significant rates when used widely (89). Fusidic Acid: Fusidic acid (available in Europe and some Western Pacific countries) is the only marketed antibiotic in a class of agents that has a unique mechanism of action and a distinct Gram-positive spectrum (excluding streptococci). It is active against both MSSA and MRSA strains (420,86). S. aureus is inhibited by fusidic acid at very low concentrations, usually between 0.03 and 0.25 µg/mL, regardless of their susceptibility to methicillin or oxacillin (244,420). Fusidic acid-resistant mutants are harbored at relatively high frequencies. However, the rates of resistance have remained low in almost all regions where fusidic acid is regularly used (415). Fosfomycin: Fosfomycin is an epoxide antibiotic that has a different structure and mode of action from other antimicrobials. The MIC 90 for MRSA is 4µg/ mL. Fosfomycin, alone or in combination with ß-lactam antibiotics, has been said to be active against MRSA in vitro (4). Rifampin: Rifampin (rifampicin) is highly active against staphylococci, with an MIC 50 for about 0.03 µg/mL. Resistant mutants can be easily selected in vitro and are thought to be naturally present in susceptible populations at frequencies of 10 -6 to 10 -7 (202,214). Because of this resistance, rifampin is almost always used in combination with other anti-staphylococcal drugs when treating established infection. Resistance to rifampin is more prevalent in classical haMRSA than in caMRSA, EMRSA-15/16 or MSSA in most regions of the world (412). Streptogramins: Streptogramins are antibiotics that are a combination of two types: streptogramins A and B. Streptogramins B share the same site of action as macrolides and lincosamides, while streptogramins A act at a separate site on the ribosome that enhances the effect of the streptogramins A (83). The original agent in the class, pristinamycin, has been available as an oral medication in France for many years. More recently, a semisynthetic injectable streptogramin combination, quinupristin-dalfopristin, has been developed particularly aimed at the treatment of multi-resistant Gram-positive infections. The advantage of these agents is that activity is usually retained against staphylococci and other gram-positives that are resistant to macrolides and lincosamides (36,184), and thus all forms of MRSA. Resistance to quinupristin-dalfopristin is currently very rare (184,185). Oxazolidinones: Oxazolidinones are synthetic agents, the original members of which were MAO inhibitors (98). One, linezolid, is now available in some countries for the treatment of resistant staphylococcal infection (79). These drugs have a novel mechanism of action on ribosomal protein synthesis, and are active against strains resistant to other classes of antibiotics (293). Linezolid can be given orally as well as parenterally. Daptomycin: Daptomycin is the first of a novel class of cyclic lipopeptides. Originally developed by Eli Lilly, its development was dropped, and was later taken up and completed by Cubist. Its unique mechanism of action involves calcium-dependent binding to the cell membrane, membrane depolarization, cessation of protein and DNA synthesis, potassium leakage and cell death. It is active against all types of S. aureus and licensed as a parenteral formulation for the treatment of serious skin and skin structure infections (236,382). Combination Drugs In Vitro Guided Medline Search Addition of gentamicin to nafcillin produces an enhanced bactericidal effect in vitro (351,354). The combinations of vancomycin-gentamicin and vancomycin-tobramycin are synergistic against most MSSA and MRSA strains (433). If synergism is defined as a decrease in colony counts of at least 100-fold at 24 h with the combination compared with that of the most active single drug, vancomycin-gentamicin synergism is not predictable for strains of MRSA with gentamicin MICs of 0.5 to more than 128 µg/mL (271). However, a gentamicin MIC above 500 µg/mL predicts a lack of vancomycin-gentamicin synergism for strains of MRSA (152). The in vitro effect of rifampin in combination with semisynthetic penicillins, vancomycin, and aminoglycosides is highly variable (163, 407, 433, 460). For both MSSA and MRSA, minocycline-rifampin synergism had been demonstrated by checkerboard evaluation (75, 363). Rifampin reduces the extracellular bactericidal activity of dicloxacillin but not fusidic acid, but neither of the latter agents reduce the intracellular killing effect of rifampin against S. aureus (289). The combination of linezolid and ampicillin/sulbactam was found to be either additive or synergistic against 48 clinical isolates of MRSA, including 10 strains with reduced susceptibility to vancomycin (200). Resistance in vitro does not appear to emerge if MRSA are exposed to a combination of fusidic acid and rifampin (257), and sub-MIC concentrations of trimethoprim also appear to be able to prevent selection of rifampin resistant-mutants (183). Similarly, rifampin can suppress the emergence of ciprofloxacin resistance in an in vitro pharmacodynamic model (193). Combinations of older agents to overcome resistance is a possibility only now being explored. An in vitro pharmacodynamic model has shown that combinations of cefepime with a wide range of other agents such as aminoglycosides, and the more recent agents linezolid, daptomycin and tigecycline (177) In Vivo (Animal Studies) Guided Medline Search Adding gentamicin to nafcillin accelerates killing of MSSA within experimentally induced cardiac vegetations in animal models (351). The addition of MiKasome, a new liposome-encapsulated formulation of conventional amikacin, enhances the in vivo bactericidal effects of oxacillin in S. aureus experimental endocarditis and may preserve selected physiological functions in target end organs (453). Some animal models favor the use of rifampin (171,190). In animal studies, rifampin has been shown to play a unique role in the complete sterilization of foreign bodies infected by S. aureus (77). Addition of rifampin to combination antibiotic regimens has proven highly effective in an animal model of MRSA osteomyelitis (171). In experimental MRSA endocarditis in rabbits, an ampicillin/sulbactam/rifampin regimen (with a high ampicillin dosage at 625-800mg/kg/day) was as effective as vancomycin (55). In animal models, use of quinolone combinations with rifampin may prevent resistance to both drugs (190,356). Several studies have evaluated the effectiveness of quinupristin-dalfopristin combined with other antimicrobial agents. In animal models of S. aureus endocarditis, the activity of quinupristin-dalfopristin combined with beta-lactam antibiotics (423,424) or vancomycin (299) was additive or synergistic against both macrolide-lincosamide-streptogramin B (MLS B) susceptible and resistant bacteria. The combination of quinupristin-dalfopristin and rifampin is highly synergistic in experimental S. aureus prosthetic joint infection (350).
ANTIMICROBIAL THERAPY Drug of Choice Guided Medline Search Smart Search Treatment of staphylococcal infections depends on the site and severity of infection, the antibiotic susceptibility pattern of the organism and the presence of any patient allergy or drug intolerance. Penicillin-Susceptible S. aureus: Benzylpenicillin ( penicillin G) is still the drug of choice for ß-lactamase-negative strains because on a weight-for-weight basis, it is more active than penicillinase-resistant penicillins. Penicillin-susceptible S. aureus strains have not been reported to become resistant to penicillin during treatment. Phenoxymethylpenicillin (penicillin V) and amoxycillin can both be used as oral agents for penicillin-sensitive S. aureus, with amoxycillin preferred when higher levels are required to achieve adequate penetration. Methicillin-Susceptible S. aureus (MSSA): Penicillinase-resistant penicillins are the preferred drugs for all S. aureus infections caused by penicillin-resistant MSSA strains. These agents have gained wide acceptance because they are bactericidal and, like other penicillins, have a low incidence of adverse reactions. A variety of penicillinase-resistant penicillins are available, including the isoxazoyl penicillins cloxacillin, dicloxacillin, flucloxacillin, and oxacillin for both oral and parenteral use and methicillin and nafcillin for parenteral use. There are no apparent differences in efficacy between these agents, and they have similar pharmacokinetic profiles. Continuous infusions of these agents are being used increasing in serious staphylococcal infection, especially as outpatient therapy, with satisfactory outcomes (223). In most countries, methicillin has been superseded by the other agents in this group because of its association with a higher incidence of adverse reactions, especially hypersensitivity and interstitial nephritis. A serious reaction to flucloxacillin, characterized by prolonged hepatic cholestasis, has been described with some frequency and has been associated with both parenteral and oral therapy (122,408). Cephalosporins, particularly those of the first generation, have proven useful alternatives to penicillinase-resistant penicillins, since they are relatively stable to staphylococcal beta-lactamase. They are most commonly used in patients with a history of allergy or intolerance to penicillins. However, it is considered imprudent to administer them to patients with a history of accelerated reactions (e.g., angioedema or anaphylaxis). Patients with this type of history should not be given beta-lactams of any class, and other antistaphylococcal agents such as clindamycin or vancomycin should be used. Cephaloridine, the cephalosporin with the greatest potency in vitro against staphylococci, has been abandoned due to the risk of nephrotoxicity and relative instability to staphylococcal beta-lactamases. Suitable first-generation cephalosporins for staphylococcal infections include cephalothin, cefazolin , cefapirin and cephradine for parenteral use and cephalexin , cefapirin and cephradine for oral use. Currently available combinations of penicillin/beta-lactamase inhibitor include amoxycillin/clavulanic acid, ticarcillin/clavulanic acid, ampicillin/sulbactam and piperacillin/tazobactam. These combinations have no in vitro or in vivo superiority over penicillinase-resistant penicillins, but they do have the advantage of possessing a broad spectrum of activity against Gram-negative bacteria, including anaerobes. Therefore, they should not be used as substitutes for penicillinase-resistant penicillins when staphylococci are the sole pathogens. However, their broader spectrum gives them a significant advantage when staphylococci are involved in mixed infections with enteric Gram-negative organisms and anaerobes. Imipenem and meropenem provide adequate staphylococcal coverage in mixed infections but, again, have no role in the specific treatment of pure staphylococcal infection. Erythromycin has been used extensively for treatment of both minor and serious staphylococcal infections. It is bacteriostatic against staphylococci and, for this reason, has generally lost favor for the management of serious and life-threatening infection, being largely supplanted by penicillinase-resistant penicillins. Moreover, the role of erythromycin in empirical treatment is further limited because of drug resistance. Nevertheless, oral erythromycin is still suitable for minor skin infections caused by S. aureus, especially in penicillin-allergic patients, provided the strain has demonstrated susceptibility on laboratory testing. For erythromycin-susceptible strains, other macrolides are likely to be equally effective. The newer macrolides, roxithromycin, clarithromycin , dirithromycin, and azithromycin , have activity similar to that of erythromycin against staphylococci. In general, strains resistant to erythromycin are also resistant to these newer macrolides. The latter two agents have very high tissue penetration, which may be of advantage in some sequestrated staphylococcal infections. Although there is good experience with the newer macrolides for staphylococcal skin infections, there is little experience with these drugs in the treatment of osteomyelitis. These drugs are not currently recommended for the treatment of bacteremia or endocarditis. Like erythromycin, lincosamides (clindamycin and lincomycin) have been available for many years and have been used frequently for treatment of staphylococcal infections. They are also bacteriostatic and have been mostly relegated to reserve agents. However, the lincosamides are perhaps more useful second-line agents than macrolides, and are gaining acceptance as first-line agents for the treatment of caMRSA (252). Clindamycin is better absorbed than lincomycin when administered orally and is the preferred agent for oral use. It demonstrates good penetration into tissues, notably bone, and oral clindamycin in particular is less likely than erythromycin to cause gastrointestinal upset at high doses. This makes clindamycin potentially useful in patients with a history of accelerated reactions to penicillins, when cephalosporins are contraindicated. Fluoroquinolones have been used but never strongly advocated for the treatment of S. aureus infections. They do provide some coverage when S. aureus is present in a mixed infection with aerobic Gram-negative bacteria, and oral therapy is adequate. It is likely that the new generation fluoroquinolones will supersede ciprofloxacin, and (lev) ofloxacin in this setting due to their greater intrinsic activity. Vancomycin and teicoplanin may occasionally be considered for treatment of life-threatening staphylococcal infection in patients with a history of accelerated allergic reactions to beta-lactams. It is not clear whether they are superior to macrolides or lincosamides in this setting, but they are usually preferred because of greater experience gained in the management of life-threatening MRSA infections. However, there is growing concern about their efficacy compared to beta-lactams. Caveats for their use are given below. The new classes of drugs, the streptogramins, oxazolidinones and lipopeptides, should be reserved for MRSA infections. There may, however, be the occasional need to use these agents for MSSA infections, especially if there is intolerance/allergy to other drug classes. Methicillin-Resistant S. aureus (MRSA): Guided Medline Search Vancomycin is the drug of choice for serious infections caused by S. aureus strains that are resistant to beta-lactam antibiotics and for patients who have potentially life-threatening allergy to the latter drugs. Recently, several anecdotal reports have questioned the efficacy of vancomycin for both MSSA and MRSA (66, 198, 232, 374). Vancomycin treatment of deep-seated staphylococcal infections such as endocarditis has been reported to clear bacteremia more slowly than beta-lactam treatment (268). Bacteremia is often prolonged more than 6 days in patients receiving vancomycin therapy (231, 232, 332, 374). In contrast, almost all the blood cultures of the patients receiving nafcillin or other beta-lactams become sterile within 6 days (56, 118, 211). Higher rates of relapse, complications, treatment failure and mortality in S. aureus bacteremia and endocarditis are associated with vancomycin therapy (57, 65, 66,132, 145, 146, 153, 152, 169, 250, 347, 374) Indeed, a recent study from Spain of bacteremic S. aureus pneumonia showed a significant mortality rate for vancomycin treatment of MSSA, compared to no mortality when patients were treated with cloxacillin (152). The slower bactericidal rate than beta-lactams has been suggested as a possible reason for the higher failure rate seen with vancomycin therapy in patients with MSSA endocarditis (374). However, it is possible that tolerance to vancomycin in some strains may also play a part (254,378). Slower eradication by vancomycin was considered an important factor in the delayed clinical response in a recent study of intensive care patients with S. aureus lower respiratory tract infections (266). It is possible that high doses may result in better efficacy. A prospective study in patients with staphylococcal lower respiratory tract infections showed that vancomycin AUC/MIC ratios of >400 gave much better clinical outcomes (78%) than ratios below that ratio (23%) (265, 266). In an effort to reduce costs and selective pressure for resistance, once daily dosing of vancomycin has been tried. However, early experience suggests that the risk of failure is high in patients with normal renal function (216, 247, 248,302). For patients with bacteremia, it is not mandatory to have chosen effective antibiotic therapy prior to blood cultures results (341). This is particularly important in settings where MRSA is prevalent. Thus, when S. aureus bacteremia is suspected, therapy can be initiated with beta-lactams without increase in mortality of those patients who subsequently prove to have MRSA (341). However, when MRSA is documented, failure to switch to an effective antimicrobial is associated with increased morbidity and mortality (178). For teicoplanin to be efficacious in the treatment of S. aureus endocarditis, the trough serum concentration should be maintained at 20 ug/mL during the first week of therapy (233, 239,444). Long-term teicoplanin therapy can rarely result in emergence of strains resistant to teicoplanin; however, these strains remain susceptible to vancomycin (191). The use of alternatives to vancomycin and teicoplanin might be considered for serious infections caused by caMRSA. Clindamycin is possibly the best choice if the organism is erythromycin susceptible. For strains with inducible clindamycin resistance, the weight of evidence suggests that there is a significant risk of failure or relapse, and clindamycin should not be used unless the infection is minor (234). There are a number of alternatives for oral therapy for less serious infection or step-down therapy. In general, combination oral therapy with two active agents is recommended for haMRSA. Most experience has been gained with rifampin, fluoroquinolones, and (in countries where it is available) fusidic acid . Chloramphenicol was used in the past but has fallen into disfavour. Linezolid is a suitable alternative as a single agent and has proven effective in the short and long term management of bone and joint (including prosthetic) infections (18, 325, 326). Pristinamycin has found a role in MRSA infections when patients are intolerant of other drugs (284). There has been a resurgence of interest in the role of trimethoprim-sulfamethoxazole for MRSA infections, given that the great majority of caMRSA are susceptible. A recent review of clinical studies suggested that while evidence of efficacy is still largely anecdotal, that in general this combination agent is effective (160). Staphylococcus aureus, glycopeptide-intermediate (GISA): Guided Medline Search The optimum treatment for vancomycin intermediate S. aureus has yet to be established. In Japan, most experience has been gained with the combinations of agents such as ampicillin-sulbactam and arbekacin, an aminoglycoside approved specifically for MRSA treatment in Japan (175). Current options include quinupristin-dalfopristin and linezolid. There is some animal model experience to support the use of ampicillin-sulbactam (14) and combinations of vancomycin and nafcillin (81). Other agents used over the years for the treatment of serious MRSA infections might also be considered, depending on their availability, such as fosfomycin (321). Some promise is being shown with lysostaphin, a peptidase produced by Staphylococcus similans, in experimental endocarditis models (82,298). [Review Article: Hiramatsu K. Vancomycin-resistant Staphylococcus aureus: a new model of antibiotic resistance. Lancet Infect Dis. 2001 Oct;1(3):147-55.] Considerations in the Choice of Regimen: The traditional approach to serious and deep staphylococcal infections has been to use high dose therapy intravenously for a number of weeks. With the exception of line sepsis with a removable focus and no evidence of metastatic seeding, where 10 to 14 days therapy is considered adequate, the general approach to deep infections has been to administer treatment for 4 to 6 weeks. In the past, this treatment has been administered intravenously and in hospital, but now many infections are managed by (i) completing the intravenous course as an outpatient (OPAT) (224, 442, 400) or (ii) by a switch to high dose oral therapy after an adequate intravenous course, such has been practiced for osteomyelitis in children for many years (158,199). Use of or combinations such as fluoroquinolones with rifampin have also been shown to be effective when introduced early in management (300, 361). Outpatient intravenous therapy is now a reasonable alternative in most situations where prolonged intravenous administration is felt to be optimum (224). For OPAT generally, drugs with prolonged half-lives such as ceftriaxone or teicoplanin are preferred as the number of administration visits is minimized. However, although ceftriaxone is thought to have only marginal activity against S. aureus, recent experience in acute and staphylococcal chronic osteomyelitis has shown acceptable cure rates (161). The short half-lives of the antistaphylococcal penicillins have made standard scheduling of these drugs impractical for outpatient use. Instead, continuous infusion is recommended for these agents and early experience with cloxacillin (422) flucloxacillin (223) and oxacillin (224) is very favourable. In many clinical circumstances it is probable that switch to high-dose oral therapy for completion of the course is adequate. Important exceptions are meningitis and endocarditis, where the high levels achieved by maximum IV doses are necessary to ensure adequate delivery of antibiotic. Apart from the published experience in pediatric osteomyelitis (158,199) and orthopedic implant infections (105, 106, 384), there are limited published data. Nevertheless, there is considerable experience with this approach, and hence Table 2 provides some suggestions for oral regimens following initial intravenous treatment. Special Situations Suggested antibiotics, doses, and duration for treatment of S. aureus infections are shown in Table 2. Some special situations are discussed below. Skin and Soft Tissue Infections Guided Medline Search Most causes of cellulitis are caused by streptococci and S. aureus, so that beta-lactam antibiotics with activity against penicillinase-producing S. aureus are the typical antibiotics of choice for uncomplicated cellulitis in an immunocompetent host. Parenteral cefazolin (1.0g intravenously every 6 to 8 hours) or nafcillin (1.0 to 1.5g intravenously every 4-6 hours) are most commonly-used. Oral antibiotics for established MSSA infections include clindamycin, dicloxacillin, doxycycline, trimethoprim-sulfamethoxazole. If MRSA is the suspected cause (nosocomial infection or infection with community acquired MRSA in areas of prevalence of such strains) and in penicillin-allergic individuals, vancomycin (1.0 g intravenously every 12 hours) is the most commonly used antibiotic. Daptomycin (4mg/kg every 24 hours intravenously) is an alternative agent for treatment of established MRSA soft tissue infections. For invasive cellulitis unresponsive to standard therapy, combination agents including rifampin and gentamicin (low dose) are often added, although clinical data supporting these additions are based on anecdotes. A switch to oral therapy (e.g. dicloxacillin, 0.5g every 6 hours) may be made after clinical responsive has occurred with objective improvement in erythema, tenderness, heat or swelling. First generation oral cephalosporins (cephradine, cephalexin, or cefadroxil) are common alternatives. linezolid 600 gm. every 12 hours, orally can be used for MRSA skin and soft tissue infections. Cellulitis in the setting of a diabetic foot infection may involve a much wider spectrum of potential pathogens and warrants broader antimicrobial coverage such as ampicillin/sulbactam (3.0g intravenously every 6 hours in adults) or other antimicrobial combinations targeting anaerobes as well as gram negative aerobes. Broadening the spectrum of initial antimicrobial therapy for additional bacterial species may be indicated in specific clinical settings. These include, human or animal bites, for which initial therapy might include ampicillin/sulbactam intravenously (1.5-3g intravenously every 6 hours) (MN) or amoxicillin/clavulanate (500 mg orally every 8 hours in an adult). In the setting of cellulitis after an abrasion or laceration occurring with salt water exposure, where Vibrio vulnificus might be the pathogen, treatment with doxycycline (200 mg intravenously per day in two divided doses) might be preferred. Doxycycline also covers MSSA. In the setting of cellulitis after an abrasion or laceration occurring with fresh water exposure, where Aeromonas hydrophila might be involved, treatment with ciprofloxacin (400 mg intravenously every 12 hours) might be preferred. Quinolones also cover MSSA, but are not the antibiotics of choice. A combination of ceftazidime plus gentamicin may be added to antistaphylococcal antibiotics for invasive infection in immunosuppressed patients. Recurrent or Persistent Staphylococcus aureus Skin and Soft Tissue Infection: A number of approaches have been devised although none of these have yet been validated in controlled trials. 1. Sustained and regular application bathing with a disinfectant. a. Chlorhexidine b. Hexachlorophene c. Bleach. Add 1 cup of household bleach per bathtub (or 20 gallons) Soak for no longer than 20 minutes twice a week for 3 months. No data exists for the comparative superiority or the ease of use for any of these three disinfectants. 2. Nasal mupirocin b.i.d. for 5 days every month or b.i.d. for 5 days every 3 months. Reculture anterior nares at 6 months (325a). 3. Keep nails clipped and short. Avoid scratching of pruritic areas. 4. Wear light clothing to minimize perspiration. 5. Use antibacterial soap for hand washing. 6. Aerate intertriginous areas. 7. Eradicate nasal carriage with Mupirocin in spouses. Administration of oral antimicrobial agents is discouraged for long-term use because of risk for emergence of S. aureus resistance to the antibiotics and the adverse effects of the antibiotics. For severe cases, an initial 10-14 day course of minocycline or trimethoprim-sulfamethoxazole (TMP-SMZ) might be given if the cutaneous infection has been persistent; addition of rifampin for combination therapy is also an option. Catheter-Related Bacteremia Guided Medline Search Several prospective studies have identified S. aureus as one of the leading causes of catheter-related bacteremia (31, 79, 85, 242, 368). More than 30% of S. aureus bacteremia is attributable to an infected intravascular device (58, 83, 118, 126, 157). It is a particular problem in certain clinical settings where central lines are frequently used such as haemodialysis (68, 250, 288, 301, 395), adult and neonatal intensive care areas (91, 335, 343) oncology (154) and coronary care units (334) Cure rates without removal of the intravascular device are generally below 20% (68,109). Risk factors for hematogenous complications include symptom duration, hemodialysis dependence, presence of a long-term catheter or non-catheter device, failure to remove the catheter and infection with MRSA (136). Most authors advocate shorter courses of antibiotic therapy for uncomplicated catheter-related S. aureus bacteremia, claiming that this infection has a low rate of complications after a 2-week course of intravenous antistaphylococcal antibiotics. On the other hand, a few studies have suggested that short-course therapy is inadequate, citing experiences with patients who had complications and relapse (181, 263, 253, 319, 434,445). In a well-reasoned meta-analysis, an average late complication rate of 6.1% was found for 11 studies totalling 132 patients treated with short-course therapy for uncomplicated catheter-related S. aureus bacteremia (181). These investigators also estimated the rate that might be observed after 4 to 6 weeks of therapy to be in the range of 0.07-0.99%. In another study (not included in the meta-analysis) of 12 patients with S. aureus bacteremia associated with intravenous catheter infection, three of eight patients who received 2-week treatment were considered failures: one each developed endocarditis 3 days and 7 weeks later, respectively, and one developed epidural abscess and meningitis after the first week of therapy and underwent 6 weeks of antibiotic therapy (318). A further recent study of 276 patients failed to show any relationship between relapse of deep-seated infection and duration of treatment (399). One study of S. aureus bacteremia in patients on chronic hemodialysis showed that less than 4 weeks of treatment was associated with a higher occurrence of primary treatment failure (definition not given) than treatment for more than 4 weeks (313), but this was not confirmed in another study (225). Ideally, catheters should be removed regardless of type (i. e. peripheral vs. central venous catheters, non-tunnelled vs. tunnelled catheters). Delayed removal of the infected catheter was associated with persistence of bacteremia (243). Only 18% of those with Hickman catheter-related S. aureus bacteremia and only 10% of patients with exit site infections are cured without catheter removal (109). In a study that controlled for other variables using logistic regression, it has been shown that patients in whom the intravascular device is not removed are 6.5 times more likely to experience complications or die than those in whom the device was removed (135). Some investigators have shown higher cure rates without catheter removal recently (345), and acidification of central lines may be able to increase the useful life of infected lines prior to removal (402). If the focus of infection has been promptly removed with rapid documented resolution of the bacteremia (< 3 days), 2 weeks of antibiotic therapy with a penicillinase-resistant penicillin, first-generation cephalosporin or glycopeptide is likely to be enough. Serious infectious complications such as septic thrombosis, deep-seated infections and sepsis-related death have often resulted from vascular catheter-related S. aureus bacteremia (315). If signs of endocarditis, metastatic infection, or prolonged bacteremia are present, longer therapy is needed (209). Large doses of oral dicloxacillin sodium (for MSSA) that can be taken at home for two weeks to supplement an initial 2-week intravenous regimen have been recommended (320). Under no circumstances should patients simply have the catheter removed without antibiotic treatment. It is likely that transesophageal echocardiography will become a standard investigation in catheter-related bacteremia to exclude complicating endocarditis, and further define the duration of therapy (2 or 4-6 weeks for negative and positive TOE respectively). A recent analysis has confirmed the cost-effectiveness of this strategy (344). [Victor L. Yu: Repeat blood cultures after 3 days in patients with Stapylococcus aureus bacteremia] Endocarditis Guided Medline Search S. aureus endocarditis is the predominant cause of endocarditis throughout the world. Among 1490 cases of Duke definite endocarditis collected from 5 sites in 4 countries, S. aureus was the single most common pathogen, accounting for 32.5% of cases (2). Endocarditis caused by S. aureus is a very serious disease associated with a high rate of morbidity and mortality. In non-injecting drug users it primarily involves valves in the left side of the heart and is associated with mortality rates of 25 to 40%. By contrast, staphylococcal endocarditis occurring in injecting drug users usually involves the tricuspid valve (right-sided endocarditis) and has a low mortality. Using traditional diagnostic means (clinical criteria and transthoracic echocardiography (TTE)), the incidence of infective endocarditis in patients with community-acquired S. aureus bacteremia ranges from 6 to 64% (23, 128, 187, 207, 219, 263, 269, 277, 338, 347, 365,445) although rates are generally below 15%. The introduction of transesophageal echocardiography (TEE) has substantially increased the rate of diagnosis of endocarditis in the setting of bacteremia. In unselected community plus hospital-acquired cases, TEE detects endocarditis in upwards of 25% of cases (132, 388). As might be expected, patients with a positive TTE have higher rates of embolism and death than those whose TTE is negative but TEE is positive (134). MSSA Endocarditis: Guided Medline Search The drugs of choice for endocarditis (Table 3) caused by MSSA are the semisynthetic penicillinase-resistant penicillins. Benzylpenicillin is preferred in the uncommon circumstance that the strain is shown to be penicillin susceptible. First-generation cephalosporins are effective alternatives, usually reserved for patients with a history of minor penicillin allergy. There is lingering concern about cefazolin , however, as highlighted by a recent case of relapse in a strain of MSSA producing type A beta-lactamase (281). The third-generation cephalosporin, ceftriaxone, has been used but the relapse rate is unacceptably high at 28%, even when combined with gentamicin for 2 weeks (97). Vancomycin is currently the drug of choice for patients with life-threatening penicillin allergy and those with endocarditis due to MRSA. Daptomycin is an alternative treatment option at a dose of 6 mg/kg once daily. The efficacy of Daptomycin in the treatment of patients with S. aureus bacteremia was evaluated in a randomized, multicenter open-label study (136b). Patients received either IV daptomycin or a comparator regimen of vancomycin, plus initial low-dose gentamicin if the patient had MRSA, or a semisynthetic penicillin plus initial low-dose gentamicin, if the patient had methicillin susceptible S. aureus. Success rate of daptomycin was 45% compared to comparator of 49% for MSSA endocarditis and 44% and 32% for MRSA endocarditis, respectively. Clindamycin has been successfully used for S. aureus endocarditis, but clinical data are limited. Relapse and treatment failure are documented (44,72,406,435). Moreover, the efficacy against strains which are erythromycin resistant and have inducible lincosamide resistance is unresolved, but accumulating evidence suggests that it is unlikely to be efficacious (234, 297, 323), and clindamycin is therefore not recommended for endocarditis caused by them. Although vancomycin is still recommended therapy for S. aureus endocarditis in patients with life-threatening penicillin allergy, recent experience suggests caution, as suboptimal outcomes have been associated with the use of this agent. One early study showed failure rates of approximately 40% have been documented in patients with S. aureus endocarditis treated with vancomycin despite right-sided involvement (374). Evidence supports the contention that patients with MSSA bacteremia and endocarditis treated with vancomycin fare worse than those treated with beta-lactams (66, 250, 145, 392 and see above) and alternative drugs or strategies are keenly sought. beta-lactam desensitization should be carefully considered for MSSA endocarditis in patients with beta-lactam allergy and suboptimal response to vancomycin (21). In patients with a questionable history of immediate-type hypersensitivity to penicillins, decision analysis confirms that skin testing prior to starting therapy has the best cost-utility (101). Alternative methods of antibiotic dosing are currently being examined. Continuous infusion beta-lactams are being used increasing in endocarditis. Early experience has been favorable (223). Left-Sided Endocarditis: A minimum of 4 weeks of intravenous treatment is recommended for left-sided endocarditis (426). This duration is also recommended for S. aureus septicemia complicated by metastatic infection, on the presumption that endocarditis may well be present. Of 20 patients with S. aureus endocarditis receiving at least 4 weeks of treatment, all were cured at 1 month follow-up (318). For endocarditis occurring on prosthetic devices, a 6-week course of a penicillin with an aminoglycoside has been recommended (426). The combination of nafcillin and gentamicin was associated with a more rapid clearance of bacteremia in MSSA endocarditis. Addition of gentamicin for the first 2 weeks of a 6-week course of intravenous nafcillin therapy led to more rapid defervescence but did not improve the cure rate (211). However, an increased risk of renal dysfunction was observed (211). Addition of gentamicin for the first 3 to 5 days of therapy may avoid nephrotoxicity and should be considered (30). A case can be made for not adding gentamicin for these reasons, and based on a recent retrospective study of mon- versus combined therapy (108). Although aminoglycosides have been added to the vancomycin regimen for MRSA endocarditis, this addition should be restricted to endocarditis caused by aminoglycoside-susceptible strains, and aminoglycoside use should be limited to 3 to 5 days (30) to minimize synergistic toxicity. Treatment of MRSA endocarditis is problematic. Vancomycin remains the treatment of choice, but the efficacy possibly suboptimal. The emergence of vancomycin-intermediate S. aureus has further compromised vancomycin, and failures are observed even with strains that are heterogeneously resistant (267, 391). For patients with MRSA endocarditis unresponsive to vancomycin, addition of rifampin and/or gentamicin (if susceptible) should strongly be considered (65,66). The use of minocycline, trimethoprim-sulfamethoxazole, linezolid, ciprofloxacin plus rifampin, daptomycin can also be considered if the strain is susceptible, although experience with these regimens is limited (19, 110, 111, 125, 136b, 249, 228). Caution should be used with the quinolone-rifampicin combinations as data from animal models suggest antagonism for some strains (59). Quinupristin/dalfopristin is also an alternative agent, although no comparative studies have been performed. The in vivo effect of rifampin in combination with nafcillin, oxacillin, vancomycin, or aminoglycosides is highly variable. Routine use of rifampin is not recommended for treatment of native valve S. aureus endocarditis. Although the addition of rifampin to vancomycin for patients with MRSA endocarditis failed to show either enhanced survival or reduced duration of bacteremia in comparison with vancomycin alone (232), it could be used as a supplemental therapy in patients who do not respond adequately to conventional antimicrobial therapy. In support of this, one recent retrospective analysis of septicemia due to EMRSA-15 suggested that rifampin may have played a role in the prevention of deaths caused by this epidemic MRSA (45). Breakthrough bacteremia was shown in patients with S. aureus endocarditis receiving teicoplanin in a graduated dosing regimen of 20mg/kg/day for 3 days, 12 mg/kg/day for 4 days, and 7 mg/kg/day thereafter (131,226). Teicoplanin is not currently recommended as initial therapy for severe S. aureus infection (i. e., septicemia and endocarditis). Right-Sided Endocarditis: Right-sided endocarditis has a high cure rate, and carefully selected regimens as short as 2 weeks have been efficacious, provided there are no other foci of infection (57, 98, 333, 405). Effective regimens have included intravenous nafcillin and tobramycin (57), intravenous cloxacillin and amikacin (405), and intravenous cloxacillin with and without gentamicin (333) and daptomycin (136b). Four weeks of oral ciprofloxacin plus rifampin was also shown to be effective (170). Prosthetic Valve Endocarditis: Because prosthetic valve staphylococcal endocarditis has a high mortality, combination therapy seems prudent. It is clearly associated with better bacteriological outcomes (108). The addition of rifampin and gentamicin to the beta-lactam or vancomycin is most commonly recommended, although controlled studies have not been done. Gentamicin should be given for 2 weeks rather than for 3 to 5 days as with native valve endocarditis. A favorable response of MRSA prosthetic valve endocarditis to minocycline therapy after unsuccessful treatment with vancomycin has been reported (220). Because of the very limited clinical data, minocycline should probably be regarded as an alternative agent for MRSA endocarditis. Daptomycin or combinations using daptomycin might be considered, although no data exist to support this option. Survival is probably more dependent on whether valve replacement surgery can be performed than on the choice of antibiotic regimen (182,455). Recent evidence favours ceasing rather than continuing with anticoagulant treatment to reduce morbidity from cerebral events (404). [Victor L. Yu: Repeat blood cultures after 3 days in patients with Stapylococcus aureus bacteremia] Meningitis Guided Medline Search Meningitis caused by S. aureus is often found early after neurosurgery or trauma in those with cerebrospinal fluid shunts (203) and can occur spontaneously as part of a staphylococcal sepsis syndrome (308). Other underlying conditions include diabetes mellitus, alcoholism, chronic renal failure requiring hemodialysis, intravenous drug abuse, and malignancy (155, 203,357). Mortality rates have ranged from 14 to 77%. Like all penicillins, penicillinase-resistant penicillins show minimal penetration into the cerebrospinal fluid in the absence of inflammation, but acceptable concentrations are seen in the presence of inflammation, despite the high level of protein binding (213). Hence, recommended treatment of meningitis caused by MSSA is a penicillinase-resistant penicillin in high dose (120, 203, 357). For MRSA and for patients allergic to penicillin, vancomycin 2 g/day for adults and 60 mg/kg daily for children |