Fusidic acid has activity against a range of pathogens but has mainly been used to treat staphylococcal infections. Fusidic acid monotherapy, especially topical preparations, has been strongly associated with the emergence of fusidic acid resistance among both methicillin-resistant Staphylococcus aureus (MRSA) and methicillin-susceptible S. aureus. Key resistance determinants include mutations in the fusA gene, which encodes elongation factor G, and plasmid-mediated resistance (i.e., acquisition of resistance gene fusB). Clonal outbreaks of fusidic acid—resistant S. aureus have been noted throughout the United Kingdom and Europe, such that the efficacy of fusidic acid is threatened. Fusidic acid in combination with other agents, such as rifampicin, has proven effective for difficult-to-treat MRSA infections and provides a convenient oral alternative to oxazolidinones. Ensuring that systemic fusidic acid is always used in combination and that the use of topical fusidic acid is either abolished or restricted will be vital if we are to prevent the loss of this potentially useful agent.
Fusidic acid is derived from the fungus Fusidium coccineum and was developed by Leo Laboratories in Copenhagen, with the most active derivative, the sodium salt (sodium fusidate), released for clinical use in the early 1960s [1]. Since that time, fusidic acid has been widely used throughout Europe and Australia, particularly for the treatment of staphylococcal infection. For reasons that are unclear, US Food and Drug Administration licensure for fusidic acid has never been sought; thus, the drug is not currently available in the United States, despite the fact that it provides a potentially useful option (usually in combination with rifampicin) for the treatment of infection with multidrug-resistant staphylococci, including methicillin-resistant Staphylococcus aureus (MRSA). Although resistance to fusidic acid was recognized as a potential problem soon after its release, clinically significant rates of resistance, in association with the widespread and often inappropriate use of topical fusidic acid (monotherapy) ointment/cream for chronic skin conditions, have emerged mainly over the past few years [2]. In the present review, while describing the in vitro and clinical efficacy of fusidic acid, we primarily focus on the factors leading to the emergence of resistance, the mechanisms of this resistance, and the strategies that may be applied to prevent the loss of this potentially useful antistaphylococcal agent.
Fusidic acid is primarily active in vitro against various strains of staphylococci, including methicillin-susceptible S. aureus (MSSA), MRSA, heterogeneous and nonheterogeneous vancomycin-intermediate S. aureus (hVISA and VISA), and most coagulase-negative staphylococci [1, 3]. Corynebacteria and gram-positive anaerobes, such as clostridia (Clostridia tetani, Clostridia perfringens, and Clostridia difficile) and Peptococcus and Peptrostreptococcus species, are susceptible, whereas fusidic acid has only limited activity against streptococci and enterococci. Most gram-negative bacteria are resistant, except for Neisseria and Moraxella species, Legionella pneumophila, and some strains of the Bacteroides fragilis group. Interestingly, the drug has good in vitro and clinical activity against Mycobacterium leprae, and some Mycobacterium tuberculosis strains are borderline susceptible. Fusidic acid also has some in vitro activity against Coxiella burnetii [1, 4].
Although there are no Clinical and Laboratory Standards Institute (CLSI)—defined breakpoints for fusidic acid, susceptibility is generally defined as an MIC of ⩽0.25 or ⩽0.5 mg/L and resistance as an MIC of ⩾2 mg/L [4, 5]. By CLSI disc susceptibility methods [4], fusidic acid susceptibility (for a 2.5-µg disc) is defined by a zone of ⩾22 mm, whereas intermediate and resistant zones are 18–21 mm and ⩽17 mm, respectively [6, 7].
In vitro studies assessing the interaction between fusidic acid and other antimicrobial agents have produced variable results with S. aureus. Fusidic acid plus β-lactams have demonstrated antagonism, indifference, and synergy, depending on the study; fusidic acid plus fluoroquinolones results in antagonism; and fusidic acid in combination with glycopeptides results in indifference, whereas, for rifampicin, indifference has been reported unless time-kill methods are used, in which case synergy has been observed [4]. The clinical relevance of these in vitro observations is unclear, although combination therapy with β-lactams or rifampicin appears to be associated with lower rates of fusidic acid resistance (see Factors Associated with the Emergence of Fusidic Acid Resistance).
A crucial stage in bacterial protein biosynthesis is the elongation phase, in which the nascent polypeptide grows as the ribosome moves along the mRNA in a stepwise fashion. Two elongation factors, EF-Tu and EF-G, are intimately involved in this process, with EF-G particularly associated with the translocation step in which the mRNA is advanced along the ribosome by 1 codon to allow the next round of polypeptide elongation to begin [8]. Fusidic acid blocks bacterial protein synthesis by binding to EF-G on the ribosome, thereby preventing release of the EF-G—guanosine diphosphate complex and effectively stalling bacterial protein synthesis by inhibiting the next stage in translation [8–10]. The action of fusidic acid is mainly bacteriostatic but, at high concentrations, may be bactericidal. The gene encoding EF-G is fusA, which is chromosomally located.
Fusidic acid is available in intravenous, oral, and topical (skin and ophthalmic) preparations and, when given systemically, is widely distributed throughout the body, including areas such as bone, joint fluid, prostate, and large abscesses. CSF penetration through noninflamed meninges is poor, but reasonable concentrations in cerebral abscesses have been reported. Oral fusidic acid is the most popular preparation (usually 500 mg po every 8–12 h), achieving serum concentrations similar to those obtained through intravenous administration [1].
Fusidic acid is mainly used to treat staphylococcal infections, for which it is usually combined with another antistaphylococcal agent to minimize the emergence of resistant clones, since naturally occurring fusidic acid—resistant strains occur at a rate of 1 in 106–108 colony-forming units (CFU) (see Factors Associated with the Emergence of Fusidic Acid Resistance). It can be particularly useful in treating infections for which there are few other oral treatment alternatives, such as infections due to fluoroquinolone-resistant MRSA, thereby avoiding the need for agents such as the oxazolidinones. In Australia, for example, the oral combination of rifampicin and fusidic acid is commonly used to treat soft tissue/skin and bone/joint infections due to MRSA, after initial effective control with vancomycin therapy. Data to support the use of fusidic acid for these indications are reasonably substantive and have been reviewed recently [11–14]. In other countries, particularly in Europe, fusidic acid is added to either β-lactams or glycopeptides for the treatment of staphylococcal bacteremia, endocarditis, and osteomyelitis [11, 13]; however, despite the recommendation of these combination regimens by some antibiotic treatment guidelines, there are only limited published clinical data to support their use in these settings [11]. Table 1 summarizes the indications for which fusidic acid has been clinically effective. A key area in which there has been a notable recent increase in fusidic acid monotherapy is topical ointments and creams for the treatment of acute skin infections, including impetigo and potentially infected atopic dermatitis (usually given in combination with topical glucocorticoids) [2]. Such topical therapy has proven effective (table 1) but has also been associated with significant emergence of resistance.
The understanding of the mechanisms of resistance to fusidic acid is rather limited and mainly focuses on S. aureus. The major resistance mechanisms appear to be related to alterations in EF-G structure, leading to reduced fusidic acid binding, or to acquisition of the fusidic acid resistance gene, fusB, which causes resistance by an undetermined mechanism. Plasmid-mediated acquisition of resistance determinants appears to be common, and a recent European study of an epidemic clone of fusidic acid—resistant S. aureus found that presence of the fusB gene was the predominant mechanism of resistance [26]. However, inactivation or sequestering of fusidic acid by enzymes has also been described in bacteria other than S. aureus [27].
fusA mutations leading to altered EF-G. Mutations in the chromosomally located fusA gene that lead to individual amino acid exchanges in EF-G are an important mechanism of fusidic acid resistance, presumably as a result of decreased affinity of the drug for the target [28]. This type of resistance is thought to be harbored naturally in S. aureus, occurring at a frequency of 1 in 106–108 CFU and being associated with a shift in fusidic acid MIC to >12 µg/mL [2, 27]. However, different mutations lead to different changes in MIC, and 2 S. aureus strains with low-level fusidic acid resistance that have nucleotide substitutions in fusA have been described recently [26].
fusB acquisition. The predominant mechanism of fuscidic acid resistance in S. aureus is acquisition of a plasmid-mediated fusB resistance determinant (a 21.9-kb plasmid pUB101) that encodes modest levels of fusidic acid resistance, as well as β-lactamase and cadmium resistance [29], and that was first described soon after fusidic acid was introduced [30]. However, in a recent European study of a large outbreak of fusidic acid—resistant S. aureus infection, the fusB gene in some isolates was found to be located on a chromosome, rather than on a plasmid [26]. These isolates had fusidic acid MICs of 4 mg/L, in contrast to S. aureus strains that harbored fusB on the pUB101 plasmid, which had MICs of 16 mg/L [26]. The mechanism of fusB-mediated resistance remains unclear; early studies that suggested that it encoded a fusidic acid permeability barrier [31] are now contradicted by studies showing no alterations in membrane composition in fusidic acid—resistant S. aureus strains [29].
Other mechanisms of resistance. Recently, 4 nonepidemic fusidic acid—resistant S. aureus strains with low-level resistance have been identified in Europe that do not harbor either the fusB resistance determinant or mutations in the fusA gene, thereby suggesting that other mechanisms of resistance may occur in S. aureus, as has been described for other species. These include binding and sequestering of fusidic acid by the type 1 chloramphenicol acetyltransferase found in enterobacteriaceae, deacetylation by an esterase produced in Streptomyces species [27], and efflux by the AcrAB efflux system in Escherichia coli [32].
A large, multicenter, worldwide 1996 study of staphylococcal resistance suggested that the rate of fusidic acid resistance among S. aureus was highly variable [33]. Overall, among the 4065 isolates collected from consecutive patients (20 hospitals; 19 countries), the mean rate of fusidic acid resistance was 5% (median, 1%; range, 0%–49%), with the highest rates of resistance reported in Greece (49%), Kuwait (20%), and New Zealand (13%), especially in dermatology (10% in all countries) and intensive care units (8%). Typing studies suggested that, in each of these countries, a single clone was primarily responsible for the high rates (MRSA in Greece and Kuwait; MSSA in New Zealand). Notably, low rates of fusidic acid resistance were observed in all the large US hospitals studied (located in Colorado, California, and New Jersey) [33]. In Australia, until 1999, fusidic acid resistance was found in ∼5% of MRSA and MSSA strains, although topical fusidic acid was only introduced into the country in the mid-1990s [34].
In 2000–2002 in the United Kingdom, there were increasing reports of clonal spread of fusidic acid—resistant strains of S. aureus, including the emergence of a multiply resistant hospital-acquired MRSA strain, UK EMRSA-17 [35]. Similarly, a number of Scandinavian and European countries have reported clonal outbreaks of childhood impetigo due to fusidic acid—resistant MSSA, with low-level resistance apparently related to the presence of chromosomal fusB [26, 36, 37]. Fusidic acid resistance has also been described among non—multidrug-resistant strains of community-associated MRSA (CA-MRSA), including 1 outbreak among intravenous drug users in Liverpool, United Kingdom [38]. Similarly, fusidic acid resistance has been a feature of 1 of the common CA-MRSA strains in Australia (WA-MRSA-1), although these account for only a small percentage of total Australian MRSA isolates [39].
In vitro and animal studies. Naturally occurring resistant subpopulations of S. aureus with mutations in the fusA gene have been isolated from patients who have never been exposed to fusidic acid [40]. O'Neill et al. [41] found that such chromosomal mutations occurred at a rate of 10-7 to 10-8 when S. aureus strains (MSSA, MRSA, and strains with reduced vancomycin susceptibility) were exposed to fusidic acid concentrations of 10 mg/L (which approximates Cmin levels after standard dosing in humans), but no mutants were detected when fusidic acid concentrations of 15 or 30 mg/L (which approximate Cmax levels after standard dosing) were used. However, a recent study that included CA-MRSA strains suggested that exposure to fusidic acid alone at 64 times the MIC was still associated with the development of resistance [42].
In an animal model of MRSA endocarditis, fusidic acid monotherapy lead to development of resistance in 5 of 12 animals treated and was associated with fusidic acid treatment failure. Development of resistance was particularly associated with high bacterial inocula, and combination therapy with vancomycin appeared to prevent such resistance [43].
Clinical studies: systemic fusidic acid therapy. Numerous clinical studies have reported the emergence of fusidic acid resistance among S. aureus strains either during fusidic acid monotherapy or after fusidic acid combination therapy; these studies have been summarized recently [27]. Overall, however, fusidic acid resistance appears to emerge less frequently when combination antistaphylococcal therapy is used (table 2).
The rate of emergence of fusidic acid resistance in >850 patients who received systemic fusidic acid monotherapy for a range of clinical conditions was ∼5.1% [27, 44], but most of these patients received short courses (<2 weeks) of fusidic acid. Among the patients with chronic osteomyelitis who received prolonged courses of fusidic acid monotherapy, a high rate (15%) of resistance was noted, whereas, among patients with burns, a very high rate (46%) of fusidic acid resistance emergence was reported despite a short duration of therapy.
In comparison, the rate of resistance among >800 patients who received fusidic acid combination therapy was ∼0.8%, despite the fact that many patients had serious staphylococcal infections (bacteremia, acute and chronic osteomyelitis, MRSA infection, and hVISA/VISA infection) that often required prolonged therapy. In many of these studies, fusidic acid was combined with a β-lactam (penicillin, flucloxacillin/cloxacillin, or methicillin), but other combinations included rifamycins, erythromycin, chloramphenicol, lincomycin, and novobiocin [3, 27].
Clinical studies: topical fusidic acid therapy. Topical preparations containing fusidic acid have been increasingly used in some regions (especially the United Kingdom and Europe) to treat a range of dermatological problems, particularly skin infections (e.g., impetigo, folliculitis, and traumatic wounds) and atopic dermatitis. In the United Kingdom, the use of topical fusidic acid almost doubled from 1995 to 2001 and currently constitutes about two-thirds of the total usage [2]. This increase has coincided with increased reports of fusidic acid—resistant S. aureus from many centers, particularly among patients receiving long-term topical fusidic acid therapy [2]. Shah et al. [45] reviewed all dermatology patients (age range, 6 months to 75 years) for whom there was a positive S. aureus culture during a 4-month period at a district hospital in the United Kingdom, and they found that 62% of patients had used topical fusidic acid during the previous 6 months and that a significantly higher proportion of isolates from dermatology patients were fusidic acid resistant, compared with those from nondermatology patients (51% vs. 9.6%). Patients with atopic eczema were most affected, with 78% of isolates being fusidic acid resistant. Among patients infected with fusidic acid—resistant or susceptible S. aureus isolates, 96% and 29%, respectively, had used topical fusidic acid therapy in the previous 6 months. Similar findings have been noted by Ravenscroft et al. [46] in the United Kingdom and by others in The Netherlands and elsewhere [47, 48].
Mason et al. [49] recently reported a significant (P = .01) association between high rates of general practice prescribing of topical fusidic acid and the emergence of fusidic acid resistance among strains of MSSA. Similarly, a retrospective case-control study performed by those same authors in 2002 suggested that exposure to topical fusidic acid during the previous 6 months was significantly associated (OR, 2.77; P = .027) with subsequent infection with fusidic acid—resistant MSSA [50].
It is difficult to determine to what extent the widespread dissemination of individual fusidic acid—resistant S. aureus clones has contributed to these findings from the United Kingdom. Other non-European studies have noted less dramatic results, but these have been confounded by a number of factors, including the use of topical combination therapy (fusidic acid and gentamicin) rather than monotherapy, and treatment courses of brief duration [51].
Nevertheless, whether the high rates of cutaneous colonization and infection with fusidic acid—resistant S. aureus are clonal or nonclonal, they appear to impact the rate of resistance among invasive S. aureus isolates, because the rate of fusidic acid resistance in MSSA bacteremia isolates in the United Kingdom increased from 2.0% to 6.4% from 1990 to 2001 (figure 1) [52]. Livermore et al. [52] also noted that the rates of fusidic acid resistance among MRSA and MSSA strains were roughly equal during this time.
Number of methicillin-susceptible Staphylococcus aureus (MSSA) and methicillin-resistant S. aureus (MRSA) bloodstream isolates and percentage of those isolates that were fusidic acid (FA) resistant in the United Kingdom, 1990–2001. Prior to 1994, the numbers of MRSA isolates were very low, and percentage rates of FA resistance were variable. For those years, MRSA isolate numbers (and the percentage that were FA resistant) were as follows: 1990, 82 (1.8%); 1991, 74 (12.5%); 1992, 131 (7.9%); and 1993, 207 (10.6%). Based on data from [52].
Given current trends regarding the spread and clinical impact of hospital- and community-acquired MRSA infection, it would seem to be wise to maintain all possible treatment options, including the potential use of fusidic acid. Especially because combination therapy with rifampicin can be a useful oral treatment option for less severe MRSA infections, such as those involving the skin or soft tissue, and for chronic osteomyelitis or prosthetic joint infections for which long-term parenteral therapy has been unsuccessful and/or for which surgery is not a practical option (e.g., in the elderly population). Thus, prevention of the emergence of fusidic acid resistance (especially when high numbers of organisms and intrinsic resistance are present) by restricting the use of systemic and topical fusidic acid monotherapy appears to be a worthwhile goal, given its documented association with the emergence of fusidic acid resistance. In particular, the use of topical fusidic acid monotherapy, especially for prolonged periods to treat chronic skin disorders, should be reconsidered, because it appears to be a major driver of resistance and clonal selection. Common sense would suggest that antibiotics used topically should be ones that are not used systemically [2]. When systemic fusidic acid is used appropriately in combination with other agents (e.g., rifampicin), resistance rates appear to remain low, even in regions where usage has been common for many years. To lose the efficacy of this useful agent because of inappropriate use would be dumb.
Potential conflicts of interest. B.P.H. and M.L.G.: no conflicts.
IDSA Members: For your free access to this journal, log in via the IDSA members area.
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy