Staphylococcus aureus bacteremia with reduced susceptibility to vancomycin

INTRODUCTION — Treatment failure for methicillin-resistant Staphylococcus aureus bacteremia with vancomycin (or with teicoplanin, a glycopeptide structurally related to vancomycin used outside the United States) should prompt consideration of infection due to S. aureus with reduced susceptibility to vancomycin.

Reduced S. aureus susceptibility to vancomycin may occur via the following mechanisms (see 'Overview of mechanisms' below):

●Increased vancomycin minimum inhibitory concentration (MIC); known as "MIC creep"

●Cell wall alterations (as in the case of vancomycin-intermediate S. aureus)

●Plasmid-mediated horizontal gene transfer (as in the case of vancomycin-resistant S. aureus)

Issues related to the definitions, mechanism, epidemiology, and treatment of S. aureus bacteremia with reduced susceptibility to vancomycin will be reviewed here. Other issues related to S. aureus infection are discussed separately. (See "Clinical approach to Staphylococcus aureus bacteremia in adults" and "Methicillin-resistant Staphylococcus aureus (MRSA) in adults: Treatment of bacteremia" and "Methicillin-resistant Staphylococcus aureus (MRSA) in adults: Prevention and control".)

DEFINITIONS

MIC breakpoints — The Clinical and Laboratory Standards Institute (CLSI) and the US Food and Drug Administration have established the following vancomycin minimum inhibitory concentration (MIC) interpretive criteria for S. aureus. The definitions were modified in response to increasing reports of vancomycin treatment failure in infections due to strains with elevated MICs (2 mcg/mL) as well as to flag those isolates that are likely to be heteroresistant [1-3].

●Vancomycin susceptible: ≤2 mcg/mL

●Vancomycin intermediate: 4 to 8 mcg/mL

●Vancomycin resistant: ≥16 mcg/mL

The acronyms for vancomycin-intermediate S. aureus (VISA), glycopeptide-intermediate S. aureus (GISA), and vancomycin-resistant S. aureus are derived from these criteria. VISA and GISA refer to the same susceptibility cutoff. The term VISA is more commonly used, although the term GISA may be more accurate since early reports indicated that most of these strains also had intermediate resistance to the glycopeptide teicoplanin.

The European Committee on Antimicrobial Susceptibility testing breakpoints differ from the CLSI breakpoints; ≤2 mcg/mL is considered susceptible and >2 mcg/mL is considered resistant [4].

Reduced vancomycin susceptibility can rarely occur in S. aureus irrespective of background methicillin susceptibility and may result in increased tolerance to several classes of antibiotics [5,6].

Heteroresistance — Heteroresistant VISA (hVISA) refers to VISA strains in which subpopulations display variable rather than uniform susceptibility to vancomycin [7,8]. Heteroresistant strains of S. aureus contain subpopulations of bacteria with vancomycin MICs in the intermediate range, while the vancomycin MIC for the entire population of the strain remains within the susceptible range [8]. Like VISA strains, hVISA populations have accumulated mutations that help withstand vancomycin by means of an unusually thickened cell wall. (See 'Overview of mechanisms' below and "Overview of antibacterial susceptibility testing", section on 'Heteroresistance'.)

The prototype heteroresistant strain was S. aureus Mu3, a clinical isolate recovered from a Japanese patient with staphylococcal pneumonia [8]. The reported frequency of heteroresistance is variable. Most studies describe a frequency of 0.5 to 1.5 percent, but frequencies as high as 20 percent have been reported in Japan [2,8,9].

It is uncertain whether heteroresistance is a cause of vancomycin treatment failure [10-12]. However, there are several reports of vancomycin failure and persistent infection due to hVISA [13-15]. In a review of 65 patients with endocarditis due to MRSA, 19 isolates (29 percent) exhibited the hVISA phenotype by population analysis profiling [15,16]. Patients with these isolates were more likely to have persistent bacteremia (68 versus 37 percent) and heart failure (47 versus 19 percent) [15]. Although screening for heteroresistance in clinical laboratories is not routinely performed, two studies noted the continued presence of these strains [17].

OVERVIEW OF MECHANISMS — Reduced S. aureus susceptibility to vancomycin can occur via the following mechanisms [18]:

●Increase in the minimum inhibitory concentration (MIC) of methicillin-resistant S. aureus (MRSA) isolates to vancomycin (known as "MIC creep") – While these isolates have not necessarily reached the level of vancomycin resistance as vancomycin-intermediate S. aureus (VISA), they have been at the upper limits of the susceptible range. This has raised concern regarding the efficacy of vancomycin for treatment of clinical infections caused by these organisms [3,19]. (See 'Borderline vancomycin susceptibility' below.)

●Vancomycin-intermediate S. aureus (VISA) – The first clinical VISA isolate described in 1997, as well as subsequent isolates, were noted to have an unusually thickened cell wall containing excessive vancomycin target dipeptides (D-Ala-D-Ala) capable of "trapping" vancomycin, thereby reducing availability of the drug for cellular targets [7,20-29]. (See 'Vancomycin-intermediate S. aureus' below.)

●Vancomycin-resistant S. aureus (VRSA) – The first clinical VRSA isolate, described in 2002, emerged via vanA gene plasmid-mediated transfer from enterococci with vancomycin resistance to S. aureus via mobile genetic element Tn1546 (similar to mechanism of vancomycin-resistant enterococci emergence) [30-34]. (See 'Vancomycin-resistant S. aureus' below.)

Borderline vancomycin susceptibility — Issues related to borderline vancomycin susceptibility (when isolates approach the limit of the susceptible range [2 mcg/mL]) are discussed further separately.

Vancomycin-intermediate S. aureus — The genetic basis for cell wall alterations usually involves the stepwise acquisition of multiple mutations in genes that are two component regulatory systems [35]. Several studies have identified mutations in selected genes, including vraR, graRS, and walRK, that appear to contribute to cell wall synthesis and, as a result, to the development of resistance [36-40]. Downregulated genes also reduce expression of autolysins reducing cross-linking and autolysis [35,41]. The group II polymorphism at the accessory gene regulator locus is present in some VISA strains (as well as some MRSA strains) [29,42]. The increased metabolic demand caused by these thickened cell walls places these strains at a fitness disadvantage and may in part explain the limited instances of nosocomial spread of these strains in health care settings. In general, these strains appear to have attenuated virulence [35].

VISA strains often develop from MRSA strains that are exposed to vancomycin for prolonged periods of time; this was suggested by the similarity between pulse-field gel electrophoresis patterns of the MRSA and subsequent VISA strains isolated from individual patients [8,43-47]. This course of events may have resulted from failure to eradicate the initial MRSA strain or subsequent reinfection with the same strain.

The colonial morphology of VISA strains may be different (eg, smaller or with different pigmentation) in part due to their slower growth rate [31]. (See 'Diagnosis and laboratory testing' below.)

Vancomycin-resistant S. aureus — A number of different plasmids have been identified that are capable of transferring the Tn1546 transposon, including those containing both enterococcal and staphylococcal plasmid deoxyribonucleic acid (DNA) [48]. VRSA resistance is due to the presence of the van gene cluster. Resistance is due to the synthesis of an alternative cell wall terminal peptide (D-ala-D-lac) rather than the normal terminal peptide (D-ala-D-ala). Vancomycin is unable to bind to the first peptide. Expression of the vanA genes is only initiated by exposure to vancomycin so that there is limited effect on the fitness of the isolate (in contrast with the VISA strains). The instability of the plasmids in these strains may be one of the factors contributing to the failure of these strains to spread [49].

EPIDEMIOLOGY — Cases of S. aureus infections with reduced susceptibility to vancomycin continue to be described worldwide [50,51].

VISA — Several vancomycin-intermediate S. aureus (VISA) strains associated with clinical infection have been described [7,10,43-47,52,53]. The first reported case was observed in 1997 in Japan and occurred in a four-month-old with a surgical site infection. This original VISA strain was designated Mu50; its vancomycin minimum inhibitory concentration (MIC) was 8 mcg/mL [7]. The patient was successfully treated with amoxicillin-clavulanate plus gentamicin. However, the first infection with VISA appears to have occurred in 1995 in France in a two-year-old with leukemia and central catheter–associated bacteremia; management was successful with surgical drainage and quinupristin-dalfopristin [52].

Subsequently, clinical infections with similar strains have been reported in the United States and around the world [10,43-46,54-56]. Features common to many of the United States cases included ongoing or recent dialysis, methicillin-resistant S. aureus (MRSA) bacteremia associated with central venous catheters or prosthetic graft material, and prolonged vancomycin exposure (6 to 18 weeks) in the three to six months preceding infection [43,45-47,53]. Contact investigation for two patients with VISA infection (including 177 contacts) yielded no VISA carriers [43]. There have also been a limited number of outbreaks associated with VISA strains, primarily in intensive care unit settings [57,58].

In general, the incidence of VISA is difficult to estimate given the rarity of infection and challenges related to laboratory detection. Surveillance data from the United States and Europe indicate that S. aureus isolates with vancomycin MICs ≥4 mcg/mL represent less than 0.3 percent of MIC values [2].

VRSA — There have been scattered reports of infection due to vancomycin-resistant S. aureus (VRSA) around the world.

Worldwide, 52 cases have been reported; in the United states, 14 cases have been reported [48,59-67]. The first United States case of VRSA was reported in 2002, isolated from a newly infected chronic foot ulcer in a patient in Michigan with diabetes, peripheral artery disease, and chronic renal failure on hemodialysis [30]. A strain of vancomycin-resistant Enterococcus faecalis (VRE) was also cultured from the ulcer, and DNA sequencing demonstrated that the vanA genes in the Staphylococcus and Enterococcus isolates were identical [59].

Additional cases were subsequently identified in Michigan as well as in Pennsylvania, New York, and Delaware. Most of the VRSA isolates were from skin and soft tissue sites, and most case patients were also colonized or infected with a VRE [65]. A comparative genomic analysis of the first 12 reported VRSA isolates concluded that resistance did not appear to be due to clonal dissemination of a single VRSA strain but rather represented separate events in which the plasmid containing the vancomycin-resistant transposon was acquired [48].

Sporadic cases have been reported elsewhere. The first case in Europe was reported in 2013 [68]. In 2014, a VRSA bloodstream infection caused by the epidemic, community-associated MRSA strain ST8 was reported in a young man in Brazil with diabetes who received several courses of antibiotics, including vancomycin, for skin infections [69]. VRSA was subsequently isolated during teicoplanin therapy for a methicillin-resistant but vancomycin-susceptible S. aureus bacteremia. Genomic analysis demonstrated that the vanA sequence in the VRSA was identical to that in a VRE isolate subsequently recovered from a rectal swab.

Because cocolonization with MRSA and VRE is common, development of further VRSA strains remains a concern [70,71]. Although the majority of VRSA isolates reported to date have been from health care-associated MRSA strains, a few have been associated with community-acquired strains (as noted above for the bloodstream VRSA infection from Brazil), indicating the possibility of transferring plasmid-mediated vancomycin resistance to various MRSA strains. Secondary spread of these strains to health care workers appears to be limited [66].

DIAGNOSIS AND LABORATORY TESTING

Clinical suspicion — In the setting of appropriate source control, repeated isolation of S. aureus from normally sterile sites despite seemingly appropriate therapy for longer than two to three days should prompt consideration of infection with a strain of S. aureus with reduced susceptibility to vancomycin that has emerged during therapy, even if the MIC of the original isolate was within the susceptible range [2,72-79]. Alternatively, persistent bacteremia may reflect an unrecognized abscess or metastatic infection (rather than reduced antibiotic susceptibility).

Persistent bacteremia may be defined as ≥2 days of bacteremia; the longer methicillin-resistant S. aureus (MRSA) bacteremia persists, the greater the risk of complications [78,80,81]. (See "Methicillin-resistant Staphylococcus aureus (MRSA) in adults: Treatment of bacteremia".)

Laboratory testing — Confirmatory MIC testing should be performed on S. aureus isolates for which the vancomycin MIC is ≥2 mcg/mL. Laboratory detection of S. aureus with reduced susceptibility to vancomycin may require special inquiry with the microbiology laboratory about susceptibility testing methods as discussed below [82].

An MIC susceptibility testing method (such as broth microdilution, agar dilution, or agar-gradient diffusion [E-test]) must be used for detection of S. aureus with reduced susceptibility to vancomycin; disk diffusion or automated methods are not sufficient [83-86]. A full 24-hour incubation period should be used with all methods. The use of matrix-assisted laser desorption/ionization time of flight mass spectrometry to distinguish vancomycin-susceptible S. aureus from vancomycin-intermediate S. aureus suggests that this newer technology may soon be able to rapidly detect these strains with reduced susceptibility to vancomycin [87].

The approach to S. aureus susceptibility testing should be tailored to the clinical scope of a particular laboratory [56,88]. Additional susceptibility testing is warranted in patients with repeated isolates of S. aureus from normally sterile sites despite seemingly appropriate therapy for longer than three to four days. Isolates should also be sent to the state public health laboratory or the United States Centers for Disease Control and Prevention for confirmatory evaluation when S. aureus with reduced susceptibility is suspected. (See 'MIC breakpoints' above.)

There is no validated laboratory test for routine detection of heteroresistant S. aureus populations; a number of methods have been described. The most accurate method consists of performing population analysis profiles [16]. As a result of this uncertainty, the Clinical and Laboratory Standards Institute lowered the intermediate vancomycin MIC breakpoint to 4 mcg/mL in order to flag isolates that are likely to be heteroresistant and/or clinically refractory to vancomycin therapy [2]. (See 'Heteroresistance' above and 'MIC breakpoints' above and "Overview of antibacterial susceptibility testing", section on 'Heteroresistance'.)

TREATMENT — General issues related to evaluation and management of S. aureus bacteremia and the duration of antibiotic therapy (which depends on the source of infection) are discussed separately. (See "Clinical approach to Staphylococcus aureus bacteremia in adults".)

General issues related to treatment of methicillin-resistant S. aureus (MRSA) infection are discussed separately. (See "Methicillin-resistant Staphylococcus aureus (MRSA) in adults: Treatment of bacteremia".)

Borderline vancomycin susceptibility — For patients with infection due to S. aureus isolates approaching the limit of the susceptible range (2 mcg/mL) who are not responsive to or are intolerant of vancomycin or daptomycin, there are several potential alternative agents. The approach is discussed further separately. (See "Methicillin-resistant Staphylococcus aureus (MRSA) in adults: Treatment of bacteremia".)

Infection due to VISA or VRSA — The optimal regimen for treatment of infection due to vancomycin-intermediate S. aureus (VISA) or vancomycin-resistant S. aureus (VRSA) infection is uncertain. Patients who respond (with clinical improvement and clearance of bacteremia) to vancomycin or daptomycin may continue treatment with one of these agents.

For patients with VISA or VRSA bacteremia who do not respond clinically to treatment with vancomycin or daptomycin, our approach is as follows:

●For patients with source control and proven absence of deep-seated infection, monotherapy is reasonable; possible regimens include telavancin, ceftaroline, or linezolid (if laboratory testing demonstrates susceptibility to the selected agent) [89-93]:

•Telavancin monotherapy may prove effective for treatment of MRSA bacteremia (thus far, available data largely reflect study of pneumonia); in a phase II trial of telavancin for treatment of bacteremia including 17 patients, cure rates were comparable for telavancin and standard therapy (88 versus 89 percent) [91,94]. Dalbavancin and oritavancin are long-acting lipoglycopeptides; data on these agents for the treatment of MRSA bacteremia is limited.

•Ceftaroline monotherapy has been successfully used for invasive MRSA infections but not VISA or VRSA infections [89,95,96]. Ceftaroline has in vitro activity against VISA strains.

•Use of linezolid is limited by toxicity [90].

Daptomycin monotherapy should not be used; this approach may increase the risk for emergence of resistance during therapy [97]. In addition, there is no role for use of quinupristin-dalfopristin, tigecycline, or fluoroquinolones for treatment of S. aureus bacteremia.

●For patients with concomitant deep-seated infection, combination therapy is warranted to minimize the likelihood of further emergence of resistance during therapy [79,98,99]. Based on the available data, we favor combination therapy with daptomycin (dosed at 8 to 10 mg/kg rather than 6 mg/kg intravenously daily) and ceftaroline, if supported by susceptibility testing. This approach is based on data discussed separately. (See "Methicillin-resistant Staphylococcus aureus (MRSA) in adults: Treatment of bacteremia", section on 'Combination therapy'.)

Other possible combination regimens include [79]:

•Daptomycin plus a beta-lactam [95,100-103]

•Vancomycin plus a hydrophilic beta-lactam [104,105]

•Daptomycin plus trimethoprim-sulfamethoxazole [106]

•Ceftaroline plus trimethoprim-sulfamethoxazole [107]

PREVENTION — Special efforts to ensure compliance with contact precautions and handwashing should be instituted when patients are infected or colonized with S. aureus strains with reduced susceptibility to vancomycin [57,108]. Issues related to nasal decolonization and chlorhexidine bathing for prevention of S. aureus infection are discussed further separately. (See "Methicillin-resistant Staphylococcus aureus (MRSA) in adults: Prevention and control".)

In addition, health care providers should be vigilant about removing temporary venous catheters when they are no longer needed and minimizing prolonged empiric antimicrobial therapy whenever possible [74,109].

Consultation with infectious disease specialists, hospital epidemiologists, the local health department, and the United States Centers for Disease Control and Prevention is appropriate for optimizing laboratory investigation, patient management, and infection control issues. (See "Infection prevention: Precautions for preventing transmission of infection".)

SUMMARY

●General principles − Failure of treatment for methicillin-resistant Staphylococcus aureus (MRSA) bacteremia with vancomycin should prompt consideration of infection due to S. aureus with reduced susceptibility to vancomycin, even if the minimum inhibitory concentration (MIC) is within the susceptible range. (See 'Introduction' above.)

●Mechanisms − Reduced S. aureus susceptibility to vancomycin may occur via increased vancomycin MIC (known as "MIC creep"), cell wall alterations (as in the case of vancomycin-intermediate S. aureus [VISA]), or plasmid-mediated gene transfer (as in the case of vancomycin-resistant S. aureus [VRSA]). (See 'Overview of mechanisms' above.)

●Definitions − The Clinical and Laboratory Standards Institute definitions for S. aureus vancomycin minimal inhibitory concentrations are as follows (see 'MIC breakpoints' above):

•Vancomycin susceptible: ≤2 mcg/mL

•Vancomycin intermediate: 4 to 8 mcg/mL

•Vancomycin resistant: ≥16 mcg/mL

●Clinical suspicion for reduced vancomycin susceptibility − In the setting of appropriate source control, repeated isolation of S. aureus from normally sterile sites despite seemingly appropriate therapy for longer than two to three days should prompt consideration of infection with a strain of S. aureus with reduced susceptibility to vancomycin that has emerged during therapy, even if the MIC of the original isolate was within the susceptible range. (See 'Clinical suspicion' above.)

●Laboratory testing − An MIC susceptibility testing method (such as broth microdilution, agar dilution, or agar-gradient diffusion) must be used for detection of S. aureus with reduced susceptibility to vancomycin; disk diffusion or automated methods are not sufficient. (See 'Laboratory testing' above.)

●Treatment − The optimal regimen for treatment of infection due to vancomycin-intermediate S. aureus or vancomycin-resistant S. aureus infection is uncertain. (See 'Infection due to VISA or VRSA' above.)

•Patients who respond (with clinical improvement and clearance of bacteremia) to vancomycin or daptomycin may continue treatment with one of these agents.

•For patients do not respond clinically to treatment with vancomycin or daptomycin, our approach is as follows:

-For patients with source control and proven absence of deep-seated infection, we suggest monotherapy (Grade 2C); possible regimens include telavancin, ceftaroline, or linezolid.

-For patients with concomitant deep-seated infection, we suggest switching to combination therapy with daptomycin plus ceftaroline (Grade 2C). We continue combination therapy until bacteremia has cleared with resolution of fever and clinical improvement, then complete treatment with daptomycin monotherapy.