Assessment of the Activity of Tigecycline against Gram-Positive and Gram-Negative Organisms Collected from Italy between 2012 and 2014, as Part of the Tigecycline Evaluation and Surveillance Trial (T.E.S.T.)

As part of the Tigecycline Evaluation and Surveillance Trial (T.E.S.T) we report the in vitro activity of tigecycline and its comparators against Gram-negative and Gram-positive organisms collected from Italian centers between 2012 and 2014. Minimum inhibitory concentrations were determined according to the broth microdilution methodology of the Clinical and Laboratory Standards Institute, and antimicrobial resistance was determined using the European Committee on Antimicrobial Susceptibility Testing interpretive criteria. Among the Enterobacteriaceae, 31% of Escherichia coli isolates, 22% of Klebsiella pneumoniae, and 1% of Klebsiella oxytoca were extended-spectrum β-lactamase producers (ESBLs). Resistance rates among ESBL-K. pneumoniae and ESBL-E. coli to meropenem were 24% and <1%, respectively. Thirty-seven percent of K. pneumoniae were multidrug resistant (MDR) strains. Resistance rates among isolates of Acinetobacter baumannii to amikacin, levofloxacin and meropenem were between 84% and 94%. Eighty percent of A. baumannii isolates were MDR strains. Methicillin-resistant Staphylococcus aureus (MRSA) accounted for 38% of S. aureus isolates. No isolates of MRSA were resistant to linezolid, tigecycline or vancomycin. Antimicrobial resistance remains a problem in Italy with increasing numbers of MDR organisms. Despite high levels, MRSA rates appear to be stabilising. Tigecycline retains its in vitro activity against the majority of organisms, including those with multidrug resistance.


Introduction
Across Europe the overuse and misuse of antibiotics has led to increasing rates of antimicrobial resistance, particularly in the southern and eastern areas [1]. Contributing factors include varying rates of infection control, incorrect prescribing behavior, and a lack of patient knowledge and understanding [2,3]. In recent years many healthcare facilities in Europe have implemented infection control programs to combat antimicrobial resistance [1]. Following this, rates of methicillin-resistant Staphylococcus aureus (MRSA) have stabilized in some areas and have decreased in others [1]. Italy is no exception to this, with reports of decreasing rates of MRSA over recent years [4].
High percentages of resistance among ESBL-K. pneumoniae were reported for ceftriaxone (99%) and cefepime (93%) ( Table 3). Resistance to amikacin, amoxicillin/clavulanate and meropenem decreased by 10%-12% between 2012 and 2014. Resistance to piperacillin/tazobactam reduced from 70% in 2012 to 44% in 2014. Between 2012 and 2014 a 10% increase in resistance among ESBL-K. pneumoniae to cefepime was seen. Tigecycline showed the lowest percentage resistance (8%) between 2012 and 2014. Between 2012 and 2014 ≥94% of MDR K. pneumoniae isolates were resistant to amoxicillin/clavulanate, cefepime, ceftriaxone, levofloxacin and piperacillin/tazobactam; 88% of isolates were resistant to meropenem. Twelve percent of MDR K. pneumoniae were resistant to tigecycline. Thirty-one percent of E. coli isolates collected between 2012 and 2014 were ESBL-producers ( Table 2). Resistance among ESBL-producing E. coli to the majority of antimicrobials in the T.E.S.T. panel remained constant between 2012 and 2014 (Table 3). One exception to this was amoxicillin/clavulanate, which showed a decrease in resistance from 71% in 2012, to 55% in 2013, and 65% in 2014. Cefepime resistance also decreased from 88% in 2012 to 67% in 2014. No isolates of ESBL-E. coli were resistant to tigecycline.

Gram-Positive Organisms
Among S. aureus isolates, 62% were methicillin-susceptible (MSSA) and 8% of these isolates were resistant to levofloxacin and 2% were resistant to minocycline (Table 4). No MSSA isolates were resistant to linezolid, tigecycline and vancomycin.
No Enterococcus faecalis isolates were resistant to ampicillin, linezolid or tigecycline (Table 4). Among E. faecium isolates, 86% were resistant to ampicillin. No isolates of E. faecium were resistant to linezolid and only one isolate (in 2013) was resistant to tigecycline.
Among S. pneumoniae isolates the highest rates of resistance were to the macrolides (36%-37%) ( Table 4). Between 2012 and 2014 resistance to azithromycin, erythromycin and clarithromycin decreased by 16%-17%. Resistance to clindamycin reduced from 36% in 2012 to 26% in 2014. No isolates of S. pneumoniae were resistant to ceftriaxone, linezolid, meropenem or vancomycin.

Gram-Positive Phenotypes
Methicillin-resistant S. aureus accounted for 38% of S. aureus isolates collected between 2012 and 2014 ( Table 2). No isolates of MRSA were resistant to linezolid, tigecycline or vancomycin; 83% of isolates were resistant to levofloxacin (Table 5) Table 2); none of these isolates were resistant to linezolid or tigecycline (Table 5). Ten isolates of vancomycin-resistant E. faecalis and six isolates of penicillin-resistant S. pneumoniae were identified between 2012 and 2014 ( Table 2). can be inferred from ampicillin [8]. "-" no EUCAST resistance or susceptibility breakpoints available.tab

Discussion
Italy has relatively high rates of antimicrobial resistance compared to other parts of Europe [1]. Reports show that despite an improvement in infection control in Italy, more effort is needed to standardize infection control procedures between regions and hospitals, as well as to ensure their effective operation [9,10]. This report on T.E.S.T. data from Italy between 2012 and 2014 is an update of the previous publication by Stefani et al. which presented data from the 2004 to 2011 time period [7]. Comparisons between the two studies are limited because the current report uses the European Committee on Antimicrobial Susceptibility Testing (EUCAST) criteria for determining susceptibility and resistance and the previous publication used the Clinical and Laboratory Standards Institute (CLSI) interpretive criteria [8,11]. Both guidelines use different methods for determining clinical breakpoints: CLSI use a variant of the error-rate-bounded method which incorporates an intermediate zone [12], whereas EUCAST define MIC breakpoints on the basis of epidemiological cut-off values, pharmacokinetic/pharmacodynamic parameters, and by correlating MIC breakpoints to zone diameter values using the "MIC-coloured zone diameter histogram technique" [13]. EUCAST breakpoints do not define an intermediate category which Marchese et al. [14], and Hombach et al. [15,16] conclude will lead to increasing numbers of resistant bacteria being reported in countries that shift from using CLSI to EUCAST criteria, such as Italy. Generally, for the organisms included in this study breakpoints are different between CLSI and EUCAST, with EUCAST typically having lower susceptibility breakpoints. The decision was taken to use the EUCAST criteria in this report as these breakpoints are now considered the European standard and their use would allow the comparison of data from this study with other contemporary studies, it was also felt that the data would be more meaningful to healthcare providers currently practicing.
Our report shows comparable rates of ESBL-producing K. pneumoniae between 2012 and 2014 to that reported in the earlier study by Stefani et al. (22% and 24% respectively) [7]. However, the number of ESBL-producing E. coli isolates has increased between the two studies, from 25% (2004-2011), to 31% (2012-2014) [7]. Tigecycline and meropenem were the most active agents against E. coli and its resistant phenotype (≥98% susceptible); this result is comparable with that by Stefani et al. [7]. A surveillance study by Jones et al. monitored antimicrobial resistance in 21 European countries, including Italy, in 2011. Similarly they also reported elevated levels of ESBL-producing E. coli and Klebsiella spp. (20.1% and 45.7%, respectively) in Europe, as well as comparable rates of susceptibility among ESBL-E. coli to tigecycline and carbapenems (>99%) [17].
Increasing numbers of carbapenem-resistant Enterobacteriaceae are a major global health concern and Italy has one of the highest levels of carbapenem resistance in Europe [1,5,[17][18][19][20]. A survey conducted in Italy by Giani et al. in 2011 identified that 2% of all Enterobacteriaceae were carbapenem-resistant, and the majority of these were K. pneumoniae (86.7%) [5]. In 2012 the European Center for Disease Prevention and Control (ECDC) launched the European survey of carbapenemase-producing Enterobacteriaceae (EuSCAPE) which aimed to monitor epidemiology, undertake surveillance and enhance laboratory capacity and diagnostics [20,21]. A recent report by Albiger et al. on data collected from the EuSCAPE project in 2015 identified that Italy was one of four European countries that classified carbapenemase-producing Enterobacteriaceae as an endemic situation [20]. A Spanish study by Palacios-Baena et al. also reported that K. pneumoniae isolates accounted for a large proportion of carbapenemase-producing Enterobacteriaceae (74%) [22]. Our report shows similar results; in total we report 8 [17]. Tigecycline is not active in vitro against P. aeruginosa. We report that 90% of A. baumannii isolates were inhibited by tigecycline at a concentration of 2 mg/L. This is comparable with reports from Italy [26], Eastern Europe [27] and Spain [28].
Limitations of this study include the varying number of participating centers and isolates submitted between years, which may cause fluctuations in antimicrobial resistance. It should be noted that despite efforts of Magiorakos et al. to standardize methodology to define MDR, variations in definitions between studies exists, which may cause limitations when comparing rates of MDR between studies [37]. We report the in vitro activity of antimicrobial agents, which limits the ability to compare the relationship of serum levels to dose and resistance at the site of infection.
This report shows that antimicrobial resistance in Italy continues to be major public health concern. There are increasing numbers of MDR organisms, particularly MDR A. baumannii. Despite levels of MRSA remaining high, this appears to have stabilized over recent years. The escalation of carbapenem-resistant Enterobacteriaceae in Italy gives cause for concern and it is therefore essential to monitor these organisms. Tigecycline continues to retain its in vitro activity against the majority of organisms including those with multidrug resistance. The results of this study show the importance of continuing surveillance of antimicrobial resistance and susceptibility to help to reduce the incidence of infection and optimize the use of antimicrobial agents.

Materials and Methods
Between 2012 and 2014 a total of 19 Italian centers submitted isolates as part of the T.E.S.T. study. All centers did not participate in all years. Ten centers participated for 3 years, four centers for 2 years, and five centers for 1 year. Details of the isolate collection and antimicrobial susceptibility testing have been previously published (e.g., Stefani et al. [7]). Minimum inhibitory concentrations were determined using broth microdilution methodology described by the CLSI [38]. Antimicrobial susceptibility was determined according to EUCAST interpretive criteria [8]. Methicillin resistance in S. aureus and ESBL-production among E. coli and Klebsiella spp. were determined by IHMA according to CLSI guidelines [39].
In this study multidrug resistance was defined as resistance to three or more classes of antimicrobial agents. Classes used to define MDR Acinetobacter baumannii and P. aeruginosa were the same as previously described by Stefani et al. [7]. Classes used to define MDR K. pneumoniae were aminoglycosides (amikacin), β-lactams (ampicillin, amoxicillin/clavulanate, cefepime, ceftriaxone or piperacillin/tazobactam), carbapenems (imipenem/meropenem), glycylcycline (tigecycline) and quinolones (levofloxacin).