• –

  In case of sepsis potentially caused by Enterobacteriaceae, clinicians should evaluate the risk of ESBL producers considering both the epidemiological setting (e.g. rate of ESBL-producing microorganisms in a given institution) and individual risk factors (BII).

• –

  The following individual risk factors should be assessed in all community-onset sepsis potentially caused by Enterobacteriaceae in order to evaluate the risk for ESBL producers: recent use of fluoroquinolones or cephalosporins; recent hospitalization; transfer from another healthcare faculty, including long-term care facilities; Charlson index >3; and age >70 years (BII). Recent travel to high endemic areas must also be considered (BIII). The same risk factors should be considered for community-onset AmpC producers (BIII).

• –

  ESBL producers should be considered for empirical therapy in patients with community-onset severe sepsis or septic shock potentially caused by Enterobacteriaceae if presenting at least one of the previous risk factors, and in patients with non-severe sepsis if more than 2 risk factors are present (CIII).

• –

  If ESBL-coverage is decided for a community-onset infection, a carbapenem is of choice (BII); however, in case of complicated urinary tract infection (cUTI) a β-lactam/β-lactamase inhibitor (BLBLI) plus an aminoglycoside (BII) or a third-generation cephalosporin (probably ceftazidime) plus an aminoglycoside (CIII) are alternatives, according to local prevalence of susceptibility to these drugs among ESBL producers.

• –

  In nosocomial sepsis potentially caused by Enterobacteriaceae, individual risk factors (longer hospital stay, exposure to mechanical ventilation, and previous receipt of cephalosporins, fluoroquinolones or carbapanems) should be considered according to local epidemiology (e.g., local prevalence, outbreak) for decision about empirical therapy in order to cover for ESBL, AmpC and/or carbapenemase-producers (BIII). Because of the current rates of ESBL producers in Spanish centers, empirical therapy against ESBL producers is recommended in all patients with nosocomial infections potentially caused by Enterobacteriaceae and presenting with severe sepsis or septic shock (CIII).

• –

  Carbapenems are the drugs of choice for invasive infections caused by ESBL- and AmpC-producing Enterobacteriaceae (BII). Ertapenem is suggested for patients without septic shock and isolates with MIC ≤0.25mg/L to avoid the selective pressure on Pseudomonas aeruginosa posed by group-2 carbapenems (CII). For other infections, imipenem or meropenem are recommended (BII); the experience with doripenem is scarce but it is probably as useful (CIII).

• –

  In vitro active BLBLI (specifically, amoxicillin/clavulanic acid and piparacillin/tazobactam) are reasonable alternatives for bacteraemic UTI or biliary tract infections caused by ESBL-producing Escherichia coli, and may be used as carbapenem-sparing regimens; the recommended doses for patients with normal renal function are: 2/0.2g/8h in 30min for amoxicillin/clavulanic acid and 4/0.5g/6h in 30min or 4/0.5g in extended infusion/8h (or/6h in critically ill patients) for piperacillin/tazobatam (CII). Data for other infection or Enterobacteriaceae are scarce.

• –

  There is insufficient data to recommend the use of active cephalosporins for invasive infections caused by ESBL producers according to EUCAST or CLSI breakpoints; however, until more data are available, we recommend caution when considering the use of these drugs; their use is only recommended as an alternative option for the treatment of non-severe UTI sepsis in low-risk patients (CIII).

• –

  There is insufficient data to recommend the use of cephamycins, fluoroquinolones, aminoglycosides, trimethoprim-sulfamethoxazole, colistin or fosfomycin. For fully susceptible isolates there is no reason to expect different efficacy than in non-ESBL producers (CIII).

• –

  Therapy must be individualized according to susceptibility results, source of infection and severity of disease (CII). According to general knowledge in management of infections, source control and support therapy are key aspects in the management of these infections (BII).

• –

  Combination therapy is recommended for severe infections caused by KPC-producing Klebsiella pneumoniae (CII), and possibly for other carbapenemase-producing Enterobacteriaceae until more data are available (CIII). There is no data to support combination therapy for patients with mild infections for which fully active drugs, useful for the specific type of infection, are available; therefore, we recommend monotherapy for such infections, and particularly for non-severe UTIs (CIII).

• –

  Monotherapy with a carbapenem is not recommended for patients with invasive infections caused by CPE but may be considered in cases of mild invasive infections if adequate source control is readily achieved and the isolate is susceptible according to EUCAST or CLSI breakpoints; the typical example would be sepsis from the urinary tract, without urinary tract obstruction or severe sepsis or septic shock (CIII).

• –

  For patients in which combination therapy is indicated, a regimen with a carbapenem (see preferred drug and recommended dose below) plus one or two fully active drugs (including colistin, tigecycline, an aminoglycoside or fosfomycin, the latter preferably as a third drug) is recommended if the carbapenem MIC is ≤8mg/L; this applies mainly to patients with severe infections caused by KPC-producing K. pneumoniae (BII). We suggest a similar approach for other carbapenemases until more data are available (CIII). No recommendation can be given for using the combination of ertapenem plus doripenem or meropenem for KPC producers (unresolved issue).

• –

  There is not enough data to recommend including a carbapenem in combination regimens if minimal inhibitory concentration (MIC) is >8mg/L; if this is the case, carbapenems are probably useless, particularly if MIC is >16mg/L; we recommend including at least two fully active drugs in the combination regimen according to susceptibility testing and source of infection (drugs to be considered: colistin, aminoglycosides, fosfomycin and tigecycline) (CIII).

• –

  Patients with less severe invasive infections and cUTIs might be treated with carbapenem-sparing combinations (drugs to be considered: colistin, aminoglycosides, fosfomycin, tigecycline – the latter not for UTI) or even monotherapy (see above) (CIII).

• –

  Dosing of all administered drugs should be optimized to increase the probability of reaching the appropriate pharmacodynamics target (BII); in the case of carbapenems, we recommend using meropenem at 2g every 8h in extended infusion (BII).

• –

  Colistin should be preserved for treating infections produced by Enterobacteriaceae strains showing resistance to all β-lactam antibiotics (CIII).

• –

  Colistin is suggested as part of the empirical treatment of patients with severe infections if participation of CPE is suspected, such as in outbreaks or colonized patients (CIII).

• –

  When using colistin and until more data are available, we recommend the use of a loading dose of 9 MU and subsequent high, extended-interval maintenance doses (4.5 MU/12h) in critically-ill patients and patients with severe sepsis or septic shock with creatinine clearance above 50mL/min; maintenance dose should be individually adjusted according to creatinine clearance according to published nomograms (CIII).

• –

  Data to recommend a loading dose in non-critically ill patients with non-severe infections are unavailable (unresolved issue); maintenance dosing with 4.5 MU/12h (or alternatively 3 MU/8h) is suggested, but renal function should be closely monitored (CIII).

• –

  No recommendations can be given for patients with extreme weights (unresolved issue).

• –

  We recommend 1–2 MU of colistin every 12h for patients undergoing intermittent hemodialysis. Whenever possible, hemodialysis should be conducted at the end of the dosing interval to minimize the clearance of colistin. If this is not possible, a supplemental dose at the end of the dialysis should be considered (CIII). For patients undergoing continuous venovenous haemodiafiltration, even though the data are not consistent, a dose of 9 MU/day is suggested (CIII).

• –

  Experience with fosfomycin for the treatment of invasive infections caused by MDR and XDR Enterobacteriaceae is limited; however, in patients with limited options, fosfomycin (4–6g every 6h to 8g every 8h) may be considered as part of a combination regimen including at least one more active agent (CIII).

• –

  Based on the experience of non-MDR Enterobacteriaceae infections, monotherapy with an active aminoglycoside may be considered for the treatment of cUTI caused by MDR and XDR Enterobacteriaceae. Toxicity should be closely monitored (BII).

• –

  Monotherapy with aminoglycosides for other invasive infections is not recommended. For such infections, combination therapy with other drugs is suggested (BII). However, aminoglycosides are to be considered as an accompanying agent in all combination regimens according to the susceptibility results with closely monitorisation of toxicity (CIII).

There is no clinical experience with aztreonam or cephalosporins for the treatment of invasive infections due to susceptible MBL- or OXA-48-producing Enterobateriaceae, respectively. Very scarce in vitro and animal model data suggest that they may be useful; if considered, we recommend using these drugs in combination except in cUTI (CIII).

• –

  Tigecycline monotherapy should be avoided whenever possible (AI); exceptions may be selected patients with mild cIAI and cSSSI infections caused by XDR Enterobacteriaceae with other few adequate alternative options (CIII).

• –

  Tigecycline should be considered as part of a combination regimen in patients with infections other than UTI caused by Enterobacteriaceae producing either ESBLs (if a carbapenem-spare regimen is to be used) or carbapenemases when tigecycline MIC is ≤1mg/L (CIII).

• –

  Higher dose of tigecycline (150mg loading dose followed by 75mg/12h, or 200mg loading dose followed by 100mg/12h) should be considered for patients in septic shock, VAP or Enterobacteriaceae with MIC ≥1mg/L, but adverse events should be carefully monitored (BIII).