Rev Esp Anestesiol Reanim. 2014;61(3):e1---e19

Revista Española de Anestesiologíay Reanimación

www.elsevier.es/redar

REVIEW

Bugs, hosts and ICU environment: Counteringpan-resistance in nosocomial microbiota and treatingbacterial infections in the critical care setting�,��

Emilio Masedaa,∗,1,2, José Mensab,2, Juan-Carlos Valíac,1,José-Ignacio Gomez-Herrerasd,1, Fernando Ramascoe,1, Enric Samsof,1,Miguel-Angel Chiveli g,1, Jorge Pereirah,1, Rafael González i,1, Gerardo Aguilar j,1,Gonzalo Tamayok,1, Nazario Ojedal,1, Jesús Ricom,1, María-José Giménezn,2,Lorenzo Aguilarn,2

a Anesthesiology and Surgical Critical Care Department, Hospital Universitario La Paz, Madrid, Spainb Infectious Diseases Department, Hospital Clínic, Barcelona, Spainc Anesthesiology and Surgical Critical Care Department, Hospital General Universitario, Valencia, Spaind Anesthesiology and Surgical Critical Care Department, Hospital Clínico Universitario, Valladolid, Spaine Anesthesiology and Surgical Critical Care Department, Hospital Universitario La Princesa, Madrid, Spainf Anesthesiology and Surgical Critical Care Department, Hospital del Mar, Barcelona, Spaing Anesthesiology and Surgical Critical Care Department, Hospital Universitario y Politécnico La Fe, Valencia, Spainh Anesthesiology and Surgical Critical Care Department, Complejo Hospitalario de Vigo, Vigo, Spaini Anesthesiology and Surgical Critical Care Department, Hospital de León, León, Spainj Anesthesiology and Surgical Critical Care Department, Hospital Clínico Universitario, Valencia, Spaink Anesthesiology and Surgical Critical Care Department, Hospital de Cruces, Bilbao, Spainl Anesthesiology and Surgical Critical Care Department, Hospital Universitario de Gran Canaria Dr. Negrín, Las Palmas de GranCanaria, Spainm Anesthesiology and Surgical Critical Care Department, Hospital Universitario Río Hortega, Valladolid, Spainn PRISM-AG, Madrid, Spain

Received 29 October 2013; accepted 4 November 2013Available online 1 February 2014

KEYWORDSMethicillin-resistant

Abstract ICUs are areas where resistance problems are the largest, and these constitute amajor problem for the intensivist’s clinical practice. Main resistance phenotypes among nosoco-

Staphylococcusaureus;Vancomycin-resistantenterococci;

mial microbiota are (i) vancomycin-resistance/heteroresistance and tolerance in grampositives(MRSA, enterococci) and (ii) efflux pumps/enzymatic resistance mechanisms (ESBLs, AmpC,metallo-betalactamases) in gramnegatives. These phenotypes are found at different ratesin pathogens causing respiratory (nosocomial pneumonia/ventilator-associated pneumonia),

� This article is part of the Anaesthesiology and Resuscitation Continuing Medical Education Program. An evaluation of the questions onthis article can be made through the Internet by accessing the Education Section of the following web page: www.elsevier.es/redar

�� According to the authors and editors, this article is published unabridged in Rev Esp Quimioter 2013;26(4).∗ Corresponding author.

E-mail addresses: emilio.maseda@gmail.com, emaseda.hulp@salud.madrid.org (E. Maseda).1 GTIPO-SEDAR: Grupo de Trabajo de Infecciones Perioperatorias-Sociedad Espanola de Anestesiología, Reanimación y Terapéutica del

Dolor.2 SEQ: Sociedad Espanola de Quimioterapia.

0034-9356/$ – see front matter © 2013 Sociedad Española de Anestesiología, Reanimación y Terapéutica del Dolor. Published by Elsevier España, S.L. All rights reserved.

http://dx.doi.org/10.1016/j.redar.2013.11.012

e2 E. Maseda et al.

Extended spectrum�-lactamase;Pseudomonasaeruginosa;Acinetobacterbaumannii;Critical care

bloodstream (primary bacteremia/catheter-associated bacteremia), urinary, intraabdominaland surgical wound infections and endocarditis in the ICU. New antibiotics are available toovercome non-susceptibility in grampositives; however, accumulation of resistance traits ingramnegatives has led to multidrug resistance, a worrisome problem nowadays. This articlereviews microorganism/infection risk factors for multidrug resistance, suggesting adequateempirical treatments. Drugs, patient and environmental factors all play a role in the decision toprescribe/recommend antibiotic regimens in the specific ICU patient, implying that intensivistsshould be familiar with available drugs, environmental epidemiology and patient factors.© 2013 Sociedad Española de Anestesiología, Reanimación y Terapéutica del Dolor. Publishedby Elsevier España, S.L. All rights reserved.

PALABRAS CLAVEStaphycoccus aureusresistentea meticilina;Enterococo resistentea vancomicina;Betalactamasas deespectro prolongado;Pseudomonasaeruginosa;Acinetobacterbaumannii;Cuidados críticos

Gérmenes, huéspedes y el entorno de la Unidad de Cuidados Intensivos:contrarrestando la panresistencia en la microbiota nosocomial para tratar lasinfecciones bacterianas en cuidados críticos

Resumen Las UCI son las áreas con mayor problema de resistencias, lo que constituye unode los principales contratiempos de los intensivistas en su práctica clínica. Los principalesfenotipos de resistencia en la microbiota nosocomial son: i) la resistencia/heterorresistenciay la tolerancia a la vancomicina en grampositivos (SARM, enterococo), y ii) las bombasde eflujo/mecanismos enzimáticos de resistencia (BLEE, AmpC, metalobetalactamasas) engramnegativos. Estos fenotipos pueden encontrarse, con distinta frecuencia, en patógenoscausantes de infecciones respiratorias (neumonía nosocomial/neumonía asociada a ventilaciónmecánica), del torrente sanguíneo (bacteriemia primaria/bacteriemia asociada a catéter), uri-narias, intraabdominales, de herida quirúrgica y endocarditis en la UCI. Hay nuevos antibióticosdisponibles para contrarrestar la no sensibilidad en grampositivos; sin embargo, la acumu-lación de factores de resistencia en gramnegativos lleva a la multirresistencia/panresistencia,un problema en nuestros días. Este artículo revisa por microorganismo/infección los factoresde riesgo de resistencia/multirresistencia, proponiendo tratamientos empíricos adecuados.Fármacos, pacientes y factores ambientales tienen todos un papel básico en la decisión deprescribir/recomendar regímenes antibióticos en el paciente específico de la UCI, implicandoque los intensivistas deben estar familiarizados con los fármacos disponibles, la epidemiologíalocal y las características del paciente crítico.© 2013 Sociedad Espanola de Anestesiología, Reanimación y Terapéutica del Dolor. Publicadopor Elsevier España, S.L. Todos los derechos reservados.

T

Eipfcggibic

ptpwatag

er

ioocaaert

rmroof

he nosocomial microbiome and resistome

volution of relationships between human and bacterias conditioned by environmental changes. Among anthro-ogenic factors changing the environment and thus, shapinguture interactions between human and bacteria,1,2 chemi-al pollution (including antibiotics and antimicrobial strate-ies) altering microbial biodiversity, new medical technolo-ies (opening the way for opportunistic infections), thencreasing number of highly susceptible hosts and control ofacterial access to host are important factors for nosocomialnfections, and theoretically, these factors counterbalanceolonization/multidrug resistance in nosocomial microbiota.

The ‘‘nosocomial human population’’, which includesatients and health care personnel, is closely linked tohe ‘‘nosocomial microbiome’’ (microbiota from health careersonnel and from non-infected and infected patients),ith its specific ‘‘resistome’’ (antibiotic resistance genesnd genetic elements that participate in resistance gene

ransfer). The horizontal gene transfer within speciesnd between different species of gram-negative andram-positive bacteria,3 facilitated when bacteria are

fmr

xposed to antibiotic stress,1,4,5 has driven to multidrugesistance.

Resistance implies the need for new antibiotics that, oncentroduced, if their mechanism of action is similar to thatf previous compounds may select pre-existing resistancesr induce new resistances in the nosocomial resistome thatould be further selected, thus implying the need for newntibiotics and closing the circle.6 Antimicrobial pressures driving engine for resistance and multidrug resistance isvident in the nosocomial environment, with a well definedelationship between antibiotic use and emergence of mul-idrug resistant strains.7---9

In the presence of antibiotic stress, antimicrobialesistance can be considered a colonization factor.1 Accu-ulation of ‘‘genotypic colonization factors’’ (phenotypic

esistance traits) drives to multidrug resistance, a hallmarkf nosocomial microbiota since the phenomena of selectionf co-resistance and co-selection of resistance are morerequent in hospitals than in the community. If resistance

avours ‘‘colonisation’’ of elements of the nosocomialicrobiota, strategies aimed at reducing resistance will

esult not only in a decrease in the resistance prevalence

wearwa

Tn

Tivivrsa(MMbIi(

tagtovwsobc1itaiogSi

Ca

Bmoitp

Bugs, hosts and ICU environment

but also in a decrease in colonization and a subsequentdecrease in nosocomial infections.

Hospital-acquired infections affect a quarter of criti-cally ill patients, and can double the risk of a patientdying,10,11 requiring rapid treatment to reduce morbid-ity and mortality.12 Nosocomial infections acquired in theintensive care unit (ICU) represent an area in which muchimprovement is still achievable.13 However it should betaken into account that infection is often the cause ofICU admission,14,15 influencing the microbiological environ-ment of the unit.16 The drugs, patient and environmentalfactors all play a role in the decision to prescribe or rec-ommend (and daily review) antibiotic dosing regimens in aspecific patient,12 mplying that personnel involved should befamiliar with available drugs and environmental (bacterialepidemiology and resistance traits) and patient factors.

The concrete battlefields in the ICU

An approach to the existing resistome can be done throughthe choice of indicator microorganisms based on their clin-ical relevance and their potential for acquisition of geneticdeterminants of resistance. Nowadays, the main resistancephenotypes among multiresistant nosocomial microbiota are(i) vancomycin resistance and tolerance in nosocomial gram-positives (MRSA and enterococci) and (ii) efflux pumps andenzymatic resistance mechanisms (ESBLs, AmpC and met-allobetalactamases) in nosocomial gram-negative bacteria.Antibiotics/antibiotic regimens for the treatment of nosoco-mial infections should counter these sometimes emerging,always diffusible and clinically worrisome resistance traits.

The methicillin-resistant Staphylococcusaureus (MRSA) case

Staphylococcal infections became treatable with theintroduction of penicillin but, soon after, production of �-lactamase by staphylococci became a reality. Penicillinase-resistant isoxazolyl penicillins were then introduced tocounter resistance mediated by �-lactamases, with the sub-sequent emergence of methicillin resistance. Nowadays,MRSA is spread in hospitals worldwide, with prevalencereaching rates of 25---50% in much of Americas, Australiaand Southern Europe.17 The global rate of evolution of MRSAamong S. aureus in Spanish ICUs from 1994 to 2008 (studyENVIN-UCI including up to 100 ICUs) shows similar rates(≈25%) in the first and last years with oscillations rangingfrom 13% in 1997 to 42.3% in 2006.18 In addition to intra-ICUtransmission dynamics of MRSA (influenced, among others,by colonization of health care workers in the ICU19), MRSAimported cases in the ICU as predictor of occurrence of noso-comial MRSA infections should be taken into account,20 withcommunity-acquired MRSA genotypes as the emerging causeof colonization among patients admitted in adult ICUs in theUSA.21

The dramatic increase in MRSA nosocomial infectionsled to a substantial increase in the use of vancomycin,

and this could be related to the appearance of differentvancomycin non-susceptible phenotypes in both entero-cocci and staphylococci. The risk of emergence of MRSAnon-susceptible to vancomycin is much higher in countries

wcis

e3

ith high prevalence of both MRSA and vancomycin-resistantnterococci.22 Published studies suggest a link betweenntibiotic usage at individual and institutional levels andesistance, showing an increase in the risk of acquiring MRSAhen using not only glycopeptides23 but also quinolones23,24

nd cephalosporins.24,25

he associated problem of vancomycinon-susceptibility and vancomycin tolerance

he first vancomycin non-susceptible strains were des-gnated as vancomycin-intermediate S. aureus withancomycin MICs of 8---16 mg/l.26 Among vancomycin-ntermediate strains, 90% of strains are heterogeneousancomycin intermediate (heteroresistant; h-VISA) cha-acterized by the presence of a selectable resistantubpopulation in an otherwise fully susceptible population,nd only 10% are homogenous vancomycin intermediatehomoresistant; VISA).27,28 The prevalence of h-VISA amongRSA is variable worldwide ranging from ≈10 to 50%.29---32

RSA strains resistant to vancomycin have been describedut, fortunately, its diffusion is unappreciable nowadays.33

ntermediate resistance to vancomycin can also be foundn coagulase-negative staphylococci at non negligible rates≈10%).34

In addition, there are MRSA isolates that are susceptibleo vancomycin but tolerant to its killing effect. Toler-nce is defined as ‘‘bacterial capability of survival withoutrowth in the presence of a current lethal concentra-ion’’,35 and is expressed as an MBC/MIC quotient of ≥16r ≥32.36 Nevertheless, a recent study has shown that evenancomycin-susceptible strains with MBC/MIC ratios of 8,hen exposed to simulated vancomycin concentrations in

erum, exhibit a pharmacodynamic behaviour similar to thatf strains with MBC/MIC ≥16, with no bactericidal activityy vancomycin despite susceptibility.37 Tolerance to van-omycin is present in 100% VISA strains, 75% h-VISA and5% vancomycin-susceptible MRSA,36 and this phenomenons extensive to other glycopeptides as teicoplanin, witheicoplanin tolerance reported in 18.8% of MRSA strains.38 Inddition, tolerance to glycopeptides has also been describedn ≈25% of coagulase negative staphylococci and ≥40%f group viridans (Streptococcus bovis, Streptococcus san-uis, Streptococcus gordonii, Streptococcus mutans andtreptococcus oralis) isolates,34 both bacterial groups beingmportant etiological agents in endocarditis.

linical impact of non-susceptibility, resistancend/or tolerance

actericidal activity is important in infections caused byethicillin-susceptible S. aureus.39 A classical study in

ur country showed a significantly higher mortalityn methicillin-susceptible S. aureus bacteremia in patientsreated with vancomycin compared with cloxacillin, inart attributable to the slow vancomycin killing.40 This

as corroborated in an in vitro study showing that van-omycin was not bactericidal within the dosing intervaln contrast to daptomycin, regardless of methicillinusceptibility/resistance of the study strains.41

e

aaatvridtiw>r

tccbctb

Vnctia4abwbst

Tc

Eacfcattfr1tIfuVCtsar

aai1aVodtctswtcp

rzio

pitierarr

Sp

CttCpicssi

t(d

fsbtiil

4

Some high-inoculum staphylococcal infections suchs bacteremia, persistent bacteremia, endocarditisnd osteomyelitis have been associated with heteroresist-nce.33,42,43 Vancomycin heteroresistance has been linkedo strains susceptible to vancomycin but with high MICalues within the susceptibility category.44,45 In turn, theelationship of MICs to clinical failure with vancomycins striking.31 In a published study, high vancomycin MICs,efined as 1.5---2.0 mg/ml, was an independent predic-or of poor response to vancomycin therapy for MRSAnfection, even when vancomycin trough levels >15 mg/mlere achieved.46 Importantly, vancomycin trough levels15 mg/ml appears to be associated with a 3-fold increasedisk of nephrotoxicity.47

Considering the current situation, it has been suggestedhat strains with vancomycin MIC of 1---2 mg/l should beonsidered h-VISA or VISA48 since even the new Clini-al and Laboratory Standards Institute (CLSI) susceptibilityreakpoint for vancomycin (≤2 mg/l) may fail to pre-isely differentiate potential responders to vancomycinherapy,36,49 suggesting that, according to clinical data, thereakpoint value should be even lowered to 1 or 0.5 mg/l.48

The spectrum of clinical disease caused by MRSA, h-ISA, VISA and tolerant isolates is similar to that caused byon-tolerant methicillin-susceptible S. aureus. Since antimi-robial treatment is empirically initiated, there is evidenceo show that less than a quarter of patients with MRSAnfections receive correct therapy within 48 h of hospitaldmission, and only ≈40% receive appropriate agents after8 h.50 Clinical implications of heteroresistance and toler-nce evidenced as poor clinical outcome, persistence ofacteremia and increased length of stay,24,51---55 togetherith the fact that these phenomena are not routinely testedy microbiologists and reported to treating physicians,34

tress the importance of therapeutic strategies to overcomehem.

he vancomycin-resistant Enterococcus (VRE)ase

nterococci, historically regarded as a second-rate pathogennd with low virulence, have become one of the mosthallenging nosocomial problems. Nowadays, Enterococcusaecium is almost as common as Enterococcus faecalis as aause of nosocomial infection.56 All enterococci show toler-nce to vancomycin.57 In addition, acquisition of resistanceo ampicillin, aminoglycosides (high level) and glycopep-ides in E. faecium is a cause of concern,22 making E.aecium infections difficult to treat. In USA vancomycinesistance increased in E. faecium isolates from 0% in mid980s to 80% in 2007.58 In Europe the vancomycin resis-ance prevalence is variable, ranging from <1% to >40%.22,59

n Spain rates of around 14.3% have been reported in E.aecium.60 At hospital level, the increase in vancomycinse to treat MRSA infections seems to be the origin ofRE. In addition, the intensive use of oral vancomycin forlostridium difficile infections in hospitals is also likely

o select and increase faecal carriage of VRE.3 In thisense, the description of multidrug resistant, hospital-dapted E. faecium clonal complexes without communityeservoir can be explained by cross-transmission, selection

hisv

E. Maseda et al.

nd diffusion by selective antibiotic pressure.61 Factorsssociated with VRE colonization in critically ill patientsnclude prolonged ICU stay (each day in the ICU increases.03 times the risk of acquisition), previous antibiotic usend carbapenem use.62---64 Risk factors for development ofRE infections include prolonged hospitalization, surgicalr intensive care units, intravascular or bladder catheterevices, proportion of colonized patients and exposureo antibiotics.65,66 Among antibiotics, in addition to van-omycin, certain compounds as ticarcillin/clavulanate andhird generation cephalosporins have demonstrated to causeelection.67,68 Although initially hospital-associated clonesere different from community-associated clones, the lat-

er have become important nosocomial pathogens,58 witholonization prior to ICU admission being associated withrevious hospitalization and, again, antibiotic exposure.69

In enterococci full resistance to daptomycin, althougheported,70 is rare, as for linezolid.58 The increase in line-olid use has been related to an increase (and to outbreaks)n VRE resistant to linezolid,71---73 also in patients not previ-usly exposed to the drug.74

Clinical outcomes are worse and mortality higher inatients with VRE infections when compared to that in thosenfected with susceptible strains.66 The classical toleranceo the killing capability of penicillins and glycopeptidesn enterococci has clinical implications, as evidenced innterococcal endocarditis where, due to the historical highecurrence rates with penicillin or glycopeptide monother-py, combined therapy (including an aminoglycoside) is theule.75 However, nowadays, due to the high aminoglycosideesistance in VRE, recurrences can occur.

trategies to overcome non-susceptibilityhenotypes in MRSA and VRE

ompromise of the bactericidal activity, among other fac-ors, by vancomycin heteroresistance/tolerance (MRSA) orolerance/resistance (VRE) may have clinical implications.onceptually, treatments achieving bactericidal activity arereferred than those only presenting bacteriostatic activ-ty, although this has not been clearly demonstrated inlinical trials. There are clinical indications where it is con-idered that bactericidal activity is absolutely necessary,uch as bacteremia, endocarditis, meningitis and infectionsn immunocompromised patients.39

Two strategies can be considered to overcome deteriora-ion of bactericidal activity by non-susceptible phenotypes:i) combined therapy obtaining synergism and (ii) bacterici-al antibiotics for initial treatment.

The addition of a new antimicrobial is outlined whenacing a poor response with vancomycin monotherapy, thusuggesting tolerance of the infecting strain,55 and haseen successfully used in the treatment of refractory bac-eremia by tolerant isolates.76 The election of drugs to bencluded in the combination is important since antagonisticnteractions between linezolid and vancomycin or betweeninezolid and gentamicin have been described,77,78 on one

and, and the commented high aminoglycoside resistancen VRE on the other. No antagonistic interactions have beenhown between daptomycin and gentamicin, linezolid orancomycin.79

(bspKhrrDcwatcw

wgclthla

iob

A

ISrgcacscidbcpimcuoiDit

wt

Bugs, hosts and ICU environment

Regarding initiation of antibiotic therapy with bacterici-dal drugs, among compounds with potential activity againstgram-positives, it should be taken into account that linezolidand tigecycline are bacteriostatic against S. aureus, andthat quinupristin/dalfopristin, although bactericidal againstS. aureus, is bacteriostatic against E. faecium and nonactive against E. faecalis.27 Bactericidal compounds to beused should present activity against gram-positive isolatesand lack of tolerance or heteroresistance, in contrast toglycopeptides, as the lipopeptide daptomycin that repre-sents an adequate option for initial treatment of nosocomialgram-positive infections such as staphylococcal bacteremia,endocarditis and skin and soft-tissue infections, but not ofpneumonia due to the inhibition of its antibacterial activityby the pulmonary surfactant.

The extended spectrum �-lactamase (ESBL),AmpC and carbapenemases case

Pan-resistance is an increasing problem among nosoco-mial gram-negatives mainly due to antibiotic inactivatingenzymes, sometimes in combination with efflux pumpsand/or porine deficits. Acinetobacter baumannii, Pseu-domonas aeruginosa and Klebsiella pneumoniae arespecifically addressed as the most problematic and oftenextensively or pan-drug resistant pathogens.80 In Spain, theproportion of A. baumannii isolates showing resistance tocarbapenems, ceftazidime, aminoglycosides and quinolonesis around 50% (for the first three) and 87% for ciprofloxacin,and in P. aeruginosa isolates the proportions are ≈20% (car-bapenems), ≈15% (ceftazidime) and ≈25% (aminoglycosidesand quinolones).80 In K. pneumoniae, resistance to third-generation cephalosporins and aminoglycosides is ≈10% and≈18% for quinolones.80

Different types of �-lactamases are increasingly appear-ing and diffusing as response to antibiotic pressure at thenosocomial level. In general, �-lactamases diffusing amonghuman microbiota may be classified into three groups:(1) Extended-spectrum �-lactamases (ESBL), (2) AmpC and(3) carbapenemases.

ESBL

After the introduction in the 1980s of extended-spectrumthird-generation cephalosporins, mutations in both blaTEM

and blaSHV genes were reported, mainly in Klebsiella spp.In the last decade, there has been a rise in the prevalenceof CTX-M �-lactamases that, unlike TEM and SHV ESBLs, didnot remain confined to Klebsiella and have proliferated inEscherichia coli.81 In Spain the prevalence of ESBL-producingE. coli has 8-fold increased from 2000 to 200682; the SMARTstudy reported a frequency of ESBL-producing isolates of≈8.5% for E. coli and for Klebsiella spp.83 Urine, followed byblood, and internal medicine, general surgery and ICUs werethe most common sites and wards of isolation, respectively,in another study.84

The huge amount of molecular variants widely diffused

around the world is creating problems in the treatmentof nosocomial infections since these enzymes are capableof conferring resistance to penicillins, to first-, second-and third-generation cephalosporins and to aztreonam

ppTg

e5

but not to cephamycins and carbapenems), but cane inhibited by �-lactamase inhibitors.85 However, non-usceptibility rates (according to EUCAST breakpoints) toiperacillin/tazobactam in CTX-M-producing E. coli and. pneumoniae were 27.4% and 38.1%, respectively, withigh resistance rates to cefepime (≈70% and ≈80%,espectively).86 In addition, in ESBL-producing strains co-esistance to aminoglycosides and quinolones is present.85

ue to this, ESBL-producing strains have been clearly asso-iated with poor outcome. In this sense, empirical therapyith cephalosporins or fluoroquinolones was associated with

higher mortality compared with mortality in patientsreated with a �-lactam/�-lactamase inhibitor or witharbapenem-based regimens in a Spanish series of patientsith bacteremia produced by ESBL-producing E. coli.87

Selection and diffusion of ESBLs have been associatedith antibiotic pressure derived from the use of third-eneration cephalosporins (with special importance foreftazidime), aminoglycosides and quinolones, but not �-actams/�-lactam inhibitors or carbapenems.85 In additiono previous antibiotic treatments, other risk factors thatave been described for infection by ESBL-producing iso-ates in ICU patients are previous hospitalization, advancedge, diabetes and use of catheters.84

Carbapenems are probably the best options for treatingnfections caused by ESBL-producing strains,84,87 but the riskf the emergence of carbapenem resistance should alwayse considered (see below).

mpC

solates of Enterobacter cloacae, Enterobacter aerogenes,erratia marcescens, Citrobacter freundii, Providenciaettgeri and Morganella morganii (known as the ESCPMroup) have the potential to produce AmpC-induciblehromosomal �-lactamases upon exposure to inducinggents: aminopenicillins, first-generation cephalosporins,ephamycins and carbapenems as strong inducers, andecond- or third-generation cephalosporins, acylureidopeni-illins or monobactams as weakly inducers.88,89 When thenducer is removed, AmpC production returns to hardlyetectable basal levels; thus, when isolated from patients,acteria are found to be susceptible to third-generationephalosporins. However, AmpC production should be sus-ected in all isolates belonging to these species. Whennducer drugs are clinically used, selection of derepressedutants (constitutively producing �-lactamase) occurs, with

ontingent clinical failure.89,90 An association between these of third-generation cephalosporins and the emergencef resistance has been established among organisms withnducible chromosomally encoded AmpC �-lactamases.90

erepressed overproduction has been described in 20%nfections by Citrobacter spp. or Enterobacter spp. duringhird-generation cephalosporin treatment.3

AmpC genes have been mobilized to plasmids and spreadorldwide, with increasing numbers in the diversity of this

ype of enzymes.3 Infections caused by plasmid AmpC-

roducing isolates significantly increase treatment failurerobably due to inadequate initial treatment therapy.91

he CLSI provides susceptibility breakpoints for third-eneration cephalosporins and AmpC producers but advice

e

tenpict

raEse

C

Mm

(

(

(

mAbdnlekbiicacRAmB

PPopocPtshA≈≈

crta

AApPlroRicattI≈atatoaev

EApmpmotnr0pcfBtVg�iiiSra

6

hat resistance can emerge, and many infectious dis-ases specialists advocate that these compounds shouldot be used for significant infections caused by AmpC-roducing enterobacteria.92,93 In addition, ESBLs have beenncreasingly described in AmpC producers, which furtheromplicate decisions related to the optimum antimicrobialherapy.93

AmpC- and ESBL-producing isolates exhibit high rates ofesistance to penicillins (including piperacillin/tazobactam)nd cephalosporins (including cefepime) according toUCAST breakpoints.86 Treatment with carbapenems repre-ents a good option but, again, concerns on the potentialmergence of carbapenem resistance arise.

arbapenemases

ost carbapenemase-producers have multiple resistanceechanisms to �-lactams and to aminoglycosides.94

Resistance to carbapenems can arise by the following:

a) Permeability alterations (efflux pumps and/or porinedeficit) plus AmpC (class C �-lactamases) or ESBL (classA) enzymes,

b) Acquisition of non- metallo-carbapenemases mainly ofthe KPC or OXA (class D �-lactamases) families, and/or

c) Acquisition of metallo-�-lactamases (MBLs; class B �-lactamases), mainly of the IMP- and VIM- families.

The heavy use of carbapenems after dissemination ofultidrug resistant Enterobacteriaceae (due to ESBL andmpC �-lactamases) raises the fear of the relationshipetween the use of these antibiotics and the selection andiffusion of carbapenemase-producing strains. Althoughowadays the prevalence of carbapenemases is relativelyow, they are sources of considerable concern due to thenzyme spectrum of activity that encompasses almost allnown �-lactams, from penicillins to carbapenems, andecause they are not susceptible to class A �-lactamasenhibitors and currently there are no clinically availablenhibitors to block MBLs action.95 The association ofarbapenemase production to resistance traits to otherntibiotic classes may lead to polymyxins and tigecy-line as last active agents, neither of them being ideal.esistance mediated by carbapenemases affects primarily. baumannii, P. aeruginosa and to lesser extent, K. pneu-oniae, although its emergence has also been described in. fragilis.96

. aeruginosaan-resistance in P. aeruginosa results from the convergencef multiple resistance mechanisms97: low outer membraneermeability, AmpC �-lactamases, efflux pumps and, lessften, production of MBLs.97,98 However, in many Europeanountries, mainly in the Mediterranean area, VIM producing. aeruginosa has currently become endemic.99 In Spainhe prevalence of carbapenemase-producing P. aeruginosatrains among bacteremic isolates resistant to imipenem

as increased 10 times in few years, reaching 4% in 2008.100

ccording to the EARSS study, non-susceptibility rates are8% to piperacillin/tazobactam, ≈15% to ceftazidime,25% to aminoglycosides and quinolones, and ≈20% to

cihd

E. Maseda et al.

arbapenems.80 In the ICU, risk factors for multidrugesistance in P. aeruginosa are previous exposure tohird-generation cephalosporins, to carbapenems or tocylureidopenicillins.101

. baumannii

. baumannii is more often resistant. A. baumanniiroduces a naturally occurring AmpC �-lactamase, like. aeruginosa, together with a naturally occurring oxacil-inase with carbapenemase properties.102 Additionally,esistance to carbapenems has been linked to the loss ofuter membrane porins and upregulated efflux pumps.80

esistance to carbapenems remained rare until 2000 despitets widespread resistance to other compounds.98 However,arbapenem resistance has increased sharply since then,nd is mediated by OXA-type, and less often by IMP- and VIM-ype, carbapenemases.80,98 Several studies have describedhe OXA-40 gene spread across the Iberian Peninsula.103,104

n our country, resistance rates are ≈35% to amikacin,40% to ceftazidime, ≈70% to piperacillin/tazobactam,nd ≈45% to carbapenems.105 It is considered that resis-ance to carbapenems is enough to define an isolates highly resistant.106 Risk factors for carbapenem resis-ance in A. baumannii are hospital size, ICUs, lengthf stay in the ICU, recent surgery, invasive proceduresnd, mainly, previous exposure to antibiotics (carbapen-ms and third-generation cephalosporins) and mechanicalentilation.107,108

nterobacteriaceaes previously described, the main multidrug resistancehenotype in enterobacteria is hyperproduction of chro-osomal AmpC �-lactamases or ESBLs. Undoubtedly, thishenotype is also represented by carbapenem resistanceainly mediated VIM- and IMP-type MBLs. In the Enter-

bacteriaceae family, K. pneumoniae is the species withhe highest rates of carbapenem resistance. In the multi-ational SENTRY study (2007---2009), overall carbapenemaseesistance in K. pneumoniae was 5.3%, while it was.3% in E. coli, mainly due to KPC �-lactamases in K.neumoniae and OXA-48 in E. coli.109 In Spain, class Barbapenemase-producers (VIM-1 and IMP-22) have beenound in specific areas (Madrid, Catalonia, Andalucia,alearic Islands) with a local prevalence <0.2%.110 Buthe situation may be changing since the description ofIM-producers outbreaks,110---114 together with the emer-ence of the KPC-3115 and the New Delhi MBL (NDM-I)-lactamases in K. pneumoniae and E. coli,116,117 confirm-

ng the dissemination of carbapenemase-producing isolatesn our country. Nonetheless, according to last EARSS datan 2011, carbapenemase resistance in K. pneumoniae inpain is 0.3%.118 Risk factors associated with carbapenemesistance in K. pneumoniae are previous exposure tontibiotics (carbapenems, cephalosporins, acylureidopeni-

illins and quinolones), mechanical ventilation, and stayn the ICU.80,119---121 Carbapenem-resistant K. pneumoniaeas been independently associated with poor outcome andeath.120,122,123

hwfcrotpcosstratmttimcot

itgiIiaImntiaoIrHwmcasv

lAdan

Bugs, hosts and ICU environment

Pharmacokinetics/pharmacodynamics (PK/PD)in the ICU setting

The choice of an antibiotic for empirical treatment of seri-ous bacterial infections in the ICU is based predominantlyin the identity and susceptibility patterns of bacteria com-monly isolated in a particular ICU. Serious infections incritically ill patients require rapid treatment to limit mor-bidity and mortality. Intravenous treatment should beginwithin the first hour after diagnosis of severe sepsis124 as themost important factor affecting outcome. However, this isnot always met since as few as 25% of the first doses of antibi-otics are administered within 1 h of prescription.12 What isoften overlooked is the optimum dose of an antibiotic,12

and to avoid empiricism, the PK/PD relation should beexploited.80 However, PK/PD parameters predicting efficacyusually rely on steady-state concentrations, avoiding eventsoccurring when the pathogen is exposed to the initial dose,which are relevant for outcome.125 Ideally, the first doseshould rapidly reach enough concentrations above the MIC toavoid resistance selection, and these concentrations shouldbe maintained throughout the treatment course. In order toescape resistance, under-dosing should be avoided and theduration of therapy should be limited, starting de-escalationof administered antibiotics as soon as culture results areready.80 Considering all these facts and the challenging sit-uation of resistances, the role of clinicians is currentlyenhanced since they are vital resource in the implementa-tion of strategies against worrisome pathogens.

From the pharmacodynamic perspective, antimicrobialsare basically classified according to the type of antibacte-rial activity (concentration dependent or time dependent)and the presence of post-antibiotic effect (time to bacte-rial regrowth after elimination of the antibiotic from themedia).126 According to this, three main groups can bedefined:

(1) Antibiotics with concentration-dependent activity andprolonged post-antibiotic effect. PK/PD parametersrelated to efficacy are Cmax/MIC and AUC/MIC. Com-monly used antibiotics in the ICU included in this groupare aminoglycosides, fluoroquinolones and daptomycin.Target values of PK/PD parameters are: Cmax/MICof 10---12 for aminoglycosides, and AUC0---24 h/MIC >125for fluoroquinolones in severe infections and ≥666 fordaptomycin.126,127

(2) Antibiotics with time-dependent activity and minimal ormoderate post-antibiotic effect. The PK/PD parameterrelated to efficacy is fT > MIC (time that free concen-trations exceed the MIC, expressed as % of the dosinginterval). Commonly used antibiotics in the ICU includedin this group are penicillins, cephalosporins, monobac-tams, carbapenems and macrolides, with target valuesof >50% for penicillins, >60---70% for cephalosporins andmonobactams, >30---40% for carbapenems and >40% formacrolides.126,127

(3) Antibiotics with concentration-independent action and

prolonged antibiotic effect. The PK/PD parameterrelated to efficacy is the AUC/MIC. Commonly usedantibiotics in the ICU included in this group are van-comycin, linezolid, azalides and tigecycline, with target

aIHm

e7

values of ≥400 for vancomycin, ≥100 for linezolid, ≥25for azalides and ≥15---20 for tigecycline.126,127

Antibiotics belonging to the first group can be used atigh doses and the prolonged post-antibiotic effect allowsider dosing intervals. In this sense there is good evidence

or extended duration of the dosing interval of aminogly-osides in critically ill patients12 that, in addition, reducesenal toxicity.127 For antibiotics in the second group, thebjective is the consecution of a long bacterial exposure tohe antibiotic; for this reason, continuous infusion (whenossible) is the best regimen since antibiotic serum con-entrations are constantly above the MIC for the durationf treatment. In an in vitro model the intermittent infu-ion of ceftazidime provided bactericidal activity againstusceptible P. aeruginosa strains, but not against resis-ant strains, and continuous infusion optimized t>MIC andesulted in bactericidal activity.128 Continuous infusion withn initial loading dose (to rapidly obtain bactericidal concen-rations) allows adequate concentrations at steady state,inimizing fluctuations of serum concentrations. However,

here are scarce clinical studies demonstrating the bet-er efficacy obtained with continuous versus intermittentnfusion; with reports using piperacillin/tazobactam129 oreropenem.130 In contrast, no significant differences in out-

omes and toxicity between bolus and continuous infusionf �-lactams are usually described, with a lack of studies inhe ICU.127

Finally, for antibiotics in the third group, the increasen concentrations only slightly increases bacterial eradica-ion, but highly increases a prolonged inhibition of bacterialrowth. One of the principal difficulties for vancomycin dos-ng is predicting future doses from trough level data in theCU, and therapeutic drug monitoring is needed.12 Admin-stration of vancomycin by continuous infusion has beendvocated to improve clinical outcome, although data fromCU patients are scarce. A published study showed lowerortality in ICU patients with ventilator-associated pneumo-

ia receiving continuous vancomycin infusion.131 However,he risk of nephrotoxicity associated with continuous-nfusion vancomycin requires further investigation132 sincecute kidney injury was frequently observed during continu-us vancomycin infusion in a study in critically ill patients.133

n the case of linezolid, both AUC/MIC and t>MIC (85%) cor-elate with eradication and clinical cure in ICU patients.134

owever, interstitial linezolid concentrations in patientsith sepsis suffer high inter-individual variability, supportingore frequent dosing schemes to avoid subinhibitory con-

entrations in infected tissues.135 Continuous infusion haslso been suggested for critically ill patients to obtain moretable linezolid levels and adequate AUC/MIC and t>MICalues.136

Colistin, a polymyxin agent, is in some cases theast option for the treatment of multidrug resistant. baumanni and P. aeruginosa. It exhibits a concentration-ependent activity with prolonged post-antibiotic effectt high concentrations.137 Due to its poor gastrointesti-al absorption and the classically reported nephrotoxicity

nd neurotoxicity of the intravenous formulation, in theCU setting colistin is usually used as a nebulized drug.owever, colistin can be the sole agent active againstulti-drug resistant gram-negatives in critical care, and

e

iocte

hflamscHuotm(tilt

acoiccotaaddbiiuafmst

I

Hcdidtamtaitam

R

AfoaacoepSdaadbPofOcSEc

B

Csiapdccibab3a1wetoasacpc

U

I

8

t has been suggested that its toxicity may have beenverestimated.137 The lack of PK/PD data results in a diffi-ulty for optimization of its daily dose aimed at maximizinghe AUC/MIC ratio, parameter best associated with colistinfficacy.138

In critically ill patients, in addition to alterations inepatic or renal functions, variations in the extravascularuid affect drug disposition. Hydrophilic drugs (�-lactams,minoglycosides and glycopeptides) and renally excretedoderately lipophilic agents (quinolones) have a con-

iderable risk of presenting daily fluctuations in plasmaoncentrations that may require dose adjustments.139

ydrophilic compounds tend to have much larger vol-me of distribution and tend to expand when the volumef extracellular water expands greatly, as occurs duringhe acute inflammatory phase, thus high starting dosesay be optimal.12 On the contrary, for lipophilic agents

as linezolid and macrolides), the dilution in intersti-ial fluids is less relevant, but they penetrate deepernto fatty tissues and thus, published evidence supportsarger doses in patients with a greater amount of adiposeissue.140

Critically ill patients are predisposed to drug inter-ctions due to the complexity of drug regimens. Inritically ill patients, interactions of antimicrobials withther pharmacological classes have been described, includ-ng immunosuppressants, statins, benzodiazepins, antipsy-hotics, antiepileptics, antiarrythmics, loop diuretics andalcium channel blockers.141 The drug interaction profilef �-lactams is typically associated with the inhibition ofheir renal secretion while interactions of macrolides andzalides depend on the inhibition of the CYP450 systemnd P-glycoprotein. Main interactions of aminoglycosideserive from additive or synergistic effects with otherrugs for nephrotoxicity, ototoxicity and neuromuscularlockade. For quinolones, in addition to chelation-relatednteractions, the risk of QTc prolongation implies mon-toring in patients with history of QT prolongation orncorrected electrolyte abnormalities and those receivingntiarrythmics. Few drug interactions have been describedor vancomycin (but its non-negligible nephrotoxicity, thatay increase with the concomitant use of aminoglycosides

hould be taken into account), daptomycin, linezolid andigecycline.141

nfections in the ICU environment

ospital-acquired infections affect a quarter of criti-ally ill patients and can double the risk of patientying,10 with more than one-quarter of all nosocomialnfections diagnosed in the ICU.142 Principal infectionsiagnosed and/or treated in ICU patients are: respira-ory tract infections (nosocomial pneumonia/ventilator-ssociated pneumonia (VAP)), bloodstream infections (pri-ary bacteremia/catheter-associated bacteremia), urinary

ract infections, intraabdominal infections, endocarditis,nd surgical wound infections. Table 1 shows the type of

nfection, microorganisms to be suspected in relation tohe presence or not of risk factors for multidrug resistance,nd suggested empirical treatments. Table 2 shows recom-ended antibiotic regimens for critically ill patients.

iomc

E. Maseda et al.

espiratory tract infections

etiology of early-onset infections may be distinguishedrom that of late-onset infections. When the disease devel-ps within 4 days of admission or intubation, core organismsre Streptococcus pneumoniae, Haemophilus influenzaend Moraxella catarrhalis, microorganisms associated withommunity-acquired pneumonia.143 When the disease devel-ps after 5 days, in addition to these core organisms,nterobacteria (K. pneumoniae, E. coli and the AmpC-roducing microorganisms included in the ESCPM group) and. aureus predominate.143 These last organisms also pre-ominate in patients with severe comorbidities and recentntimicrobial therapy, thus the distinction between earlynd late onset is far from absolute. In addition, longeruration of mechanical ventilation and treatments withroad-spectrum antimicrobial therapy increase the risk for. aeruginosa, Acinetobacter spp. and MRSA,143 with enter-bacteria and non-fermentative gram-negatives being morerequent in VAP versus non-VAP nosocomial pneumonia.144

f relevance is that 20 to 50% of VAP cases have polymi-robial aetiology143 and that ESKAPE organisms (E. faecium,. aureus, K. pneumoniae, A. baumannii, P. aeruginosa andnterobacter spp.), with their associated resistance profile,onstitute 80% of VAP episodes.145

loodstream infections

ritically ill patients carry much higher rates of blood-tream infections than patients in general wards, with anncidence in the ICU ranging from 3 to 10 episodes/100 ICUdmissions.146 Staphylococci seems to predominate both inrimary bloodstream infections and in those associated withevices,147---149 and although S. aureus is a frequent cause,oagulase-negative staphylococci has become the mostommon cause in the last decades.150 However, a significantncrease in the incidence of bloodstream infections causedy gram-negatives and fungi has been described.151 In

recent multinational study including 162 ICUs, ≈58%loodstream infections were caused by gram-negatives,2.8% by gram-positives, 7.8% by fungi and 1.2% by strictnaerobes.152 The rate of polymicrobial infections was2%,152 but in another study in our country the rateas considerably higher (20%).153 The increase in thempirical use of broad-spectrum antibiotics has increasedhe rate of non-classical bacterial isolates such as enter-bacteria, non-fermenters and fungi in infusion-relatednd cannula-related infections.150 Studies in the ICU havehown that Pseudomonas, Acinetobacter and enterococci inddition to staphylococci (including MRSA) are the commonause of bloodstream infections.147,148 In addition, ESBL-roducing E. coli should not be forgotten as the commonause of nosocomial bloodstream infections.154

rinary tract infections (UTIs)

t has been estimated that UTIs represent 20---50% of all ICU

nfections,155 the majority of them associated with the usef urethral catheters.156 Duration of catheterization is theain risk factor, with short-term (<30 days) duration asso-

iated with a prevalence of 30% and long-term (≥30 days)

Bugs, hosts

and ICU

environment

e9

Table 1 By type of infection, microorganisms to be suspected in relation to the presence or not of risk factors for multidrug resistance and suggested empirical treatments [VAP:ventilador-associated pneumonia; MDR: multidrug resistance; ESBL: extended-spectrum �-lactamase; ESCPM group (Enterobacter cloacae, Enterobacter aerogenes, Serratiamarcescens, Citrobacter freundii, Providencia rettgeri and Morganella morganii); MRSA: methicillin-resistant S. aureus; HACEK (Haemophilus spp., Aggregatibacter --- formerlyActinobacillus --- actinomycetemcomitans, Cardiobacterium hominis, Eikenella corrodens, Kingella spp)].

Infection type Suspected pathogens Empirical treatment Comments

Pneumonia No risk factors for MDRbacteriaS. pneumoniaeH. influenzaeS. aureus (methicillinsusceptible)EnterobacteriaceaeLegionella

Cefotaxime or ertapenem±

Azithromycin or levofloxacin

IV antibiotic treatment should not exceed >7 days

Addition of macrolides/azalides improves theprognosis of pneumococcal pneumonia

Presence of risk factors forfirst level of resistancea

Above microorganisms plus:

ESBL-producingenterobacteriaPenicillin-resistantS. pneumoniaeP. aeruginosaMRSA

Piperacillin/tazobactam orcefepime or meropenemor doripenemPLUSLevofloxacin or amikacin

±

Linezolid

ESBL-producing isolates are involved in ≈10%pneumonia caused by enterobacteria. Whenconfirmed, monotherapy with carbapenems(meropenem, imipenem, ertapenem) is indicated

Suspicion of infection by P. aeruginosa:recommended the association of twoantipseudomonal compounds is recommended

In bacteremic infections by MRSA, consider theassociation of linezolid + daptomycin

Presence of risk factors forsecond level of resistanceb

Above microorganisms plus:

Non-fermenter gramnegative bacilliAmpC and/orcarbapenemase-producingenterobacteriaMultidrug-resistantP. aeruginosa

Antipseudomonalbetalactam different fromthose previously used, withpreference for carbapenemsPLUSLevofloxacin or amikacinPLUSLinezolid

Treatment election should consider localepidemiology, previous antibiotic treatments andsusceptibility of isolates in surveillance culturesof colonizing flora

Consider administration of an inhaled antibiotic

Consider associations with colimycin, fosfomycinand tigecycline

e10

E. M

aseda et

al.

Table 1 (Continued)

Infection type Suspected pathogens Empirical treatment Comments

Bloodstream infections: primarybacteremia/catheter-associatedbacteremia

Coagulase-negativestaphylococciS. aureus (including MRSA)Enterococcus spp.E. coliKlebsiella spp.ESCPM groupP. aeruginosaAcinetobacter spp.

DaptomycinPLUSCefepime orpiperacillin/tazobactam ormeropenem or doripenem

±

Amikacin

Gram-negative bacteria should always besuspected in the critically ill patient regardlesssite of central venous catheter

If methicillin susceptibility in staphylococci isconfirmed, change to cloxacillin

In persistent (>5---7 days) or recurrent (withoutendovascular foci) bacteremia by S. aureus, asecond anti-staphylococcal drug (with or withoutrifampicin) should be added.If the patient is under cloxacillin treatment, adddaptomycinwith or without rifampicin.If the patient is under daptomycin treatment, addlinezolid or fosfomycin or cloxacillin,with or without rifampicin.If the patient is under vancomycin treatment,change to daptomycin + cloxacillin,with or without rifampicin

Candida spp. Echinocandin or fluconazol An antifungal drug with activity against Candidaspp. should be considered in critically ill patientswith central venous catheter in the femoral veinand/or parenteral nutrition, severe sepsisor recent abdominal surgery

Urinary tract infections With criteria for severesepsis or presence of riskfactors for first-levelof resistancea

Meropenem or doripenem±

Due to its high frequency, ESBL-producingenterobacteria should be considered in patientswith severe sepsis or septic shock

ESBL-producingenterobacteria

Amikacin

Presence of risk factors forsecond-level of resistanceb

Above microorganisms plus:

ESCPM groupMultidrug-resistantP.aeruginosa,Enterococcus spp.Acinetobacter spp.Candida spp.

Meropenem orDoripenem + Amikacin±

Fluconazol

Treatment election should consider localepidemiology, previous antibiotic treatments andsusceptibility of isolates in surveillance culturesof colonizing flora

Use of colimycin or tigecyclin may be necessary.Although tigecycline concentrations in urine arenot high, it may be useful in case of pyelonephritis

Bugs, hosts

and ICU

environment

e11

Table 1 (Continued)

Infection type Suspected pathogens Empirical treatment Comments

Intraabdominal infections No risk factors for MDRbacteriaE. coliK. pneumoniaeB. fragilis

Ertapenem orcefotaxime + metronidazole

In case of lack of control of the infectious foci,follow treatment recommendations in thepresence of risk factors for first-level resistance

Presence of risk factors forfirst-level of resistancea

Above microorganisms plus:

ESBL-producingenterobacteriaP. aeruginosaEnterococcus spp.MRSA

Meropenem or imipenemor ertapenem±Daptomycin or linezolidor vancomycinORTigecycline±Piperacillin/tazobactamor cefepime or amikacin

In case of lack of control of the infectious foci,follow treatment recommendations in thepresence of risk factors for second-levelresistance

Presence of risk factors forsecond-level of resistanceb

All the abovemicroorganisms plus:

Non-fermenter gramnegative bacilliAmpC and/orcarbapenemase-producingenterobacteriaMultidrug-resistantP. aeruginosaCandida spp.

Meropenem ordoripenem + daptomycinor linezolid or vancomycinORTigecycline + piperacillin/tazobactamor cefepime

±Amikacin

Treatment election should consider localepidemiology, previous antibiotic treatments andsusceptibility of isolates in surveillance culturesof colonizing flora

±Echinocandin

In critically ill patients, echinocandins are theelective treatment for Candida antifungal therapy

EndocarditisNative valveProsthetic valve>12 months post-surgery

S. aureusCoagulase-negativestaphylococciViridans group streptococciEnterococcus spp.Streptococcus bovisHACEK group

Ampicillin + cloxacillin±Gentamicin (3---5 days)

If glomerular filtrate is <40 ml/min or concomitanttreatment with potentially neurotoxic drugs,change gentamicin by daptomycin

e12

E. M

aseda et

al.

Table 1 (Continued)

Infection type Suspected pathogens Empirical treatment Comments

Risk for MRSA (includingintravenous drug usersand healthcare facilities)

Ampicillin + daptomycin + fosfomycin±Gentamicin (3---5 days)

OR

Ampicillin + vancomycin

If vancomycin MIC ≥1 mg/l, severe sepsis orbacteremia for >5 days, consider heteroresistanceor tolerance and change to daptomycin

Addition of gentamicin should be avoided ifglomerular filtrate is <40 ml/min. Consider changeto cotrimoxazole.Addition of fosfomycin should be avoided if MIC≥32 mg/l. Consider change to cotrimoxazole

Prosthetic valve <12 monthspost-surgery

MRSACoagulase-negativestaphylococciViridans group streptococciEnterococcus spp.Streptococcus bovisHACEK groupE. coliK. pneumoniaeSalmonella enteritidisP. aeruginosa

Daptomycin + rifampicin(3---5days) ± fosfomycin ± gentamicinor amikacin

PLUSMeropenem

Vancomycin could be considered when MIC≤1 mg/lfor the MRSA and normal renal function

Addition of gentamicin should be avoided ifglomerular filtrate is <40 ml/min. Consider changeto cotrimoxazole.Addition of fosfomycin should be avoided if MIC≥32 mg/l. Consider change to cotrimoxazole

Consider gentamicin if Enterococcus spp. isisolated

Skin and soft tissueinfectionsNecrotizing fasciitis(Fournier’s gangrene,early surgical woundinfection 24---48 hpost-surgery)

Group A streptococciClostridium perfringesClostridium septicumStaphylococcus aureus

Piperacillin/tazobactamor meropenem

PLUS

In infections by S. aureus producing PantonValentine leukocidin or superantigens, theantibiotic regimen should include linezolidor clindamicin

Mixed polymicrobialinfection:Enterococcus spp.Bacillus cereusE. coliP. aeruginosaKlebsiella spp.Proteus spp.Peptostreptococcus spp.Bacteroides spp.

Daptomycin or linezolidor clindamycin

OR

Tigecycline

Consider high doses of tigecycline in moderatelysevere polymicrobial infections involving MRSAand in patients with allergy to �-lactams

a Risk factors for first level of resistance: Significant comorbidities and/or antibiotic treatment for >3---5 days.b Risk factors for second level of resistance: Hospital admission and/or prolonged antibiotic treatment (>7 days).

Bugs, hosts and ICU environment e13

Table 2 Doses of common antibiotics for the treatment of infections in the critically ill patient.

Drug Dose (iv) Comments

Amikacin 20---30 mg/kg/24 hAmpicillin 2 g/6 h 1---2 g as initial dose followed by 8 g in 24 h continuous

infusionAzithromycin 500 mg/24 hCefepime 2 g/8 h 1---2 g as initial dose followed by 6 g in 24 h continuous

infusionCeftazidime 2 g/8 h 1---2 g as initial dose followed by 6 g in 24 h continuous

infusionCefotaxime 2 g/6---8 h 1---2 g as initial dose followed by 6 g in 24 h continuous

infusionCiprofloxacin 400 mg/8 hCloxacillin 2 g/4 h 1---2 g as initial dose followed by 12 g in 24 h continuous

infusionCotrimoxazole 5 mg/kg of trimetropin/8 hColimycin 9 × 106 U followed by 4.5 × 106 U/12 hDaptomycin 10 mg/kg/day May be administered as bolusDoripenem 1 g/8 h Administered as intermittent slow infusion (4 h)Ertapenem 1 g/12 hFosfomycin 4---8 g/8 h Administered as intermittent slow infusion (4 h)

or continuous infusionGentamicin 7---9 mg/kg/d (1 dose) referred to adjusted body weight; body weight = ideal

body weight + 0.4 × (total weight ---ideal weight)Imipenem 1 g/8 h Intermittent slow infusion (2 h)Levofloxacin 500 mg/12 hLinezolid 600 mg/8---12 h 1200 mg in 24 h continuous infusionMeropenem 2 g/8 h Intermittent slow infusion (3 h)Metronidazole 500 mg/8 hPiperacillin-tazobactam 4---0.5 g/6 h 2 g as initial dose followed by 16 g in 24 h continuous

infusionRifampicin 600 mg/12---24 h hTigecycline 100---200 mg followed by

50---100 mg/12 hVancomycin 15---20 mg/kg/8 h (in 1---2 h) 35 mg/kg

followed by 35 mg/kg/dayin continuous infusion

kg referred to total body weight

iortuaotrgEsiqm

duration with a 90% prevalence of UTI.157 E. coli, Klebsiella,Pseudomonas and enterococci are target bacteria associ-ated with short-term duration of catheterization whereaslong-term duration is associated, in addition to the previ-ously cited microorganisms, with members of the ESCPMgroup (with their AmpC production), and with the possibilityof polymicrobial infection.157 It should be considered thatthe most frequent source of bacteremia caused by ESBL-producing bacteria was UTI in a Dutch multicenter study,158

and that multidrug resistant UTIs may be very frequentamong patients with sepsis admitted in the ICU.159

Intraabdominal infections

Core microorganisms are enterobacteria, such as E. coli or

K. pneumoniae, and Bacteroides spp. (mainly Bacteroidesfragilis) in infections in patients with less than 5 days ofhospitalization. There are discussions about the role of Ente-rococcus spp., which in some studies plays a minor role

nsii

n secondary peritonitis160 but in others increases the ratef morbidity.161 In a published study on secondary bacte-ial peritonitis, higher rates of isolation were found whenhere was a nosocomial onset of the disease, higher val-es of Charlson and APACHE II scores, rapidly fatal diseasend ICU admission.162 When the onset of the infectionccurs in patients with >5 days of hospitalization, and thushere are risks for infection by multidrug resistant bacte-ia, in addition to core microorganisms, non-fermentativeram-negatives (P. aeruginosa, Acinetobacter spp.) andSBL-producing E. coli and K. pneumoniae should also beuspected. P. aeruginosa is more frequently isolated inntraabdominal infections of nosocomial origin and the fre-uency of ESBL producers in intraabdominal infections in aulticenter study in our country was ≈8.5% for K. pneumo-

iae and E. coli83. Importantly, the second most frequentource of bacteremia caused by ESBL-producing bacterian the Dutch multicenter study previously commented wasntraabdominal infection (after UTI).158

e

E

ImScsCSubgnetfdfaasimim

S

Bcsscttcwvri

qossrTts

C

AfTIowIcb

imehtctm

C

Aa

A

Tm

R

14

ndocarditis

nfective endocarditis still carries high morbidity andortality for the subset of patients requiring ICU admission.

taphylococci and streptococci account for the majority ofases, with trends towards a rising prevalence of cases bytaphylococcal skin flora of nosocomial iatrogenic origin.163

ommon blood cultures in infective endocarditis include. aureus (with special importance in intravenous drugsers), viridans streptococci (among them Streptococcusovis in the elderly is often associated with underlyingastrointestinal neoplasm), enterococci and coagulase-egative staphylococci.163---165 Culture-negative infectivendocarditis may account for one-third cases,166 andhe HACEK group (Haemophilus spp., Aggregatibacter ---ormerly Actinobacillus --- actinomycetemcomitans, Car-iobacterium hominis, Eikenella corrodens, Kingella spp.)or 5---10% of all cases of infective endocarditis.167 Percent-ges for each etiological agent may differ if endocarditisffects native valves or intracardiac devices. While viridanstreptococci is more frequent in native valve endocarditisn non-drug users, coagulase-negative staphylococci isore frequent in infective endocarditis in patients with

ntracardiac devices168 but, in all cases, S. aureus is theost frequent pathogen.168

urgical wound infections

acterial contamination of surgical wounds is inevitable, butommon wound pathogens depend on clean/contaminatedurgical procedures.169 For clean surgical procedures,taphylococci are the most common cause of wound infe-tions, and the patient’s microbiota has been implicated ashe most likely source. S. aureus colonization appears to behe major risk factor for developing S. aureus wound infe-tion. This has particular importance in selected populationshere colonization rates exceed 50%, such as diabetic indi-iduals and hemodyalized patients.169 Considering the highates of methicillin resistance among S. aureus, the possibil-ty of infection by MRSA isolates should always be suspected.

For contaminated procedures, wound pathogens fre-uently are among those species that comprise normal floraf the viscus entered during the surgical procedure. In thisense, polymicrobial infections are common in digestiveurgery involving colorectal procedures, with enterobacte-ia (E. coli and Klebsiella) and B. fragilis as target bacteria.he possibility of a high prevalence of intestinal coloniza-ion with ESBL-producing enterobacteria on ICU admissionhould always be considered in this context.170

onclusions

ntibiotic treatment and use of medical devices are highlyrequent in severely ill patients requiring specialized care.his fact and the concentration of high-risk patients in

CUs constitute accumulative factors for multidrug antibi-tic resistance. In hospitals, ICUs are considered areas

here antibiotic resistance problems are the largest, and

CU physicians feel that this problem is major and signifi-ant in their clinical practice. Recently new antibiotics haveeen available to overcome non-susceptibility phenotypes

E. Maseda et al.

n gram-positive microorganisms. However, the plethora ofechanisms of resistance in gram-negative bacteria, new

merging mechanisms and accumulation of resistance traitsave led to multidrug resistance, a worrisome problem in thereatment of gram-negative infections. Intensive care physi-ians should be aware of the local epidemiology of resistanceo select the most appropriate drugs in the antibiotic regi-en.

onflict of interest

ll authors complied with ethical responsibilities anduthors’ requirements, and declare no conflict of interest.

cknowledgement

he authors acknowledge Novartis S.A. for supporting theeeting of the GTIPO-SEDAR held in Valencia in May 2012.

eferences

1. Martínez JL, Baquero F. Interactions among strategies associ-ated with bacterial infection: pathogenicity, epidemicity, andantibiotic resistance. Clin Microbiol Rev. 2002;15:647---79.

2. Baquero F. Environmental stress and evolvability in microbialsystems. Clin Microbiol Infect. 2009;15 Suppl. 1:5---10.

3. Hawkey PM. The growing burden of antimicrobial resistance.J Antimicrob Chemother. 2008;62 Suppl. 1:i1---9.

4. Baquero MR, Galán JC, del Carmen Turrientes M, Cantón R,Coque TM, Martínez JL, et al. Increased mutation frequenciesin Escherichia coli isolates harboring extended-spectrum beta-lactamases. Antimicrob Agents Chemother. 2005;49:4754---6.

5. Ubeda C, Maiques E, Knecht E, Lasa I, Novick RP, Penadés JR.Antibiotic-induced SOS response promotes horizontal dissem-ination of pathogenicity island-encoded virulence factors instaphylococci. Mol Microbiol. 2005;56:836---44.

6. Giménez M, García-Rey C, Barberán J, Aguilar L. Clinicalexperience with tigecycline in the treatment of nosocomialinfections caused by isolates exhibiting prevalent resistancemechanisms. Rev Esp Quimioter. 2009;22:48---56 [Article inSpanish].

7. Graffunder EM, Preston KE, Evans AM, Venezia RA. Risk factorsassociated with extended-spectrum beta-lactamase-producingorganisms at a tertiary care hospital. J Antimicrob Chemother.2005;56:139---45.

8. Lautenbach E, Weiner MG, Nachamkin I, Bilker WB, Sheridan A,Fishman NO. Imipenem resistance among Pseudomonas aerugi-nosa isolates: risk factors for infection and impact of resistanceon clinical and economic outcomes. Infect Control Hosp Epi-demiol. 2006;27:893---900.

9. Martínez JA, Aguilar J, Almela M, Marco F, Soriano A, LópezF, et al. Prior use of carbapenems may be a significantrisk factor for extended-spectrum beta-lactamase-producingEscherichia coli or Klebsiella spp. in patients with bacter-aemia. J Antimicrob Chemother. 2006;58:1082---5.

10. Vincent JL, Rello J, Marshall J, Silva E, Anzueto A, Martin CD,et al. International study of the prevalence and outcomes ofinfection in intensive care units. JAMA. 2009;302:2323---9.

11. Bueno-Cavanillas A, Delgado-Rodríguez M, López-Luque A,Schaffino-Cano S, Gálvez-Vargas R. Influence of nosocomial

infection on mortality rate in an intensive care unit. Crit CareMed. 1994;22:55---60.

12. McKenzie C. Antibiotic dosing in critical illness. J AntimicrobChemother. 2011;66 Suppl. 2:ii25---31.

Staphylococcus aureus infections: efficacy and toxicity. Arch

Bugs, hosts and ICU environment

13. Pittet D. Infection control and quality health care in the newmillennium. Am J Infect Control. 2005;33:258---67.

14. Vincent JL, Sakr Y, Sprung CL, Ranieri VM, Reinhart K, GerlachH, et al. Sepsis in European intensive care units: results of theSOAP study. Crit Care Med. 2006;34:344---53.

15. Malacarne P, Langer M, Nascimben E, Moro ML, Giudici D,Lampati L, et al. Building a continuous multicenter infectionsurveillance system in the intensive care unit: findings fromthe initial data set of 9,493 patients from 71 Italian intensivecare units. Crit Care Med. 2008;36:1105---13.

16. Malacarne P, Boccalatte D, Acquarolo A, Agostini F, AnghileriA, Giardino M, et al. Epidemiology of nosocomial infec-tion in 125 Italian intensive care units. Minerva Anestesiol.2010;76:13---23.

17. Grundmann H, Aires-de-Sousa M, Boyce J, Tiemersma E. Emer-gence and resurgence of meticillin-resistant Staphylococcusaureus as a public-health threat. Lancet. 2006;368:874---85.

18. Olaechea PM, Insausti J, Blanco A, Luque P. Epidemiologíae impacto de las infecciones nosocomiales. Med Intensiva.2010;34:256---67 [Article in Spanish].

19. Hall IM, Barrass I, Leach S, Pittet D, Hugonnet S. Transmis-sion dynamics of methicillin-resistant Staphylococcus aureusin a medical intensive care unit. J R Soc Interface.2012;9:2639---52.

20. Schweickert B, Geffers C, Farragher T, Gastmeier P, Behnke M,Eckmanns T, et al. The MRSA-import in ICUs is an importantpredictor for the occurrence of nosocomial MRSA cases. ClinMicrobiol Infect. 2011;17:901---6.

21. Nair N, Kourbatova E, Poole K, Huckabee CM, Murray P, HuskinsWC, et al. Molecular epidemiology of methicillin-resistantStaphylococcus aureus (MRSA) among patients admitted toadult intensive care units: the STAR*ICU trial. Infect ControlHosp Epidemiol. 2011;32:1057---63.

22. Leclercq R. Epidemiological and resistance issues in multidrug-resistant staphylococci and enterococci. Clin Microbiol Infect.2009;15:224---31.

23. Tacconelli E. Antimicrobial use: risk driver of multidrug resis-tant microorganisms in healthcare settings. Curr Opin InfectDis. 2009;22:352---8.

24. Dancer SJ. The effect of antibiotics on methicillin-resistant Staphylococcus aureus. J Antimicrob Chemother.2008;61:246---53.

25. Ginn AN, Wiklendt AM, Gidding HF, George N, O’Driscoll JS,Partridge SR, et al. The ecology of antibiotic use in the ICU:homogeneous prescribing of cefepime but not tazocin selectsfor antibiotic resistant infection. PLoS ONE. 2012;7:e38719.

26. Appelbaum PC. The emergence of vancomycin-intermediateand vancomycin-resistant Staphylococcus aureus. Clin Micro-biol Infect. 2006;12 Suppl. 1:16---23.

27. French GL. Bactericidal agents in the treatment of MRSAinfections----the potential role of daptomycin. J AntimicrobChemother. 2006;58:1107---17.

28. Howe RA, Monk A, Wootton M, Walsh TR, Enright MC.Vancomycin susceptibility within methicillin-resistant Staphy-lococcus aureus lineages. Emerg Infect Dis. 2004;10:855---7.

29. Liu C, Chambers HF. Staphylococcus aureus with hetero-geneous resistance to vancomycin: epidemiology, clinicalsignificance, and critical assessment of diagnostic methods.Antimicrob Agents Chemother. 2003;47:3040---5.

30. Khosrovaneh A, Riederer K, Saeed S, Tabriz MS, Shah AR, HannaMM, et al. Frequency of reduced vancomycin susceptibilityand heterogeneous subpopulation in persistent or recurrentmethicillin-resistant Staphylococcus aureus bacteremia. ClinInfect Dis. 2004;38:1328---30.

31. Tenover FC, Moellering Jr RC. The rationale for revising

the Clinical and Laboratory Standards Institute vancomycinminimal inhibitory concentration interpretive criteria forStaphylococcus aureus. Clin Infect Dis. 2007;44:1208---15.

e15

32. Campanile F, Borbone S, Perez M, Bongiorno D, Cafiso V,Bertuccio T, et al. Heteroresistance to glycopeptides in Italianmeticillin-resistant Staphylococcus aureus (MRSA) isolates. IntJ Antimicrob Agents. 2010;36:415---9.

33. Holmes NE, Johnson PD, Howden BP. Relationship betweenvancomycin-resistant Staphylococcus aureus, vancomycin-intermediate S. aureus, high vancomycin MIC, and out-come in serious S. aureus infections. J Clin Microbiol.2012;50:2548---52.

34. Aguilar L, Giménez MJ, Barberán J. Glycopeptide heterore-sistance and tolerance in hospital grampositive isolates:‘‘invisible’’ phenomena to the clinician with clinical implica-tions? Rev Esp Quimioter. 2009;22:173---9 [Article in Spanish].

35. Bourgeois I, Pestel-Caron M, Lemeland JF, Pons JL, Caron F.Tolerance to the glycopeptides vancomycin and teicoplaninin coagulase-negative staphylococci. Antimicrob AgentsChemother. 2007;51:740---3.

36. Jones RN. Microbiological features of vancomycin in the 21stcentury: minimum inhibitory concentration creep, bacterici-dal/static activity, and applied breakpoints to predict clinicaloutcomes or detect resistant strains. Clin Infect Dis. 2006;42Suppl. 1:S13---24.

37. Gonzalez N, Sevillano D, Alou L, Cafini F, Gimenez MJ,Gomez-Lus ML, et al. Influence of the MBC/MIC ratio onthe antibacterial activity of vancomycin versus linezolidagainst methicillin-resistant Staphylococcus aureus isolates ina pharmacodynamic model simulating serum and soft tissueinterstitial fluid concentrations reported in diabetic patients.J Antimicrob Chemother. 2013;68:2291---5.

38. Traczewski MM, Katz BD, Steenbergen JN, Brown SD. Inhibitoryand bactericidal activities of daptomycin, vancomycin,and teicoplanin against methicillin-resistant Staphylococcusaureus isolates collected from 1985 to 2007. Antimicrob AgentsChemother. 2009;53:1735---8.

39. Alder J, Eisenstein B. The advantage of bactericidal drugs inthe treatment of infection. Curr Infect Dis Rep. 2004;6:251---3.

40. González C, Rubio M, Romero-Vivas J, González M, PicazoJJ. Bacteremic pneumonia due to Staphylococcus aureus:a comparison of disease caused by methicillin-resistantand methicillin-susceptible organisms. Clin Infect Dis.1999;29:1171---7.

41. Cafini F, Aguilar L, González N, Giménez MJ, Torrico M, AlouL, et al. In vitro effect of the presence of human albuminor human serum on the bactericidal activity of daptomycinagainst strains with the main resistance phenotypes in Gram-positives. J Antimicrob Chemother. 2007;59:1185---9.

42. Maor Y, Hagin M, Belausov N, Keller N, Ben-David D, Rahav G.Clinical features of heteroresistant vancomycin-intermediateStaphylococcus aureus bacteremia versus those of methicillin-resistant S. aureus bacteremia. J Infect Dis. 2009;199:619---24.

43. Howden BP, Johnson PD, Ward PB, Stinear TP, Davies JK. Iso-lates with low-level vancomycin resistance associated withpersistent methicillin-resistant Staphylococcus aureus bac-teremia. Antimicrob Agents Chemother. 2006;50:3039---47.

44. Soriano A, Marco F, Martínez JA, Pisos E, Almela M, Dimova VP,et al. Influence of vancomycin minimum inhibitory concentra-tion on the treatment of methicillin-resistant Staphylococcusaureus bacteremia. Clin Infect Dis. 2008;46:193---200.

45. Picazo JJ, Betriu C, Culebras E, Rodríguez-Avial I, Gómez M,López F, et al. Activity of daptomycin against staphylococci col-lected from bloodstream infections in Spanish medical centers.Diagn Microbiol Infect Dis. 2009;64:448---51.

46. Hidayat LK, Hsu DI, Quist R, Shriner KA, Wong-BeringerA. High-dose vancomycin therapy for methicillin-resistant

Intern Med. 2006;166:2138---44.47. Bosso JA, Nappi J, Rudisill C, Wellein M, Bookstaver PB,

Swindler J, et al. Relationship between vancomycin trough

e

16

concentrations and nephrotoxicity: a prospective multicentertrial. Antimicrob Agents Chemother. 2011;55:5475---9.

48. Gould IM. The problem with glycopeptides. Int J AntimicrobAgents. 2007;30:1---3.

49. Jones RN. Key considerations in the treatment of complicatedstaphylococcal infections. Clin Microbiol Infect. 2008;14 Suppl.2:3---9.

50. Schramm GE, Johnson JA, Doherty JA, Micek ST, Kollef MH.Methicillin-resistant Staphylococcus aureus sterile-site infec-tion: the importance of appropriate initial antimicrobialtreatment. Crit Care Med. 2006;34:2069---74.

51. Charles PG, Ward PB, Johnson PD, Howden BP, Grayson ML.Clinical features associated with bacteremia due to heteroge-neous vancomycin-intermediate Staphylococcus aureus. ClinInfect Dis. 2004;38:448---51.

52. Howden BP. Recognition and management of infections causedby vancomycin-intermediate Staphylococcus aureus (VISA) andheterogenous VISA (hVISA). Intern Med J. 2005;35 Suppl.2:S136---40.

53. Bae IG, Federspiel JJ, Miró JM, Woods CW, Park L, Rybak MJ,et al. Heterogeneous vancomycin-intermediate susceptibilityphenotype in bloodstream methicillin-resistant Staphylococ-cus aureus isolates from an international cohort of patientswith infective endocarditis: prevalence, genotype, and clinicalsignificance. J Infect Dis. 2009;200:1355---66.

54. May J, Shannon K, King A, French G. Glycopeptide tol-erance in Staphylococcus aureus. J Antimicrob Chemother.1998;42:189---97.

55. Sakoulas G, Moise-Broder PA, Schentag J, Forrest A, Moeller-ing Jr RC, Eliopoulos GM. Relationship of MIC and bactericidalactivity to efficacy of vancomycin for treatment of methicillin-resistant Staphylococcus aureus bacteremia. J Clin Microbiol.2004;42:2398---402.

56. Hidron AI, Edwards JR, Patel J, Horan TC, Sievert DM, Pol-lock DA, et al. NHSN annual update: antimicrobial-resistantpathogens associated with healthcare-associated infections:annual summary of data reported to the National Health-care Safety Network at the Centers for Disease Controland Prevention, 2006---2007. Infect Control Hosp Epidemiol.2008;29:996---1011.

57. Saribas S, Bagdatli Y. Vancomycin tolerance in enterococci.Chemotherapy. 2004;50:250---4.

58. Arias CA, Murray BE. The rise of the Enterococcus:beyond vancomycin resistance. Nat Rev Microbiol. 2012;10:266---78.

59. Werner G, Coque TM, Hammerum AM, Hope R, Hryniewicz W,Johnson A, et al. Emergence and spread of vancomycin resis-tance among enterococci in Europe. Euro Surveill. 2008;13.

60. Sader HS, Watters AA, Fritsche TR, Jones RN. Daptomycinantimicrobial activity tested against methicillin-resistantstaphylococci and vancomycin-resistant enterococci isolatedin European medical centers (2005). BMC Infect Dis. 2007;7:29.

61. Top J, Willems R, Blok H, de Regt M, Jalink K, Troelstra A, et al.Ecological replacement of Enterococcus faecalis by multiresis-tant clonal complex 17 Enterococcus faecium. Clin MicrobiolInfect. 2007;13:316---9.

62. Warren DK, Kollef MH, Seiler SM, Fridkin SK, Fraser VJ. The epi-demiology of vancomycin-resistant Enterococcus colonizationin a medical intensive care unit. Infect Control Hosp Epidemiol.2003;24:257---63.

63. Pan SC, Wang JT, Chen YC, Chang YY, Chen ML, Chang SC.Incidence of and risk factors for infection or colonization ofvancomycin-resistant enterococci in patients in the intensivecare unit. PLoS ONE. 2012;7:e47297.

64. Batistão DW, Gontijo-Filho PP, Conceicão N, Oliveira AG,

Ribas RM. Risk factors for vancomycin-resistant enterococcicolonisation in critically ill patients. Mem Inst Oswaldo Cruz.2012;107:57---63.

E. Maseda et al.

65. Sydnor ER, Perl TM. Hospital epidemiology and infection con-trol in acute-care settings. Clin Microbiol Rev. 2011;24:141---73.

66. Rivera AM, Boucher HW. Current concepts in antimicrobialtherapy against select gram-positive organisms: methicillin-resistant Staphylococcus aureus, penicillin-resistant pneumo-cocci, and vancomycin-resistant enterococci. Mayo Clin Proc.2011;86:1230---43.

67. Donskey CJ, Schreiber JR, Jacobs MR, Shekar R, Salata RA,Gordon S, et al. A polyclonal outbreak of predominantlyVanB vancomycin-resistant enterococci in northeast Ohio.Northeast Ohio Vancomycin-Resistant Enterococcus Surveil-lance Program. Clin Infect Dis. 1999;29:573---9.

68. Kolar M, Urbanek K, Vagnerova I, Koukalova D. The influenceof antibiotic use on the occurrence of vancomycin-resistantenterococci. J Clin Pharm Ther. 2006;31:67---72.

69. Song JY, Cheong HJ, Jo YM, Choi WS, Noh JY, Heo JY,et al. Vancomycin-resistant Enterococcus colonization beforeadmission to the intensive care unit: a clinico-epidemiologicanalysis. Am J Infect Control. 2009;37:734---40.

70. Kelesidis T, Tewhey R, Humphries RM. Evolution of high-leveldaptomycin resistance in Enterococcus faecium during dap-tomycin therapy is associated with limited mutations in thebacterial genome. J Antimicrob Chemother. 2013;68:1926---8.

71. Scheetz MH, Knechtel SA, Malczynski M, Postelnick MJ,Qi C. Increasing incidence of linezolid-intermediate or -resistant, vancomycin-resistant Enterococcus faecium strainsparallels increasing linezolid consumption. Antimicrob AgentsChemother. 2008;52:2256---9.

72. Dobbs TE, Patel M, Waites KB, Moser SA, Stamm AM, HoesleyCJ. Nosocomial spread of Enterococcus faecium resistant tovancomycin and linezolid in a tertiary care medical center.J Clin Microbiol. 2006;44:3368---70.

73. Kainer MA, Devasia RA, Jones TF, Simmons BP, Melton K,Chow S, et al. Response to emerging infection leading tooutbreak of linezolid-resistant enterococci. Emerg Infect Dis.2007;13:1024---30.

74. Ntokou E, Stathopoulos C, Kristo I, Dimitroulia E, LabrouM, Vasdeki A, et al. Intensive care unit dissemination ofmultiple clones of linezolid-resistant Enterococcus faecalisand Enterococcus faecium. J Antimicrob Chemother. 2012;67:1819---23.

75. Moellering Jr RC. The Garrod Lecture. The Enterococcus:a classic example of the impact of antimicrobial resistance ontherapeutic options. J Antimicrob Chemother. 1991;28:1---12.

76. Safdar A, Rolston KV. Vancomycin tolerance, a potential mech-anism for refractory gram-positive bacteremia observationalstudy in patients with cancer. Cancer. 2006;106:1815---20.

77. Lentino JR, Narita M, Yu VL. New antimicrobial agents as ther-apy for resistant gram-positive cocci. Eur J Clin Microbiol InfectDis. 2008;27:3---15.

78. Singh SR, Bacon 3rd AE, Young DC, Couch KA. In vitro 24-hourtime-kill studies of vancomycin and linezolid in combinationversus methicillin-resistant Staphylococcus aureus. AntimicrobAgents Chemother. 2009;53:4495---7.

79. Steenbergen JN, Mohr JF, Thorne GM. Effects of daptomycinin combination with other antimicrobial agents: a review ofin vitro and animal model studies. J Antimicrob Chemother.2009;64:1130---8.

80. Souli M, Galani I, Giamarellou H. Emergence of extensivelydrug-resistant and pandrug-resistant Gram-negative bacilli inEurope. Euro Surveill. 2008;13, pii: 19045.

81. Livermore DM. Fourteen years in resistance. Int J AntimicrobAgents. 2012;39:283---94.

82. Díaz MA, Hernández-Bello JR, Rodríguez-Bano J, Martínez-Martínez L, Calvo J, Blanco J, et al. Diversity of

Escherichia coli strains producing extended-spectrum beta-lactamases in Spain: second nationwide study. J Clin Microbiol.2010;48:2840---5.

1

1

1

1

1

1

1

1

1

1

1

1

1

1

Bugs, hosts and ICU environment

83. Cantón R, Loza E, Aznar J, Calvo J, Cercenado E, Cisterna R,et al. Antimicrobial susceptibility of Gram-negative organismsfrom intraabdominal infections and evolution of isolates withextended spectrum �-lactamases in the SMART study in Spain(2002---2010). Rev Esp Quimioter. 2011;24:223---32.

84. Rubio-Perez I, Martin-Perez E, Garcia DD, Calvo ML, Barrera EL.Extended-spectrum beta-lactamase-producing bacteria in atertiary care hospital in Madrid: epidemiology, risk factors andantimicrobial susceptibility patterns. Emerg Health Threats J.2012:5.

85. Paterson DL, Bonomo RA. Extended-spectrum beta-lactamases: a clinical update. Clin Microbiol Rev.2005;18:657---86.

86. Hombach M, Mouttet B, Bloemberg GV. Consequences ofrevised CLSI and EUCAST guidelines for antibiotic suscep-tibility patterns of ESBL- and AmpC �-lactamase-producingclinical Enterobacteriaceae isolates. J Antimicrob Chemother.2013;68:2092---8.

87. Rodríguez-Bano J, Navarro MD, Romero L, Muniain MA, deCueto M, Ríos MJ, et al. Bacteremia due to extended-spectrum beta -lactamase-producing Escherichia coli in theCTX-M era: a new clinical challenge. Clin Infect Dis. 2006;43:1407---14.

88. Dunne Jr WM, Hardin DJ. Use of several inducer and sub-strate antibiotic combinations in a disk approximation assayformat to screen for AmpC induction in patient isolates of Pseu-domonas aeruginosa, Enterobacter spp., Citrobacter spp., andSerratia spp. J Clin Microbiol. 2005;43:5945---9.

89. Livermore DM, Brown DF. Detection of beta-lactamase-mediated resistance. J Antimicrob Chemother. 2001;48 Suppl.1:59---64.

90. Chow JW, Fine MJ, Shlaes DM, Quinn JP, Hooper DC, John-son MP, et al. Enterobacter bacteremia: clinical features andemergence of antibiotic resistance during therapy. Ann InternMed. 1991;115:585---90.

91. Park YS, Yoo S, Seo MR, Kim JY, Cho YK, Pai H. Risk factorsand clinical features of infections caused by plasmid-mediatedAmpC beta-lactamase-producing Enterobacteriaceae. Int JAntimicrob Agents. 2009;34:38---43.

92. Livermore DM, Brown DF, Quinn JP, Carmeli Y, Paterson DL,Yu VL. Should third-generation cephalosporins be avoidedagainst AmpC-inducible Enterobacteriaceae? Clin MicrobiolInfect. 2004;10:84---5.

93. Harris PN, Ferguson JK. Antibiotic therapy for inducible AmpC�-lactamase-producing Gram-negative bacilli: what are thealternatives to carbapenems, quinolones and aminoglycosides?Int J Antimicrob Agents. 2012;40:297---305.

94. Mushtaq S, Irfan S, Sarma JB, Doumith M, Pike R, Pitout J,et al. Phylogenetic diversity of Escherichia coli strains pro-ducing NDM-type carbapenemases. J Antimicrob Chemother.2011;66:2002---5.

95. Palzkill T. Metallo-�-lactamase structure and function. Ann NY Acad Sci. 2013;1277:91---104.

96. Wybo I, De Bel A, Soetens O, Echahidi F, Vandoorslaer K, VanCauwenbergh M, et al. Differentiation of cfiA-negative andcfiA-positive Bacteroides fragilis isolates by matrix-assistedlaser desorption ionization-time of flight mass spectrometry.J Clin Microbiol. 2011;49:1961---4.

97. Bonomo RA, Szabo D. Mechanisms of multidrug resistancein Acinetobacter species and Pseudomonas aeruginosa. ClinInfect Dis. 2006;43 Suppl. 2:S49---56.

98. Livermore DM. Has the era of untreatable infections arrived?J Antimicrob Chemother. 2009;64 Suppl. 1:i29---36.

99. Walsh TR. Clinically significant carbapenemases: an update.

Curr Opin Infect Dis. 2008;21:367---71.

100. Suarez C, Pena C, Campo A, Murillas J, Almirante B,Pomar V, et al. Impact of carbapenem resistance on Pseu-domonas aeruginosa (PA) bloodstream infections outcome.

1

e17

In: 49th interscience conference on antimicrobial agents andchemotherapy. 2009 [Abstract K-332].

01. Georges B, Conil JM, Dubouix A, Archambaud M, Bonnet E,Saivin S, et al. Risk of emergence of Pseudomonas aeruginosaresistance to beta-lactam antibiotics in intensive care units.Crit Care Med. 2006;34:1636---41.

02. Girlich D, Naas T, Nordmann P. Biochemical characterizationof the naturally occurring oxacillinase OXA-50 of Pseudomonasaeruginosa. Antimicrob Agents Chemother. 2004;48:2043---8.

03. Lopez-Otsoa F, Gallego L, Towner KJ, Tysall L, Woodford N, Liv-ermore DM. Endemic carbapenem resistance associated withOXA-40 carbapenemase among Acinetobacter baumannii iso-lates from a hospital in northern Spain. J Clin Microbiol.2002;40:4741---3.

04. Da Silva GJ, Quinteira S, Bértolo E, Sousa JC, Gal-lego L, Duarte A, et al. Long-term dissemination of anOXA-40 carbapenemase-producing Acinetobacter baumanniiclone in the Iberian Peninsula. J Antimicrob Chemother.2004;54:255---8.

05. Picazo JJ, Betriu C, Rodríguez-Avial I, Culebras E, Gómez M,López F, et al. Antimicrobial resistance surveillance: VIRASTUDY 2006. Enferm Infecc Microbiol Clin. 2006;24:617---28[Article in Spanish].

06. Kluytmans-Vandenbergh MF, Kluytmans JA, Voss A. Dutchguideline for preventing nosocomial transmission of highlyresistant microorganisms (HRMO). Infection. 2005;33:309---13.

07. Cisneros JM, Rodríguez-Bano J, Fernández-Cuenca F, RiberaA, Vila J, Pascual A, et al. Risk-factors for the acqui-sition of imipenem-resistant Acinetobacter baumannii inSpain: a nationwide study. Clin Microbiol Infect. 2005;11:874---9.

08. del Mar Tomas M, Cartelle M, Pertega S, Beceiro A, LlinaresP, Canle D, et al. Hospital outbreak caused by a carbapenem-resistant strain of Acinetobacter baumannii: patient prognosisand risk-factors for colonisation and infection. Clin MicrobiolInfect. 2005;11:540---6.

09. Castanheira M, Mendes RE, Woosley LN, Jones RN. Trendsin carbapenemase-producing Escherichia coli and Klebsiellaspp. from Europe and the Americas: report from the SENTRYantimicrobial surveillance programme (2007---09). J AntimicrobChemother. 2011;66:1409---11.

10. Cantón R, Akóva M, Carmeli Y, Giske CG, Glupczynski Y, Gniad-kowski M, et al. Rapid evolution and spread of carbapenemasesamong Enterobacteriaceae in Europe. Clin Microbiol Infect.2012;18:413---31.

11. Sánchez-Romero I, Asensio A, Oteo J, Munoz-Algarra M, IsidoroB, Vindel A, et al. Nosocomial outbreak of VIM-1-producingKlebsiella pneumoniae isolates of multilocus sequence type15: molecular basis, clinical risk factors, and outcome. Antimi-crob Agents Chemother. 2012;56:420---7.

12. Miró E, Segura C, Navarro F, Sorlí L, Coll P, Horcajada JP,et al. Spread of plasmids containing the bla(VIM-1) andbla(CTX-M) genes and the qnr determinant in Enterobactercloacae, Klebsiella pneumoniae and Klebsiella oxytoca iso-lates. J Antimicrob Chemother. 2010;65:661---5.

13. Tato M, Coque TM, Ruíz-Garbajosa P, Pintado V, Cobo J, SaderHS, et al. Complex clonal and plasmid epidemiology in thefirst outbreak of Enterobacteriaceae infection involving VIM-1 metallo-beta-lactamase in Spain: toward endemicity? ClinInfect Dis. 2007;45:1171---8.

14. Cendejas E, Gómez-Gil R, Gómez-Sánchez P, Mingorance J.Detection and characterization of Enterobacteriaceae pro-ducing metallo-beta-lactamases in a tertiary-care hospital in

Spain. Clin Microbiol Infect. 2010;16:181---3.

15. Curiao T, Morosini MI, Ruiz-Garbajosa P, Robustillo A, BaqueroF, Coque TM, et al. Emergence of bla KPC-3-Tn4401a asso-ciated with a pKPN3/4-like plasmid within ST384 and ST388

e

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

18

Klebsiella pneumoniae clones in Spain. J AntimicrobChemother. 2010;65:1608---14.

16. Solé M, Pitart C, Roca I, Fàbrega A, Salvador P, Munoz L,et al. First description of an Escherichia coli strain producingNDM-1 carbapenemase in Spain. Antimicrob Agents Chemother.2011;55:4402---4.

17. Struelens MJ, Monnet DL, Magiorakos AP, Santos O’ConnorF, Giesecke J. European NDM-1 Survey Participants. NewDelhi metallo-beta-lactamase 1-producing Enterobacteri-aceae: emergence and response in Europe. Euro Surveill.2010;15.

18. European Center for Disease Prevention and Control. Antimi-crobial resistance surveillance in Europe; 2011. Availableat: http://ecdc.europa.eu/en/publications/Publications/antimicrobial-resistance-surveillance-europe-2011.pdf

19. Falagas ME, Rafailidis PI, Kofteridis D, Virtzili S, ChelvatzoglouFC, Papaioannou V, et al. Risk factors of carbapenem-resistantKlebsiella pneumoniae infections: a matched case controlstudy. J Antimicrob Chemother. 2007;60:1124---30.

20. Patel G, Huprikar S, Factor SH, Jenkins SG, Calfee DP.Outcomes of carbapenem-resistant Klebsiella pneumoniaeinfection and the impact of antimicrobial and adjunctive ther-apies. Infect Control Hosp Epidemiol. 2008;29:1099---106.

21. Gupta N, Limbago BM, Patel JB, Kallen AJ. Carbapenem-resistant Enterobacteriaceae: epidemiology and prevention.Clin Infect Dis. 2011;53:60---7.

22. Schwaber MJ, Klarfeld-Lidji S, Navon-Venezia S, SchwartzD, Leavitt A, Carmeli Y. Predictors of carbapenem-resistantKlebsiella pneumoniae acquisition among hospitalized adultsand effect of acquisition on mortality. Antimicrob AgentsChemother. 2008;52:1028---33.

23. Daikos GL, Petrikkos P, Psichogiou M, Kosmidis C, Vryonis E,Skoutelis A, et al. Prospective observational study of theimpact of VIM-1 metallo-beta-lactamase on the outcome ofpatients with Klebsiella pneumoniae bloodstream infections.Antimicrob Agents Chemother. 2009;53:1868---73.

24. Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, OpalSM, et al. Surviving Sepsis Campaign: international guidelinesfor management of severe sepsis and septic shock, 2012. Inten-sive Care Med. 2013;39:165---228.

25. Martinez MN, Papich MG, Drusano GL. Dosing regimen matters:the importance of early intervention and rapid attainmentof the pharmacokinetic/pharmacodynamic target. AntimicrobAgents Chemother. 2012;56:2795---805.

26. Sociedad Espanola de Enfermedades Infecciosas y Micro-biología Clínica. Procedimientos en Microbiología Clínica.Análisis farmacocinéticofarmacodinámico en Microbiología:herramienta para evaluar el tratamiento antimicro-biano http://www.seimc.org/documentos/protocolos/microbiologia/

27. Scaglione F, Paraboni L. Pharmacokinetics/pharmacodynamicsof antibacterials in the Intensive Care Unit: setting appropriatedosing regimens. Int J Antimicrob Agents. 2008;32:294---301.

28. Alou L, Aguilar L, Sevillano D, Giménez MJ, EcheverríaO, Gómez-Lus ML, et al. Is there a pharmacodynamicneed for the use of continuous versus intermittent infu-sion with ceftazidime against Pseudomonas aeruginosa? Anin vitro pharmacodynamic model. J Antimicrob Chemother.2005;55:209---13.

29. Lorente L, Jiménez A, Martín MM, Iribarren JL, Jiménez JJ,Mora ML. Clinical cure of ventilator-associated pneumoniatreated with piperacillin/tazobactam administered by con-tinuous or intermittent infusion. Int J Antimicrob Agents.2009;33:464---8.

30. Lorente L, Lorenzo L, Martín MM, Jiménez A, Mora ML.

Meropenem by continuous versus intermittent infusion inventilator-associated pneumonia due to gram-negative bacilli.Ann Pharmacother. 2006;40:219---23.

E. Maseda et al.

31. Rello J, Sole-Violan J, Sa-Borges M, Garnacho-Montero J,Munoz E, Sirgo G, et al. Pneumonia caused by oxacillin-resistant Staphylococcus aureus treated with glycopeptides.Crit Care Med. 2005;33:1983---7.

32. DiMondi VP, Rafferty K. Review of continuous-infusion van-comycin. Ann Pharmacother. 2013;47:219---27.

33. Spapen HD, Janssen van Doorn K, Diltoer M, Verbrugghe W,Jacobs R, Dobbeleir N, et al. Retrospective evaluation ofpossible renal toxicity associated with continuous infusionof vancomycin in critically ill patients. Ann Intensive Care.2011;1:26.

34. Rayner CR, Forrest A, Meagher AK, Birmingham MC, Schen-tag JJ. Clinical pharmacodynamics of linezolid in seriously illpatients treated in a compassionate use programme. Clin Phar-macokinet. 2003;42:1411---23.

35. Buerger C, Plock N, Dehghanyar P, Joukhadar C, KloftC. Pharmacokinetics of unbound linezolid in plasma andtissue interstitium of critically ill patients after multipledosing using microdialysis. Antimicrob Agents Chemother.2006;50:2455---63.

36. Adembri C, Fallani S, Cassetta MI, Arrigucci S, Ottaviano A,Pecile P, et al. Linezolid pharmacokinetic/pharmacodynamicprofile in critically ill septic patients: intermittent versus con-tinuous infusion. Int J Antimicrob Agents. 2008;31:122---9.

37. Ratjen F, Rietschel E, Kasel D, Schwiertz R, Starke K, BeierH, et al. Pharmacokinetics of inhaled colistin in patients withcystic fibrosis. J Antimicrob Chemother. 2006;57:306---11.

38. Michalopoulos AS, Falagas ME. Colistin: recent data on phar-macodynamics properties and clinical efficacy in critically illpatients. Ann Intensive Care. 2011;1:30.

39. Pea F, Viale P, Furlanut M. Antimicrobial therapy in critically illpatients: a review of pathophysiological conditions responsiblefor altered disposition and pharmacokinetic variability. ClinPharmacokinet. 2005;44:1009---34.

40. Erstad BL. Dosing of medications in morbidly obese patientsin the intensive care unit setting. Intensive Care Med.2004;30:18---32.

41. Pereira JM, Paiva JA. Antimicrobial drug interactions in thecritically ill patients. Curr Clin Pharmacol. 2013;8:25---38.

42. Palomar M, Vaque J, Alvarez Lerma F, Pastor V, Olaechea P,Fernández-Crehuet J. Nosocomial infection indicators. MedClin (Barc). 2008;131 Suppl. 3:48---55 [Article in Spanish].

43. Strausbaugh LJ. Nosocomial respiratory infections. In: MandellGL, Bennett JE, Dolin R, editors. Mandell, Douglas, Bennett’sprinciples and practice of infectious diseases. 6th ed. Philadel-phia: Elsevier Inc.; 2005. p. 3362---70.

44. Esperatti M, Ferrer M, Theessen A, Liapikou A, Valencia M,Saucedo LM, et al. Nosocomial pneumonia in the intensive careunit acquired by mechanically ventilated versus nonventilatedpatients. Am J Respir Crit Care Med. 2010;182:1533---9.

45. Sandiumenge A, Rello J. Ventilator-associated pneumoniacaused by ESKAPE organisms: cause, clinical features, andmanagement. Curr Opin Pulm Med. 2012;18:187---93.

46. Maseda Garrido E, Alvarez J, Garnacho-Montero J, Jerez V,Lorente L, Rodríguez O. Update on catheter-related blood-stream infections in ICU patients. Enferm Infecc Microbiol Clin.2011;29 Suppl. 4:10---5.

47. Olaechea PM, Palomar M, Álvarez-Lerma F, Otal JJ, InsaustiJ, López-Pueyo MJ, et al. Morbidity and mortality associatedwith primary and catheter-related bloodstream infections incritically ill patients. Rev Esp Quimioter. 2013;26:21---9.

48. Orsi GB, Franchi C, Marrone R, Giordano A, Rocco M, VendittiM. Laboratory confirmed bloodstream infection aetiology in anintensive care unit: eight years study. Ann Ig. 2012;24:269---78.

49. Diekema DJ, Beekmann SE, Chapin KC, Morel KA, Munson

E, Doern GV. Epidemiology and outcome of nosocomial andcommunity-onset bloodstream infection. J Clin Microbiol.2003;41:3655---60.

1

1

1

1

1

1

1

1

1

1

1

Bugs, hosts and ICU environment

150. Beekmann SE, Henderson DK. Infections caused by percuta-neous intravascular devices. In: Mandell GL, Bennett JE, DolinR, editors. Mandell, Douglas, Bennett’s principles and practiceof infectious diseases. 6th ed. Philadelphia: Elsevier Inc; 2005.p. 3347---62.

151. Rodríguez-Créixems M, Munoz P, Martín-Rabadán P, CercenadoE, Guembe M, Bouza E. Evolution and aetiological shift ofcatheter-related bloodstream infection in a whole institution:the microbiology department may act as a watchtower. ClinMicrobiol Infect. 2012;19:845---51.

152. Tabah A, Koulenti D, Laupland K, Misset B, Valles J, Bruzzide Carvalho F, et al. Characteristics and determinants of out-come of hospital-acquired bloodstream infections in intensivecare units: the EUROBACT International Cohort Study. IntensiveCare Med. 2012;38:1930---45.

153. Sancho S, Artero A, Zaragoza R, Camarena JJ, González R,Nogueira JM. Impact of nosocomial polymicrobial bloodstreaminfections on the outcome in critically ill patients. Eur J ClinMicrobiol Infect Dis. 2012;31:1791---6.

154. Rodríguez-Bano J, Picón E, Gijón P, Hernández JR, Cis-neros JM, Pena C, et al. Risk factors and prognosis ofnosocomial bloodstream infections caused by extended-spectrum-beta-lactamase-producing Escherichia coli. J ClinMicrobiol. 2010;48:1726---31.

155. López MJ, Cortés JA. Urinary tract colonization and infectionin critically ill patients. Med Intensiva. 2012;36:143---51.

156. Richards MJ, Edwards JR, Culver DH, Gaynes RP. Nosocomialinfections in medical intensive care units in the United States.National Nosocomial Infections Surveillance System. Crit CareMed. 1999;27:887---92.

157. Warren JW. Nosocomial urinary tract infections. In: MandellGL, Bennett JE, Dolin R, editors. Mandell, Douglas, Bennett’sprinciples and practice of infectious diseases. 6th ed. Philadel-phia: Elsevier Inc; 2005. p. 3370---81.

158. Frakking FN, Rottier WC, Dorigo-Zetsma JW, van Hattem JM,van Hees BC, Kluytmans JA, et al. Appropriateness of empiricaltreatment and outcome in bacteremia caused by extended-spectrum �-lactamase producing bacteria. Antimicrob AgentsChemother. 2013;57:3092---9.

159. Mojtahedzadeh M, Panahi Y, Fazeli MR, Najafi A, Pazouki M,

Navehsi BM, et al. Intensive care unit-acquired urinary tractinfections in patients admitted with sepsis: etiology, risk fac-tors, and patterns of antimicrobial resistance. Int J Infect Dis.2008;12:312---8.

e19

60. Röhrborn A, Wacha H, Schöffel U, Billing A, Aeberhard P, Geb-hard B, et al. Coverage of enterococci in community acquiredsecondary peritonitis: results of a randomized trial. Surg Infect(Larchmt). 2000;1:95---107.

61. Gauzit R, Péan Y, Barth X, Mistretta F, Lalaude O, TopStudy Team. Epidemiology, management, and prognosis ofsecondary non-postoperative peritonitis: a French prospec-tive observational multicenter study. Surg Infect (Larchmt).2009;10:119---27.

62. Cercenado E, Torroba L, Cantón R, Martínez-Martínez L, ChavesF, García-Rodríguez JA, et al. Multicenter study evaluating therole of enterococci in secondary bacterial peritonitis. J ClinMicrobiol. 2010;48:456---9.

63. Farinas MC, Llinares P, Almirante B, Barberán J, de DiosColmenero J, Garau J, et al. New trends in infective endo-carditis. Enferm Infecc Microbiol Clin. 2011;29 Suppl. 4:22---35.

64. Fernández Guerrero ML, Goyenechea A, Verdejo C, RoblasRF, de Górgolas M. Enterococcal endocarditis on native andprosthetic valves: a review of clinical and prognostic fac-tors with emphasis on hospital-acquired infections as a majordeterminant of outcome. Medicine (Baltimore). 2007;86:363---77.

65. Pierce D, Calkins BC, Thornton K. Infectious endocarditis: diag-nosis and treatment. Am Fam Phys. 2012;85:981---6.

66. Naber CK, Erbel R. Infective endocarditis with negative bloodcultures. Int J Antimicrob Agents. 2007;30 Suppl. 1:S32---6.

67. Tornos P, Gonzalez-Alujas T, Thuny F, Habib G. Infectiveendocarditis: the European viewpoint. Curr Probl Cardiol.2011;36:175---222.

68. Murdoch DR, Corey GR, Hoen B, Miró JM, Fowler Jr VG, BayerAS, et al. Clinical presentation, etiology, and outcome ofinfective endocarditis in the 21st century: the InternationalCollaboration on Endocarditis-Prospective Cohort Study. ArchIntern Med. 2009;169:463---73.

69. Talbot TR, Kaiser AB. Post-operative infections and antimi-crobial prophylaxis. In: Mandell GL, Bennett JE, Dolin R,editors. Mandell, Douglas, Bennett’s principles and practice ofinfectious diseases. 6th ed. Philadelphia: Elsevier Inc.; 2005.p. 3533---47.

70. Razazi K, Derde LP, Verachten M, Legrand P, Lesprit P,

Brun-Buisson C. Clinical impact and risk factors for coloniza-tion with extended-spectrum �-lactamase-producing bacteriain the intensive care unit. Intensive Care Med. 2012;38:1769---78.