Leitthema

Anaesthesist 2019 · 68 (Suppl 1):S40–S62https://doi.org/10.1007/s00101-017-0396-zPublished online: 30 January 2018© Springer Medizin Verlag GmbH, ein Teil vonSpringer Nature 2017

D. C. Richter1 · A. Heininger2 · T. Brenner1 · M. Hochreiter1 · M. Bernhard3 ·J. Briegel4 · S. Dubler1 · B. Grabein5 · A. Hecker6 · W. A. Kruger7 · K. Mayer8 ·M. W. Pletz9 · D. Storzinger8 · N. Pinder8 · T. Hoppe-Tichy2 · S. Weiterer1 ·S. Zimmermann2 · A. Brinkmann10 · M. A. Weigand1 · C. Lichtenstern1

1 Department of Anesthesiology, Heidelberg University Hospital, Heidelberg, Germany2 Zentrum für Infektiologie, Sektion für Krankenhaus- und Umwelthygiene, UniversitätsklinikumHeidelberg, Heidelberg, Germany

3 Zentrale Notaufnahme, UniversitätsklinikumLeipzig, Leipzig, Germany4 Klinik für Anästhesiologie, Klinikum der Universität München, München, Germany5 Stabsstelle “KlinischeMikrobiologie und Krankenhaushygiene”, Klinikum der Universität München,München, Germany

6 Klinik für Allgemein-, Viszeral-, Thorax-, Transplantations- und Kinderchirurgie, UniversitätsklinikumGießen und Marburg, Standort Gießen, Gießen, Germany

7 Klinik für Anästhesiologie und operative Intensivmedizin, Gesundheitsverbund Landkreis Konstanz,Klinikum Konstanz, Konstanz, Germany

8 Apotheke, UniversitätsklinikumsHeidelberg, Heidelberg, Germany9 Zentrum für Infektionsmedizin und Krankenhaushygiene, Universitätsklinikum Jena, Jena, Germany10 Klinik für Anästhesie, operative Intensivmedizin und spezielle Schmerztherapie, KlinikumHeidenheim,Heidenheim, Germany

Bacterial sepsisDiagnostics and calculated antibiotictherapy

Principles

Diagnostic criteria

Early diagnosis and rapid initiationof treatment are crucial factors in thetreatment of sepsis and septic shock.Despite detailed criteria, establishmentof diagnosis based on the classical sys-temic inflammatory response syndrome(SIRS) concept was associated with rel-evant problems and gray areas, whichled to an underestimation of the dis-ease [1]. For example, the retrospectiveanalysis conducted by Kaukonen et al.[2], including a total of 109,663 patients,revealed that 12.5% of patients—despitesevere infection and new-onset organdisfunction—did not fulfil the necessarySIRS criteria (according to SEPSIS-1 [3])and the “sepsis” diagnosis was not estab-

The German version of this article can befound under https://doi.org/10.1007/s00101-017-0363-8.

Contribution available free of charge by “FreeAccess” (https://link.springer.com/article/10.1007/s00101-017-0396-z).

lished. Vice versa, an array of patientsfulfilled the SIRS criteria at some pointduring their period of hospitalization,without ever developing a relevant infec-tion or showing an associated increasedmortality [4, 5]. Due to the poor validityof the SIRS concept (SEPSIS-1/2) and thedemand for more sensitive diagnosticcriteria, during The Third InternationalConsensus Conference on the Definitionof Sepsis and Septic Shock (SEPSIS-3),simplified and (presumably) more feasi-ble diagnostic criteria were established.According to the SEPSIS-3 task force,high-risk patients can be identified (e. g.,in an outpatient setting and the emer-gency department) by applying a simplemodified quick SOFA score (qSOFA,SOFA: sequential organ failure assess-ment; [5]). The objective of qSOFAscreening is identification of patientswith probable sepsis, who then undergofurther diagnostic tests and intensivemonitoring. In patients fulfilling twoof the three qSOFA criteria (respiratoryrate >22/min, altered mentation, systolicblood pressure <100mmHg), intensive

medical care and further diagnostic testsare indicated (including measurementof lactate levels). During the subsequentcourse, the classical SOFA score andnew-onset organ dysfunction form thebasis of a definitive “sepsis” diagnosis(SOFA score ≥2). Organ dysfunction isthus as of now the decisive diagnosticrequirement.

Alongside hemodynamic, symp-tomatic treatment, the proclaimed goalsof SEPSIS-3 are also rapid identificationof the source of sepsis (and treatment ofthe infection) and earliest possible initia-tion of a calculated anti-infective therapy[5–7]. Whether or not SEPSIS-3 can ful-fil all expectations remains questionable.On a critical note, the introduction oforgan dysfunction as an obligatory di-agnostic criterium may, under certaincircumstances, delay definitive diagnosisuntil a later, possibly more severe diseasestage [8]. A consequence of this wouldbe, e. g., a delayed (and potentially lesseffective) initiation of treatment. More-over, diagnosis of sepsis based on theSOFA score is not undisputed. For in-

S40 Der Anaesthesist · Suppl 1 · 2019

Table 1 Typical pathogens of selected intensive caremedicine infections

CAP HAP/VAP Skin/soft tissue Catheter-asso-ciated BSI

Intestine (sec-ondary/tertiaryperitonitis)

Streptococcus pneu-moniae

Enterobacteri-aceae

Streptococcuspyogenes

CoNS Enterobacteriaceae

Mycoplasma pneu-moniae

Haemophilusinfluenzae

Staphylococcusaureus

S. aureus Anaerobes

H. influenzae MSSA Anaerobes Gram-negativebacilli

Enterococci

S. aureus S. pneumoniae Enterobacteriaceae(+ Clostridiaceae)

Corynebacteriumjeikeium

Nosocomial/postoperative

Rarer CAPpathogens

CommonMDRpathogens

Pseudomoasaeruginosa

Propionibacteria

Nosocomial/pretreatment

Candida species Enterobacteriaceae(ESBL)

Enterobacteriaceae MRSA Nosocomial/pretreatment

Enterococci (VRE)

Chlamydophilaspecies

Enterobacteri-aceae (ESBL)

MRSA MRSA

Legionella species Pseudomonasaeruginosa

MRGN MRSA Anaerobes

– Acinetobacterbaumannii

– MRGN Pseudomonasspecies

Stenotropho-monasmal-tophilia

Enterococci (VRE) Candida species

BSI bloodstream infection (bacteremia), CAP community-acquired pneumonia, ESBL extend-ed-spectrum β-lactamases, HAP hospital-acquired pneumonia (nosocomial pneumonia), CoNS co-agulase-negative staphylococci, MDR multidrug-resistant pathogens, MRSA multiresistant/methicillin-resistant Staphylococcus aureus, MRGN multiresistant gram-negative pathogens,MSSA multisensitive Staphylococcus aureus, VAP ventilator-associated pneumonia, VRE van-comycin-resistant enterococci

stance, patients with chronic diseasesaffecting one or more axes (e. g., ter-minal renal failure; chronic obstructivepulmonary disease, COPD; hepatic in-sufficiency) already have a high SOFAscore at baseline, which can hardly beincreased by the presence of sepsis.

Conclusion. Sepsis is a life-threateningorgan dysfunction caused by an infec-tion. Diagnosis is established accordingto the SEPSIS-3 criteria. As of now, thenewly introduced qSOFA score serves toidentify high-risk patients outside of theintensive care unit (ICU); subsequently,definitive diagnosis is established usingthe SOFA score and the additional cri-terium of new-onset organ dysfunction.The objective is an increased sepsis de-tection rate.

Epidemiology

Pathophysiologically, sepsis is a dysreg-ulated host response to infection [5, 9,10]. Septic shock is a particularly se-rious course of sepsis with most severecellular-metabolic and cardiocirculatoryproblems, as well as an associated veryhigh rate of patient mortality (up to 50%;[9, 11]). Similar to the situation in stroke,myocardial infarction, and very severelyinjured patients (polytrauma), early di-agnosis and treatment play a crucial rolein sepsis. Increasing incidences duringthe past decade [12–14] have renderedsepsis a medical challenge in intensivecare medicine [15]. Solid epidemiologicdata on sepsis differ according to geog-raphy and due to the diversity of studieson the topic. For example, in retrospec-tive works by Fleischmann et al. [12]between 2007 and 2013, an annual in-crease in sepsis incidence of 5.7%was ob-served. Patient mortality was very high,

up to 55% (47% without and 62% withshock; [12, 13, 16]). The multicentricprospective Incidence of Severe Sepsisand Septic Shock in German IntensiveCare Units (INSEP; [17]) trial publishedin 2016 delivered similar results: 12.6%of participants had severe sepsis or sep-tic shock (1503 from 11,883 patients).The calculated incidence rate of severesepsis and septic shock reached 11.64%per 1000 treatment days (95%confidenceinterval, 95%CI, 10.51–12.86) in INSEP[17]. Mortality in this collective was34.3% during the stay in the ICU and40.4% for the hospital stay. Data frominternational studies vary within a com-parable range [14, 18–22]. Whereas sep-sis is the third most common cause ofdeath in non-surgical ICUs [23, 24], itis the primary cause of death in surgicalICUs [15]. However, in addition to thehigh mortality rates and still poor ther-apeutic outcomes, sepsis also generatesrelevant costs for the health care systemevery year [21].

Sepsis begins with an initial infectiousstimulus and a dysregulated immune re-sponse to this infection. From an infec-tious diseases (ID) perspective, nosoco-mial infections play a special role. InINSEP [17], 57.2% of the detected infec-tionswere ofnosocomial origin; in25.7%of these nosocomial infections, infectionoccurred in the ICU. Upon consideringthe entire ICU population, the infectionis in the region of the airways in about60–63% of patients, is intraabdominal in25–30%, and affects skin and soft tissuesin up to 10% [24–26]. In surgical inten-sive medicine, intraabdominal and softtissue infections are the primary start-ing points for sepsis. Important and fre-quently occurring “problem pathogens”causing selected diseases are presentedin . Table 1.

The spectrum of pathogens includesboth gram-positive cocci such as Staphy-lococcus aureus and Streptococcus pneu-moniae, as well as gram-negative bacilli,e. g., the Enterobacteriaceae Escherichiacoli andKlebsiellapneumoniae, and Pseu-domonas aeruginosa [27–29]. Since thenumber of infections with multidrug-resistant (MDR)pathogens has increasedsignificantly during recent years, theseare now also more frequently the trigger

Der Anaesthesist · Suppl 1 · 2019 S41

Abstract · Zusammenfassung

Anaesthesist 2019 · 68 (Suppl 1):S40–S62 https://doi.org/10.1007/s00101-017-0396-z© Springer Medizin Verlag GmbH, ein Teil von Springer Nature 2017

D. C. Richter · A. Heininger · T. Brenner · M. Hochreiter · M. Bernhard · J. Briegel · S. Dubler · B. Grabein · A. Hecker · W. A. Kruger · K. Mayer ·M. W. Pletz · D. Storzinger · N. Pinder · T. Hoppe-Tichy · S. Weiterer · S. Zimmermann · A. Brinkmann · M. A. Weigand · C. Lichtenstern

Bacterial sepsis. Diagnostics and calculated antibiotic therapy

AbstractThe mortality of patients with sepsis andseptic shock is still unacceptably high. Aneffective calculated antibiotic treatmentwithin 1 h of recognition of sepsis is animportant target of sepsis treatment. Delayslead to an increase in mortality; therefore,structured treatment concepts form a rationalfoundation, taking relevant diagnostic andtreatment steps into consideration. In additionto the assumed infection and individual risksof each patient, local resistance patterns andspecific problem pathogens must be takeninto account during the selection of anti-infective treatment. Many pathophysiologicalterations influence the pharmacokinetics(PK) of antibiotics during sepsis. The principleof standard dosing should be abandoned andreplaced by an individual treatment approachwith stronger weighting of the pharmaco-kinetics/pharmacodynamics (PK/PD) indexof the substance groups. Although this isnot yet the clinical standard, prolonged (orcontinuous) infusion of β-lactam antibiotics

and therapeutic drug monitoring (TDM)can help to achieve defined PK targets.Prolonged infusion is sufficient without TDM,but for continuous infusion, TDM is generallynecessary. A further argument for individualPK/PD-oriented antibiotic approaches isthe increasing number of infections due tomultidrug-resistant (MDR) pathogens in theintensive care unit. For effective treatment,antibiotic stewardship teams (ABS teams) arebecomingmore established. Interdisciplinarycooperation of the ABS teamwith infectiousdisease (ID) specialists, microbiologists,and clinical pharmacists leads not only torational administration of antibiotics, butalso has a positive influence on treatmentoutcome. The gold standards for pathogenidentification are still culture-based detectionand microbiologic resistance testing for thevarious antibiotic groups. Despite the rapidinvestigation time, novel polymerase chainreaction(PCR)-based procedures for pathogenidentification and resistance determination

are currently only an adjunct to routine sepsisdiagnostics, due to the limited number ofstudies, high costs, and limited availability. Incomplicated septic courses with multiple anti-infective therapies or recurrent sepsis, PCR-based procedures can be used in addition totreatment monitoring and diagnostics. Novelantibiotics represent potent alternatives inthe treatment of MDR infections. Due to theoften defined spectrum of pathogens and thepractically (still) absent resistance, they aresuitable for targeted treatment of severe MDRinfections (therapy escalation). (Contributionavailable free of charge by “Free Access”[https://link.springer.com/article/10.1007/s00101-017-0396-z].)

KeywordsDrug resistance, multiple, bacterial · Lactams ·Prolonged and continuous β-lactam infusion ·Therapeutic drug monitoring · Patient carebundles

Bakterielle Sepsis. Diagnostik und kalkulierte Antibiotikatherapie

ZusammenfassungDie Sterblichkeit von Patientenmit Sepsisund septischem Schock ist weiterhininakzeptabel hoch. Eine effektive, kalkulierteAntibiotikatherapiebinnen der ersten Stundenach Erkennen der Sepsis ist ein wichtigesZiel der effektiven Sepsistherapie. Verzöge-rungen führen zum deutlichen Anstieg derSterblichkeit. Daher bilden strukturierte Be-handlungskonzepte eine rationale Grundlageunter Beachtung relevanter Diagnose- undBehandlungsschritte. Neben dem vermutetenFocus und individuellen Risiken einzelnerPatientenmüssen lokale Resistenzmusterund spezifische Problemerreger bei der Wahlder antiinfektiven Therapie berücksichtigtwerden. Vielfältige pathophysiologischeVeränderungen beeinflussen im Rahmen derSepsis die substanzspezifische Pharmakoki-netik (PK) vieler Antibiotika. Daher sollte dasPrinzip der „Standarddosierung“ verlassen unddurch einen individuelleren Therapieansatzmit stärkerer Gewichtung der Pharmakokine-tik(PK)-/Pharmakodynamik(PD)-Indizes derSubstanzgruppen ersetzt werden. Wenngleichdies noch nicht der klinische Standard

ist, können Applikationsformen wie dieprolongierte (oder kontinuierliche) Infusionvon β-Lactamen und ein therapeutischesDrugmonitoring (TDM) helfen, definierte PK-Ziele zu erreichen. Während die prolongierteInfusion auch ohne TDM auskommt, ist TDMbei kontinuierlicher Infusion grundsätzlichnotwendig. Ein weiteres Argument fürden individuellen, PK/PD-orientiertenAntibiotikaeinsatz ist die Zunahme kom-plizierter Infektionen durch multiresistenteErreger (MRE) auf Intensivstationen. Zureffektiveren Behandlung etablieren sichdort zunehmend „antibiotic stewardshipteams“ (ABS-Team). Die interprofessionelleZusammenarbeit des Behandlungsteamsmit Infektiologen/Mikrobiologen undklinischen Pharmazeuten führt nicht nurzum rationaleren Antibiotikaeinsatz, sondernbeeinflusst das Behandlungsergebnis positiv.Den Goldstandard der Erregerdiagnostikstellenweiterhin der kulturbasierte Nachweisaus Probenmaterial und die mikrobiologischeResistenztestung auf die verschiedenenAntibiotikagruppen dar. Neue Polymerase-

Kettenreaktion(PCR)-basierte Verfahrender Er-regeridentifikation und Resistenzbestimmungergänzen trotz hoher Untersuchungsge-schwindigkeit aufgrund der limitiertenaktuellen Studienlage, der hohen Kosten undder eingeschränkten Verfügbarkeit derzeitdie Sepsisroutinediagnostik lediglich. Beikomplizierten, septischenKrankheitsverläufenmit mehrfacher, antiinfektiver Vorbehandlungoder rekurrenter Sepsis können PCR-basierteVerfahren ergänzend zu Therapie-Monitoringund Diagnostik eingesetzt werden. NeueAntibiotika stellen potente Alternativenin der Behandlung von MRE-Infektionendar. Aufgrund des oftmals definiertenErregerspektrums und der praktisch (noch)nicht vorhandenen Resistenzen sind diesezur gezielten Behandlung schwerer MRE-Infektionen geeignet (Therapieeskalation).

SchlüsselwörterMedikamentenresistenz,multipel, bakteriell ·Lactame · Prolongierte und kontinuierlicheβ-Lactam-Infusion · Therapeutisches Drugmo-nitoring · Patientenversorgungsbündel

S42 Der Anaesthesist · Suppl 1 · 2019

for sepsis—particularly in the case ofnosocomial infections [17]. The past fewyears have seen an increase in infectionswith MDR gram-negative (MDRGN)pathogens inGermany[30]. Eventhoughthis development has reached a consid-erably more serious extent in variousother European countries [31–33], an-tibiotic-resistant pathogens are beingincreasingly isolated as the cause of se-vere, barely manageable (nosocomial)infections with high mortality in Ger-man ICUs. ClassicalMDRGNpathogensinclude Enterobacteriaceae (e. g., E. colispecies, Klebsiella species) and non-fermenters (e. g., Pseudomonas species,Acinetobacter-species). These groups arefrequently resistant to broad-spectrumantibiotics, in which enzymatic inacti-vation (e. g., β-lactamases) of antibioticsplays a particularly important role. MDRgram-positive pathogens, such as van-comycin-resistant enterococci (VRE)or methicillin-resistant staphylococci(MRSA), represent another problematiccollective. Whereas the proportion ofasymptomatic VRE carriers and VREinfections continues to rise [30, 34], theclinical relevance of MRSA as the triggerfor sepsis has decreased in recent years[30].

The impact of global tourism andlong-distance travel to areas with a highprevalence of MDR pathogens andMDRpathogen carriers on the distribution ofMRD pathogens has only come intofocus during the past few years [35].A prospective study from 2017 [36]showed that up to 75% of the inves-tigated travelers returning from, e. g.,India, had acquired Enterobacteriaceae,which produce broad-spectrum β-lacta-mases. Although these tourists are pri-marily symptom-free carriers of MDRpathogens, these bacteria could becomerelevant in the case of an infection (e. g.,abdominal infection). In light of thewor-risome development of increasing MDRinfection rates and the continued oc-currence of new resistance phenotypes,the World Health Organization (WHO)published a global action plan for cur-tailing and preventing MDR pathogensand MDR infections in 2015 [37].

Conclusion.With increasing incidences,sepsis and septic shock represent the pri-mary cause of patient mortality in surgi-cal ICUs. Mortality remains high. Noso-comial infections and/or infections withdifficult-to-manageMDRpathogensplayan important role in thedevelopmentandtreatment of sepsis, and also contributeto the high mortality of this disease.

General therapeutic principles:sepsis bundles

For many years, the gold standard ofsepsis treatment was “early goal-directedtherapy” (EGDT), as postulatedbyRiverset al. [38]. Intensive care medicinephysicians of today are still versed inEGDT. By dictating a strict treatmentalgorithm and strictly adhering to thispath, Rivers et al. were able to signifi-cantly reduce mortality in sepsis patientsfor the first time. However, subsequentstudies on sepsis treatment optimizationrevealed that more flexible “sepsis bun-dles” were not inferior to strict EGDT[18, 19, 39, 40]. Consequently, treat-mentmoved away fromEGDTaccordingtoRivers et al., and the sepsis bundles de-fined by the Surviving Sepsis Campaign(SSC)were implemented [7]. In additionto the definition of sepsis as a medicalemergency, hemodynamic optimization(30ml/kg fluid) of the patient, and thesearch for the source/pathogen-diagnos-tic tests, the sepsis bundles also includerapid initiation of a calculated antibi-otic treatment within the first hour afterrecognition of sepsis [1, 7, 41–43]. Thatadherence to defined bundle measuresreduces mortality among sepsis patientswas demonstrated by Damiani et al. ina meta-analysis incorporating 50 stud-ies (odds ratio, OR, 0.66; 95%CI 06–072;[44]). Prospective investigations, e. g.,the International Multicentre PrevalenceStudy on Sepsis (IMPreSS) by Rhodeset al., achieved a reduction in in-hospitalmortality of up to 36–40%by implement-ing bundle measures [45]. This has alsobeen confirmed by other studies [18, 19,40].

The fact that themeasures in the sepsisbundles do not all have the same weightand the same priority was demonstratedby Seymour et al. [46] in a recent ret-

rospective study. In this investigation,it was rapid completion of the 3-hourbundle and earliest possible commence-ment of antibiotic therapy that were thedecisive factors for reducing mortality.Surprisingly, completion of the fluid bo-lus and thus hemodynamic stabilizationin the acute phase seemed to be (at leastin this study) less relevant for a sepsis pa-tient’s survival. Integrationof ICUnursesinto the treatment team is of high rele-vance, and studies have shown that thisalso reduces mortality [47, 48].

Conclusion. The SSC bundle measuresare the new therapeutic standard in sep-sis treatment. Adherence to these recom-mendations leads to a reduction in in-hospital mortality. In addition to hemo-dynamic optimization of the patients,earliest possible (≤1 h) initiation of a cal-culatedantibiotic therapyandrapid treat-ment of the infection are the central el-ements.

Biomarkers and pathogendiagnostics

According to SEPSIS-3 [5], diversebiomarkers continue to be used fordiagnosis, treatment management, andprognostic estimation. The relevanceof the different sepsis markers remains,however, unclear. Many new biomarkershave been investigated over the years, ofwhich none had the power to accuratelyrecognize or definitively exclude sepsiswith sufficient specificity and sensitivity[49, 50]. In the following sections, sev-eral important infection markers that arefrequently assessed in clinical routineare discussed in terms of their valueand prognostic advantages in the con-text of sepsis. Moreover, experimentalbiomarkers that have not yet found entryinto clinical routine are also presented.

Biomarkers

LactateAs a product of anaerobic glycolysis,lactate (the anion resulting from dissoci-ation of lactic acid) serves as a biomarkerfor tissue hypoxia. The acceleratedglycolysis and reduced mitochondrialmetabolism occurring during shock

Der Anaesthesist · Suppl 1 · 2019 S43

Leitthema

lead to increased lactate production.Lactate levels are further increased inthis situation by decreased eliminationand enzyme induction. In the context ofsepsis, lactate can be used for progno-sis as well as for treatment monitoring(stagnation or diminishment of plasmaconcentrations; [51–55]). After years ofcontroversial discussion, the new def-inition of septic shock [5] states thecriterium of hyperlactatemia >2mmol/l(≈18mg/dl) as obligatory for diagno-sis [10]. In an analysis of data frommore than 280,000 patients of the SSCcollective, Casserly et al. [53] showedthat hyperlactatemia >4mmol/l and lowlactate clearance during the first 6 h wereassociated with a significant increase inin-hospital mortality.

ProcalcitoninOver the past years, procalcitonin (PCT)has become established in clinical rou-tine as a sensitive and rapid parameterof bacterial infection. Due to its kinetics(response half-time, response-HT, up to4–6 h), PCT is viewed as the diagnos-tic gold standard for bacterial infections[56, 57]. In spite of these advantages, par-ticularly locally limited infections, e. g.,ventilator-associated pneumonia (VAP)or abscesses, can occurred without anincrease in PCT [58, 59]. Interpretationof the PCT value in critically ill patientswith relevant liver diseases or after liversurgery is sometimes equally difficult:a PCT increasemay bemeasured in thesepatients in the absence of a bacterial in-fection. Whether PCT can be considereda sufficiently reliable marker for differen-tiating between different grades of sepsisseverity and noninfectious SIRS is con-troversially discussed [58, 60, 61]. Thespecificity and sensitivity of PCT for sep-sis diagnosis are under 90%, irrespectiveof the cutoff value used by the measur-ing laboratory [62, 63]. Therefore, thePCT value should always be interpretedunder consideration of the patient’s clin-ical condition and the limitations of thismarker [58]. With an induction timeof approximately 4 h, it is, in principle,possible for fulminant septic shock to de-velop without a relevant increase in PCTlevels.

Interleukin 6A biomarker that is particularly fre-quently measured in pediatric IDs isinterleukin 6 (IL-6). IL-6 is released atan early stage of the immune reactionand reaches its peak in plasmawithin 2 h(induction time approximately 20min).The peak level correlates positively withthe severity and course of the infection[64, 65]. A limitation to the value ofIL-6 as a biomarker is the fact that anarray of other diseases are also associatedwith in an increase in IL-6 levels ([59];e. g., autoimmune diseases, surgery, andtrauma). In patients with severe sepsisor septic shock, a reduction in IL-6 levelscan be used as a prognostic factor [66].The kinetics of the IL-6 level can be usedto monitor the effectiveness of an anti-infective therapy [66]. Despite theseadvantages and the increasing trend to-ward chemical laboratory measurementof IL-6 in emergency departments andICUs, IL-6 is not a routine parameterfor diagnosis of sepsis in adult patients.

C-reactive proteinC-reactive protein (CRP) is one of thehepatically synthesized acute-phase pro-teins, and has been measured as a rou-tine parameter for diagnosis and treat-ment monitoring for many years. Thespecificity of CRP for diagnosis of an in-fection is low [56]. Increases in CRP arenot only observed during infection, butalso after surgery, trauma, and burns, orin patients with acute (aseptic) pancre-atitis. Particularly for diagnosis of sepsismust CRP be viewed very critically. Infulminant disease courses, the CRP in-duction time (response-HT after stimu-lus 6–10 h, relative increase 10–100-fold)means that septic shock can already bepresent before laboratory analysis detectsan increase in the CRP value. CRP is,however, of value for evaluation of dis-ease course and monitoring of treatmentefficiency [59, 67]. Persistent increasesin CRP level or a secondary increasein plasma concentration in patients onanti-infective therapy may indicate in-adequate control of the infection (e. g.,infectious complications, secondary in-fections, polymicrobial infections, MDRpathogens, ineffective antibiotic, inade-quate antibiotic dosage; [59]). In this

situation, the CRP concentration coursecan lead to re-evaluation of the overallclinical situation, the assumed source,and the treatment regimen [50, 68].

Proadrenomedullin and solublesubtype of the CD14 cell surfacereceptorNew experimental biomarkers for sepsisinclude the anti-inflammatory and an-timicrobial peptide proadrenomedullin(proADM) and the soluble subtype ofthe CD14 cell surface receptor (sCD14-ST, presepsin; [59]).

In clinical studies, proADM was ofhigher prognostic value for sepsis pa-tients with nosocomial pneumonia (hos-pital-acquired pneumonia, HAP) thanthe classical biomarkers CRP and PCT[69, 70]. However, since raised plasmalevels of ADM are seen in diverse dis-eases, e. g., cardiovascular and autoim-mune diseases [71], ADM is unsuitablefor initial detection of sepsis. In clini-cal studies, the ADM plasma level cor-related very well with disease course/disease severity and the mortality of thepatient collective [69]. Therefore, ADMmay become a suitable marker for prog-nostic estimations in sepsis patients inthe future. Furthermore, in one clinicalinvestigation, ADM had a significantlyhigherprognostic significance for in-hos-pital mortality than the classical markersCRP and PCT [70].

The sCD14-ST (presepsin) concentra-tions in plasma increase within 6 h af-ter bacterial infection [72]. Due to therelatively strong correlation with bacte-rial infection[73], presepsinmaybecomea specific marker for detection of sepsisand its discrimination from noninfec-tious SIRS in the future [74, 75]. Sincethe height of the measured plasma lev-els correlates poorly with the grade ofsepsis severity, presepsin appears to beunsuitable as prognostic marker [75].

Other chemical laboratory testsThe complete battery of chemical labo-ratory tests used in clinical routine ob-viously includes leucocyte count (leuko-cytosis and leucopenia) and the classi-cal differential blood count. These candeliver additional information in sepsis

S44 Der Anaesthesist · Suppl 1 · 2019

patients but should not be used for di-agnosis.

Conclusion.No sufficiently sensitive andspecific sepsis biomarkers are currentlyavailable. The PCT, IL-6, and lactatemarkers can substantiate the diagnosisand may be drawn upon for monitor-ing treatment efficiency and/or estimat-ing prognosis (lactate). It is important tobe aware of the fact that during fulminantdisease courses, the initial values may bewithin the normal range and only in-crease in subsequence stages. Therefore,diagnosis or exclusion of sepsis shouldnever be based solely on these parame-ters.

Blood cultures

An implication of the earliest-possibleinitiation of a calculated antibiotic ther-apy [7, 76] is immediate evaluation of thepossible infection and expected causalpathogen. Whereas the search for theinfection is based primarily on clinicalexamination and imaging, isolation ofthe causal pathogen requires that a con-siderable number of samples be taken. Inaddition to other samples (e. g., deep tra-cheal secretions/bronchoalveolar lavage,BAL; abscess punctures; cerebrospinalfluid collection; tissue samples), bloodcultures play an important role. Two tothree paired blood culture samples [77,78] should be collected aseptically fromdifferent areas prior to commencementof antibiotic therapy. A fresh puncturesite should be used at least once, andif a catheter-associated infection is sus-pected, a blood sample should also betaken from the corresponding catheter(e. g., central venous catheter, CVC; Shal-don catheter; arterial catheter) and sentfor microbiologic testing. Collection ofthree pairedblood culture samples is pos-siblewithinminutes [79], and is sufficientto identify the pathogen in 90% of casesof bacteremia [80, 81]. Even one applica-tion/the initial application of antibioticcan render identification of the pathogenimpossible in blood culture medium [82,83]. Although pathogen identification isof elementary importance in sepsisman-agement (reduction of health care costs,possibility of de-escalation or pathogen-

specific treatment, reduction of selectivepressureonpathogens, reductionofmor-tality; [84, 85]) and the likelihood of iso-lating the pathogen in the presence ofbacteremia is very high, the rate of sam-ple taking in Germany—approximately55blood cultures per 1000patient days inGermanICUs—isconsiderablybelowtheinternational average (100–200/1000 pa-tient days; [86]). An increase in this rateto around 100/1000 patient days mustthus be promoted as a matter of urgency[87].

When handling blood cultures, thepreanalytical phase is of great impor-tance. Not only sterile sample collec-tion but also the inoculation volume(8–10ml/tube) and immediate (withas little delay as possible) delivery tothe microbiologic laboratory are criticalfactors. The once-propagated collectionof blood samples for in vitro pathogendiagnosis during the fever onset phasedoes not increase detection rates, nordoes inoculation of the blood culturemedium with large volumes [88, 89]. Itis important to note that coagulase-neg-ative staphylococci (CoNS) are classicalblood culture contaminants. It these aredetected, it must be critically consideredwhether this result is relevant to infectionor due to non-sterile handling.

The decisive disadvantage of bloodcultures in the context sepsis is the la-tency period between sample collectionand culture results (18–24 h to pathogenidentification, up to 72 h for resistancetesting).

Conclusion. Immediate initiation ofthe search for the infection and thepathogen—before starting the calculatedantibiotic therapy—formsthe foundationof successful sepsis treatment. Identifi-cation of the infection delivers importantinformation concerning the spectrum ofpathogens, and thus influences the calcu-lated antibiotic therapy. Blood culturesremain the gold standard for detectionof bacteremia, and collection of samplesis thus essential for sepsis management.

New techniques

Molecular genetic and culture-indepen-dent methods for pathogen detection

have been commercially available formany years now. These techniques ap-pear rational, since thephase of empiricalantibiotic therapy without differentiatedmicrobiologic finding should be kept asshort as possible—not only in the inter-ests of the individual patient, but also inthe interests of specific and economicaluse of antibiotics [90]. Furthermore, inaddition to identifying the pathogen, itis also important to know its resistanceto various antibiotics, particularly inintensive care medicine. Whereas de-tection of MRSA (mecA or mecC gene)poses no technical difficulties, adequatedetermination of β-lactam resistancein gram-negative bacteria is a chal-lenge, due to the hundreds of differenttypes of β-lactamases [91]. Most mul-tiplex polymerase chain reaction (PCR)approaches are thus limited to a fewfrequently occurring resistance deter-minants. However, molecular genetictechniques have only been able to iden-tify a few resistance mechanisms to date[92]. Whether the shorter latency periodbetween sample collection and results[93] propagated by the manufacturersof these products is realizable in clini-cal routine appears questionable. Sincea eubacterial PCR with subsequent se-quencing of the amplification producttakes many hours, a clear time advantageof PCR-based pathogen diagnostics isprimarily achievable when searching fordefined pathogens or within a limitedpathogen spectrum, e. g., when lookingfor Neisseria meningitidis and Listeriamonocytogenes in patients with menin-goencephalitis, or for viral infectiondiagnostics. An undeniable advan-tage of PCR-based techniques is thatpathogens (DNA fragments) are stilldetectable when the blood culture iscompromised, e. g., due to prior anti-infective treatment.

To what extent PCR-based diagnostictesting has an influence on treatmentdecisions which consecutively improvepatent outcome has hardly been in-vestigated. In an observational studyinvestigating implementation of multi-plex PCR in the clinical routine of anICU, an influence on treatment coursewas only observed when a pathogen wasdetected that was not susceptible to the

Der Anaesthesist · Suppl 1 · 2019 S45

Leitthema

initial calculated therapy—specifically,MDR (VRE and MRSA) bacteria andCandida species [94]. Only two ran-domized studies have been published onthis topic so far: In one study comparingAspergillus PCR to standard diagnostictesting in patients with fever and neu-tropenia, mortality could be significantlyreduced by the PCR-directed treatment[95]. In another investigation conductedin an neonatal ICU in India, a signifi-cant survival advantage was conferred bymultiplex PCR-based diagnostics com-pared to blood cultures [96]. However,concerning the latter study, it must becritically noted that in an environmentwith a very high rate of extended-spec-trum β-lactamases (ESBL), an initialcalculated therapy with cephalosporinsis standard.

In the following sections, three newtypes of pathogen diagnostic test are pre-sented.

Combination of PCR and massspectrometryThe IRIDICA® system (Abbott, Chicago,IL, USA) is based on a combinationof PCR and mass spectrometry (PCR/electrospray ionization mass spectrom-etry, PCR/ESI-MS). By using distinctprimers in a target-oriented analysis,800 bacterial, fungal, and viral pathogenscan be identified in about 6 h. In a sec-ond step, the amplification productsyielded in PCR undergo a fully auto-mated database comparison and can beassigned to a specific pathogen on thebasis of their molecular weight. Possiblesample materials include blood but also,e. g., sputum or tissue particles. Theprospective Rapid Diagnosis of Infec-tions in the Critically Ill (RADICAL;[97]) study compared results of conven-tional blood cultures with those of theIRIDICA® system in 616 patient sam-ples. Use of the PCR/ESI-MS techniqueachieved a sensitivity of 81% and a speci-ficity of 69%. The negative predictivevalue of the test was 97% in this investi-gation. IRIDICA® is also able to detectthe presence of some resistance genes(mecA, vanA/B, KPC). In addition tothe potential time advantage, pathogendiagnosis can also be successfully ac-complished with this method even after

initiation of antibiotic therapy, whenconventional/culture-based techniquesare no longer possible. Further prospec-tive data on this technique are currentlylacking.

Next-generation sequencingAnother technique is next-generationsequencing (NGS). In this method, cell-free DNA (cfDNA) is extracted fromplasma samples, and all pathogens aredetected and sequenced. Discriminationbetween infection and colonization ismade possible by calculating the Sep-sis Indicating Quantifier (SIQ) score[98]. Analogous to PCR-based tech-niques, a further advantage of NGS isthe prediction of particular resistancephenotypes [98]. NGS will become in-creasingly interesting in the future, sincewith this method—analogous to bloodcultures—there exists the possibility ofa widescale search for the pathogenwhich is independent of the pathogengrowth rate.

Pathogen identification anddetermination of the specificminimum inhibitory concentrationA method for pathogen identificationand determination of the specific min-imum inhibitory concentration (MIC)is the Accelerate PhenoTest™ BC Kit(Accelerate Diagnostics™ Inc., Tucson,AZ, USA). Using fully automated test-ing of pathogen-positive blood samples,a large number of gram-positive andgram-negative pathogens (as well asCandida species) can be identified, andthe MCI of classical broad-spectrumantibiotics can be tested. According tothe manufacturer, the identification ofthe pathogen in positive blood samplessucceeds within 90min; the results ofsusceptibility testing are available within7 h. This technique therefore appearssuitable for significantly reducing thetime interval between empirical and tar-geted pathogen-specific (MIC-adapted)treatment, and thus also for reducingthe increased mortality of patients whoreceive inadequate initial treatment ofbacteremia (bloodstream infection, BSI;[99, 100]). The possibility of early adap-tation of the anti-infective therapy isdrawing nearer.

Future perspectivesThe abovementioned PCR- and NGS-based techniques provide interestingoptions for accelerating pathogen iden-tification and resistance phenotyping,in order to positively influence sepsistreatment. However, these methodsare currently not routinely implemented[93]. The costs of these investigations arepresently very high [101, 102], whichlimits their use primarily to studies.In addition to pathogen identification,applications such as the Accelerate Phe-noTest™ BC Kit may also rapidly deliverMIC data in the future, thus enablingearly individualization of treatment.

Conclusion. The PCR-based techniquesfor pathogen diagnosis have been estab-lished formany years now. Newmethodsbased on molecular genetics presentlyonly detect defined pathogens and spe-cific resistance types. They are currentlynot applied in clinical routine. Due to thepossibility of covering a very broad spec-trum of pathogens (analogous to bloodcultures),NGSisahighlypromisingtech-nique for the future.

Calculated antibiotic therapy

Principles of treatment

Surgical intensive care patients withsepsis and septic shock are nowadaysmanaged by an interdisciplinary, inter-professional treatment team comprisingintensive care physicians, surgeons, clin-ical pharmacists/pharmacologists, mi-crobiologists/hygienists, and consultantphysicians from other disciplines (e. g.,neurology, cardiology).

Thecentralmaximofa successful anti-infective therapy was formulated in 1913by Paul Ehrlich at the 17th InternationalCongress of Medicine by the statement[103] “Frapper fort et frapper vite” (“hithard and fast”)—a principle which stillapplies today.

Ehrlich’s principle of fast antibiotictherapy was taken up at the end of the1990s by Kollef et al. They showed [104]that delayed application of antibioticswas associated with a significant increasein the mortality of sepsis patients. Thefact that in addition to the time factor

S46 Der Anaesthesist · Suppl 1 · 2019

the effectiveness of the antibiotic therapyis also critical was summed up by Ku-mar and Kethireddy 100 years after PaulEhrlich, in an editorial of the year 2013[105]: “Speed is life but a hammer helpstoo.” This principle is underlined by ananalysis of the first (approximately 1000)patients of the Medical Education forSepsis Source Control and Antibiotics(MEDUSA; [6]) study. In MEDUSA,sepsis patients with an inadequate ini-tial therapy had a significantly increased28-day mortality. On one hand, delaying(>6 h) surgical treatment of the infectionand source control increased mortalityby 16.2% (42.9 vs. 26.7%, p < 0.001); onthe other, an inadequate antibiotic ther-apy (irrespective of the time point of thefirst application) was also associatedwitha significant increase in mortality (30.3vs. 40.9%, p < 0.001). After completionof the study, Bloos et al. [106] evaluatedthe entire MEDUSA dataset with a to-tal of 4000 patients, investigating, amongother things, the effect of the timepoint ofthe first application on patient mortality.These authors found that with every hourthe antibiotic therapy was delayed, mor-tality increasedby2%. Delayedtreatmentof the source infection was also associ-ated with an increase in mortality of 1%per hour [106]. Diverse other studieswere also able to prove the association ofa delayed first application with increasedmortality [6, 107–111] and worsening ofsecondary endpoints (e. g., acute kidneydamage; development of acute respira-tory distress syndrome, ARDS; increaseof the SOFA score; [112–114]) among theinvestigated patient collective. It is there-fore not surprising that rash initiation ofa calculated antibiotic therapy is one ofthemost importantcornerstonesofsepsistreatment [1, 7, 76, 115–119] and a cen-tral element of current guidelines [7]. In2006, Kumar et al. [115] demonstratedin a retrospective analysis that the factor“time” is an independent predictor forsurvival in septic shock patients. Refer-ring to the recommended timeframe fortreatment initiation in very seriously in-jured patients, these authors paraphrasedtheir results as the “golden hour of sep-sis.” Subsequent studies showed similarresults [76, 107, 120]. Although the timefactor and the mortality increase result-

ing from delayed administration of an-tibiotics appear to be very well proven,the “golden hour of sepsis” concept andthe implied necessity of a calculated an-tibiotic therapy within the first hour issubject to justified criticism [7, 115]. Ina very recent editorial, Singer [121] sum-marizes these criticisms and relativizesthe “golden hour of sepsis” dogma withrespect to the problem of overuse of an-tibiotics in critically ill ICU patients andthe phenomenon of “incestuous ampli-fication.”

Modern sepsis treatment is nowadaysan interprofessional challenge. The con-stant reevaluation of all therapeutic as-pects is too complex for the individualspecialists or indeed a single discipline.Therefore, treatment today is generallymanaged by an antimicrobial steward-ship (ABS) team [122]. This team com-prises ID specialists/microbiologists andclinical pharmacists. A current meta-analysis of the effectiveness of ABS teams(based on 145 studies addressing 14 ABSmeasures) was able to show that im-proved adherence to guidelines was asso-ciated with reduced mortality (mortalityreduced by 35%; relative risk, RR, 0.65;95%CI054–080; p<0.0001)andstringentde-escalation (mortality reduced by 65%;RR 0.44; 95%CI 0.30–0.66; p < 0.0001;[123]). The ABS measures are now anintegral component of ICUs and are in-cluded in the Deutsche Gesellschaft fürInfektiologie e. V. (DGI; “German Soci-ety of Infectious Diseases”) S3 guideline[124].

Conclusion. Calculated antibiotic ther-apy of sepsis should be initiated as soonas possible (ideally within the first hourafter recognition of sepsis); the interpro-fessional ABS team takes over furthertreatment management.

Tarragona strategy

A concrete treatment concept is repre-sented by the Tarragona strategy [125].Originally established for VAP, this strat-egycanalsobeapplied to the initial calcu-lated antibiotic therapy in sepsis patients,and incorporates the most important di-agnostic and therapeutic considerations.

The key elements of the Tarragona strat-egy are presented in . Table 2.

Thecentralaspecthere isonceagainanearly, high-dose, sufficiently broad-spec-trum antibiotic therapy. Again, the as-sumed infection, the individual patient’srisk profile, the possibility of an MDRpathogen, and local antibiotic resistancerates are incorporated into the concept.Particularly in ICU and/or invasive ven-tilation patients, nosocomial infectionscaused by MDR pathogens are highlyrelevant [21, 30] and must be consid-ered during the antibiotic selection pro-cess. A frequent consequence of deviat-ing from established treatment conceptsis treatment failure [126]. It should benoted that the Tarragona strategy incor-porates the possibility of constant reeval-uation and treatment monitoring. Reg-ular checking of the concept’s key ques-tions at fixed time intervals and in theevent of the patient’s condition changingis obligatory.

Treatment duration and treatmentmanagement

The persistent high mortality of patientswith sepsis and septic shock [116], aswellas the associationwith delayed and/or in-effective antibiotic therapy, led to the rec-ommendation for earliest possible (<1 h)calculated first application of a broad-spectrum antibiotic [7]. Since pathogenidentification and, perhaps even moreimportantly, resistance phenotyping cantake many hours to days, extreme vig-ilance and caution must be taken withthe antibiotic therapy. The selection ofsubstance class and the decision for cal-culated mono- or combination therapyare based on the (assumed) source andthe clinical conditionof the patient (pres-ence of septic shock, immune suppres-sion, etc.). Therefore, despite the criti-cal time-dependent nature of a “sepsis”emergency, the following basic questionsmust be clarified:4 Which infection is the source of

sepsis, and how can this infection betreated (surgically or interventionally,e. g., using CT-guided drainage)?Which pathogens can be expected?

4 Does the patient exhibit septic shock?

Der Anaesthesist · Suppl 1 · 2019 S47

Leitthema

Table 2 Tarragona strategy. (Modified from Sandiumenge et al. [125])

Core element Explanation

1 Look at yourpatient

Infection?→ search for the infection/pathogen diagnostics

Preexisting disease?

Immune status?

Risk of an MDR pathogen? (Antibiotic pretreatment, frequent hospi-talization, knownMDR colonization, travel in MDR pathogen endemicregions)

2 Listen to yourhospital

Surveillance: Knowledge of local pathogens/resistance phenotypes,ABS measures

3 Hit hard andearly

High-dose application of a broad-spectrum antibiotic (≤1 h), combina-tion therapy as appropriate

4 Get to the point Assumed infection→ PK/PD profile and tissue penetration of theantibiotic

5 Focus Reevaluation of treatment at regular intervals

De-escalate if possible

Escalate if clinically and/or microbiologically necessary

With a total of five relevant points, the Tarragona strategy represents a structured diagnostic andtherapeutic pathway for themanagement of infections. Although originally developed as a guidelinefor treatment of ventilation-associated pneumonia, the concept is also suitable for treatment ofsepsisABS antibiotic stewardship program, MDR multidrug-resistant gram-negative pathogens,PK/PD profile substance-specific pharmacokinetic and pharmacodynamic profile of an antibiotic

4 How likely is the patient to have aninfection with an MDR pathogen?Existing risk factors: previous anti-infective treatment, frequent hospi-talization, known colonization withMDR bacteria, or frequent travel inendemic regions?

Based on these considerations, the ne-cessity of primary combination versusmonotherapy can be derived. During thecourse of continued therapeutic reevalu-ation and after microbiologic identifica-tionof thepathogen, treatment is adaptedin terms of de-escalation or escalation[102].

In this way, it is possible to pre-serve the patient’s chance of survival [7].A general combination therapy com-prising two (or more) broad-spectrumantibiotics cannot be recommendeddue to the inconsistency of currentdata. Some studies have demonstratedreduced mortality achieved by combi-nation therapy [127–132], whereas therandomized multicentric comparison oftwo antibiotic regimens (meropenem vs.meropenem+moxifloxacin) in the treat-ment of severe sepsis and septic shock byBrunkhorst et al. (Max-Sep study; [133])found no superiority of combinationtreatment (meropenem + moxifloxacin)

over monotherapy (meropenem) forsevere sepsis. In this study, mortality28 days after randomization was 23.9%(95%CI 19–29.4%) in the combinationgroup and 21.9% (95%CI 17.1–27.4%)in the monotherapy group. Treatmentoutcomes at day 90 did not differ signif-icantly [133]. To summarize, it must bestated that primary combination ther-apy of sepsis was never superior whenthe antibiotic used as monotherapy wasa carbapenem. Moreover, in retrospec-tive analyses it could be shown that inonly about 30% of patients did the initialcalculated antibiotic therapy actuallyencompass the pathogen, and de-escala-tion of therapy was only undertaken ina thirdofpatients [127–133]. Combiningbroad-spectrum antibiotics simultane-ously potentiates the selective pressureon pathogens and leads to increaseddevelopment of resistance [134].

On the basis of contradictory evi-dence, according to the SSC [7], mono-or combination therapy can be used asthe primary treatment for septic shock(recommendation grade: high; evidence:moderate). Whereas primarymonother-apy is to be preferred for sepsis withoutsignsofsepticshock(exception: catheter-associated BSI), a primary combinationtherapy to broaden the spectrum may

be urgently required in, e. g., septicshock patients and/or upon suspicionof infection with an MDR pathogen.Carbapenems such as meropenem andimipenem/cilastatin are often usefulcombination partners for treatment es-calation or combination therapy in thepresence of septic shock, since theyare also effective against gram-negative,broad-spectrum β-lactamase (ESBL)-producing pathogens. Gram-positive“problem” pathogens such as MRSAhave shown stable resistance types forsome time now (as well as a constant in-cidence of infections), and are generallysusceptible to vancomycin, teicoplanin,tigecycline, and linezolid. While theincidence of VRE is clearly increasing[34], an increase in linezolid-resistantenterococci has also been registered[135]. Increasing VRE incidences arethus likely to represent a problem in thefuture.

When the source of sepsis is known,it is often possible to limit the spectrumof possible pathogens. However, sep-sis resulting from an unknown infectionrepresents a challenge.

Principally, as part of a structuredtreatment, therapy must be reevaluatedafter 48–72 h of treatment at the latest;adaptation of the anti-infective therapyshould be initiated as appropriate (esca-lation if necessary, de-escalation if possi-ble). Similarly, in patients with an insuf-ficient response to treatment, the infec-tion should be reexamined. As a surro-gate parameter for assessing the effective-ness of therapy, as well as for treatmentmanagement and monitoring of diseasecourse, the course of plasma PCT levelscan be used. Many studies on the topichave shown that a decrease in plasmaPCT correlates with an effective antibi-otic therapy (a bacterial infection is veryunlikely with PCT < 0.5 μg/l), and thatthis is a useful parameter for treatmentmanagement [79, 136–140]. In the Mul-ticenter Procalcitonin Monitoring SepsisStudy (MOSES; [141]), Schuetz et al. ad-ditionally showed that a decrease in PCTlevel of less than 80% of the initial levelwithin 4 days of therapy was an indepen-dent predictor for death of the patient.In their study on PCT-guided de-escala-tion of anti-infective therapy, Jong et al.

S48 Der Anaesthesist · Suppl 1 · 2019

[138] observed—in addition to short-ened antibiotic exposure—significantlyreduced patient mortality among the pa-tient collective; however, this survival ad-vantage could not be confirmed by a cur-rent Cochrane analysis [142] including10 studies and over 100 patients. In sum-mary, PCT-guided treatment monitor-ing should take precedence over otherbiomarkers [79, 140, 143, 144]. Even ifa survival advantage has not yet beenproven beyond all doubt, this approachcan shorten the exposure to antibioticsand thus reduce side effects and resis-tance development. In patients exhibit-ing clinical and chemical indications ofan insufficient therapy, not only mustthe antibiotic regimen be reevaluated,but imaging must be performed anewand low-threshold second-look surgeryshould be considered. Furthermore, inthe absence of a response, the presenceof an MDR infection or invasive mycosismust be considered. Vice versa, in pa-tients with a response to treatment, withplausible microbiologic pathogen find-ings and resistance testing results, de-es-calation should be strived for [145]: the“never change a winning team” principleis not valid here.

Although investigations assessingquestions related to treatment durationand prompt de-escalation did not yieldconsistent results, Leone et al. [146]showed in their study published in 2014,e.g., that antibiogram-directed adapta-tion of the calculated antibiotic therapydid not have a negative influence onmortality compared to continuation ofthe empirical therapy. Similar results onde-escalation strategies were deliveredby the studies of Turza and Havey et al.[147, 148]. Neither clinical and mi-crobiologic recovery, nor survival werenegatively influenced by a shorter treat-ment duration compared to groups withlonger therapy. This was the conclusionalso reached by Sawyer et al. in thetreatment of complicated intraabdom-inal infections (cIAI): After successfulsurgical treatment of the infection and4 days of antibiotic therapy, there was nodifference to an 8-day therapy in termsof therapeutic success [149]. Chastreet al. [150] were also unable to show anadvantage of long (15 days) over short

therapy duration (8 days) in the treat-ment of VAP.The principle of pathogen-specific adaptation and de-escalationof therapy after identification and resis-tance testing is thus supported by currentdata [151, 152] and incorporated intocurrent SSC guidelines [7]. Treatmentduration should only exceed 7–10 daysin exceptional cases (e. g., endocarditis).Modern treatment of severe infections inICUs should always bemultidisciplinary.Strategies such as ABS teams have be-come established and found entry intospecialist society guidelines [124].

Conclusion. The decision on whether toinitiate mono- or combination therapyis made based on individual patient-spe-cific risk factors, the (assumed) infec-tion, and the likelihood that sepsis wastriggered by an MDR pathogen. De-escalation of therapy following clinicalimprovement or pathogen identificationis essential and does not negatively in-fluence therapeutic outcomes. For treat-mentmanagement andmonitoring, PCTmeasurements can be used. Only in ex-ceptional cases should treatment dura-tion exceed 7–10 days.

Examples

Sepsis of unknown source withoutpretreatmentSepsis of unknown source without pre-treatment initially permits monotherapy([7, 153]; exceptions: catheter-asso-ciated BSI in particular). Primarilybroad-spectrum β-lactam with β-lacta-mase inhibitors (BLI; e. g., piperacillin/tazobactam) shouldbe administered. Al-ternatively, carbapenems (e. g., merope-nem, imipenem/cilastatin) can be used,particularly if ESBLs are likely. Combi-nation with group 2/3 fluoroquinolonesis possible. In addition to considering thepatient’s characteristics and risk factors,when the source of sepsis is unclear,it is vital to have knowledge of localpathogens, as well as of the resistanceepidemiology of “problem pathogens.”ESBL-synthesizing pathogens can inac-tivate almost all cephalosporins and insome cases also piperacillin/tazobactam.If ESBL producers are to be expected,carbapenems should generally be pre-

ferred, even though several clinicallyrelevant ESBL variants (temoneira β-lac-tamase, TEM; sulfhydryl variable typeβ-lactamase, SHV; cefotaxime-Munichβ-lactamase, CTX-M) may retain sus-ceptibility to piperacillin/tazobactam[154]. If the antibiogram reveals anESBL pathogen to be susceptible topiperacillin/tazobactam, treatment ofurinary tract infections is usually un-problematic; however, forbacteremia, noconsensushasbeenreachedregardingtheadequacy of therapy due to the inoculumeffect [155] and the serious implicationsof altered β-lactam pharmacokineticsin sepsis patients [156, 157]. Whetherantibiogram-directed treatment againstESBL-synthesizing pathogens should beconducted with piperacillin/tazobactamthusremainsadecisionthatmustbemadeon the basis of an individual risk–benefitassessment. High-risk groups (majorsurgery, long-term intensive care, anti-infective pretreatment, ventilation) andfrequent complications due to MRSAinfections at the particular site, or colo-nizationof thepatientwithMRSA, justifycombination with a glycopeptide (van-comycin, teicoplanin). Alternatively,in some constellations, combinationwith a lipopeptide (daptomycin) may beappropriate (caveat: not in MRSA pneu-monia). In septic shock of unknownorigin, combination therapy should beadministered primarily [7]. In additionto a carbapenem, recommendable com-bination partners included vancomycin(broadening of the spectrum againstMRSA, CoNS, and vancomycin-sensi-tive enterococci) or a fluoroquinolone.An exemplary algorithm for calculatedtherapy of sepsis and septic shock ofunknown source is presented in . Fig. 1.The therapeutic objective in the ini-tial phase is rapid, calculated antibiotictherapy (>1 h). The decision on mono-versus combination therapy should bebased on the clinical status of the patient(presence of septic shock), a possibleanti-infective pretreatment, and the riskof infection with an MRD pathogen. Aswith every algorithm, this is not bindingand can of course be adapted or aban-doned at any time in light of the clinicalsituation or local problem pathogens.

Der Anaesthesist · Suppl 1 · 2019 S49

Leitthema

Emergency: sepsis

Diagnosis/suspicion

Search for the infection

Immediate antibiosis

Sepsis→ e.g., piperacillin/tazobactam

(alternative: carbapenem)

Septic shock/antibioticpretreatment→ carbapenem

(alternative: piperacillin/tazobactam)

Septic shock with a high risk of an MDR pathogen→ primary combination therapy

<1 h (golden hour)

• qSOFA ≥ 2?• SOFA score? Increase ≥ 2 points?• Lactate > 2mmol/l (18mg/dl)?

• Blood cultures• Cultures, intraoperative swabs,

aspirate, deep tracheal secretion/BAL, cerebrospinal fluid/urine where appropriate

• Imaging• Interdisciplinary workup

• Initial bolus• Prolonged infusion preferred for

β-lactamas

• Bolus: PIP/TAZ short infusion over 20min• Then prolonged infusion over 3ha

• PIP/TAZ depending on local resistance situation• Combination according to infection and assumed

pathogena

• e.g., carbapenem (initial bolus, then prolonged)• Combination according to infection and assumed

pathogen (see above)a

o MRSA: + a) glycopeptide, b) lipopeptide (catheter infection, BSI), c) linezolid (CAP/HAP)

o P. aeruginosa: + fluoroquinolone 2/3 or + aminoglycoside

o Assumed candidemia: + echinocandin

AB regimen:

• β-lactams → initial bolus, then prolonged infusion

• Continuous infusion of β-lactams ONLYwith TDM and known pathogen-specific MIC!

Further treatment:

• Further treatment in ABS team• Tight consultation/collaboration with

infectious diseases and pharmacy specialists

• Reevaluation of treatment even when sepsis symptoms improve (at least every 72 h)

• Adaptation and de-escalation if a) the infection is sufficiently treated, b) positive clinical course, c) resistogram available

• Discontinuation of antibiotic in cases of positive PCT course (reduction ≥ 80%, absolute <0.5 μg/l)

Lack of therapeutic success in the initial phase:

• Infection correct? Multiple infections?• In doubt: renewed search for infection!• Pathogen correct? (MDR pathogen?

Multibacterial infection? Assumed candidemia? Viral coinfection?)

• Infection adequately treated? Surgical reevaluation?

• Consider TDM (particularly for β-lactams)

Fig. 18 Example algorithm for initial calculated therapy of sepsis of unclear origin.AB antibiotics,ABS antibiotic steward-ship program, BAL bronchoalveolar lavage, BSI bloodstream infection (bacteremia),CAP community-acquired pneumo-nia,HAPhospital-acquired pneumonia,MICminimum inhibitory concentration,MDRmultidrug resistant gram-negativepathogen,MRSAmultiresistant/methicillin-resistant S. aureus, PIP/TAZpiperacillin/tazobactam, qSOFAQuick SequentialOrgan Failure Assessment Score, PCT procalcitonin, TDM therapeutic drugmonitoring. aCatheter-associated sepsis/BSI:combination therapywith vancomycin or daptomycin

Catheter-associated infectionIf patients with a catheter-associatedinfection develop bacteremia (catheter-associated BSI) and sepsis, immediatetreatmentof thesource, i. e., changeof thecatheter, is indicated. In thewidest sense,not only classical catheters, but also portsystems, pacemaker probes, or dialysiscatheters are possible causal infectionloci. Removed catheters must be sub-jected to microbiologic investigation. Ofparamount importance for identificationof a catheter-associated BSI pathogenis inoculation of multiple multilocularpaired blood cultures [80, 81]. SinceCoNS comprise the majority of catheterinfection pathogens, and because theseare generally resistant to penicillin andcephalosporin, glycopeptides are themost important elements of the initialtreatment of catheter-associated BSI. Inorder to also target the significantly rarergram-negative pathogens, combinationwith a β-lactam antibiotic (piperacillin/tazobactam or carbapenem) is neces-

sary. Since intermittent administrationof a vancomycin bolus is often asso-ciated with nephrotoxic side effects,regular measurement of vancomycinconcentrations are a necessary part oftreatmentmonitoring (target: with inter-mittent infusion, trough level: 15 μg/ml;[158, 159]). Nephrotoxicity appearsto be reduced with continuous van-comycin infusion [158]. Alternatively,instead of vancomycin, teicoplanin orthe lipopeptide daptomycin can alsobe used (6–12mg/kg/day; [153]). Ac-cording to current data, daptomycinappears to be safe up to a maximal dailydose of 12mg/kg [160]. This varianttherapy also seems to be effective whenVRE (E. faecium) are likely. If infectionwith a resistant gram-negative pathogenis probable, the primary combinationpartner should be a carbapenem.

In patients without a septic diseasecourse and without antibiotic pretreat-ment, if regression of symptoms isobserved following catheter change,

a watchful waiting strategy may beindicated or therapy commenced af-ter identification of the pathogen. Incases of bacteremia caused by methi-cillin-sensitive S. aureus (MSSA) species,the rationality of de-escalation to flu-cloxacillin or the cephalosporin cefa-zoline has been demonstrated [161].Other cephalosporins or β-lactam/BLIcombinations cannot be recommendedon the basis of current evidence [162]. Ifvancomycin treatmentofa catheter-asso-ciated MRSA, CoNS, or enterococci BSIis unsuccessful, a combination of high-dose daptomycin (10–12mg/kg/day)with gentamicin or rifampicin can beadministered [136]. Case studies havedescribed effective use of ceftarolineas a treatment option for daptomycin-refractory persistent MRSA bacteremiain patients with infectious endocarditisor osteomyelitis [164, 165]. In severalstudies on catheter-associated BSI withgram-positive pathogens, the new lipo-glycopeptide dalbavancin showed signif-

S50 Der Anaesthesist · Suppl 1 · 2019

icantly better outcomes than vancomycinin terms of clinical and microbiologicresponse rates [166, 167]. The durationof therapy depends on the course ofinfection. An uncomplicated S. aureusinfection is represented by bacteremiawithout evidence of endocarditis, theabsence of septic foci/metastatic infec-tions, the absenceof intravascular foreignbodies (vascular catheters, heart valveprostheses, vascular prostheses, pace-maker, etc.), a sterile follow-up bloodculture collected 2–4 days after the initialpositive blood culture, anddefervescencewithin 72 h. Uncomplicated infectionsshould be treated for at least 2 weeks.In patients with a complicated infection,treatment for 4–6 weeks is indicated[163]. Management of Staphylococcusbacteremia includesnotonly treatmentofthe infection and control blood culturesduring therapy, but also transesophagealechocardiography (TEE) to exclude valvevegetation/endocarditis [163, 168, 169].

Nosocomial and ventilator-associated pneumoniaHAP and VAP are frequent causes ofsepsis in intensive care medicine. Ac-cording to their definitions, HAP isdiagnosed when the symptoms of pneu-monia arise >48 h after hospitalization.The terms established by the InfectiousDiseases Society of America (IDSA) andthe American Thoracic Society (ATS;[110]) for the US American region in2005, i. e., healthcare-associated pneu-monia (HCAP) and nursing home-acquired pneumonia (NHAP), have notbecome established in Germany due totheir lack of predictive value for infec-tion with MDR pathogens. The termNHAP has now also been revised inthe IDSA guideline. According to thedefinition, one speaks of VAP when thesymptoms of pneumonia arise after atleast 48 h of ventilation. The symptomsof VAP in ventilated patients are notvery specific (purulent tracheal secre-tion, fever, newly appearing opacity onthe chest x-ray, and compromised gasexchange), and it is not always possibleto definitively differentiate VAP fromother entities. The risk of VAP increasescontinuously during the first 7 days ofventilation, but declines again thereafter.

HAP and VAP constitute up to 22% ofall nosocomial infections [171]. Withmortality rates of about 13% [172] andnumerous other negative effects [173,174] for patients, VAP is considered tobe a more complex infection than HAP;however, with complication rates of 50%([175]; e. g., empyema formation, renalfailure, and sepsis), HAP is anything buta trivial challenge for ID specialists, andis also associated with mortality ratescomparable to those of VAP [175, 176].

Calculated therapy of nosocomialpneumonia in sepsis patients withoutseptic shock or anti-infective pretreat-ment, and who are also at a low risk ofhaving an MRD pathogen, can initiallycomprisemonotherapywithpiperacillin/tazobactam [177]. Group 1 carbapen-ems (imipenem/cilastatin, meropenem)cover a similar pathogen spectrum aspiperacillin/tazobactam, but are superiorto piperacillin/tazobactam in the pres-ence of ESBL-producing pathogens. Dueto their activity against many P. aerug-inosa isolates, cefepime or ceftazidimepreparations represent an important el-ement in the treatment of pneumonia.However, monotherapy of nosocomialpneumonia is not recommended in theGerman S3 guidelines because of lim-ited effectiveness against gram-positivepneumonia pathogens [187]. In con-trast, the IDSA mentions cefepime (notceftazidime) as a substance suitable formonotherapy [170]. Initial treatmentwith fluoroquinolones (moxifloxacinor levofloxacin) is possible in princi-ple, but must be critically evaluated.Monotherapy with an aminoglycosideis to be avoided [170]. Primary combi-nation therapy of HAP/VAP with twosubstances potentially active againstP. aeruginosa (and possibly extended bya preparation with action against MRSA)should be considered when the risk ofinfection with an MRD pathogen ishigh and/or the patient has reached thestage of septic shock. In this situation,German and American standards [170,177] dictate therapy comprising a β-lactam (e. g., piperacillin/tazobactam,ceftazidime/cefepime, or a group 1 car-bapenem) in combination with a fluo-roquinolone active against Pseudomonasspecies (ciprofloxacin, levofloxacin),

an aminoglycoside, or aztreonam [178].These combinations sufficiently cover themost important and frequently highlyresistant problem pathogens such asP. aeruginosa, as well as ESBL-pro-ducing gram-negative bacteria (E. coli,K. pneumoniae, Acinetobacter bauman-nii). Particularly the combination ofcarbapenem/fluoroquinolone generallyachieves good effectiveness against resis-tant pathogens such as ESBL-producingE. coli or K. pneumoniae, as well asP. aeruginosa. For patients with anti-infective pretreatment, a change of sub-stance class is also recommended inorder to prevent and avoid possibleresistance [178]. The IDSA recommen-dations consider the combination oftwo β-lactams to be of little use [170].In cases where the MRSA risk is high(e. g., known colonization of the pa-tient), an substance with activity againstMRSA is required—at least in patientswith nosocomial pneumonia with septicshock [177, 179]. Linezolid or van-comycin (not daptomycin; see below)can be administered for this indica-tion. Individual patient characteristicsand pathogen identification then lead tospecific monotherapy (MRSA, P. aerug-inosa, ESBL-producing Enterobacteri-aceae, etc.). Therefore, for successful,rational treatment of VAP/HAP, knowl-edge of the pathogen and its specificresistance phenotype is highly impor-tant. Particularly in VAP/HAP caused bygram-negative MDR pathogens that are,e. g., susceptible to polymyxins (colistin,polymyxin B) or aminoglycosides, thecombination of an intravenous (i. v.) andan inhalation therapy can be considered[170]. Although the evidence for aninhalation therapy is low, it does lead toa high local concentration of antibioticat the site of infection (i. e., the lungs).If carbapenemases are detected in theisolatedpathogen, treatmentmust be ori-ented towards the antibiogram and maynecessitate administration of polymyx-ins (i. v. and inhaled; [170]). The fixedcombination of ceftazidime/avibactamhas also recently been approved fortreatment of HAP. This combinationhas good in vitro efficacy against ESBLproducers and carbapenemase synthe-sizers (Klebsiella pneumoniae carbapen-

Der Anaesthesist · Suppl 1 · 2019 S51

Leitthema

emase, KPC; and oxacillinase, OXA-48), as well as against P. aeruginosa;it is therefore very interesting for tar-geted treatment of HAP caused by MPDpathogens with these resistance pheno-types. The combination of ceftolozane/tazobactam may receive approval fortreatment of VAP and HAP in the nearfuture. Results of recent studies are stillawaited (NCT01853982, NCT02387372,NCT02070757). Stenotrophomonas mal-tophilia is normally resistant to car-bapenems but generally susceptible tocotrimoxazole, which thus representsthe typical therapeutic option—providedthe pathogens are considered to be thecause of HAP/VAP. If this approach isnot possible, other preparations mustbe evaluated (e. g., ceftazidime, moxi-floxacin, levofloxacin). Carbapenemsaregenerally effective against Acinetobacterspecies and are indicated in instances ofhigh intrinsic resistance to other β-lac-tams. Panresistance to β-lactams, whichis frequently accompanied by resistanceto fluoroquinolones, renders combina-tion of colistin with another substancedictated by resistance testing the onlypossible option. MRSA pneumonia canin principle be treated by vancomycin(alternatively teicoplanin) or linezolid[170, 178]. As a modern alternative fortreatment of MRSA pneumonia, tedi-zolid can also be employed successfully:In the Linezolid in the Treatment ofSubjects with Nosocomial PneumoniaProven to be Due to Methicillin-Re-sistant Staphylococcus aureus (ZEPHyR;[179]) study, tedizolid was shown to bemore effective than vancomycin. Sincedaptomycin is inactivated by surfactants,its administration in MRSA pneumoniais not rational [178]. A further optionhas been available since 2014, namelyceftobiprole (only HAP; [180]), which isalso effective against MRSA. The dura-tion of VAP and HAP treatment shouldonly exceed 7 days in isolated cases.Clinical studies and meta-analyses ofVAP/HAP treatment duration found nodifferences in terms of mortality, pneu-monia recurrence, treatment failure, orduration of invasive ventilation whencomparing short (7–8 days) and longer(up to 15 days) treatment durations [150,170, 181, 182]. In a subgroup analysis of

the study by Chastre et al. [150] on VAP,it was shown that in pneumonias causedby non-fermenters such as P. aeruginosaand A. baumannii, for equal mortalityrates, the group with shorter treatment(8 days) had more recurrent infections(40.6 vs. 25.5%). However, this aspectis now judged by the IDSA to be soweak (at least for P. aeruginosa), thatit is recommended against generalizedprolongation of treatment [170]. Theduration of antibiotic therapy of VAPand HAP is always based on clinical andradiologic criteria.

Identification of Candida species, en-terococci, or CoNS in tracheal secretionfrom a non-neutropenic patient does notrepresent an indication for treatment. Inthe vast majority of cases, this is not aninvasive infection but rather representsthe resident local flora or sample con-tamination.

Intraabdominal infectionsIntraabdominal infections (IAI) are thecause of severe sepsis or septic shock inabout 30% of patients, and are one ofthe most frequent diagnoses in surgicalICUs [26]. In 90% of cases, the IAIrequires primary surgical treatment andantibiotic therapy. Close collaborationwith surgical colleagues is vital fromthe outset. The calculated antibiotictherapy should always cover aerobic andanaerobic gram-negative pathogens, aswell as gram-positive pathogens of thegastrointestinal tract [183]. Althoughone still differentiates between ambulantand nosocomial IAI, it must be kept inmind that patients these days are fre-quently colonized with MDR pathogensdue to long-distance travel, antibioticpretreatment, or immunosuppression.Sepsis of intraabdominal origin withoutseptic shock or pretreatment can initiallyundergo anti-infective treatment withpiperacillin/tazobactam or carbapenemmonotherapy. Basedonknowledgeof thelocal pathogen spectrum, combinationof, e. g., a cephalosporin or a fluoro-quinolone with metronidazole is alsopossible. In particularly severe diseasecourses, tigecycline can serve as thecombination partner, which has activityagainst Enterobacteriaceae and VRE. Inrefractory IAI, an invasiveCandida infec-

tion and antimycotic treatment must beconsidered; echinocandins are a possibletreatment option [153]. If the responseto treatment is inadequate, surgical inter-ventionmust bediscussed. PostoperativeIAIs (e. g., anastomotic insufficiency af-ter anterior rectal reconstruction) areassociatedwith a high risk ofMDR infec-tion. In these cases, a selected pathogenspectrumencompassing enterococci (in-cluding VRE), gram-negative problempathogens (carbapenemase synthesiz-ers), fungi, and rarer KPC producersmust be expected. In this situation,administration of second substance ad-ditional to a carbapenem must thereforebe considered—at least in patients withseptic shock—in order to broaden theempirically addressed spectrum. Along-side carbapenems (such as meropenem,imipenem/cilastatin, ertapenem), e. g.,tigecycline and fosfomycin (never asmonotherapy) can be used. A treatmentwith activity against MRSA is indicatedwhen a patient with a nosocomial cIAI iscolonized with MRSA or has a high riskof an MRSA infection [183]. MRSA-cIAI is then additionally treated withvancomycin, tigecycline, or linezolid[183]. New escalation variations includeceftolozane/tazobactam or ceftazidime/avibactam [184, 185] in combinationwith metronidazole for cIAI with re-sistant gram-negative pathogens [186].Provided effective surgical control of theinfection is achieved and the patient’sclinical condition improves, a treatmentduration of 4–7 days is sufficient [149,153, 183].

Particularly in esophageal surgery ismediastinitis due to the anastomotic situ-ation a possible infection focus. Frequentpathogens after esophageal surgery aremainly gram-positive cocci, anaerobes,and Candida species. Primary treat-ment of mediastinitis should comprisea group 1/2 carbapenem (imipenem/cilastatin, ertapenem, meropenem). Al-ternatively, an acyl-aminopenicillin/BLIor group 3/4 cephalosporinwithmetron-idazole can be administered. If there isa risk of MRSA, linezolid, tigecycline,or daptomycin is indicated [153]. Withrespect to the possibility of invasivemycosis, an antimycotic may be indi-cated. In patients with septic shock or

S52 Der Anaesthesist · Suppl 1 · 2019

AUC> MIC

< MIC

MICf T>MIC

t

CFig. 29 Influence of aug-mented renal clearance ofa β-lactamon the course ofserum levels (blue curve) incomparison to the courseof serum levels with phys-iologic elimination (blackcurve), with an arrow in-dicating the reduction inthe interval duringwhichserum levels above areaboveMIC.MICminimuminhibitory concentration,AUC area under the curve,C serum concentration,fT>MIC time interval duringwhich the serum concen-trationofantibiotic isabovethe pathogen-specificMIC,t time

a complicated disease course, surgicalintervention must also be considered.In addition to endoscopic exploration,modern concepts include insertion ofa vacuum suction system (e. g., Endo-SPONGE®, B. Braun, Melsungen, Ger-many) or an interventional drainagecatheter. As a last resort, surgical re-vision of the area/anastomosis must beperformed.

Necrotizing fasciitis/Fournier gan-grene is a potentially life-threatening softtissue infection that frequently exhibitsa fulminant disease course. Along-side classical streptococcal infections,predominantly polymicrobial infectionswith gram-negative pathogens and S. au-reus also exist in necrotizing fasciitis.In light of the pathogen spectrum andfrequently polymicrobial infection, pri-mary treatment must comprise a suf-ficiently broad combination therapy.Generally, acyl-aminopenicillins/BLI ora carbapenem (e. g., meropenem), peni-cillin G, and clindamycin (or linezolid)are combined. Clindamycin is a suf-ficiently tissue-penetrating lincosamidewith activity against anaerobes and S. au-reus; in combination with a β-lactam,it also inhibits protein biosynthesis ingram-positive pathogens. The latter ef-fect can reduce the rate of complicationsdue to microbial exotoxins [187]. Inpatients with clindamycin intolerance,linezolid (or tedizolid) represents analternative (caveat: hemotoxicity, opticneuritis; [187]). Penicillin G has thegreatest effectivity against streptococci.

If infection with MRSA is likely, in thecontext of a soft tissue infection, line-zolid should be preferred to vancomycin[153]. For some time now, therapeuticalternatives have been available in theform of tedizolid, or more recently, oxa-zolidinone. Due to the fulminant diseasecourse and frequently fatal symptomsupon inadequate control of the infection,the necessity of surgical reevaluation andthe indication for revision surgery mustbe checked regularly.

Other causesOther causes of sepsis include endo-carditis, community-acquired pneumo-nia, complicated urinary tract infections,osteomyelitis, erysipelas, and meningi-tis. For treatment recommendations,the authors refer to the guidelines ofthe respective specialist societies (www.awmf.org).

Pharmacokinetic concepts

Calculated antibiotic therapy of sepsisaims to achieve a sufficiently high drugconcentration. Wherever possible, thisobjective should be reached with the firstdose of antibiotic, i. e., with the loadingdose (LD). This is particularly importantfor the time-dependent β-lactams [188].Independent of organ dysfunction, aug-mented renal clearance (ARC), or organsupport therapy, sufficient plasma levelsmust be maintained during the course oftherapy in order to maximize the anti-infective potency of β-lactams. Whereas

the necessary plasma concentration ofan antibiotic is determined primarily bythe pathogen-specificMIC, estimation ofdrug elimination rates and the volume ofdistribution is very difficult in criticallyill patients. Particularly in patients withsepsis and septic shock, almost all anti-infective substances lead to considerablechanges in substance-specific pharma-cokinetics. The example in. Fig. 2 showstheeffectofARConthekineticsofplasmaconcentrations of a β-lactam antibiotic.ARC of β-lactam results in prematuredecline of the serum level to below thepathogen-specific MIC, therefore reduc-ing the time for which the concentrationremains above the MIC (fT>MIC)—a fac-tor which is crucial for the bactericidaleffect of the preparation.

In order for the calculated antibiotictherapy to cover thewidest possible spec-trum of potential pathogens when theidentity of the pathogen is unknown,broad-spectrum antibiotics in the high-est possible concentrations are requiredat the site of infection. Dosage recom-mendations and microbiologic determi-nation of the susceptibility of a bacterialstrain to particular antibiotics are basedon the assumption that the pharmacoki-netics of the drug correspond to those ofa “normal patient.” However, the distri-bution and elimination capacity of var-ious antibiotic groups is highly variablein sepsis patents, and difficult to pre-dict [189–196]. Although specialist soci-eties such as the Paul EhrlichGesellschaft(PEG) or IDSA allow for this in their

Der Anaesthesist · Suppl 1 · 2019 S53

Leitthema

Sepsis/septic shock

Hyperdynamic circulation Abnormal fluid balance Hepatic and/or renal dysfunction

Cardiac output↑(organ perfusion)

Capillary leak, abnormal protein binding

Renal clearance↑ Volume of distribution↑ Renal clearance↓

Low serum concentration Low serum concentration Increased serumconcentration

Fig. 39 Pathophysiologicinfluences of sepsis on theserum concentration of an-tibiotics

recommendations [153, 178, 183], thesefacts are still paid too little attentionin clinical routine—with the effect ofincreased mortality in seriously ill pa-tients [104, 197–200]. Numerous organdysfunctions are at the center of sub-stance-specific changes in pharmacoki-netics. The resultant changes in antibi-otic concentrations in various compart-mentscanleadtotreatment failureononehand, and development of resistance onthe other; moreover, these changes maypredispose the patient to potential toxiceffects [159]. The time-dependent β-lac-tam antibiotics are particularly problem-atic in this context. A summary of theeffects of various importantpathophysio-logic changes occurring during sepsis onthe serum concentrations of antibioticsis presented in . Fig. 3.

Whereas a considerably shorter timeabove the MIC is sufficient for bacte-riostasis, maximal bactericide with peni-cillinsandcephalosporinsrequiresanun-bound-drug concentrationexceeding theMIC for at least 50–70% of the dosinginterval (fT>MIC). Carbapenems requirean fT>MIC of about 40% [153, 198]. ThefT>MIC is lower for carbapenems becauseamore pronounced post-antibiotic effectis assumed for this substance class thanfor other substances [198]. These rec-ommendations are derived from animal

experiments; clinical investigations un-derscore the fact that in sepsis patients,an fT>MIC of 100% elicits a more effec-tive anti-infective effect, and is thus rele-vant for outcome [190, 201–203]. Espe-cially on the basis of striving to achieveadequate tissue concentrations even indeep-lying compartments (pneumonia;bone infections; infections of the centralnervous system, CNS)—frequently in thecontext of disturbed microcirculation insepsis patients [204–206] and also wish-ing to avoid development of resistance[207, 208]—many experts recommendaiming for a concentration of β-lactamantibiotics in the primary compartment(serum/plasma) that exceeds the MIC by4-times (up to 6-times) for 60–100% ofthe dosing interval [209]. In addition toindividual dosing of antibiotics, the cur-rent DGI guideline [124] names thera-peutic drug monitoring (TDM) as a pos-sibility for treatment management andreduction of undesirable side effects ofantibiotics.

How difficult it can be to reach ef-fective serum/plasma concentrationsin sepsis patients was illustrated byRoberts et al. [202] in a prospectivestudy from 2014: Of 248 patients witha severe infection, 16% did not evenachieve the minimum pharmacokinet-ics/pharmacodynamics (PK/PD) goal

of 50% of the dosing interval abovethe MIC (50% fT>MIC). This result wasassociated with a significantly reducedprobability of a good clinical outcome.That insufficient plasma concentrationsrepresent a frequent problem in sepsispatients was demonstrated by Tacconeet al. [209] and Udy et al. [210], but alsoin the prospective study by Roberts et al.(Defining Antibiotic Levels in IntensiveCare Unit Patients, DALI; [202]) usingthe example of β-lactams. In the latterstudy, the dose of β-lactam antibiotichad to be adjusted in 175 of 236 pa-tients (75%); in 119 cases (50.4%) thisadjustment comprised a dosage increase[211]. The results of the study byTacconeet al. showed that following the initialadministration of the standard dose incritically ill patients, only meropenemachieved a sufficient pharmacokineticprofile in 75%of patients (12/16). In con-trast, ceftazidime (28%), cefepime (16%),and piperacillin/tazobactam (44%) onlyachieved an unsatisfactory PK profile,with correspondingly insufficient effects[209]. These data illustrate not only theproblems associatedwith antibiotic ther-apy, but also showhow important preciseknowledge of the pharmacodynamic andpharmacokinetic characteristics of theemployed antibiotic is.

S54 Der Anaesthesist · Suppl 1 · 2019

> 1MIC

< 1MIC

4 MIC

f T>1MIC

Bolu

s

f T>4MIC

t

1 2

Prolonged infusion over 3 h

3

Fig. 49 Effect of pro-longed infusion of β-lac-tams in sepsis. fT>xMIC timeinterval duringwhich theserum concentration ofantibiotic is 1-or x-timesabove the pathogen-spe-cificminimum inhibitoryconcentration (MIC)

For amore effective treatment of criti-cally ill sepsis patients, the authors of thecurrent article recommend an infusionregimen that it is on one hand feasible,while on the other hand capable of atleast partially counteracting the afore-mentioned pharmacokinetic problems:prolonged infusion of β-lactam antibi-otics. In this scenario, an initial high-dose bolus application over 20–30minfirst generates a peak plasma level, whichshould ensure a sufficient concentrationof antibiotic in the target compartment.Thereafter, subsequent doses are admin-istered as a prolonged infusion over 3 hvia an infusion pump. In this manner,fT>MIC can be maximized and the effec-tiveness of β-lactams increased via op-timal exposure of the antibiotic to thepathogen(. Fig. 4). After reachingapeakplasma level, intermittent bolus applica-tions (. Fig. 4 black curve) are charac-terized by a rapid decline of the plasmaconcentrations tobelowtheMIC.This re-sults in relatively short time spans duringwhich the effective concentrations (forsepsis and septic shock) are maintained(fT>1–4MIC). After an initial bolus (. Fig. 4blue curve), prolonged infusion over 3 h(up to 4 h) achieves a significant exten-sion of the time interval during whichthe effective concentration is maintained(fT>MIC and fT>4MIC).

A beneficial effect of prolongedpiperacillin/tazobactam infusion in pa-

tients with P. aeruginosa infections wasshown in 2007 by Lodise et al. [212].One year later, Lee et al. demonstratedusing pharmacokinetic models that pro-longed application of meropenem andimipenem/cilastatin increased the cumu-lative probability of a fT>MIC sufficient forclinically relevant pathogens [213]. Afterthe DALI study was able to show regularinsufficient β-lactam concentrations incritically ill patients [202], a subgroupanalysis of these data by Abdul-Azizet al. demonstrated that prolonged/continuous infusion of piperacillin/tazobactam and meropenem led to sig-nificant improvement of 30-day survival(p < 0.05) in patients with pneumoniaand/or SOFA score >9, and the cure rateof patients with SOFA >9 was also signif-icantly higher with prolonged infusion[214].

Linezolid also showed a high variabil-ity of plasma levels in patientswith severeinfections, with insufficient blood con-centrations using standard dosing con-cepts. Current data from Taubert et al.[215] show that in this case—similar tovancomycin [158]—continuous infusioncan help to attain effective PK/PD targetsand reduce toxicity.

In summary, for the treatment ofsepsis, prolonged infusion of β-lactamantibiotics after an initial bolus applica-tion should be preferred to intermittentbolus therapy. For improved treatment

management as well as in order toavoid over- and under-dosing, thera-peutic concentration monitoring maybe a rational complementary module.Particularly in the context of sepsis,clinicians lack a parameter in the highlyacute phase (24–48 h) by which they canjudge the success or lack of success ofthe calculated antibiotic therapy. Here,TDM of β-lactams can at least demon-strate that effective therapeutic plasmaconcentrations have been reached. Forthis reason, TDM is now mentionedin the DGI recommendations as a ra-tional complementary and treatmentoptimizing module [124, 153]. De-spite the fact that a positive influenceof TDM on patient mortality remainsunproven as yet, there are a varietyof pathophysiologic conditions underwhich TDM would appear essential, i. e.,high variability of drug distribution,changes in metabolism and elimina-tion, organ dysfunction, organ supporttherapy/extracorporeal support systems,interactions of diverse drug groups, ordangerous over-/under-dosing of the an-tibiotic, and this is clearlypreferred in thecorresponding studies [159, 216–223].TARGET was the first German prospec-tive multicenter study to investigatethe influence of TDM of piperacillin/tazobactam in sepsis patients (https://www.clinicaltrialsregister.eu/ctr-search/trial/2016-000136-17/DE). Prolonged

Der Anaesthesist · Suppl 1 · 2019 S55

Leitthema

infusion of β-lactams is, however, notapproved for any of the commerciallyavailable preparations, and should onlybe performed if the substance is atleast sufficiently stable at room temper-ature over the application interval of3–4 h. This holds true for piperacillin/tazobactam, meropenem, ceftazidime,and cefepime.

Conclusions

Rapid diagnosis and treatment of sepsisis of critical importance for the patients.For this reason, a new and not uncon-troversial definition of sepsis was estab-lished in 2016, which is based primarilyon the presence of infection-associatedorgan dysfunction. The presence of or-gan dysfunction should be established inthe first instance using a modified SOFAscore (qSOFA).

Blood cultures remain the gold stan-dard of ID diagnostics. The time point ofsampling, sample site, and correct sam-pling technique can eliminate a numberof potential sources of error. Molecu-lar genetic techniques for pathogen di-agnosis represent highly promising ap-proaches; however, these cannot cur-rently replace blood cultures and are notyet established in clinical routine. Thesemethods may gain importance in the fu-ture.

Antimicrobial therapy is one of thecornerstones of successful sepsis treat-ment. High sepsismortality rates and theoften unknown causal pathogen renderinitial application of a high-dose broad-spectrum mono- or combination ther-apy essential. Commencement of treat-ment within 1 h of recognition of sepsismust be strived for, since every delayleads to a significant increase in mortal-ity. Regular reevaluation of the infectionand the antibiotic therapy should be dic-tated by practical guidelines and coor-dinated within a multidisciplinary team(ABS team).

The grave pathophysiologic changesaccompanying sepsis (changes in the vol-ume of distribution, ARC, organ sup-port therapies) continue to frequently re-ceive inadequate consideration, althoughthese can have a profound influence onthe pharmacokinetics of antibiotics. In

light of these changes and in order toavoid incorrect dosing and suboptimaldrug concentrations in sepsis patients,treatment is increasingly diverging fromstandard dosing concepts and being ori-ented toward the PK/PD index of theparticular antibiotic. For β-lactams thismeans that, e. g., a prolonged (or con-tinuous) infusion should be preferred insepsispatients, suchthat theoptimaldrugconcentration (fT>MIC) can be achievedin the “blood” compartment through-out the dosing interval. For treatmentmanagement in sepsis, TDM is available.Studies assessing thebenefitofTDMhavedemonstrated positive results in terms ofmore effective treatment, improved sta-bility of plasma levels of the antibiotic,and reductions in treatment duration.

New antibiotics cover a defined spec-trum of pathogens and are characterizedpredominantly by their highly potent ac-tivity against MDR pathogens. In orderto delay development of resistance, appli-cation of these new preparations shouldbe limited to specific escalation therapyof MDR infections. The administrationof new antibiotics should be discussedwithinanddecideduponbyanABS team.Only in this way can the rational use ofnew antibiotics be ensured.

Practical conclusion

4 In-depth knowledge of the patho-physiologic changes associated withsepsis and their influence on thepharmacokinetics of antibiotics (par-ticularly β-lactams) has led to revisionof the use of β-lactams in particular.

4 With respect to PK/PD targets, pro-longed infusion of β-lactams (afteran initial bolus) should be preferredto intermittent application in sepsis.

4 Studies now exist which prove theadvantages of prolonged infusionover intermittent bolus application.Irrespective of the dosing strategy,an initial LD (high-dose bolus appli-cation) is required.

4 TDM can be useful for sepsis treat-ment management.

Corresponding address

Dr. D. C. RichterDepartment of Anesthesiology, HeidelbergUniversity HospitalIm Neuenheimer Feld 110, 69120 Heidelberg,Germanydaniel1.richter@med.uni-heidelberg.de

Compliance with ethicalguidelines

Conflict of interest. D. C. Richter, A. Heininger,T. Brenner,M. Hochreiter,M. Bernhard, J. Briegel,S. Dubler, B. Grabein, A. Hecker,W. A. Kruger, K.Mayer,M.W. Pletz, D. Storzinger, N. Pinder, T. Hoppe-Tichy, S.Weiterer, S. Zimmermann, A. Brinkmann,M. A.Weigand, andC. Lichtenstern declare that theyhave no competing interests.

This article does not contain any studieswith humanparticipants or animals performedby anyof the au-thors. Local ethical guidelines applied to all studiesincluded in the current review.

References

1. Dellinger RP, Levy MM, Rhodes A, Annane D,Gerlach H, Opal SM et al (2013) SurvivingSepsis Campaign: international guidelines formanagement of severe sepsis and septic shock,2012. Intensive CareMed 39(2):165–228. https://doi.org/10.1007/s00134-012-2769-8

2. Kaukonen KM, Bailey M, Pilcher D, Cooper DJ,BellomoR (2015) Systemic inflammatory responsesyndrome criteria in defining severe sepsis. N EnglJ Med 372(17):1629–1638. https://doi.org/10.1056/NEJMoa1415236

3. Dellinger RP, Carlet JM, Masur H, Gerlach H,Calandra T, Cohen J et al (2004) Surviving sepsiscampaign guidelines for management of severesepsis and septic shock. Intensive Care Med30(4):536–555. https://doi.org/10.1007/s00134-004-2210-z

4. ChurpekMM,Zadravecz FJ,WinslowC,HowellMD,EdelsonDP (2015) Incidence andprognostic valueof the systemic inflammatory response syndromeand organ dysfunctions in ward patients. Am JRespirCritCareMed192(8):958–964

5. SingerM, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M et al (2016) The thirdinternational consensus definitions for sepsis andseptic shock (sepsis-3). JAMA 315(8):801–810.https://doi.org/10.1001/jama.2016.0287

6. Bloos F, Thomas-Rüddel D, Rüddel H, EngelC, Schwarzkopf D, Marshall JC et al (2014)Impactof compliancewith infectionmanagementguidelines on outcome in patients with severesepsis: a prospective observational multi-centerstudy. Crit Care 18(2):R42. https://doi.org/10.1186/cc13755

7. Rhodes A, Evans LE, Alhazzani W, Levy MM,Antonelli M, Ferrer R et al (2017) SurvivingSepsis Campaign: international guidelines formanagement of sepsis and septic shock: 2016.Intensive CareMed43(3):34–377. https://doi.org/10.1007/s00134-017-4683-6

S56 Der Anaesthesist · Suppl 1 · 2019

8. Schmoch T, Bernhard M, Uhle F, Gründling M,Brenner T, Weigand M (2017) Neue SEPSIS-3-Definition. Anaesthesist1(8):614–621. https://doi.org/10.1007/s00101-017-0316-2

9. Seymour CW, Liu VX, Iwashyna TJ, Brunkhorst FM,Rea TD, Scherag A et al (2016) Assessment ofclinicalcriteria forsepsis: fortheThirdInternationalConsensus Definitions for sepsis and septic shock(Sepsis-3). JAMA315(8):762–774. https://doi.org/10.1001/jama.2016.0288

10. Shankar-HariM, PhillipsGS, LevyML, SeymourCW,Liu VX, Deutschman CS et al (2016) Developinganewdefinitionandassessingnewclinical criteriafor septic shock: for the third internationalconsensus definitions for sepsis and septic shock(sepsis-3). JAMA315(8):775–787. https://doi.org/10.1001/jama.2016.0289

11. Seymour CW, Rosengart MR (2015) Septic shock:advances in diagnosis and treatment. JAMA314(7):708–717. https://doi.org/10.1001/jama.2015.7885

12. FleischmannC, ScheragA, Adhikari NK, HartogCS,TsaganosT, SchlattmannPetal (2016)Assessmentof global incidence and mortality of hospital-treated sepsis. Current estimates and limitations.AmJRespir Crit CareMed193(3):259–272. https://doi.org/10.1164/rccm.201504-0781OC

13. FleischmannC,Thomas-RueddelDO,HartmannM,HartogCS,Welte T,HeubleinSetal (2016)Hospitalincidence and mortality rates of sepsis. DtschArztebl Int 113(10):159–166. https://doi.org/10.3238/arztebl.2016.0159

14. Martin GS, Mannino DM, Eaton S, Moss M(2003) The epidemiology of sepsis in the UnitedStates from 1979 through 2000. N Engl J Med348(16):1546–1554. https://doi.org/10.1056/NEJMoa022139

15. Vincent J-L, Rello J, Marshall J, Silva E, Anzueto A,Martin CD et al (2009) International study of theprevalence andoutcomes of infection in intensivecare units. JAMA 302(21):2323–2329. https://doi.org/10.1001/jama.2009.1754

16. Friedman G, Silva E, Vincent J-L (1998) Has themortality of septic shock changedwith time? CritCareMed26(12):2078–2086

17. Group SCCT (2016) Incidence of severe sepsis andseptic shock in German intensive care units: theprospective, multicentre INSEP study. IntensiveCare Med 42(12):1980–1989. https://doi.org/10.1007/s00134-016-4504-3

18. Mouncey PR, Osborn TM, Power GS, Harrison DA,Sadique MZ, Grieve RD et al (2015) ProtocolisedManagement In Sepsis (ProMISe): a multicentrerandomised controlled trial of the clinical effec-tiveness and cost-effectiveness of early, goal-directed, protocolised resuscitation for emergingseptic shock. Health Technol Assess. https://doi.org/10.3310/hta19970

19. Delaney AP, Peake SL, Bellomo R, Cameron P,Holdgate A, Howe B et al (2013) The AustralasianResuscitation in Sepsis Evaluation (ARISE) trialstatistical analysis plan. Crit Care Resusc15(3):162–171

20. Pro CI, Yealy DM, Kellum JA, Huang DT, BarnatoAE,Weissfeld LA et al (2014) A randomized trial ofprotocol-based care for early septic shock. N EnglJ Med 370(18):1683–1693. https://doi.org/10.1056/NEJMoa1401602

21. Gaieski DF, Edwards JM, Kallan MJ, Carr BG (2013)Benchmarking the incidence and mortality ofsevere sepsis in the United States. Crit Care Med41(5):1167–1174

22. Angus DC, Linde-ZwirbleWT, Lidicker J, ClermontG, Carcillo J, Pinsky MR (2001) Epidemiology of

severe sepsis in the United States: analysis ofincidence, outcome, and associated costs of care.CritCareMed29(7):1303–1310

23. Graf J, Reinhold A, Brunkhorst FM, Ragaller M,Reinhart K, Loeffler M et al (2010) Variabilityof structures in German intensive care units—arepresentative, nationwide analysis. Wien KlinWochenschr 122(19–20):572–578. https://doi.org/10.1007/s00508-010-1452-8

24. Engel C, Brunkhorst FM, Bone HG, BrunkhorstR, Gerlach H, Grond S et al (2007) Epidemiologyof sepsis in Germany: results from a nationalprospectivemulticenter study. IntensiveCareMed33(4):606–618. https://doi.org/10.1007/s00134-006-0517-7

25. EngelC,BrunkhorstFM,LöfflerM,ReinhartK(2007)DiagnoseundEpidemiologieder Sepsis. MedWelt58:307–310

26. DeFrances CJ, Lucas CA, Buie VC, Golosinskiy A(2006) National hospital discharge survey. NatlHealthStatReport2008(5):1–20

27. Angus DC, van der Poll T (2013) Severe sepsis andsepticshock. NEnglJMed369(9):840–851. https://doi.org/10.1056/NEJMra1208623

28. Opal SM, Garber GE, LaRosa SP, Maki DG,Freebairn RC, Kinasewitz GT et al (2003) Systemichost responses in severe sepsis analyzed bycausative microorganism and treatment effectsof drotrecogin alfa (activated). Clin Infect Dis37(1):50–58. https://doi.org/10.1086/375593

29. Ranieri VM, Thompson BT, Barie PS, Dhainaut J-F,Douglas IS, Finfer S et al (2012) Drotrecogin alfa(activated) in adults with septic shock. N EnglJ Med 366(22):2055–2064. https://doi.org/10.1056/NEJMoa1202290

30. Geffers C, Maechler F, Behnke M, Gastmeier P(2016) Multiresistente Erreger – Epidemiologie,Surveillance und Bedeutung. Anasthesiol Inten-sivmedNotfallmed Schmerzther 51(02):104–111.https://doi.org/10.1055/s-0041-103348

31. Woerther P-L, Burdet C, Chachaty E, Andremont A(2013)Trends inhumanfecal carriageofextended-spectrumβ-lactamases in the community: towardthe globalization of CTX-M. Clin Microbiol Rev26(4):744–758. https://doi.org/10.1128/CMR.00023-13

32. Munoz-Price LS, Poirel L, Bonomo RA, SchwaberMJ, Daikos GL, Cormican M et al (2013) Clinicalepidemiologyof theglobal expansionof klebsiellapneumoniae carbapenemases. Lancet Infect Dis13(9):785–796. https://doi.org/10.1016/S1473-3099(13)70190-7

33. KaramG, Chastre J, Wilcox MH, Vincent J-L (2016)Antibiotic strategies in the era of multidrugresistance. Crit Care 20(1):136. https://doi.org/10.1186/s13054-016-1320-7

34. Gastmeier P, Schroder C, Behnke M, Meyer E,Geffers C (2014)Dramatic increase in vancomycin-resistant enterococci in Germany. J AntimicrobChemother 69(6):1660–1664. https://doi.org/10.1093/jac/dku035

35. van der Bij AK, Pitout JD (2012) The role ofinternational travel in the worldwide spread ofmultiresistant enterobacteriaceae. J AntimicrobChemother 67(9):2090–2100. https://doi.org/10.1093/jac/dks214

36. ArcillaMS,vanHattemJM,HaverkateMR,BootsmaMC,vanGenderenPJ,GoorhuisAetal(2017)Importand spread of extended-spectrum β-lactamase-producing enterobacteriaceae by internationaltravellers (COMBAT study): a prospective, multi-centre cohort study. Lancet InfectDis 17(1):78–85.https://doi.org/10.1016/S1473-3099(16)30319-X

37. Mendelson M (2015) The World Health Organiza-tionglobalactionplan forantimicrobial resistance:guest editorial. SAfrMedJ105(5):325. https://doi.org/10.7196/SAMJ.9644

38. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A,Knoblich B et al (2001) Early goal-directed therapyin the treatment of severe sepsis and septic shock.N Engl JMed345(19):1368–1377. https://doi.org/10.1056/NEJMoa010307

39. Investigators A, Group ACT, Peake SL, Delaney A,Bailey M, Bellomo R et al (2014) Goal-directedresuscitation for patients with early septic shock.N Engl JMed371(16):1496–1506. https://doi.org/10.1056/NEJMoa1404380

40. Lilly CM (2014) The ProCESS trial—a newera of sepsis management. N Engl J Med370(18):1750–1751. https://doi.org/10.1056/NEJMe1402564

41. Bernhard M, Brenner T, Brunkhorst FM, WeigandMA (2015) Frühe innerklinische Sepsistherapie.NotfRettungsmed18(7):595–605. https://doi.org/10.1007/s10049-015-0098-5

42. Dellinger RP (2015) The future of sepsisperformance improvement. Crit Care Med43(9):1787–1789. https://doi.org/10.1097/CCM.0000000000001231

43. SeymourCW,Kahn JM,Martin-Gill C, CallawayCW,Yealy DM, Scales D et al (2017) Delays from firstmedical contact to antibiotic administration forsepsis. Crit Care Med 45(5):759–765. https://doi.org/10.1097/CCM.0000000000002264

44. Damiani E, Donati A, Serafini G, Rinaldi L, AdrarioE, Pelaia P et al (2015) Effect of performanceimprovementprogramsoncompliancewithsepsisbundles and mortality: a systematic review andmeta-analysis of observational studies. PLOSONE10(5):e125827. https://doi.org/10.1371/journal.pone.0125827

45. Rhodes A, Phillips G, Beale R, Cecconi M, ChicheJD, De Backer D et al (2015) The surviving sepsiscampaign bundles and outcome: results fromthe international Multicentre Prevalence Studyon Sepsis (the IMpreSS study). Intensive CareMed 41(9):1620–1628. https://doi.org/10.1007/s00134-015-3906-y

46. Seymour CW, Gesten F, Prescott HC, FriedrichME, Iwashyna TJ, Phillips GS et al (2017) Timeto treatment and mortality during mandatedemergency care for sepsis. N Engl J Med376(23):2235–2244. https://doi.org/10.1056/NEJMoa1703058

47. Jones SL, Ashton CM, Kiehne L, Gigliotti E, Bell-Gordon C, Disbot M et al (2015) Reductionsin sepsis mortality and costs after design andimplementationofanurse-basedearlyrecognitionand responseprogram. JtCommJQual PatientSaf41(11):483

48. Levy MM, Rhodes A, Phillips GS, Townsend SR,Schorr CA, Beale R et al (2015) Surviving SepsisCampaign: association between performancemetrics andoutcomes ina7.5-year study. Crit CareMed43(1):3–12. https://doi.org/10.1007/s00134-014-3496-0

49. Pierrakos C, Vincent J-L (2010) Sepsis biomarkers:a review. Crit Care 14(1):R15. https://doi.org/10.1186/cc8872

50. Lichtenstern C, Brenner T, Bardenheuer HJ,WeigandMA(2012)Predictors of survival in sepsis:what is thebest inflammatorymarker tomeasure?Curr Opin Infect Dis 25(3):328–336. https://doi.org/10.1097/QCO.0b013e3283522038

51. Jones AE, Shapiro NI, Trzeciak S, Arnold RC,Claremont HA, Kline JA et al (2010) Lactateclearance vs central venous oxygen saturation as

Der Anaesthesist · Suppl 1 · 2019 S57

Leitthema

goals of early sepsis therapy: a randomizedclinicaltrial. JAMA 303(8):739–746. https://doi.org/10.1001/jama.2010.158

52. Thomas-Rueddel DO, Poidinger B,WeissM, Bach F,DeyK,HäberleHetal (2015)Hyperlactatemia is anindependent predictor of mortality and denotesdistinctsubtypesof severesepsisandsepticshock.J Crit Care 30(2):439.e1–439.e6. https://doi.org/10.1016/j.jcrc.2014.10.027

53. Casserly B, Phillips GS, Schorr C, DellingerRP, Townsend SR, Osborn TM et al (2015)Lactate measurements in sepsis-induced tissuehypoperfusion: results from the Surviving SepsisCampaigndatabase. Crit CareMed43(3):567–573.https://doi.org/10.1097/CCM.0000000000000742

54. Reichert M, Hecker M, Hörbelt R, Lerner S, HöllerJ, Hecker C et al (2015) Die Rolle von BiomarkerninderDiagnostikder akutenMesenterialischämie.Chirurg 86(1):47–55. https://doi.org/10.1007/s00104-014-2906-8

55. Reichert M, Hecker M, Hörbelt R, Lerner S,Holler J, Hecker C et al (2015) Erratum zu: DieRolle von Biomarkern in der Diagnostik derakutenMesenterialischämie. Chirurg86(1):55–55.https://doi.org/10.1007/s00104-014-2906-8

56. Henriquez-Camacho C, Losa J (2014) Biomarkersfor sepsis. Biomed Res Int 2014:547818. https://doi.org/10.1155/2014/547818

57. Limper M, De Kruif M, Duits A, Brandjes D, vanGorp E (2010) The diagnostic role of procalcitoninand other biomarkers in discriminating infectiousfromnon-infectious fever. J Infect 60(6):409–416.https://doi.org/10.1016/j.jinf.2010.03.016

58. Wacker C, Prkno A, Brunkhorst FM, SchlattmannP (2013) Procalcitonin as a diagnostic marker forsepsis: a systematic review and meta-analysis.Lancet Infect Dis 13(5):426–435. https://doi.org/10.1016/S1473-3099(12)70323-7

59. Siegler BH, Weiterer S, Lichtenstern C, StumppD, Brenner T, Hofer S et al (2014) Use ofbiomarkers in sepsis. Update and perspectives.Anaesthesist 63(8–9):678–690. https://doi.org/10.1007/s00101-014-2347-2

60. Uzzan B, Cohen R, Nicolas P, CucheratM, Perret GY(2006) Procalcitonin as a diagnostic test for sepsisin critically ill adults and after surgery or trauma:a systematic review and meta-analysis. Crit CareMed34(7):1996–2003

61. Liu D, Su L, Han G, Yan P, Xie L (2015) Prognosticvalueofprocalcitonin inadultpatientswith sepsis:a systematic review andmeta-analysis. PLOSONE10(6):e129450. https://doi.org/10.1371/journal.pone.0129450

62. Ugarte H, Silva E, Mercan D, De Mendonca A,VincentJ-L (1999)Procalcitoninusedasamarkerofinfection in the intensive care unit. Crit Care Med27(3):498–504

63. Suprin E, Camus C, Gacouin A, Le Tulzo Y, LavoueS, Feuillu A et al (2000) Procalcitonin: a valuableindicator of infection in a medical ICU? IntensiveCare Med 26(9):1232–1238. https://doi.org/10.1007/s001340000580

64. Panacek EA, Kaul M (1999) IL-6 as a marker of ex-cessive TNF-α activity in sepsis. Sepsis 3(1):65–73.https://doi.org/10.1023/A:1009878726176

65. Wakefield C, Barclay G, Fearon K, Goldie A, RossJ, Grant I et al (1998) Proinflammatory mediatoractivity, endogenousantagonistsandthesystemicinflammatory response in intra-abdominal sepsis.Br JSurg85(6):818–825

66. JekarlDW,LeeS-Y, LeeJ,ParkY-J,KimY,ParkJHetal(2013)Procalcitoninas adiagnosticmarker and IL-6asaprognosticmarkerforsepsis. DiagnMicrobiol

InfectDis 75(4):342–347. https://doi.org/10.1016/j.diagmicrobio.2012.12.011

67. Oliveira CF, Botoni FA, Oliveira CR, Silva CB, PereiraHA, Serufo JC et al (2013) Procalcitonin versusC-reactive protein for guiding antibiotic therapyin sepsis: a randomized trial. Crit Care Med41(10):2336–2343. https://doi.org/10.1097/CCM.0b013e31828e969f

68. Welsch T, Frommhold K, Hinz U, Weigand MA,Kleeff J, Friess Het al (2008) Persisting elevation ofC-reactive protein after pancreatic resections canindicate developing inflammatory complications.Surgery 143(1):20–28. https://doi.org/10.1016/j.surg.2007.06.010

69. Suberviola B, Castellanos A, Astudillo LG, IglesiasD, Melon FO (2011) Prognostic value of proad-renomedullin in severe sepsis and septic shockpatients with community-acquired pneumonia.CritCare15(Suppl1):276. https://doi.org/10.1186/cc9696

70. Suberviola B, Castellanos-Ortega A, Ruiz AR,Lopez-Hoyos M, Santibanez M (2013) Hospitalmortality prognostication in sepsis using thenew biomarkers suPAR and proADM in a singledetermination on ICU admission. Intensive CareMed 39(11):1945–1952. https://doi.org/10.1007/s00134-013-3056-z

71. Morgenthaler NG, Struck J, Alonso C, BergmannA (2005) Measurement of midregional proad-renomedullin in plasma with an immunolumi-nometric assay. Clin Chem 51(10):1823–1829.https://doi.org/10.1373/clinchem.2005.051110

72. Yaegashi Y, Sato N, Suzuki Y, Kojika M, Imai S,Takahashi G et al (2005) Evaluation of a newlyidentified soluble CD14 subtype as a marker forsepsis. J InfectChemother11(5):234–238. https://doi.org/10.1007/s10156-005-0400-4

73. Shozushima T, Takahashi G, Matsumoto N, KojikaM, Endo S, Okamura Y (2011) Usefulness ofpresepsin (sCD14-ST) measurements as amarkerfor the diagnosis and severity of sepsis thatsatisfied diagnostic criteria of systemic inflam-matory response syndrome. J Infect Chemother17(6):764–769. https://doi.org/10.1007/s10156-011-0254-x

74. Endo S, Suzuki Y, Takahashi G, Shozushima T,Ishikura H, Murai A et al (2012) Usefulnessof presepsin in the diagnosis of sepsis ina multicenter prospective study. J InfectChemother 18(6):891–897. https://doi.org/10.1007/s10156-012-0435-2

75. UllaM,PizzolatoE, LucchiariM,LoiaconoM,SoardoF, Forno D et al (2013) Diagnostic and prognosticvalue of presepsin in the management of sepsisin the emergency department: a multicenterprospective study. Crit Care 17(4):R168. https://doi.org/10.1186/cc12847

76. Ferrer R, Martin-Loeches I, Phillips G, OsbornTM, Townsend S, Dellinger RP et al (2014)Empiric antibiotic treatment reduces mortalityin severe sepsis and septic shock from thefirst hour: results from a guideline-basedperformance improvement program. Crit CareMed 42(8):1749–1755. https://doi.org/10.1097/CCM.0000000000000330

77. Asfar P, Meziani F, Hamel JF, Grelon F, MegarbaneB, Anguel N et al (2014) High versus low blood-pressure target in patients with septic shock.N Engl JMed370(17):1583–1593. https://doi.org/10.1056/NEJMoa1312173

78. Reinhart K, Brunkhorst F, Bone H, BardutzkyJ, Dempfle C, Forst H et al (2010) Prävention,Diagnose, Therapie und Nachsorge der Sepsis.1. Revision der S-2k Leitlinien der Deutschen

Sepsis-Gesellschaft eV (DSG) und der DeutschenInterdisziplinären Vereinigung für Intensiv-undNotfallmedizin (DIVI).Anaesthesist59(4):347–370

79. Surviving Sepsis Campaign (2015) Updatedbundles in response tonewevidence. http://www.survivingsepsis.org/SiteCollectionDocuments/SSC_Bundle.pdf. Accessed22Sept2017

80. Cockerill Fr WJ, Vetter E, Goodman K, TorgersonC, Harmsen W et al (2004) Optimal testingparameters for blood cultures. Clin Infect Dis38(12):1724–1730. https://doi.org/10.1086/421087

81. Lee A, Mirrett S, Reller LB, Weinstein MP (2007)Detectionofbloodstreaminfections inadults: howmany blood cultures are needed? J ClinMicrobiol45(11):3546–3548. https://doi.org/10.1128/JCM.01555-07

82. Zadroga R, Williams DN, Gottschall R, Hanson K,Nordberg V, Deike M et al (2013) Comparison of 2blood culturemedia shows significant differencesin bacterial recovery for patients on antimicrobialtherapy. Clin InfectDis 56(6):790–797. https://doi.org/10.1093/cid/cis1021

83. Kanegaye JT, Soliemanzadeh P, Bradley JS(2001) Lumbar puncture in pediatric bacterialmeningitis: defining the time interval forrecovery of cerebrospinal fluid pathogens afterparenteral antibiotic pretreatment. Pediatrics108(5):1169–1174

84. Cardoso T, Carneiro AH, Ribeiro O, Teixeira-PintoA, Costa-Pereira A (2010) Reducing mortality insevere sepsis with the implementation of a core6-hour bundle: results from the Portuguesecommunity-acquiredsepsis study (SACiUCIstudy).Crit Care 14(3):R83. https://doi.org/10.1186/cc9008

85. de Sousa AG, Junior CJF, Santos GPD, Laselva CR,Polessi J, Lisboa LF et al (2008) The impact ofeach action in the Surviving Sepsis Campaignmeasures on hospital mortality of patients withsevere sepsis/septic shock. Einstein (Sao Paulo)6(3):323–327

86. HansenS, SchwabF, BehnkeM,CarsauwH,HeczkoP, Klavs I et al (2009) National influences oncatheter-associated bloodstream infection rates:practices among national surveillance networksparticipating in the European HELICS project.J Hosp Infect 71(1):66–73. https://doi.org/10.1016/j.jhin.2008.07.014

87. Karch A, Castell S, Schwab F, Geffers C, BongartzH, Brunkhorst FM et al (2015) Proposing anempirically justified reference threshold for bloodculture sampling rates in intensive care units.J ClinMicrobiol 53(2):648–652. https://doi.org/10.1128/JCM.02944-14

88. Li J, Plorde JJ, Carlson LG (1994) Effects of volumeand periodicity on blood cultures. J ClinMicrobiol32(11):2829–2831

89. Baron EJ, Miller JM, Weinstein MP, Richter SS,Gilligan PH, Thomson RB et al (2013) A guideto utilization of the microbiology laboratoryfor diagnosis of infectious diseases: 2013recommendations by the Infectious DiseasesSociety of America (IDSA) and the AmericanSociety for Microbiology (ASM). Clin Infect Dis57(4):e22–e121. https://doi.org/10.1093/cid/cit278

90. Munson EL, Diekema DJ, Beekmann SE, ChapinKC, Doern GV (2003) Detection and treatmentof bloodstream infection: laboratory reportingand antimicrobial management. J Clin Microbiol41(1):495–497. https://doi.org/10.1128/JCM.41.1.495-497.2003

S58 Der Anaesthesist · Suppl 1 · 2019

91. Brandt C, Braun SD, Stein C, Slickers P, Ehricht R,PletzMWet al (2017) In silico serine β-lactamasesanalysis reveals a huge potential resistome inenvironmental and pathogenic species. Sci Rep.https://doi.org/10.1038/srep43232

92. Pletz MW, Wellinghausen N, Welte T (2011)Will polymerase chain reaction (PCR)-baseddiagnostics improveoutcome in septicpatients?Aclinical view. IntensiveCareMed37(7):1069–1076.https://doi.org/10.1007/s00134-011-2245-x

93. Skvarc M, Stubljar D, Rogina P, Kaasch AJ(2013) Non-culture-based methods to diagnosebloodstream infection: does it work? Eur JMicrobiol Immunol (Bp) 3(2):97–104. https://doi.org/10.1556/EuJMI.3.2013.2.2

94. BloosF, SachseS,KortgenA,PletzMW,LehmannM,Straube E et al (2012) Evaluation of a polymerasechain reaction assay for pathogen detectionin septic patients under routine condition: anobservational study. PLOS ONE 7(9):e46003.https://doi.org/10.1371/journal.pone.0046003

95. Hebart H, Klingspor L, Klingebiel T, Loeffler J,Tollemar J, Ljungman P et al (2009) A prospectiverandomized controlled trial comparing PCR-based and empirical treatment with liposomalamphotericin B in patients after allo-SCT. BoneMarrow Transplant 43(7):553–561. https://doi.org/10.1038/bmt.2008.355

96. Bhat BV, PrasadP, Kumar VBR, Harish B, Krishnaku-mari K, Rekha A et al (2016) Syndrome EvaluationSystem (SES) versus BloodCulture (BACTEC) in thediagnosis andmanagementofneonatal sepsis—arandomized controlled trial. Indian J Pediatr83(5):370–379. https://doi.org/10.1007/s12098-015-1956-3

97. Vincent JL, Brealey D, Libert N, Abidi NE, O’DwyerM, Zacharowski K et al (2015) Rapid diagnosis ofinfection in the critically ill, a multicenter studyofmolecular detection in bloodstream infections,pneumonia, and sterile site infections. Crit CareMed 43(11):2283–2291. https://doi.org/10.1097/CCM.0000000000001249

98. Grumaz S, Stevens P, Grumaz C, Decker SO,WeigandMA,Hofer S et al (2016)Next-generationsequencing diagnostics of bacteremia in septicpatients. GenomeMed8(1):73. https://doi.org/10.1186/s13073-016-0326-8

99. Retamar P, López-Cerero L, Muniain MA, PascualÁ, Rodríguez-Baño J (2013) Impact of the MIC ofpiperacillin-tazobactam on the outcome of pa-tientswithbacteremiaduetoextended-spectrum-β-lactamase-producing escherichia coli. Antimi-crobAgentsChemother57(7):3402–3404. https://doi.org/10.1128/AAC.00135-13

100. Retamar P, Portillo MM, López-Prieto MD, Ro-dríguez-LópezF,deCuetoM,GarcíaMVetal (2012)Impact of inadequate empirical therapy on themortality of patientswith bloodstream infections:a propensity score-based analysis. AntimicrobAgents Chemother 56(1):472–478. https://doi.org/10.1128/AAC.00462-11

101. Cohen J, Vincent J-L, Adhikari NK, MachadoFR, Angus DC, Calandra T et al (2015) Sepsis:a roadmap for future research. Lancet Infect Dis15(5):581–614. https://doi.org/10.1016/S1473-3099(15)70112-X

102. VincentJ-L,BassettiM,FrançoisB,KaramG,ChastreJ, Torres A et al (2016) Advances in antibiotictherapy in the critically ill. Crit Care 20(133):1.https://doi.org/10.1186/s13054-016-1285-6

103. Ehrlich P (1913) Chemotherapeutics: scien-tific principles, methods and results. Lancet2(445):353–359

104. Kollef MH, Sherman G, Ward S, Fraser VJ (1999)Inadequate antimicrobial treatment of infections:a risk factor for hospital mortality among criticallyill patients. Chest115(2):462–474. https://doi.org/10.1378/chest.115.2.462

105. Kumar A, Kethireddy S (2013) Emerging conceptsin optimizing antimicrobial therapy of septicshock: speed is life but a hammer helps too. CritCare17(1):104. https://doi.org/10.1186/cc11890

106. Bloos F, Rüddel H, Thomas-Rüddel D, SchwarzkopfD, Pausch C, Harbarth S et al (2017) Effect ofa multifaceted educational intervention for anti-infectious measures on sepsismortality: a clusterrandomized trial. Intensive CareMed. https://doi.org/10.1007/s00134-017-4782-4

107. Scheer CS, Fuchs C, Kuhn S-O, VollmerM, RehbergS, Friesecke S et al (2017) Quality improvementinitiative for severesepsisandsepticshockreduces90-day mortality: a 7.5-year observational study.Crit Care Med 45(2):241–252. https://doi.org/10.1097/CCM.0000000000002069

108. Ibrahim EH, Sherman G, Ward S, Fraser VJ,Kollef MH (2000) The influence of inadequateantimicrobial treatmentofbloodstreaminfectionson patient outcomes in the ICU setting. Chest118(1):146–155. https://doi.org/10.1378/chest.118.1.146

109. Spanish Collaborative Group for Infections inIntensive Care Units of Sociedad Española deMedicina Intensiva, Crítica yUnidades Coronarias,Vallés J, Rello J, Ochagavía A, Garnacho J, AlcaláM (2003) Community-acquired bloodstreaminfection in critically ill adult patients: Impactof shock and inappropriate antibiotic therapy onsurvival. Chest123(5):1615–1624

110. Society AT, IDSo A (2005) Guidelines for themanagement of adults with hospital-acquired,ventilator-associated, and healthcare-associ-ated pneumonia. Am J Respir Crit Care Med171(4):388–416. https://doi.org/10.1164/rccm.200405-644ST

111. Bowers DR, Liew Y-X, Lye DC, Kwa AL, HsuL-Y, Tam VH (2013) Outcomes of appropriateempiric combination versus monotherapy forpseudomonasaeruginosabacteremia. AntimicrobAgents Chemother 57(3):1270–1274. https://doi.org/10.1128/AAC.02235-12

112. Bagshaw SM, Lapinsky S, Dial S, Arabi Y, Dodek P,Wood G et al (2009) Acute kidney injury in septicshock: clinical outcomes and impact of durationof hypotension prior to initiation of antimicrobialtherapy. Intensive Care Med 35(5):871–881.https://doi.org/10.1007/s00134-008-1367-2

113. Iscimen R, Yilmaz M, Cartin-Ceba R, Hubmayr R,Afessa B, Gajic O et al (2008) Risk factors for thedevelopment of acute lung injury in patients withseptic shock: an observational cohort study. CritCare. https://doi.org/10.1186/cc6708

114. Garnacho-Montero J, Aldabo-Pallas T, Garnacho-Montero C, Cayuela A, Jiménez R, Barroso S et al(2006) Timing of adequate antibiotic therapy isagreaterdeterminantofoutcomethanareTNFandIL-10 polymorphisms in patients with sepsis. CritCare10(4):R111. https://doi.org/10.1186/cc4995

115. Kumar A, Roberts D, Wood KE, Light B, Parrillo JE,Sharma S et al (2006) Duration of hypotensionbefore initiation of effective antimicrobial therapyis the critical determinant of survival in humansepticshock. CritCareMed34(6):1589–1596

116. KumarA,EllisP,ArabiY,RobertsD,LightB,ParrilloJEetal(2009)Initiationofinappropriateantimicrobialtherapy results in a fivefold reduction of survivalin human septic shock. Chest 136(5):1237–1248.https://doi.org/10.1378/chest.09-0087

117. Gaieski DF, Mikkelsen ME, Band RA, Pines JM,Massone R, Furia FF et al (2010) Impact oftime to antibiotics on survival in patients withsevere sepsis or septic shock in whom early goal-directed therapy was initiated in the emergencydepartment. Crit Care Med 38(4):1045–1053.https://doi.org/10.1097/CCM.0b013e3181cc4824

118. BariePS,HydoLJ,ShouJ,LaroneDH,EachempatiSR(2005) Influenceof antibiotic therapyonmortalityof critical surgical illness caused or complicatedby infection. Surg Infect (Larchmt) 6(1):41–54.https://doi.org/10.1089/sur.2005.6.41

119. BarochiaAV,CuiX,VitbergD,SuffrediniAF,O’GradyNP, Banks SM et al (2010) Bundled care for septicshock: an analysis of clinical trials. Crit Care Med38(2):668–678. https://doi.org/10.1097/CCM.0b013e3181cb0ddf

120. Ferrer R, Artigas A, Suarez D, Palencia E, LevyMM,ArenzanaAetal (2009)Effectivenessof treatmentsfor severe sepsis: a prospective, multicenter,observational study. Am J Respir Crit Care Med180(9):861–866. https://doi.org/10.1164/rccm.200812-1912OC

121. Singer M (2017) Antibiotics for Sepsis—DoesEach Hour Really Count? Or is it IncestuousAmplification? Am J Respir Crit CareMed. https://doi.org/10.1164/rccm.201703-0621ED

122. Pletz M, Tacconelli E, Welte T (2017) AntibioticStewardship 2.0. Internist (Berl) 58(7):657–665.https://doi.org/10.1007/s00108-017-0258-4

123. Schuts EC, Hulscher ME, Mouton JW, Verduin CM,Stuart JWC, Overdiek HW et al (2016) Currentevidence on hospital antimicrobial stewardshipobjectives: a systematic reviewandmeta-analysis.Lancet Infect Dis 16(7):847–856. https://doi.org/10.1016/S1473-3099(16)00065-7

124. Allerberger F, Amann S, Apfalter P, Brodt H-R,Eckmanns T, Fellhauer M et al (2016) Strategiesto enhance rational use of antibiotics in hospital:a guideline by the German Society for InfectiousDiseases. Infection 44(3):395–439. https://doi.org/10.1007/s15010-016-0885-z

125. Sandiumenge A, Diaz E, Bodí M, Rello J (2003)Therapy of ventilator-associated pneumonia.Intensive Care Med 29(6):876–883. https://doi.org/10.1007/s00134-003-1715-1

126. Garcin F, LeoneM, Antonini F, Charvet A, AlbaneseJ, Martin C (2010) Non-adherence to guidelines:an avoidable cause of failure of empiricalantimicrobial therapy in the presence of difficult-to-treat bacteria. Intensive CareMed36(1):75–82.https://doi.org/10.1007/s00134-009-1660-8

127. Kumar A, Safdar N, Kethireddy S, Chateau D(2010)A survival benefitof combinationantibiotictherapy for serious infections associated withsepsis and septic shock is contingent only on therisk of death: a meta-analytic/meta-regressionstudy. Crit CareMed38(8):1651–1664. https://doi.org/10.1097/CCM.0b013e3181e96b91

128. Kumar A, Zarychanski R, Light B, Parrillo J, Maki D,Simon D et al (2010) Early combination antibiotictherapy yields improved survival compared withmonotherapy in septic shock: a propensity-matchedanalysis. Crit CareMed38(9):1773–1785.https://doi.org/10.1097/CCM.0b013e3181eb3ccd

129. Al-Hasan MN, Wilson JW, Lahr BD, Thomsen KM,Eckel-PassowJE,VetterEAetal(2009)β-lactamandfluoroquinolone combination antibiotic therapyfor bacteremia caused by gram-negative bacilli.Antimicrob Agents Chemother 53(4):1386–1394.https://doi.org/10.1128/AAC.01231-08

130. Delannoy P-Y, Boussekey N, Devos P, AlfandariS, Turbelin C, Chiche A et al (2012) Impact ofcombinationtherapywithaminoglycosidesonthe

Der Anaesthesist · Suppl 1 · 2019 S59

Leitthema

outcome of ICU-acquired bacteraemias. Eur J ClinMicrobiol Infect Dis 31(9):2293–2299. https://doi.org/10.1007/s10096-012-1568-z

131. Díaz-Martín A, Martínez-González ML, Ferrer R,Ortiz-Leyba C, Piacentini E, Lopez-Pueyo MJ etal (2012) Antibiotic prescription patterns in theempiric therapy of severe sepsis: combinationof antimicrobials with different mechanisms ofaction reduces mortality. Crit Care 16(6):R223.https://doi.org/10.1186/cc11869

132. Martin-Loeches I, Lisboa T, Rodriguez A, PutensenC, Annane D, Garnacho-Montero J et al (2010)Combination antibiotic therapy with macrolidesimproves survival in intubated patients withcommunity-acquired pneumonia. Intensive CareMed 36(4):612–620. https://doi.org/10.1007/s00134-009-1730-y

133. BrunkhorstFM,OppertM,MarxG,BloosF, LudewigK, Putensen C et al (2012) Effect of empiricaltreatment with moxifloxacin and meropenem vsmeropenem on sepsis-related organ dysfunctionin patients with severe sepsis: a randomized trial.JAMA 307(22):2390–2399. https://doi.org/10.1001/jama.2012.5833

134. Braykov NP, Morgan DJ, Schweizer ML, Uslan DZ,Kelesidis T,WeisenbergSAetal (2014)Assessmentof empirical antibiotic therapy optimisation in sixhospitals: an observational cohort study. LancetInfect Dis 14(12):1220–1227. https://doi.org/10.1016/S1473-3099(14)70952-1

135. Bender JK, Fleige C, Klare I, Fiedler S, MischnikA, Mutters NT et al (2016) Detection of a cfr (B)variant in German enterococcus faecium clinicalisolates and the impact on linezolid resistancein enterococcus spp. PLOS ONE 11(11):e167042.https://doi.org/10.1371/journal.pone.0167042

136. HeylandDK, Johnson AP, Reynolds SC, MuscedereJ (2011) Procalcitonin for reduced antibioticexposure in the critical care setting: a systematicreviewandaneconomicevaluation. Crit CareMed39(7):1792–1799. https://doi.org/10.1097/CCM.0b013e31821201a5

137. Bouadma L, Luyt C-E, Tubach F, Cracco C,Alvarez A, Schwebel C et al (2010) Use ofprocalcitonin to reduce patients’ exposure toantibiotics in intensive care units (PRORATA trial):amulticentre randomised controlled trial. Lancet375(9713):463–474. https://doi.org/10.1016/S0140-6736(09)61879-1

138. de Jong E, van Oers JA, Beishuizen A, Vos P,Vermeijden WJ, Haas LE et al (2016) Efficacy andsafety of procalcitonin guidance in reducing theduration of antibiotic treatment in critically illpatients: a randomised, controlled, open-labeltrial. Lancet Infect Dis 16(7):819–827. https://doi.org/10.1016/S1473-3099(16)00053-0

139. Schuetz P, Muller B, Christ-CrainM, Stolz D, TammM, Bouadma L et al (2013) Procalcitonin to initiateordiscontinueantibiotics in acute respiratory tractinfections. Evid Based Child Health. https://doi.org/10.1002/14651858.CD007498

140. SchuetzP,BrielM,Christ-CrainM,StolzD,BouadmaL, Wolff M et al (2012) Procalcitonin to guideinitiation and duration of antibiotic treatment inacute respiratory infections: an individual patientdatameta-analysis. Clin Infect Dis 55(5):651–662.https://doi.org/10.1093/cid/cis464

141. Schuetz P, Birkhahn R, Sherwin R, Jones AE, SingerA, Kline JA et al (2017) Serial Procalcitonin predictsmortality in severe sepsis patients: results fromthe Multicenter Procalcitonin MOnitoring SEpsis(MOSES) study. Crit Care Med 45(5):781–789.https://doi.org/10.1097/CCM.0000000000002321

142. Andriolo BN, Andriolo RB, Salomão R, Atallah ÁN(2017) Effectiveness and safety of procalcitoninevaluation for reducing mortality in adults withsepsis, severesepsisor septicshock. CochraneLibr.https://doi.org/10.1002/14651858.CD010959.pub2

143. Westwood M, Ramaekers B, Whiting P, Tomini F,Joore M, Armstrong N et al (2015) Procalcitonintesting to guide antibiotic therapy for thetreatment of sepsis in intensive care settings andfor suspected bacterial infection in emergencydepartment settings: a systematic review andcost-effectiveness analysis. Health Technol Assess19(96):1–236. https://doi.org/10.3310/hta19960

144. Matthaiou DK, Ntani G, Kontogiorgi M, PoulakouG, Armaganidis A, Dimopoulos G (2012) An ESICMsystematic review and meta-analysis of procal-citonin-guided antibiotic therapy algorithms inadult critically ill patients. Intensive Care Med38(6):940–949. https://doi.org/10.1007/s00134-012-2563-7

145. Gomes Silva BN, Andriolo RB, AtallahÁN, SalomãoR (2010) De-escalation of antimicrobial treatmentforadultswithsepsis, severesepsisor septicshock.CochraneLibr. https://doi.org/10.1002/14651858.CD007934.pub2

146. Leone M, Bechis C, Baumstarck K, Lefrant JY,Albanese J, Jaber S et al (2014) De-escalationversus continuation of empirical antimicrobialtreatment in severe sepsis: a multicenter non-blinded randomizednoninferiority trial. IntensiveCare Med 40(10):1399–1408. https://doi.org/10.1007/s00134-014-3411-8

147. Turza KC, Politano AD, Rosenberger LH, RiccioLM, McLeod M, Sawyer RG (2016) De-escalationof antibiotics does not increase mortality incritically ill surgical patients. Surg Infect (Larchmt)17(1):48–52. https://doi.org/10.1089/sur.2014.202

148. Havey TC, Fowler RA, DanemanN (2011) Durationof antibiotic therapy for bacteremia: a systematicreview and meta-analysis. Crit Care 15(6):R267.https://doi.org/10.1186/cc10545

149. Sawyer RG, Claridge JA, Nathens AB, Rotstein OD,Duane TM, Evans HL et al (2015) Trial of short-course antimicrobial therapy for intraabdominalinfection. N Engl J Med 372(21):1996–2005.https://doi.org/10.1056/NEJMoa1411162

150. Chastre J, Wolff M, Fagon J-Y, Chevret S, ThomasF, Wermert D et al (2003) Comparison of 8 vs 15daysof antibiotic therapy for ventilator-associatedpneumonia in adults: a randomized trial. JAMA290(19):2588–2598. https://doi.org/10.1001/jama.290.19.2588

151. Garnacho-Montero J, Gutiérrez-Pizarraya A,Escoresca-Ortega A, Corcia-Palomo Y, Fernández-Delgado E, Herrera-Melero I et al (2014) De-escalation of empirical therapy is associated withlowermortality in patients with severe sepsis andseptic shock. Intensive Care Med 40(1):32–40.https://doi.org/10.1007/s00134-013-3077-7

152. Weiss CH, Persell SD, Wunderink RG, Baker DW(2012) Empiric antibiotic, mechanical ventilation,and central venous catheter duration as potentialfactorsmediating the effect of a checklist prompt-ing intervention on mortality: an exploratoryanalysis. BmcHealth Serv Res. https://doi.org/10.1186/1472-6963-12-198

153. Bodmann K, Grabein B (2010) Empfehlungenzur kalkulierten parenteralen InitialtherapiebakteriellerErkrankungenbeiErwachsenen

154. Sartelli M, Weber DG, Ruppé E, Bassetti M,Wright BJ, Ansaloni L et al (2016) Antimicrobials:a global alliance for optimizing their rational use

in intra-abdominal infections (AGORA). World JEmergSurg. https://doi.org/10.1186/s13017-016-0089-y

155. Brook I (1989) Inoculum effect. Reviews ofinfectious diseases. Clin Infect Dis 11(3):361–368.https://doi.org/10.1093/clinids/11.3.361

156. Ruppé É, Woerther P-L, Barbier F (2015) Mecha-nismsofantimicrobial resistance ingram-negativebacilli. Ann Intensive Care. https://doi.org/10.1186/s13613-015-0061-0

157. Perez F, Bonomo RA (2012) Can we really useß-lactam/ß-lactam inhibitor combinations forthe treatment of infections caused by extended-spectrumß-lactamase—producingbacteria? ClinInfectDis 54(2):175–177. https://doi.org/10.1093/cid/cir793

158. Hanrahan T, Whitehouse T, Lipman J, Roberts JA(2015) Vancomycin-associated nephrotoxicity: ameta-analysis of administration by continuousversus intermittent infusion. Int J AntimicrobAgents 46(3):249–253. https://doi.org/10.1016/j.ijantimicag.2015.04.013

159. Brinkmann A, Röhr A, Köberer A, Fuchs T, Preisen-berger J, Krüger W et al (2016) TherapeutischesDrug Monitoring und individualisierteDosierungvon Antibiotika bei der Sepsis. Med Klin Not-fallmed. https://doi.org/10.1007/s00063-016-0213-5

160. Benvenuto M, Benziger DP, Yankelev S, ViglianiG (2006) Pharmacokinetics and tolerability ofdaptomycin at doses up to 12 milligrams perkilogram of body weight once daily in healthyvolunteers. Antimicrob Agents Chemother50(10):3245–3249. https://doi.org/10.1128/AAC.00247-06

161. Schweizer ML, Furuno JP, Harris AD, JohnsonJK, Shardell MD, McGregor JC et al (2011)Comparative effectiveness of nafcillin or cefazolinversus vancomycin in methicillin-susceptiblestaphylococcusaureusbacteremia. BMCInfectDis.https://doi.org/10.1186/1471-2334-11-279

162. Nissen JL, Skov R, Knudsen JD, Østergaard C,Schønheyder HC, Frimodt-Møller N et al (2013)Effectiveness of penicillin, dicloxacillin and ce-furoximeforpenicillin-susceptible staphylococcusaureus bacteraemia: a retrospective, propensity-score-adjusted case—control and cohort analy-sis. J Antimicrob Chemother 68(8):1894–1900.https://doi.org/10.1093/jac/dkt108

163. Liu C, Bayer A, Cosgrove SE, Daum RS, FridkinSK, Gorwitz RJ et al (2011) Clinical practiceguidelines by the Infectious Diseases Society ofAmerica for the treatment of methicillin-resistantstaphylococcus aureus infections in adults andchildren. Clin InfectDis52(3):e18–e55. https://doi.org/10.1093/cid/ciq146

164. JongsmaK, Joson J, Heidari A (2013) Ceftaroline inthe treatmentofconcomitantmethicillin-resistantand daptomycin-non-susceptible staphylococcusaureus infective endocarditis and osteomyeli-tis: case report. J Antimicrob Chemother68(6):1444–1445. https://doi.org/10.1093/jac/dkt009

165. Tattevin P, Boutoille D, Vitrat V, Van GrunderbeeckN, Revest M, Dupont M et al (2014) Salvagetreatment of methicillin-resistant staphylococcalendocarditis with ceftaroline: a multicentreobservational study. J Antimicrob Chemother69(7):2010–2013. https://doi.org/10.1093/jac/dku085

166. Raad I, DarouicheR, Vazquez J, LentnekA,HachemR, Hanna H et al (2005) Efficacy and safety ofweekly dalbavancin therapy for catheter-relatedbloodstream infection caused by gram-positive

S60 Der Anaesthesist · Suppl 1 · 2019

pathogens. Clin InfectDis 40(3):374–380. https://doi.org/10.1086/427283

167. Roberts KD, Sulaiman RM, Dalbavancin Orita-vancin RMJ (2015) An innovative approach to thetreatment of gram-positive infections. Pharma-cotherapy 35(10):935–948. https://doi.org/10.1002/phar.1641

168. López-Cortés LE, del Toro MD, Gálvez-Acebal J,Bereciartua-Bastarrica E, FariñasMC, Sanz-FrancoMetal (2013) Impactof anevidence-basedbundleintervention in the quality-of-care managementand outcome of staphylococcus aureus bac-teremia. Clin Infect Dis 57(9):1225–1233. https://doi.org/10.1093/cid/cit499

169. Habib G, Lancellotti P, Antunes MJ, BongiorniMG, Casalta J-P, Del Zotti F et al (2015) 2015Esc Guidelines for the management of infectiveendocarditis: The Task Force for theManagementof Infective Endocarditis of the European Societyof Cardiology (esc). Endorsed by: EuropeanAssociation forCardio-thoracicSurgery (eacts), theEuropeanAssociationofNuclearMedicine (eanm).Eur Heart J 36(44):3075–3128. https://doi.org/10.1093/eurheartj/ehv319

170. Kalil AC, Metersky ML, Klompas M, Muscedere J,Sweeney DA, Palmer LB et al (2016)Managementof adults with hospital-acquired and ventilator-associated pneumonia: 2016 clinical practiceguidelines by the Infectious Diseases Society ofAmerica and the American Thoracic Society. ClinInfect Dis 63(5):e61–e111. https://doi.org/10.1093/cid/ciw353

171. Magill SS, Edwards JR, Bamberg W, Beldavs ZG,DumyatiG,KainerMAetal (2014)Multistatepoint-prevalence survey of health care—associatedinfections. N Engl J Med 370(13):1198–1208.https://doi.org/10.1056/NEJMoa1306801

172. MelsenWG,RoversMM,GroenwoldRH, BergmansDC, Camus C, Bauer TT et al (2013) Attributablemortality of ventilator-associated pneumonia:a meta-analysis of individual patient data fromrandomised prevention studies. Lancet Infect Dis13(8):665–671. https://doi.org/10.1016/S1473-3099(13)70081-1

173. KollefMH, HamiltonCW, Ernst FR (2012) Economicimpact of ventilator-associated pneumonia ina large matched cohort. Infect Control HospEpidemiol 33(03):250–256. https://doi.org/10.1086/664049

174. Muscedere JG,DayA,HeylandDK (2010)Mortality,attributable mortality, and clinical events as endpoints for clinical trials of ventilator-associatedpneumonia and hospital-acquired pneumonia.Clin Infect Dis 51(Supplement 1):S120–S125.https://doi.org/10.1086/653060

175. Sopena N, Sabrià M (2005) Multicenter study ofhospital-acquiredpneumonia innon-ICUpatients.Chest 127(1):213–219. https://doi.org/10.1378/chest.127.1.213

176. Esperatti M, Ferrer M, Giunta V, Ranzani OT,Saucedo LM, Bassi GL et al (2013) Validationof predictors of adverse outcomes in hospital-acquired pneumonia in the ICU. Crit Care Med41(9):2151–2161. https://doi.org/10.1097/CCM.0b013e31828a674a

177. Dalhoff K, Abele-Horn M, Andreas S, Bauer T,von Baum H, Deja M et al (2012) Epidemiologie,Diagnostik und Therapie erwachsener Patientenmit nosokomialer Pneumonie. Pneumologie66(12):707–765. https://doi.org/10.1055/s-0032-1325924

178. Dalhoff K, Abele-Horn M, Andreas S, Bauer T,von Baum H, Deja M et al (2012) Epidemiology,diagnosis and treatment of adult patients with

nosocomial pneumonia. S-3 Guideline of theGermanSociety forAnaesthesiologyand IntensiveCare Medicine, the German Society for InfectiousDiseases, the German Society for Hygiene andMicrobiology, the German Respiratory Societyand the Paul-Ehrlich-Society for Chemotherapy.Pneumologie 66(12):707–765. https://doi.org/10.1055/s-0032-1325924

179. Chavanet P (2013) The ZEPHyR study: a ran-domized comparison of linezolid and vancomycinfor MRSA pneumonia. Med Mal Infect 43(11-12):451–455. https://doi.org/10.1016/j.medmal.2013.09.011

180. Liapikou A, Cilloniz C, Torres A (2015) Ceftobiprolefor the treatment of pneumonia: a Europeanperspective. Drug Des Devel Ther 9:4565–4572.https://doi.org/10.2147/DDDT.S56616

181. PughR,GrantC,CookeRP,DempseyG(2011)Short-courseversusprolonged-courseantibiotic therapyfor hospital-acquired pneumonia in critically illadults. Cochrane Libr. https://doi.org/10.1002/14651858.CD007577.pub2

182. DimopoulosG,PoulakouG,Pneumatikos IA,Arma-ganidisA,KollefMH,MatthaiouDK(2013)Short- vslong-duration antibiotic regimens for ventilator-associated pneumonia: a systematic review andmeta-analysis. Chest 144(6):1759–1767. https://doi.org/10.1378/chest.13-0076

183. Solomkin JS, Mazuski JE, Bradley JS, Rodvold KA,Goldstein EJ, Baron EJ et al (2010) Diagnosis andmanagement of complicated intra-abdominalinfection in adults and children: guidelines bythe Surgical Infection Society and the InfectiousDiseases SocietyofAmerica. Surg Infect (Larchmt)11(1):79–109. https://doi.org/10.1089/sur.2009.9930

184. Torrens G, Cabot G, Ocampo-Sosa AA, ConejoMC, Zamorano L, Navarro F et al (2016) Activityof ceftazidime-avibactam against clinical andisogenic laboratory pseudomonas aeruginosaisolates expressing combinations ofmost relevantβ-lactam resistance mechanisms. AntimicrobAgentsChemother60(10):6407–6410. https://doi.org/10.1128/AAC.01282-16

185. Mazuski JE, Gasink LB, Armstrong J, Broadhurst H,StoneGG, RankDetal (2016) Efficacyandsafetyofceftazidime-avibactamplusmetronidazole versusmeropeneminthe treatmentofcomplicated intra-abdominal infection: results from a randomized,controlled, double-blind, phase 3 program. ClinInfect Dis 62(11):1380–1389. https://doi.org/10.1093/cid/ciw133

186. Solomkin J, Hershberger E, Miller B, Popejoy M,Friedland I, SteenbergenJetal (2015)Ceftolozane/tazobactam plus metronidazole for complicatedintra-abdominal infections in an era of multidrugresistance: results from a randomized, double-blind, phase 3 trial (ASPECT-cIAI). Clin Infect Dis60(10):1462–1471. https://doi.org/10.1093/cid/civ097

187. Schöfer H et al (2011) Diagnostik und TherapieStaphylococcus aureus bedingter Infektionen derHaut und Schleimhäute. J Deutsch Dermatol Ges9(11):953–968

188. Rhodes NJ, MacVane SH, Kuti JL, Scheetz MH(2014) Impact of loading doses on the time toadequate predicted beta-lactam concentrationsin prolonged and continuous infusion dosingschemes. Clin Infect Dis 59(6):905–907. https://doi.org/10.1093/cid/ciu402

189. CraigWA(1998)Pharmacokinetic/pharmacodynamicparameters: rationale for antibacterial dosing ofmice andmen. Clin Infect Dis 26(1):1–10. https://doi.org/10.1086/516284

190. Roberts JA, Abdul-Aziz MH, Lipman J, Mouton JW,Vinks AA, Felton TW et al (2014) Individualisedantibiotic dosing for patients who are criticallyill: challenges and potential solutions. LancetInfectDis 14(6):498–509. https://doi.org/10.1016/S1473-3099(14)70036-2

191. Pletz MW, Bloos F, Burkhardt O, Brunkhorst FM,Bode-BogerSM,Martens-LobenhofferJetal(2010)Pharmacokinetics ofmoxifloxacin in patientswithsevere sepsis or septic shock. Intensive Care Med36(6):979–983. https://doi.org/10.1007/s00134-010-1864-y

192. van Zanten AR, Polderman KH, van GeijlswijkIM, van der Meer GY, Schouten MA, Girbes AR(2008) Ciprofloxacin pharmacokinetics in criticallyill patients: a prospective cohort study. J CritCare 23(3):422–430. https://doi.org/10.1016/j.jcrc.2007.11.011

193. Blot S, Koulenti D, AkovaM, Bassetti M, De WaeleJJ, Dimopoulos G et al (2014) Does contemporaryvancomycin dosing achieve therapeutic targetsin a heterogeneous clinical cohort of criticallyill patients? Data from the multinational DALIstudy. Crit Care 18(3):R99. https://doi.org/10.1186/cc13874

194. MooreRD,SmithCR,LietmanPS(1984)Associationof aminoglycoside plasma levels with therapeuticoutcome in gram-negative pneumonia. AmJMed77(4):657–662. https://doi.org/10.1016/0002-9343(84)90358-9

195. MenP, LiH-B, Zhai S-D, ZhaoR-S (2016)Associationbetween the AUC 0-24/MIC ratio of vancomycinand its clinical effectiveness: a systematic reviewand meta-analysis. PLOS ONE 11(1):e146224.https://doi.org/10.1371/journal.pone.0146224

196. Chelluri L, Jastremski MS (1987) Inadequacy ofstandard aminoglycoside loadingdoses in acutelyillpatients. CritCareMed15(12):1143–1145

197. Kohanski MA, DePristo MA, Collins JJ (2010)Sublethal antibiotic treatment leads tomultidrugresistance via radical-induced mutagenesis. MolCell 37(3):311–320. https://doi.org/10.1016/j.molcel.2010.01.003

198. Drusano GL (2004) Antimicrobial pharmacody-namics: critical interactions of ‘bug and drug’. NatRev Microbiol 2(4):289–300. https://doi.org/10.1038/nrmicro862

199. Olofsson SK, Cars O (2007) Optimizing drugexposure to minimize selection of antibioticresistance. Clin Infect Dis 45(Supplement2):S129–S136. https://doi.org/10.1086/519256

200. UdyAA, Lipman J, Jarrett P, Klein K,Wallis SC, PatelK et al (2015) Are standard doses of piperacillinsufficient for critically ill patients with augmentedcreatinine clearance? CritCare. https://doi.org/10.1186/s13054-015-0750-y

201. Mouton JW, Ambrose PG, Canton R, Drusano GL,Harbarth S, MacGowan A et al (2011) Conservingantibiotics for the future: new ways to use oldand new drugs from a pharmacokinetic andpharmacodynamicperspective. DrugResistUpdat14(2):107–117. https://doi.org/10.1016/j.drup.2011.02.005

202. Roberts JA, Paul SK, Akova M, Bassetti M, DeWaele JJ, Dimopoulos G et al (2014) DALI:defining antibiotic levels in intensive care unitpatients: are current beta-lactam antibiotic dosessufficient for critically ill patients? Clin Infect Dis58(8):1072–1083. https://doi.org/10.1093/cid/ciu027

203. McKinnon PS, Paladino JA, Schentag JJ (2008)Evaluation of area under the inhibitory curve(AUIC) and time above the minimum inhibitoryconcentration (T>MIC) as predictors of outcome

Der Anaesthesist · Suppl 1 · 2019 S61

Leitthema

for cefepime and ceftazidime in serious bacterialinfections. Int JAntimicrobAgents 31(4):345–351.https://doi.org/10.1016/j.ijantimicag.2007.12.009

204. Drusano G, Lodise T, Melnick D, Liu W, OliverA, Mena A et al (2011) Meropenem penetrationinto epithelial lining fluid in mice and humansand delineation of exposure targets. AntimicrobAgents Chemother 55(7):3406–3412. https://doi.org/10.1128/AAC.01559-10

205. Lodise T, Sorgel F, Melnick D, Mason B, Kinzig M,DrusanoG (2011) Penetration ofmeropenem intoepithelial lining fluid of patients with ventilator-associated pneumonia. Antimicrob AgentsChemother 55(4):1606–1610. https://doi.org/10.1128/AAC.01330-10

206. Lodise T, Nau R, Kinzig M, Drusano G, Jones R,Sörgel F (2007)Pharmacodynamicsof ceftazidimeand meropenem in cerebrospinal fluid: resultsof population pharmacokinetic modelling andMonte Carlo simulation. J Antimicrob Chemother60(5):1038–1044. https://doi.org/10.1093/jac/dkm325

207. Lodise TP, Butterfield J (2011) Use of pharmaco-dynamic principles to inform β-lactam dosing:“S” does not always mean success. J Hosp Med6(S1):S16–S23

208. Tam VH, Schilling AN, Neshat S, Poole K, MelnickDA, Coyle EA (2005) Optimization of meropenemminimum concentration/MIC ratio to suppressin vitro resistance of pseudomonas aeruginosa.AntimicrobAgentsChemother49(12):4920–4927.https://doi.org/10.1128/AAC.49.12.4920-4927.2005

209. Taccone FS, Laterre P-F, Dugernier T, Spapen H,Delattre I, Wittebole X et al (2010) Insufficientβ-lactam concentrations in the early phaseof severe sepsis and septic shock. Crit Care14(4):R126. https://doi.org/10.1186/cc9091

210. Udy AA, Varghese JM, Altukroni M, BriscoeS, McWhinney BC, Ungerer JP et al (2012)Subtherapeutic initial beta-lactamconcentrationsin select critically ill patients: associationbetweenaugmented renal clearance and low trough drugconcentrations. Chest 142(1):30–39. https://doi.org/10.1378/chest.11-1671

211. HuttnerA,HarbarthS,HopeWW,LipmanJ,RobertsJA(2015)Therapeuticdrugmonitoringofthebeta-lactamantibiotics: what is theevidenceandwhichpatients should we be using it for? J AntimicrobChemother70(12):3178–3183. https://doi.org/10.1093/jac/dkv201

212. Lodise TP, Lomaestro B, Drusano GL (2007)Piperacillin-tazobactam for Pseudomonas aerug-inosa infection: clinical implications of anextended-infusion dosing strategy. Clin Infect Dis44(3):357–363. https://doi.org/10.1086/510590

213. Lee LS, Kinzig-Schippers M, Nafziger AN, Ma L,Sorgel F, Jones RN et al (2010) Comparison of 30-min and 3-h infusion regimens for imipenem/cilastatin and for meropenem evaluated byMonte Carlo simulation. Diagn Microbiol InfectDis 68(3):251–258. https://doi.org/10.1016/j.diagmicrobio.2010.06.012

214. Abdul-Aziz MH, Lipman J, Akova M, BassettiM, De Waele JJ, Dimopoulos G et al (2016) Isprolongedinfusionofpiperacillin/tazobactamandmeropenemincritically ill patientsassociatedwithimproved pharmacokinetic/pharmacodynamicand patient outcomes? An observation from theDefining Antibiotic Levels in Intensive care unitpatients (DALI) cohort. J Antimicrob Chemother71(1):196–207. https://doi.org/10.1093/jac/dkv288

215. Taubert M, Zander J, Frechen S, Scharf C, Frey L,Vogeser M et al (2017) Optimization of linezolidtherapy in the critically ill: the effect of adjustedinfusion regimens. J Antimicrob Chemother72(8):2304–2310. https://doi.org/10.1093/jac/dkx149

216. Carlier M, Carrette S, Roberts JA, Stove V,Verstraete A, Hoste E et al (2013)Meropenemandpiperacillin/tazobactam prescribing in criticallyill patients: does augmented renal clearanceaffect pharmacokinetic/pharmacodynamic targetattainment when extended infusions are used?Crit Care 17(3):R84. https://doi.org/10.1186/cc12705

217. Patel BM, Paratz J, See NC, Muller MJ, RuddM, Paterson D et al (2012) Therapeutic drugmonitoring of beta-lactam antibiotics in burnspatients—a one-year prospective study. TherDrug Monit 34(2):160–164. https://doi.org/10.1097/FTD.0b013e31824981a6

218. Dulhunty JM, Roberts JA, Davis JS, Webb SA,Bellomo R, Gomersall C et al (2015) Amulticenterrandomizedtrialofcontinuousversus intermittentβ-lactam infusion in severe sepsis. Am JRespir CritCareMed 192(11):1298–1305. https://doi.org/10.1164/rccm.201505-0857OC

219. Frey OR, Köberer A, Röhr AC, PreisenbergerJA, Brinkmann A (2013) Optimale Dosierungund Applikation von Antiinfektiva. intensiv21(05):264–267. https://doi.org/10.1055/s-0033-1355148

220. Hayashi Y, Lipman J, Udy AA, Ng M, McWhinneyB, Ungerer J et al (2013) β-lactam therapeuticdrug monitoring in the critically ill: optimisingdrug exposure in patients with fluctuating renalfunctionandhypoalbuminaemia. Int JAntimicrobAgents 41(2):162–166. https://doi.org/10.1016/j.ijantimicag.2012.10.002

221. Nosseir N, Michels G, Pfister R, Adam R,WiesenM,Müller C (2014) Therapeutisches DrugMonitoring(TDM) von Antiinfektiva in der Intensivmedizin.Dtsch Med Wochenschr 139(38):1889–1894.https://doi.org/10.1055/s-0034-1387215

222. Roberts JA, Norris R, PatersonDL,Martin JH (2012)Therapeutic drugmonitoring of antimicrobials. BrJ Clin Pharmacol 73(1):27–36. https://doi.org/10.1111/j.1365-2125.2011.04080.x

223. WongG, SimeFB, Lipman J, Roberts JA (2014)Howdoweuse therapeuticdrugmonitoringto improveoutcomes from severe infections in critically illpatients? BMC InfectDis. https://doi.org/10.1186/1471-2334-14-288

S62 Der Anaesthesist · Suppl 1 · 2019