UBC Faculty Research and Publications

Designing phase 3 sepsis trials: application of learned experiences from critical care trials in acute… Mebazaa, Alexandre; Laterre, Pierre F; Russell, James A; Bergmann, Andreas; Gattinoni, Luciano; Gayat, Etienne; Harhay, Michael O; Hartmann, Oliver; Hein, Frauke; Kjolbye, Anne L; Legrand, Matthieu; Lewis, Roger J; Marshall, John C; Marx, Gernot; Radermacher, Peter; Schroedter, Mathias; Scigalla, Paul; Stough, Wendy G; Struck, Joachim; Van den Berghe, Greet; Yilmaz, Mehmet B; Angus, Derek C Mar 31, 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-40560_2016_Article_151.pdf [ 675.51kB ]
JSON: 52383-1.0361987.json
JSON-LD: 52383-1.0361987-ld.json
RDF/XML (Pretty): 52383-1.0361987-rdf.xml
RDF/JSON: 52383-1.0361987-rdf.json
Turtle: 52383-1.0361987-turtle.txt
N-Triples: 52383-1.0361987-rdf-ntriples.txt
Original Record: 52383-1.0361987-source.json
Full Text

Full Text

REVIEW Open AccessDesigning phase 3 sepsis trials: applicationof learned experiences from critical caretrials in acute heart failureAlexandre Mebazaa1,2,3*, Pierre François Laterre4, James A. Russell5, Andreas Bergmann6, Luciano Gattinoni7,Etienne Gayat8, Michael O. Harhay9, Oliver Hartmann6, Frauke Hein6, Anne Louise Kjolbye10, Matthieu Legrand11,Roger J. Lewis12, John C. Marshall13, Gernot Marx14, Peter Radermacher15, Mathias Schroedter6, Paul Scigalla6,Wendy Gattis Stough16, Joachim Struck6, Greet Van den Berghe17, Mehmet Birhan Yilmaz18 and Derek C. Angus19,20AbstractSubstantial attention and resources have been directed to improving outcomes of patients with critical illnesses, inparticular sepsis, but all recent clinical trials testing various interventions or strategies have failed to detect a robustbenefit on mortality. Acute heart failure is also a critical illness, and although the underlying etiologies differ, acuteheart failure and sepsis are critical care illnesses that have a high mortality in which clinical trials have been difficult toconduct and have not yielded effective treatments. Both conditions represent a syndrome that is often difficult todefine with a wide variation in patient characteristics, presentation, and standard management across institutions.Referring to past experiences and lessons learned in acute heart failure may be informative and help frame research inthe area of sepsis. Academic heart failure investigators and industry have worked closely with regulators for many yearsto transition acute heart failure trials away from relying on dyspnea assessments and all-cause mortality as the primarymeasures of efficacy, and recent trials have been designed to assess novel clinical composite endpoints assessingorgan dysfunction and mortality while still assessing all-cause mortality as a separate measure of safety. Applying thelessons learned in acute heart failure trials to severe sepsis and septic shock trials might be useful to advance the field.Novel endpoints beyond all-cause mortality should be considered for future sepsis trials.Keywords: Sepsis, Clinical trials as topic, Heart failure, Mortality, Multiple organ failureIntroductionSepsis, defined as "life-threatening organ dysfunction dueto a dysregulated host response to infection" [1], is a majorcause of mortality and morbidity worldwide [2–6]. Litera-ture estimates of sepsis incidence vary widely [7]. One USstudy reported an absolute incidence ranging from 300 to1031 cases per 100,000 population [7, 8]. The annual inci-dence of sepsis globally has been roughly estimated at 15to 19 million [7, 9]. A systematic review of 33 studies ori-ginating in North America, Europe, Asia, and Australiafound a population incidence for hospital-treated sepsis of256 cases per 100,000 person-years [10]. The authors ex-trapolated these findings to estimate a global incidence forsepsis of 30.7 million cases, contributing to an estimated 6million deaths each year [10].Sepsis mortality has declined over the last decadefrom ~40 to ~20 % [11]. Improved processes of care (e.g.,earlier diagnosis; timely resuscitation with appropriatetherapies; low tidal volume during mechanical ventilation)may explain this observation [12–15]. However, neuro-muscular, psychological, metabolic, cardiovascular, andrenal complications persist and lead to impaired long-term outcomes among sepsis survivors [16, 17]. Inaddition, many sepsis patients are elderly and have otherlife-limiting comorbidities. Survival may be less importantto these patients than measures reflecting independenceand quality of life [18]. The long-term outcome and mor-bidity burden of sepsis survivors is an emerging, import-ant research and clinical care concern.* Correspondence: alexandre.mebazaa@lrb.aphp.fr1University Paris Diderot, Sorbonne Paris Cité, Paris, France2U942 Inserm, APHP, Paris, FranceFull list of author information is available at the end of the article© 2016 Mebazaa et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Mebazaa et al. Journal of Intensive Care  (2016) 4:24 DOI 10.1186/s40560-016-0151-6Effective therapies are needed to better manage sepsispatients [19]. Therapeutic goals include not only im-proving survival but also reducing morbidity, preventingorgan failure, and shortening convalescence [2, 20]. Sub-stantial attention has been directed at reducing mortalityin sepsis, but all recent multinational trials have failed toimprove survival [21–26].Other critical care conditions (e.g., acute heart failure)have faced similar challenges with efforts to prolong sur-vival in clinical trials. Acute heart failure and sepsis areboth critical care illnesses with high mortality. Both condi-tions represent syndromes with wide variation in patientcharacteristics, presentation, and standard management.Further, the underlying pathophysiology in both condi-tions is related to many processes, but pharmacologicinterventions generally target single pathways and havenot translated into survival benefits. All-cause mortality isusually the primary endpoint chosen for phase 3 pivotaltrials in sepsis and acute heart failure, but no treatmentsto date have effectively reduced the high mortality associ-ated with either of these conditions (Fig. 1). In a survey ofacute heart failure experts, most felt it was unlikely thatimprovements in short-term mortality could be shown asa single primary endpoint in acute heart failure trials [27].Thus, recent and ongoing acute heart failure trials havebeen designed with composite clinical primary endpoints,reserving all-cause mortality assessments for safety [28].This approach recently adopted in some acute heart fail-ure trials may help frame research in sepsis, since both ofthese critical care illnesses have faced similar challenges inclinical research.The European Drug Development Hub brought to-gether experts in critical care/sepsis and acute heart fail-ure with the objective of sharing the collective clinicalresearch experience in these critical care illnesses. Theultimate goal was to discuss better approaches to con-ducting clinical trials in critical care illnesses with highmortality (i.e., sepsis and acute heart failure) to promoteadvances in the care of these patients (Paris, France,January 2015). This paper summarizes the key develop-ments from the meeting, focusing on clinical trialdesigns and endpoints that should be considered for usein future sepsis trials.ReviewClinical trials in sepsisWhy have outcomes failed to improve in clinical trials?An overview of the results from a selection of recentlarge, rigorously designed and conducted sepsis clinicaltrials reveals a consistent theme (Table 1). All trials weredesigned with short-term all-cause mortality as the pri-mary endpoint, but none of the interventions has im-proved short-term survival for a variety of possiblereasons (Table 2).Mortality rates due to sepsis are declining but remainhigh [5]. Statistical power is dependent on severalparameters, including the population’s baseline risk, themodifiable mortality, and on the treatment effect sizeand its variability within the study sample. In some re-cent sepsis trials, all-cause mortality ranged from 19 to45 % depending on the study population and follow-upduration (Table 1). Achieving lower than expected eventFig. 1 In-hospital mortality rates for septicemia, respiratory failure, and acute heart failure. Acute coronary syndrome included as an example of acritical care cardiovascular condition where reductions in in-hospital mortality have been realized. Rates are per 100 discharges for acute coronarysyndrome, septicemia, and respiratory failure and were extracted from National Hospital Discharge Survey [66–68]. Rates for acute heart failurewere based on published registry data [69] and represent percent of patients in the registries who died in the hospital. Data shown are fromADHERE [70] and OPTIMIZE [71] (2000), EHFS II (2004) [72], ALARM (2007) [73], AHEAD (2010) [74], and ATTEND (2011) [75]. The acute heart failuredata should be interpreted considering the differences in registry populations and severity of illnessMebazaa et al. Journal of Intensive Care  (2016) 4:24 Page 2 of 11Table 1 Overview of key recent critical care sepsis trialsTrial Design Intervention Study population Mean SOFAfscoreEndpoint Length offollow-upN of deaths Primary endpoint resultsALBIOS [22] Multicenter,open-label,randomized,controlled20 % albumin + crystalloidvs. crystalloid alone for 28days or until ICU dischargeN = 1818≥18 years,clinical criteria forsevere sepsis [76]Albumin 8 (6–10)vs. crystalloid 8(5–10)a, median(interquartilerange)All-cause mortality 28 days 285 albumin vs.288 crystalloid31.8 % albumin vs. 32 %crystalloid (RR 1.00, 95 %CI 0.87–1.14, P = 0.94)SEPSISPAM [21] Multicenter,open-label,randomizedVasopressor treatmentadjusted to maintain MAPof 80–85 mmHg (hightarget) vs. 65–70 mmHg(low target) for 5 days oruntil vasopressor supportweanedN = 776Septic shock(Table 2) refractoryto fluid resuscitation,requiring vasopressorsLow target 10.8± 3.1 vs. hightarget 10.7 ± 3.1bAll-cause mortality 28 days 142 high target vs.132 low target36.6 % high target vs.34 % low target (HR forhigh target 1.07, 95 % CI0.84–1.38, P = 0.57)ProCESS [26] Multicenter,randomizedProtocol-based EGDT vs.protocol-based standardtherapy vs. usual careN = 1341Suspectedsepsis with ≥2 criteriafor SIRS, [76] andrefractory hypotensionor serum lactate≥4 mmol/LNot reported All-cause in-hospitaldeath60 days 92 EGDT vs. 81standard therapy vs.86 usual care21 % EGDT vs. 18.2 %standard therapy vs.18.9 % usual careCombined protocol-based groups vs. usualcare RR 1.04, 95 %CI 0.82–1.31, P = 0.83Rosuvastatin forARDSe [25]Multicenter,randomized,placebo-controlled,double-blindEnteral rosuvastatin vs.placeboN = 745Positive pressuremechanical ventilation,PaO2 to FIO2 ratio ≤300,bilateral infiltrates onCXR without evidenceof left atrial hypertension,known or suspectedinfection, and ≥1 criteriafor SIRS (Table 2)Not reported All-cause mortalitybefore hospitaldischarge home oruntil study day 6060 days 108 rosuvastatin vs.91 placebo28.5 % rosuvastatin vs.24.9 % placebo;difference 4.0 (−2.3 to10.2), P = 0.21; enrollmentstopped prematurely forfutilityTRISS [23] Multicenter,randomized,parallel-groupLeuko-reduced bloodtransfusion at lower (≤7g/dL) vs. higher (≤9 g/dL)Hgb thresholdsN = 1000ICU, fulfilledseptic shock criteria(Table 2), Hgb ≤9 g/dLBoth groups 10(8-12)c, median(interquartilerange)All-cause mortality 90 days 216 lower Hgb vs.223 higher Hgb43 % lower threshold vs.45 % higher threshold(RR 0.94, 95 % CI 0.78 to1.09, P = 0.44)ARISE [24] Multicenter,randomized,parallel-groupEGDT vs. usual care for 6 h N = 1600Suspected orconfirmed infection, ≥2criteria for SIRS (Table 2),refractory hypotension orhypoperfusion, identifiedin the ED within 6 h ofpresentationNot reported All-cause mortality 90 days 147 EGDT vs. 150usual care18.6 % EGDT vs. 18.8 %usual care (RR 0.98, 95 %CI 0.80 to 1.21, P = 0.9)Mebazaaetal.JournalofIntensiveCare (2016) 4:24 Page3of11Table 1 Overview of key recent critical care sepsis trials (Continued)PROMISE [31] Pragmatic,open, multicenter,parallel-group,randomized,controlled trial6-h EGDT resuscitationprotocol vs. usual careN = 1260Known orpresumed infection,≥2 SIRS criteria,and either refractoryhypotension orhyperlactatemiawithin 6 h after EDpresentationEGDT 4.2 ± 2.4vs. usual care4.3 ± 2.4dAll-cause mortality 90 days 184 EGDT vs. 181usual care29.5 % EGDT vs. 29.2 %usual care (RR 1.01, 95 %CI 0.85 to 1.20, P = 0.9)ICU intensive care unit, MAP mean arterial pressure, EGDT early goal-directed therapy, SIRS systemic inflammatory response syndrome, CXR chest radiography, Hgb hemoglobin, ED emergency departmentaIncludes subscores ranging from 0 to 4 for each of five components (respiratory, coagulation, liver, cardiovascular, and renal components), with higher scores indicating more severe organ dysfunction. The scoringwas modified by excluding the assessment of cerebral failure (the Glasgow Coma Scale), which was not performed in these patients, and by decreasing to 65 mmHg the mean arterial pressure threshold for acardiovascular subscore of 1, for consistency with the hemodynamic targets as defined according to the early goal-directed therapybIncludes subscores ranging from 0 to 4 for each of five components (circulation, lungs, liver, kidneys, and coagulation). Aggregated scores range from 0 to 20, with higher scores indicating more severe organ failurecSubscores ranging from 0 to 4 for each of six organ systems (cerebral, circulation, pulmonary, hepatic, renal, and coagulation). The aggregated score ranges from 0 to 24, with higher scores indicating more severeorgan failure. One variable was missing for 51 patients in the higher-threshold group and for 64 in the lower-threshold group, so their values were not includeddScores range from 0 to 24, with higher scores indicating a greater degree of organ failure. The SOFA score was calculated on the basis of the last recorded data before randomization. The SOFA renal score was basedon the plasma creatinine level only and did not include urine outputeARDS acute respiratory distress syndromefSOFA sequential organ failure assessmentMebazaaetal.JournalofIntensiveCare (2016) 4:24 Page4of11rates in trials (e.g., due to declining overall mortality, unin-tended enrollment of a lower-risk population, intentionalexclusion of patients with an imminent risk of death)reduces the likelihood of identifying true treatmenteffects. As the overall mortality rate declines in thegeneral sepsis population, the potential absolute effectof any given treatment is attenuated, if by nothingelse, a lower fraction of modifiable mortality [19, 29].At the same time, if baseline risk is higher than esti-mated, more patients will be needed as the expectedtreatment effect decreases (Fig. 2).Over- or under-estimating treatment effects should beavoided when designing clinical trials [30]. Researchershave struggled and often over-estimated control groupmortality when planning sample size and power esti-mates. For example, usual care group mortality rateswere over-estimated by 5.1 % in Protocolized Care forEarly Septic Shock (PROCESS) [26], 9.7 % in Sepsis andMean Arterial Pressure (SEPSISPAM) [21], 19.4 % inAustralasian Resuscitation in Sepsis Evaluation (ARISE)[24], and 10.8 % in Protocolized Management in Sepsis(ProMISE) [31], similar to previous over-estimates inseptic shock trials (Vasopressin and Septic Shock Trial(VASST) over-estimate was 11 %) [32].Sepsis is a complex syndrome characterized by theinterplay of many pathways and systems. Sepsis therap-ies must either (1) control several pathways with severalinterventions or (2) hit “upstream” nodes that control anumber of pathways. The treatment approach for sepsishas ranged from inhibiting the uncontrolled, inflamma-tory host response to enhancing the host immune re-sponse [33]. These seemingly conflicting approachesillustrate the complexity of the process and the signifi-cant (and ongoing) evolution in the understanding ofsepsis pathophysiology. Analogously, the failure of posi-tive inotropes to improve outcomes in clinical heartfailure trials [34] was initially unexpected, but it wasbetter understood as the knowledge of heart failurepathophysiology evolved.Whether sepsis treatments targeting a single aspect ofthis complex syndrome could be reasonably expected toreduce all-cause mortality is uncertain. All-cause mortal-ity is a robust endpoint because it reflects the net benefitof an intervention [28]. A benefit on all-cause mortalityshows that the effect of the intervention is strongenough to overcome the influence of events on whichthe treatment has no or minimal effect [28]. Whilethis approach works well when most deaths are dir-ectly related to the disease being studied, it may beless informative when heterogeneity in cause of deathis common and mortality is often attributable to fac-tors indirectly related to the disease such as occurs insepsis [35].In sepsis trials, significant patient heterogeneity exists intime to presentation and diagnosis, organisms(s), type andsource of infection, organ involvement, degree of organimpairment, severity of illness, location of enrollment(e.g., emergency department vs. ICU), pre-existing condi-tions, and differences in standard of care across institu-tions or geographical regions (Additional file 1: Table S1)[36]. Recent consensus definitions for sepsis and septicshock should help to reduce this variation in future clin-ical trials (Additional file 1: Table S1) [1, 37]. The selectionof sites participating in a clinical trial can substantially influ-ence endpoints (e.g., variation in comorbidities or applica-tion of background therapies can impact event rates acrosshigh and low enrolling centers) and make interpretation ofTable 2 Reasons for lack of survival improvements in sepsisclinical trials• Declining mortality rates over time• Over-estimated treatment effects• Suboptimal pre-clinical models• Knowledge of pathophysiology is still evolving, makingpathophysiologic targets difficult to identify• Incorrect treatment targets• Heterogeneity of the syndrome• Heterogeneity of the patient population• Improbability that a single treatment can impact key pathophysiologicprocesses that influence all-cause mortalityFig. 2 Estimated sample sizes by baseline mortality and absolutemortality reduction. This figure examines the total sample sizeneeded to identify an absolute mortality reduction of 3 to 15 %assuming three control group mortality rates (30, 20, and 10 %).The assumptions in this figure is that power is 80 % for a two-sidedtest and that 1:1 randomization will be employed (for example, atotal N of 3000 on the y-axis implies a n = 1500 in each treatmentarm). Source: author calculations (MOH)Mebazaa et al. Journal of Intensive Care  (2016) 4:24 Page 5 of 11trial results difficult, a challenge that has been experiencedin acute heart failure trials [38]. Genetic variants also ap-pear to influence severity [39]. Treatment responses mightvary, perhaps considerably, within such a group of patientsaccording to clinical and genetic heterogeneity. Recent ob-servational cohort studies highlighted the wide variation inmortality rates according to infection source [40]. Atpresent, most trials do not consider heterogeneous treat-ment effects when estimating sample sizes. As a result, sub-group analyses, though often employed, are likely to missimportant signals from treatments and interventions [19].Approaches to design clinical trials in sepsisCharacterization of pathophysiology: matching thetreatment to the diseaseAnimal models used in sepsis do not accurately reflectthe presentation of sepsis in humans [41, 42], in largepart because there is no single presentation of sepsis inhuman disease. Validated and more clinically relevantanimal models are needed to understand the diseaseprocess and enable therapy selection targeting specificpathophysiologic mechanisms. These models should rep-licate the duration of clinical intensive care treatment[42], integrate standard intensive care measures and ad-vanced supportive care [42, 43], investigate higher orderspecies to minimize the physiological and immunologicaldifferences between small animal species and humans[44–46], and investigate older animals with chroniccomorbidities to better reflect real-world patient popula-tions [42]. An alternate approach that might be more in-formative is to use the heterogeneity of animal modelsto understand predictors of treatment response, andthen seek to replicate the predictors in a human trial.This approach has been explored in a systematic re-view of anti-tumor necrosis factor (anti-TNF) animalstudies [47]. Biomarkers may play a role if they aid indiagnosis, prognosis (e.g., troponin in acute coronarysyndrome [48] or N-terminal brain natriuretic peptide(NT-proBNP) in heart failure [49]), or identify patientsubsets likely to respond to specific interventions (i.e.,predictive biomarkers). Multi-biomarker approachesmay be promising [50].Although many advances in cardiovascular medicinewere realized using the concept of large, simple trials,moving towards precision medicine has been proposed[51] (e.g., targeting patients with elevated systolic bloodpressure for vasodilator trials in acute heart failure [52]).A similar approach has been suggested for sepsis trials,with emphasis on defining pathophysiology through betterpre-clinical models, targeting drug development to spe-cific pathophysiologic abnormalities, and selecting pa-tients with clinical features likely to respond to a specifictherapeutic approach or who are at sufficient risk for pooroutcomes based on validated risk scores [42].Appropriate endpoints for sepsis clinical trials: insightsfrom acute heart failure clinical trialsAll-cause mortalityReducing the morbidity burden in surviving patients isan important therapeutic goal that is not reflected in anall-cause mortality endpoint [53]. All-cause mortality isan appropriate endpoint when the population has a sig-nificant mortality risk and minimal competing risks andthe intervention has the potential to alter the mortalityrisk. Short-term survival should predict longer-term sur-vival with an acceptable quality of life. Sepsis satisfiesthe first criterion, but it performs poorly on the others.First, patients with sepsis die from many causes, but it isoften impossible to determine which is primary (e.g.,renal, hepatic, pulmonary, cardiac) [33]. Death occursvia many pathways, some of which are unrelated to thetherapy being studied and will not be impacted by thetreatment (e.g., a decision to withdraw support in manyICU cases [2]). The “noise” of non-response can obscurea beneficial effect on disease-specific death (i.e., thedeath that the intervention is able to impact). Thus,cause-specific mortality is a more informative endpointto determine the benefit of a drug or intervention,whereas all-cause mortality is more meaningful wheninformation on the net benefit of an intervention (i.e.,benefit in the context of adverse events or non-response) is being sought [54]. In sepsis, cause-specificmortality is difficult to define but perhaps could beachieved in a clinical trial by increasing the “signal” (e.g.,enrolling patients with the abnormality targeted by theintervention and exclude patients at low risk of death)and decreasing the “noise” (e.g., excluding patients withcompeting mortality risks from conditions unrelated tothe sepsis episode). Cause-specific mortality might beuseful in sepsis trials to identify agents with a significanttreatment effect on specific components of the illness.Similar to sepsis, patients with acute heart failure havehigh short-term mortality, a factor which usually makesmortality trials easier to conduct. However, in the caseof acute heart failure, most therapies primarily targetsymptoms rather than the underlying pathophysiologythat leads to death. Additionally, acute heart failuredrugs are administered for a short-duration; both ofthese factors reduce the likelihood that all-cause mortal-ity will be influenced over the intermediate or long-term(e.g., 180 days). Although the European MedicinesAgency guideline still specifies all-cause mortality as thepreferred primary endpoint in acute heart failure trials,it states that symptomatic improvement might beacceptable as a primary endpoint for short-term trialsprovided mortality is not adversely affected [55]. Regula-tory agencies have recently agreed to a primary hierarch-ical clinical composite endpoint in an acute heart failuretrial that combines a global assessment of symptoms,Mebazaa et al. Journal of Intensive Care  (2016) 4:24 Page 6 of 11persistent or worsening heart failure requiring an inter-vention, and all-cause mortality assessed at 6, 24, and48 h. Patients are categorized as improved (moderate ormarked improvement in clinical status at all plannedassessments without hospitalization for heart failure ordeath), unchanged (modest improvement or worseningin clinical status), or worsened (moderate or markedworsening of clinical status at any planned assessment,hospitalization for heart failure requiring intravenous ormechanical interventions, or death). The distribution ofpatients in each category is compared between treatmentgroups to assess the treatment effect [56, 57]. This end-point has the advantage of reflecting considerations thatare important to patients (both symptoms and out-comes), and it allows for a short-term assessment ofmorbidity and mortality during the period when thepharmacologic effect is present. Importantly, long-termall-cause mortality should still be assessed for safety, andthe study should be powered to demonstrate that long-term mortality is not increased by a pre-specified safetymargin [52].Regulatory agencies might consider a similar clinicalcomposite endpoint adapted for sepsis trials, whereendpoints describing end-organ function, need formechanical support, or need for other interventions arecombined with short-term mortality (ideally sepsis-related mortality if consensus can be reached on a stand-ard definition) as a primary endpoint, with longer-termall-cause mortality assessed for safety. This approachalso has the advantage of reflecting relevant factors otherthan survival that are important to patients. Rigorousdefinitions for such endpoints are keys to ensureconsistency and to reduce bias in the results and toensure that the endpoint can be translated into a metricthat is important to patients.Non-fatal endpointsTotal or ICU length of stay has been considered as anendpoint for sepsis trials. It is relevant because ICU staysare costly, but it is dependent on external factors thatare unrelated to drug therapy (e.g., physician judgment,no accepted standards for discharge readiness, availabil-ity of step-down beds, payer influence, local standards ofcare). These same limitations have been recognized inacute heart failure trials [58]. Thus, the length of stay isunsuitable as a primary endpoint for pivotal trials,but it can be useful as a secondary endpoint or to in-form health technology and economic (cost/benefit)assessments. Other problems with using non-fatalendpoints include ascertainment bias, competing risks,and informative dropout when comparing treatmentand control groups (i.e., patients who die cannot behospitalized and patients who die early have de-creased length of stay) [19].Organ dysfunction is a relevant endpoint for sepsistrials. Multiple organs are impaired in sepsis [42], but all-cause mortality is insensitive to determine which organ ororgan(s) are the primary driver of death. Conceptually,integrating a measure of organ dysfunction into a mortal-ity endpoint (e.g., days alive and free of organ dysfunction)would provide a more comprehensive assessment of mor-bidity and mortality. Organ dysfunction is theoretically amore sensitive measure of the effect of an intervention onprogression of the sepsis syndrome, but this concept hasnot yet been validated in trials. Since short-termorgan dysfunction is associated with long-term out-come [17, 59], it is plausible that improvements inorgan function might translate into improved survival, butthis relationship has not yet been shown and the hypoth-esis still requires confirmation. The primary value ofmeasuring organ dysfunction at the current time is to gainan understanding of how an intervention impacts physi-ology and organ function. Correlations between change inshort-term organ dysfunction and long-term sepsis-associated morbidity could also be derived from largerobust registries that include long-term follow-up andoutcomes. If used as an endpoint, organ dysfunctionshould be pre-defined in the protocol and statistical ana-lysis plan. Ideally, consensus about how to define organdysfunction should be sought so that definitions are usedconsistently across clinical trials.Days alive and free from mechanical ventilation, renalreplacement therapy, or vasopressors (i.e., organ failurefree days) has also been proposed. These endpoints areclinically meaningful, and widespread use of the Surviv-ing Sepsis Campaign guidelines has led to more consist-ent timing and application of life support interventions.Nonetheless, the decision to institute supportive therap-ies is often subjective and can be influenced by externalfactors (e.g., reimbursement incentives, interactions ofvarious medical specialists (e.g., intensivists and nephrol-ogists)), which introduces increased variability (i.e., ran-dom noise) in the study and possibly bias if the study isnot blinded. Other complex issues also warrant consid-eration, including whether patients value more event-free days equally regardless of when they occur (e.g.,moving from 0 to 1 day is the same/better/worse thanmoving from 29 to 30 days), handling inclusion of mul-tiple organs (i.e., are all organs of equal value or shouldfailure in some organs be weighted more heavily thanothers), and methodology to account for pre-existingorgan dysfunction. Interventions can be effective inpreventing organ dysfunction (in patients who do nothave organ dysfunction) and/or preventing progression oforgan dysfunction (in patients who already have somedegree of organ dysfunction). An adequate organ dys-function scoring system must capture both of thesepossibilities.Mebazaa et al. Journal of Intensive Care  (2016) 4:24 Page 7 of 11In general, there are no accepted surrogates for safety[60], although death is not the only safety measure.Safety is difficult to assess in sepsis trials because of thehigh incidence of organ dysfunction in sepsis. Differ-ences in organ dysfunction scores between treatmentgroups could also be seen as a safety outcome (e.g., pre-vention of organ dysfunction due to side effects of exces-sive vasopressor doses and duration). Other events (e.g.,anaphylaxis) might be relevant for specific drugs. Even ifa beneficial effect was shown on organ dysfunction orother non-fatal endpoint, adequate assurance of safetywould still have to be demonstrated, either in a pivotalclinical trial, in the entirety of the drug’s database, orbased on experience with similar drugs or interventions[60]. Consultation with regulatory agencies is needed todetermine the size of the safety database and the confi-dence level required to rule out an adverse effect onmortality; these decisions are often dependent on theseverity of illness in the population studied and the spe-cific benefit of the drug (e.g., a drug that improves aclinically important outcome vs. a drug that improvescontrol of a biomarker).Role of alternative study designsAdaptive designsAdaptive designs or seamless phase II/III designs havethe potential to improve the efficiency of clinical trials.Adaptive designs can be particularly useful in fields inwhich data are limited to inform trial planning assump-tions in the areas of expected event rates, anticipatedeffect sizes, heterogeneity of treatment effect, variance,safety, or drop-outs [61, 62]. In sepsis, many uncertain-ties exist at the time of trial design, and adaptive designis a promising approach for both exploratory and con-firmatory stages of drug development, especially in thecontext of moving towards exploration of novel end-points for sepsis trials. These designs are well acceptedfor feasibility and early phase studies, but as experiencewith their use has increased, they are becoming moreaccepted for pivotal trials as well [63]. Potential chal-lenges include maintaining confidentiality and blindingof interim ongoing results and avoiding the introductionof bias resulting from the adaptations [63]. Strict controlof type I error risk and understanding the potentialbiases are important issues; rapid progress is being madearound these issues [64, 65].Realistic trial simulation is the key tool to addressthese challenges and advance the field. Trial simulationof traditional and adaptive trial designs furthers under-standing of strengths and weaknesses of proposed trialdesigns and will illustrate vulnerabilities from minordeviations to study design assumptions (e.g., event rates,missing data).Ideally, trial design should be a multi-step, collabora-tive, and interactive multidisciplinary process betweenscientific, clinical, and statistical domain experts to in-crease the quality and chance of success. This conceptapplies to all types of trials, but it is particularly import-ant for adaptive design. Early interaction with regulatorsis highly recommended when using adaptive designs inthe later stages of a drug development program [61, 62].ConclusionsSepsis is a major burden with high mortality, and thelack of progress in identifying effective treatments isdiscouraging for researchers and industry. The clinicalresearch challenges that have been encountered in sepsistrials closely resemble those experienced by investigatorsin acute heart failure trials. After decades of research, ithas become clear in the acute heart failure communitythat the substantial patient heterogeneity contributes tothe difficulties in identifying effective therapies for thecondition. The recent consensus definitions for sepsis andseptic shock are important advances in this regard [1, 37].Table 3 Priorities for future sepsis clinical trials1. Develop more informative studies using animal models2. Emphasize study of pathophysiology3. Identify biomarkers, molecular signals, or genetic markers to identifypatients having an underlying causal process that might respond tothe specific treatment being studied4. Develop networks of sepsis investigators experienced in clinical trialconduct5. Apply the recent Third International Consensus Definitions for Sepsisand Septic Shock [1, 37] when determining eligibility criteria6. Conduct targeted clinical trials in relatively homogeneous groups ofpatients with characteristics suggestive of treatment response7. Consider the addition of pre-specified covariate adjustment of theprimary endpoint to address the issue of heterogeneity8. Exclude low-risk patients if appropriate for the intervention beingstudied9. Standardize care to reduce variability and random noise but not tothe extent that results are not generalizable10. Develop realistic expectations for treatment effect and power trialsaccordingly11. Apply adaptive designs, especially when key variables are uncertain(e.g., event rates, expected treatment effect)12. Consider targeted primary endpoints with all-cause mortalityreserved for safety13. Develop consensus in the field for standard trial definitions/criteriafor interventions if used as endpoints (e.g., vasopressors, mechanicalventilation, renal replacement therapy)14. Collaborate with regulators to modify approach to clinical trialdesign in this field15. Develop robust registries to test external validity of the results oftrials in broader patient populations16. Discovery and development of a diagnostic that predicts a higherchance of response to a specific interventionMebazaa et al. Journal of Intensive Care  (2016) 4:24 Page 8 of 11Additionally, assessing all-cause mortality alone is insuffi-cient to fully characterize the burden of disease because itomits important aspects of symptoms and functional sta-tus. Academic heart failure investigators and industry haveworked closely with regulators for many years to transi-tion acute heart failure trials away from relying on short-term symptoms and all-cause mortality as the primary ef-ficacy measures, and ongoing trials are assessing novelclinical composite endpoints reflecting organ dysfunctionand mortality while still evaluating all-cause mortality as aseparate safety measure. Applying the lessons learned inacute heart failure trials to sepsis trials might be useful toadvance the field (Table 3). Selecting high-risk patientswith clinical phenotypes considered likely to respond tothe intervention under study may help to reduce patientheterogeneity within clinical trials and enable signals ofbenefit to be more readily detected. Additionally, novelendpoints beyond all-cause mortality should be consid-ered for future sepsis trials.Additional fileAdditional file 1: Table S1. This table describes the definitions of sepsisand septic shock that have been used in pivotal sepsis trials. (DOCX 21 KB)AbbreviationsADHERE: Acute Decompensated Heart Failure National Registry;AHEAD: Acute Heart Failure Database; ALARM: Acute Heart Failure GlobalRegistry of Standard Treatment; ALBIOS: Albumin Italian Outcome SepsisStudy; Anti-TNF: anti-tumor necrosis factor; ARDS: acute respiratory distresssyndrome; ARISE: Australasian Resuscitation in Sepsis Evaluation;ATTEND: Acute Decompensated Heart Failure Syndrome; CXR: chestradiography; ED: emergency department; EGDT: early goal-directed therapy;EHFS II: Second EuroHeart Failure Survey; Hgb: hemoglobin; ICU: intensivecare unit; MAP: mean arterial pressure; NT-proBNP: N-terminal pro brainnatriuretic peptide; OPTIMIZE: Organized Program to Initiate LifesavingTreatment in Hospitalized Patients with Heart Failure; PaCO2: partial pressureof arterial carbon dioxide; PROCESS: Protocolized Care for Early Septic Shock;ProMISE: Protocolized Management in Sepsis; SEPSISPAM: Sepsis and MeanArterial Pressure; SIRS: systemic inflammatory response syndrome;TRISS: Transfusion Requirements in Septic Shock; VASST: Vasopressin andSeptic Shock Trial.Competing interestsAlexandre Mebazaa received a speaker’s honoraria from The MedicinesCompany, Novartis, Orion, Roche, and Servier and fees for Advisory Boardsand Steering Committees from Cardiorentis, The Medicines Company,Adrenomed, MyCartis, ZS Pharma, and Critical Diagnostics.James A. Russell patents owned by the University of British Columbia (UBC)that are related to PCSK9 inhibitor(s) and sepsis and related to the use ofvasopressin in septic shock. Dr. Russell is an inventor on these patents. Dr.Russell is a founder, Director, and shareholder in Cyon Therapeutics Inc.(developing a sepsis therapy). Dr. Russell has share options in LeadingBiosciences Inc. Dr. Russell reports receiving consulting fees from CubistPharmaceuticals (formerly Trius Pharmaceuticals) (developing antibiotics),Ferring Pharmaceuticals (manufactures vasopressin and is developingselepressin), Grifols (sells albumin), MedImmune (regarding sepsis), LeadingBiosciences (developing a sepsis therapeutic), La Jolla Pharmaceuticals(developing a sepsis therapeutic), CytoVale Inc. (developing a sepsisdiagnostic), and Sirius Genomics Inc. (now closed; had donepharmacogenomics research in sepsis). Dr. Russell reports having receivedgrant support from Sirius Genomics and Ferring Pharmaceuticals that wasprovided to and administered by UBC.Andreas Bergmann is an employee of Adrenomed AG.Research reported in this publication was supported by the National Heart,Lung, and Blood Institute of the National Institutes of Health under AwardNumber F31HL127947 awarded to Michael O. Harhay. The content is solelythe responsibility of the authors and does not necessarily represent theofficial views of the National Institutes of Health.Oliver Hartmann is an employee of Adrenomed AG.Frauke Hein is a cofounder and employee of Adrenomed AG.Anne Louise Kjolbye is an employee of Ferring Pharmaceuticals.Dr. Lewis serves as the senior medical scientist at Berry Consultants, LLC,a statistical consulting group that specializes in the design and supportof adaptive clinical trials, including adaptive clinical trials focused onthe evaluation of treatments for severe sepsis and septic shock.John C. Marshall received personal fees (DSMB member) from Asahi Kasei.Gernot Marx received research grants from EU project THALEA, BBraunMelsungen AG, personal fees (honoraria for lecturing and consulting) fromBBraun Melsungen AG, Adrenomed, Philips and a patent pending formodulation of TLR4-signaling pathway.Peter Radermacher received research grants from Adrenomed AG,Boehringer Ingelheim Pharma GmbH & Co., German Ministry of Defense, andPoxel SA and personal fees from Boehringer Ingelheim Pharma GmbH & Co.Mathias Schroedter is an employee of Adrenomed AG.Wendy Gattis Stough is a consultant to European Drug Development Hub,Relypsa, CHU Nancy, European Society of Cardiology, Heart FailureAssociation of the European Society of Cardiology, Heart Failure Society ofAmerica, Overcome, Stealth BioTherapeutics, Covis Pharmaceuticals,University of Gottingen, and University of North Carolina.Joachim Struck is an employee of Adrenomed AG.Derek C. Angus received personal fees from Bayer HealthCare, FerringPharmaceuticals, GlaxoSmithKline, Ibis Biosciences, and Medimmune(consulting, advisory boards).All other authors declare that they have no competing interests.Authors’ contributionsAM conceived, planned, and organized the meeting where discussions onthis manuscript topic took place. AM and WGS drafted the manuscript.All authors presented and participated in the discussions during the meetingheld in Paris France, January 2015. All authors critically revised themanuscript for important intellectual content. All authors have read andapproved the final manuscript.AcknowledgementsThe authors thank EDDH - Fondation Transplantation for the logistical andadministrative support.Author details1University Paris Diderot, Sorbonne Paris Cité, Paris, France. 2U942 Inserm,APHP, Paris, France. 3APHP, Department of Anesthesia and Critical Care,Hôpitaux Universitaires Saint Louis-Lariboisière, Paris, France. 4Department ofCritical Care Medicine, St. Luc University Hospital, Université Catholique deLouvain (UCL), Brussels, Belgium. 5Center for Heart Lung Innovation and theDivision of Critical Care Medicine, St. Paul’s Hospital, University of BritishColumbia, Vancouver, Canada. 6Adrenomed AG, Hennigsdorf, Germany.7Università di Milano, Fondazione IRCCS Ca’ Granda, Ospedale MaggiorePoliclinico, Milan, Italy. 8Département d’Anesthésie – Réanimation – SMUR,Hôpitaux Universitaires Saint Louis – Lariboisière, INSERM – UMR 942,Assistance Publique – Hôpitaux de Paris, Université Paris Diderot, Paris,France. 9Division of Epidemiology, Department of Biostatistics andEpidemiology, Perelman School of Medicine, University of Pennsylvania,Philadelphia, PA, USA. 10Ferring Pharmaceuticals, Copenhagen, Denmark.11Department of Anesthesiology, Critical Care and Burn Unit, St. LouisHospital, University Paris 7 Denis Diderot, UMR-S942, Inserm, Paris, France.12Department of Emergency Medicine, Harbor-UCLA Medical Center,Torrance, CA, USA. 13Department of Surgery, Interdepartmental Division ofCritical Care Medicine, University of Toronto, St. Michael’s Hospital, Toronto,Ontario, Canada. 14Department of Intensive Care and Intermediate Care,University Hospital RWTH Aachen, Aachen, Germany. 15Institut fürAnästhesiologische Pathophysiologie und Verfahrensentwicklung,Universitätsklinikum, Ulm, Germany. 16Campbell University College ofPharmacy and Health Sciences, Buies Creek, NC, USA. 17Clinical Departmentand Laboratory of Intensive Care Medicine, Division of Cellular and MolecularMebazaa et al. Journal of Intensive Care  (2016) 4:24 Page 9 of 11Medicine, KU Leuven, Leuven, Belgium. 18Department of Cardiology,Cumhuriyet University Faculty of Medicine, Sivas, Turkey. 19CRISMA Center,Department of Critical Care Medicine, McGowan Institute for RegnerativeMedicine, Clinical and Translational Science Institute, University of PittsburghSchools of the Health Sciences, Pittsburgh, PA, USA. 20Department of HealthPolicy and Management, McGowan Institute for Regnerative Medicine,Clinical and Translational Science Institute, University of Pittsburgh Schools ofthe Health Sciences, Pittsburgh, PA, USA.Received: 3 January 2016 Accepted: 17 March 2016References1. Seymour CW, Liu VX, Iwashyna TJ, Brunkhorst FM, Rea TD, Scherag A et al.Assessment of Clinical Criteria for Sepsis: For the Third InternationalConsensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016;315:762–74.2. Kahn JM, Le T, Angus DC, Cox CE, Hough CL, White DB, et al. The epidemiologyof chronic critical illness in the United States*. Crit Care Med. 2015;43:282–7.3. Mayr FB, Yende S, Angus DC. Epidemiology of severe sepsis. Virulence. 2014;5:4–11.4. Ogura H, Gando S, Saitoh D, Takeyama N, Kushimoto S, Fujishima S, et al.Epidemiology of severe sepsis in Japanese intensive care units: aprospective multicenter study. J Infect Chemother. 2014;20:157–62.5. Quenot JP, Binquet C, Kara F, Martinet O, Ganster F, Navellou JC, et al.The epidemiology of septic shock in French intensive care units: theprospective multicenter cohort EPISS study. Crit Care. 2013;17:R65.6. Taniguchi LU, Bierrenbach A, Toscano CM, Schettino G, Azevedo L.Sepsis-related deaths in Brazil: an analysis of the national mortalityregistry from 2002 to 2010. Crit Care. 2014;18:608.7. Cohen J, Vincent JL, Adhikari NK, Machado FR, Angus DC, Calandra T, et al.Sepsis: a roadmap for future research. Lancet Infect Dis. 2015;15:581–614.8. Gaieski DF, Edwards JM, Kallan MJ, Carr BG. Benchmarking the incidenceand mortality of severe sepsis in the United States. Crit Care Med. 2013;41:1167–74.9. Adhikari NK, Fowler RA, Bhagwanjee S, Rubenfeld GD. Critical care and theglobal burden of critical illness in adults. Lancet. 2010;376:1339–46.10. Fleischmann C, Scherag A, Adhikari N. Global burden of sepsis: a systematicreview [abstract]. Crit Care. 2015;19:P21.11. Kaukonen KM, Bailey M, Suzuki S, Pilcher D, Bellomo R. Mortality related tosevere sepsis and septic shock among critically ill patients in Australia andNew Zealand, 2000-2012. JAMA. 2014;311:1308–16.12. Eisner MD, Thompson T, Hudson LD, Luce JM, Hayden D, Schoenfeld D, etal. Efficacy of low tidal volume ventilation in patients with different clinicalrisk factors for acute lung injury and the acute respiratory distresssyndrome. Am J Respir Crit Care Med. 2001;164:231–6.13. Funk DJ, Kumar A. Antimicrobial therapy for life-threatening infections:speed is life. Crit Care Clin. 2011;27:53–76.14. Gao F, Melody T, Daniels DF, Giles S, Fox S. The impact of compliance with6-hour and 24-hour sepsis bundles on hospital mortality in patients withsevere sepsis: a prospective observational study. Crit Care. 2005;9:R764–70.15. Jones AE, Focht A, Horton JM, Kline JA. Prospective external validation ofthe clinical effectiveness of an emergency department-based early goal-directed therapy protocol for severe sepsis and septic shock. Chest. 2007;132:425–32.16. Linder A, Guh D, Boyd JH, Walley KR, Anis AH, Russell JA. Long-term(10-year) mortality of younger previously healthy patients with severesepsis/septic shock is worse than that of patients with nonseptic criticalillness and of the general population. Crit Care Med. 2014;42:2211–8.17. Linder A, Fjell C, Levin A, Walley KR, Russell JA, Boyd JH. Small acuteincreases in serum creatinine are associated with decreased long-termsurvival in the critically ill. Am J Respir Crit Care Med. 2014;189:1075–81.18. Fried TR, Bradley EH, Towle VR, Allore H. Understanding the treatmentpreferences of seriously ill patients. N Engl J Med. 2002;346:1061–6.19. Harhay MO, Wagner J, Ratcliffe SJ, Bronheim RS, Gopal A, Green S, et al.Outcomes and statistical power in adult critical care randomized trials.Am J Respir Crit Care Med. 2014;189:1469–78.20. Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al.Surviving sepsis campaign: international guidelines for management ofsevere sepsis and septic shock: 2012. Crit Care Med. 2013;41:580–637.21. Asfar P, Meziani F, Hamel JF, Grelon F, Megarbane B, Anguel N, et al. Highversus low blood-pressure target in patients with septic shock. N Engl JMed. 2014;370:1583–93.22. Caironi P, Tognoni G, Masson S, Fumagalli R, Pesenti A, Romero M, et al.Albumin replacement in patients with severe sepsis or septic shock. N EnglJ Med. 2014;370:1412–21.23. Holst LB, Haase N, Wetterslev J, Wernerman J, Guttormsen AB, Karlsson S, etal. Lower versus higher hemoglobin threshold for transfusion in septicshock. N Engl J Med. 2014;371:1381–91.24. Peake SL, Delaney A, Bailey M, Bellomo R, Cameron PA, Cooper DJ, et al.Goal-directed resuscitation for patients with early septic shock. N Engl JMed. 2014;371:1496–506.25. Truwit JD, Bernard GR, Steingrub J, Matthay MA, Liu KD, Albertson TE, et al.Rosuvastatin for sepsis-associated acute respiratory distress syndrome.N Engl J Med. 2014;370:2191–200.26. Yealy DM, Kellum JA, Huang DT, Barnato AE, Weissfeld LA, Pike F, et al.A randomized trial of protocol-based care for early septic shock. N Engl JMed. 2014;370:1683–93.27. Gheorghiade M, Adams KF, Cleland JG, Cotter G, Felker GM, Filippatos GS, etal. Phase III clinical trial end points in acute heart failure syndromes: a virtualroundtable with the Acute Heart Failure Syndromes International WorkingGroup. Am Heart J. 2009;157:957–70.28. Zannad F, Garcia AA, Anker SD, Armstrong PW, Calvo G, Cleland JG, et al.Clinical outcome endpoints in heart failure trials: a European Society ofCardiology Heart Failure Association consensus document. Eur J Heart Fail.2013;15:1082–94.29. Yusuf S, Collins R, Peto R. Why do we need some large, simple randomizedtrials? Stat Med. 1984;3:409–22.30. Stevenson EK, Rubenstein AR, Radin GT, Wiener RS, Walkey AJ. Two decadesof mortality trends among patients with severe sepsis: a comparative meta-analysis*. Crit Care Med. 2014;42:625–31.31. Mouncey PR, Osborn TM, Power GS, Harrison DA, Sadique MZ, Grieve RD, etal. Trial of early, goal-directed resuscitation for septic shock. N Engl J Med.2015;372:1301–11.32. Russell JA, Walley KR, Singer J, Gordon AC, Hebert PC, Cooper DJ, et al.Vasopressin versus norepinephrine infusion in patients with septic shock.N Engl J Med. 2008;358:877–87.33. Hotchkiss RS, Karl IE. The pathophysiology and treatment of sepsis. N Engl JMed. 2003;348:138–50.34. Cohn JN, Goldstein SO, Greenberg BH, Lorell BH, Bourge RC, Jaski BE, et al.A dose-dependent increase in mortality with vesnarinone among patientswith severe heart failure. Vesnarinone Trial Investigators. N Engl J Med. 1998;339:1810–6.35. Gayat E, Lemasle L, Payen D. Drotrecogin alfa (activated) in severe sepsis.Lancet Infect Dis. 2013;13:109–10.36. Gattinoni L, Ranieri VM, Pesenti A. Sepsis: needs for defining severity.Intensive Care Med. 2015;41:551–2.37. Shankar-Hari M, Phillips GS, Levy ML, Seymour CW, Liu VX, Deutschman CSet al. Developing a New Definition and Assessing New Clinical Criteria forSeptic Shock: For the Third International Consensus Definitions for Sepsisand Septic Shock (Sepsis-3). JAMA. 2016;315:775–87.38. Gheorghiade M, Vaduganathan M, Greene SJ, Mentz RJ, Adams Jr KF,Anker SD, et al. Site selection in global clinical trials in patients hospitalized forheart failure: perceived problems and potential solutions. Heart Fail Rev. 2014;19:135–52.39. Toubiana J, Courtine E, Pene F, Viallon V, Asfar P, Daubin C, et al. IRAK1functional genetic variant affects severity of septic shock. Crit Care Med.2010;38:2287–94.40. Leligdowicz A, Dodek PM, Norena M, Wong H, Kumar A, Kumar A.Association between source of infection and hospital mortality in patientswho have septic shock. Am J Respir Crit Care Med. 2014;189:1204–13.41. Opal SM, Dellinger RP, Vincent JL, Masur H, Angus DC. The next generationof sepsis clinical trial designs: what is next after the demise of recombinanthuman activated protein C?*. Crit Care Med. 2014;42:1714–21.42. Angus DC, van der Poll T. Severe sepsis and septic shock. N Engl J Med.2013;369:840–51.43. Fink MP, Heard SO. Laboratory models of sepsis and septic shock. J SurgRes. 1990;49:186–96.44. Seok J, Warren HS, Cuenca AG, Mindrinos MN, Baker HV, Xu W, et al.Genomic responses in mouse models poorly mimic human inflammatorydiseases. Proc Natl Acad Sci U S A. 2013;110:3507–12.Mebazaa et al. Journal of Intensive Care  (2016) 4:24 Page 10 of 1145. Osuchowski MF, Remick DG, Lederer JA, Lang CH, Aasen AO, Aibiki M, et al.Abandon the mouse research ship? Not just yet! Shock. 2014;41:463–75.46. Radermacher P, Haouzi P. A mouse is not a rat is not a man: species-specificmetabolic responses to sepsis—a nail in the coffin of murine models forcritical care research? Intensive Care Med Exp. 2013;1:7.47. Lorente JA, Marshall JC. Neutralization of tumor necrosis factor in preclinicalmodels of sepsis. Shock. 2005;24 Suppl 1:107–19.48. Amsterdam EA, Wenger NK, Brindis RG, Casey Jr DE, Ganiats TG, Holmes JrDR, et al. 2014 AHA/ACC guideline for the management of patients withnon-ST-elevation acute coronary syndromes: a report of the AmericanCollege of Cardiology/American Heart Association Task Force on PracticeGuidelines. J Am Coll Cardiol. 2014;64:e139–228.49. Maisel AS, Krishnaswamy P, Nowak RM, McCord J, Hollander JE, Duc P, et al.Rapid measurement of B-type natriuretic peptide in the emergencydiagnosis of heart failure. N Engl J Med. 2002;347:161–7.50. Wong HR, Walley KR, Pettila V, Meyer NJ, Russell JA, Karlsson S, et al.Comparing the prognostic performance of ASSIST to interleukin-6 andprocalcitonin in patients with severe sepsis or septic shock. Biomarkers.2015;20:132–5.51. Jackson N, Atar D, Borentain M, Breithardt G, van Eickels M, Endres M, et al.Improving clinical trials for cardiovascular diseases: a position paper fromthe Cardiovascular Roundtable of the European Society of Cardiology.Eur Heart J. 2015. doi: 10.1093/eurheartj/ehv213.52. Mebazaa A, Longrois D, Metra M, Mueller C, Richards AM, Roessig L, et al.Agents with vasodilator properties in acute heart failure: how to designsuccessful trials. Eur J Heart Fail. 2015;17:652–64.53. Iwashyna TJ, Angus DC. Declining case fatality rates for severe sepsis:good data bring good news with ambiguous implications. JAMA. 2014;311:1295–7.54. Zannad F, Stough WG, Pitt B, Cleland JG, Adams KF, Geller NL, et al.Heart failure as an endpoint in heart failure and non-heart failurecardiovascular clinical trials: the need for a consensus definition.Eur Heart J. 2008;29:413–21.55. European Medicines Agency (9-20-2012) Guideline on clinical investigationof medicinal products for the treatment of acute heart failure (CHMP/EWP/2986/03 Rev. 1). http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2012/10/WC500133497.pdf. Accessed 5/1/2015.56. Packer M. Proposal for a new clinical end point to evaluate the efficacy ofdrugs and devices in the treatment of chronic heart failure. J Card Fail.2001;7:176–82.57. (10-1-2014) Efficacy and Safety of Ularitide for the Treatment of AcuteDecompensated Heart Failure (TRUE-AHF). http://www.clinicaltrials.gov/ct2/show/NCT01661634. Accessed 11/19/2014.58. Allen LA, Hernandez AF, O’Connor CM, Felker GM. End points for clinicaltrials in acute heart failure syndromes. J Am Coll Cardiol. 2009;53:2248–58.59. Nfor TK, Walsh TS, Prescott RJ. The impact of organ failures and theirrelationship with outcome in intensive care: analysis of a prospectivemulticentre database of adult admissions. Anaesthesia. 2006;61:731–8.60. Temple R. Are surrogate markers adequate to assess cardiovascular diseasedrugs? JAMA. 1999;282:790–5.61. European Medicines Agency (10-18-2007) Reflection paper onmethodological issues in confirmatory clinical trials planned with anadaptive design. http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2009/09/WC500003616.pdf.62. Food and Drug Administration (2-1-2010) guidance for industry: adaptivedesign clinical trials for drugs and biologics. http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm201790.pdf.63. Bretz F, Koenig F, Brannath W, Glimm E, Posch M. Adaptive designs forconfirmatory clinical trials. Stat Med. 2009;28:1181–217.64. Mehta C, Gao P, Bhatt DL, Harrington RA, Skerjanec S, Ware JH. Optimizingtrial design: sequential, adaptive, and enrichment strategies. Circulation.2009;119:597–605.65. Muller HH, Schafer H. Adaptive group sequential designs for clinical trials:combining the advantages of adaptive and of classical group sequentialapproaches. Biometrics. 2001;57:886–91.66. CDC/NCHS National Hospital Discharge Survey (12-31-2007) Number, rate,and standard error of deaths for discharges from short-stay hospitals, byage and selected first-listed diagnosis: United States, 2007. http://www.cdc.gov/nchs/data/nhds/5hospital/2007hos5_numberrate.pdf.67. Kozak LJ, Hall MJ, Owings MF. National hospital discharge survey: 2000annual summary with detailed diagnosis and procedure data. Vital HealthStat 13. 2002;153:1–194.68. Kozak LJ, DeFrances CJ, Hall MJ. National hospital discharge survey: 2004annual summary with detailed diagnosis and procedure data. NationalCenter for Health Statistics. Vital Health Stat. 2006;13:34.69. Ambrosy AP, Fonarow GC, Butler J, Chioncel O, Greene SJ, Vaduganathan M,et al. The global health and economic burden of hospitalizations for heartfailure: lessons learned from hospitalized heart failure registries. J Am CollCardiol. 2014;63:1123–33.70. Adams Jr KF, Fonarow GC, Emerman CL, LeJemtel TH, Costanzo MR,Abraham WT, et al. Characteristics and outcomes of patients hospitalizedfor heart failure in the United States: rationale, design, and preliminaryobservations from the first 100,000 cases in the Acute DecompensatedHeart Failure National Registry (ADHERE). Am Heart J. 2005;149:209–16.71. Fonarow GC, Abraham WT, Albert NM, Gattis Stough W, Gheorghiade M,Greenberg BH, et al. Influence of a performance-improvement initiative onquality of care for patients hospitalized with heart failure: results of theOrganized Program to Initiate Lifesaving Treatment in Hospitalized PatientsWith Heart Failure (OPTIMIZE-HF). Arch Intern Med. 2007;167:1493–502.72. Nieminen MS, Brutsaert D, Dickstein K, Drexler H, Follath F, Harjola VP, et al.EuroHeart Failure Survey II (EHFS II): a survey on hospitalized acute heartfailure patients: description of population. Eur Heart J. 2006;27:2725–36.73. Follath F, Yilmaz MB, Delgado JF, Parissis JT, Porcher R, Gayat E, et al. Clinicalpresentation, management and outcomes in the Acute Heart FailureGlobal Survey of Standard Treatment (ALARM-HF). Intensive Care Med.2011;37:619–26.74. Spinar J, Parenica J, Vitovec J, Widimsky P, Linhart A, Fedorco M, et al.Baseline characteristics and hospital mortality in the Acute Heart FailureDatabase (AHEAD) Main registry. Crit Care. 2011;15:R291.75. Sato N, Kajimoto K, Keida T, Mizuno M, Minami Y, Yumino D, et al. Clinicalfeatures and outcome in hospitalized heart failure in Japan (from theATTEND Registry). Circ J. 2013;77:944–51.76. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, et al.Definitions for sepsis and organ failure and guidelines for the use ofinnovative therapies in sepsis. The ACCP/SCCM Consensus ConferenceCommittee. American College of Chest Physicians/Society of Critical CareMedicine. Chest. 1992;101:1644–55.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Mebazaa et al. Journal of Intensive Care  (2016) 4:24 Page 11 of 11


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items